ABSTRACT Title of Document: AN OCCUPATION WITH DEMOCRATIZATION: A MARGINAL VALUE APPROACH TO UNDERSTANDING THE CONSOLIDATION OF IMPOSED DEMOCRATIC REGIMES Jesse-Douglas Mathewson, Doctor of Philosophy, 2013 Directed By: Professor Karol Soltan Department of Government and Politics The dissertation attempts to understand the causes and correlates of democratic consolidation in occupied territories. A Marginal Value Model attempts to explain the consolidation of democracy in these cases as a function of international threat dynamics and the relationship between the occupiers and the occupied regime. The dissertation tests the Marginal Value Model and its corresponding hypotheses against four case studies: post-WWI Germany, post-WWII Germany, Japan and Korea. The study finds that democracies are more likely to consolidate when there is an external threat, when the occupier credibly protects the new regime against this threat, and when the occupier provides additional goods to the domestic population. These tests find support for the Marginal Value Model and its corresponding hypotheses. AN OCCUPATION WITH DEMOCRATIZATION: A MARGINAL VALUE APPROACH TO UNDERSTANDING THE CONSOLIDATION OF IMPOSED DEMOCRATIC REGIMES By Jesse-Douglas Robert Mathewson Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park, in partial fulfillment of the requirements for the degree of Doctorate of Philosophy 2013 Advisory Committee: Professor Karol Soltan, Chair Professor George Quester Professor Virginia Haufler Professor Scott Kastner Professor I. M. Destler, Dean?s Representative (Public Policy) ? Copyright by Jesse-Douglas Robert Mathewson 2013 Acknowledgements This work would not have been possible without the guidance of the members of my dissertation committee, Drs. Soltan, Quester, Haufler, Kastner and Destler. Additional thanks belong to Dr. Miranda Schreurs, in whose class the idea for this project took root. Moreover, I could not have survived this experience without the constant help and support of Ann Marie Clark and Cissy Roberts. Finally, I express the utmost gratitude to my fianc?e Jennifer, the great woman rolling her eyes behind the man. ii Dedication For Mike iii Table of Contents Acknowledgements .................................................................................................................................. ii Dedication .............................................................................................................................................. iii Table of Contents .....................................................................................................................................iv List of Tables .......................................................................................................................................... vii List of Figures ........................................................................................................................................ viii Chapter 1: Introduction and Literature Review ..................................................................................... 1 The Big Picture .................................................................................................................................... 1 Introduction ..................................................................................................................................... 1 Significance ..................................................................................................................................... 2 Literature Review ................................................................................................................................ 4 Introduction ..................................................................................................................................... 4 What is Democratic Consolidation? ............................................................................................. 6 Inherent Instability .......................................................................................................................... 8 The Role of Time .......................................................................................................................... 12 The Role of Institutions ................................................................................................................ 13 The Political Economy of Consolidation ................................................................................... 18 The Roles of Culture, Learning and History .............................................................................. 24 The Role of Interstate Politics ..................................................................................................... 26 Summation of Democratization Literature ................................................................................ 29 Quo Vadis? ........................................................................................................................................ 33 Chapter 2: Research Design .................................................................................................................. 35 Introduction ........................................................................................................................................ 35 Narrowing the Spectrum of Political Contestation ........................................................................ 36 Occupations as Caretaker States ...................................................................................................... 38 Marginal Value of Occupation and Modes of Political Contestation .......................................... 42 Specification of Terms ...................................................................................................................... 45 Democracy .................................................................................................................................... 45 Regime Survival ........................................................................................................................... 45 Consolidation and Instability ....................................................................................................... 46 Occupation and Regime Imposition ........................................................................................... 47 Optimal, Semi-optimal and Suboptimal Winning Coalitions .................................................. 47 Marginal Value of Occupation .................................................................................................... 48 The Marginal Value Model .............................................................................................................. 50 Threat Processes ............................................................................................................................ 50 Variables ........................................................................................................................................ 53 iv Hypothesis ..................................................................................................................................... 55 Methodology ...................................................................................................................................... 56 Case Selection and Justification ....................................................................................................... 60 Selecting on the Dependent Variable ......................................................................................... 61 Organization of the Thesis ............................................................................................................... 62 Chapter 3: West Germany, 1945-1975 ................................................................................................ 63 Introduction ........................................................................................................................................ 63 Rival Hypotheses versus Case History ........................................................................................... 66 Rival Hypothesis: Demilitarization ............................................................................................. 66 Rival Hypothesis: Political Re-education ................................................................................... 69 Rival Hypothesis: Institutional Design ....................................................................................... 72 Rival Hypothesis: Delegitimation via Defeat ............................................................................ 77 Rival Hypothesis: Geographic Division ..................................................................................... 80 Summation .................................................................................................................................... 84 Test of Working Hypotheses versus Case History ........................................................................ 86 Threat Environment ...................................................................................................................... 86 Signaling Security Value ............................................................................................................. 92 Facilitating Regime Consolidation ............................................................................................. 99 Domestic Outcomes ................................................................................................................... 104 Summation .................................................................................................................................. 110 Reconciliation of Competing Hypotheses .................................................................................... 111 Economic Drivers ....................................................................................................................... 113 Summation of Findings and Conclusion ...................................................................................... 120 Chapter 4: Japan, 1945-1972 .............................................................................................................. 124 Introduction ...................................................................................................................................... 124 Rival Hypotheses versus Case History ......................................................................................... 126 Rival Hypothesis: Delegitimation and Demilitarization ......................................................... 126 Rival Hypothesis: Retention of Institutions ............................................................................. 129 Rival Hypothesis: Prewar Modernization ................................................................................ 132 Test of Working Hypotheses versus Case History ...................................................................... 135 Threat Environment .................................................................................................................... 135 Signaling Security Value ........................................................................................................... 140 Facilitating Regime Consolidation ........................................................................................... 146 Domestic Outcomes ................................................................................................................... 155 Summation .................................................................................................................................. 163 Reconciliation of Competing Hypotheses .................................................................................... 163 Economic Drivers ....................................................................................................................... 165 Summation of Findings and Conclusion ...................................................................................... 170 Chapter 5: Weimar Germany, 1919-1933 ......................................................................................... 172 Introduction ...................................................................................................................................... 172 Rival Hypotheses versus Case History ......................................................................................... 174 Rival Hypothesis: Institutional Design ..................................................................................... 174 Rival Hypothesis: Demilitarization ........................................................................................... 177 Rival Hypothesis: Prewar Modernization (or lack thereof) ................................................... 178 Summation .................................................................................................................................. 179 v Test of Working Hypotheses versus Case History ...................................................................... 180 Threat Environment .................................................................................................................... 180 Signaling Security Value ........................................................................................................... 184 Facilitating Regime Consolidation ........................................................................................... 191 Domestic Outcomes ........................................................................................................................ 194 Summation .................................................................................................................................. 209 Reconciliation of Competing Hypotheses .................................................................................... 210 Economic Drivers ....................................................................................................................... 212 Summation of Findings and Conclusion ...................................................................................... 217 Chapter 6: South Korea, 1945-1950 .................................................................................................. 221 Introduction ...................................................................................................................................... 221 Rival Hypotheses versus Case History ......................................................................................... 222 Rival Hypothesis: Leadership Failure ...................................................................................... 222 Rival Hypothesis: Prewar Modernization (or lack thereof) ................................................... 225 Test of Working Hypotheses versus Case History ...................................................................... 227 Threat Environment .................................................................................................................... 227 Signaling Security Value ........................................................................................................... 234 Facilitating Domestic Consolidation ........................................................................................ 248 Domestic Outcomes ................................................................................................................... 258 Summation .................................................................................................................................. 266 Reconciliation of Competing Hypotheses .................................................................................... 267 Economic Drivers ....................................................................................................................... 270 Summation of Findings and Conclusion ...................................................................................... 272 Chapter 7: Review of Findings and Conclusion ............................................................................... 274 Review of Hypotheses and Findings ............................................................................................. 274 Reviewing Rival Hypotheses ......................................................................................................... 279 Demilitarization .......................................................................................................................... 279 Political Re-education ................................................................................................................ 279 Institutional Design ..................................................................................................................... 280 Delegitimation via Defeat .......................................................................................................... 281 Geographic Division .................................................................................................................. 282 Retention of Institutions ............................................................................................................. 283 Prewar Modernization ................................................................................................................ 284 Economic Growth ....................................................................................................................... 285 Summation .................................................................................................................................. 286 Brief Application of the Marginal Value Model to Ongoing Cases .......................................... 287 Palestine ....................................................................................................................................... 287 Afghanistan ................................................................................................................................. 289 Iraq ................................................................................................................................................ 292 Conclusion ....................................................................................................................................... 294 Contributions to the Literature ................................................................................................... 295 Going Forward ............................................................................................................................ 297 Bibliography ......................................................................................................................................... 299 vi List of Tables Table 2.1: Effect of Component Factors on the Marginal Security Value of Occupation .......... 54 Table 2.2: Summary of Cases ............................................................................................................ 61 Table 3.1: Translation of Votes to Seats in Federal Elections to the Bundestag .......................... 76 Table 3.2: Influences on the Threat Environment in West Germany ............................................ 86 Table 3.3: Indicators of Occupation Security Value in West German Occupation ..................... 93 Table 3.4: Evidence of Facilitating Regime Consolidation in West Germany ............................. 99 Table 4.1: Influences on the Threat Environment in Japan .......................................................... 135 Table 4.2: Indicators of Occupation Security Value in Japan ...................................................... 139 Table 4.3: Evidence of Facilitating Regime Consolidation in Japan ........................................... 147 Table 5.1: Influences on the Threat Environment in Weimar Germany ..................................... 181 Table 5.2: Indicators of Occupation Security Value in Weimar Germany ................................. 182 Table 5.3: Evidence of Facilitating Regime Consolidation in Weimar Germany ..................... 191 Table 6.1: Influences on the Threat Environment in South Korea............................................... 227 Table 6.2: Indicators of Occupation Security Value in South Korea ........................................... 234 Table 6.3: Evidence of Facilitating Regime Consolidation in South Korea ............................... 248 vii List of Figures Figure 2.1: Expected Effect of Marginal Security on Radical and Moderate Political Preferences ......................................................................................................................... 42 Figure 3.1: Various Indicators of Civic Engagement in West Germany ........................................ 71 Figure 3.2: Performance of the Optimal and Suboptimal Coalitions in Federal Elections to the Bundestag, 1949-1972 ...................................................................................................... 79 Figure 4.1: Expansion of the Japanese Armed Forces .................................................................... 129 Figure 4.2: Performance of the Optimal and Suboptimal Coalitions in National Elections to the Lower House of the Japanese Diet, 1946-1967 ..................................................... 159 Figure 4.3: Inflation in Occupied Japan: Yen to Dollar Exchange Rate, 1945-1950 .................. 167 Figure 5.1: Performance of the Optimal and Suboptimal Coalitions in Federal Elections to the Reichstag, 1918-1933 ..................................................................................................... 196 Figure 5.2: Succession of Governments and Chancellors in Weimar Germany, 1918-1933 ..... 196 viii Chapter 1: Introduction and Literature Review The Big Picture Introduction The phenomenon of the foreign-imposed democratic regime is one of great importance at this point in history. Rare but consequential, forced liberalization is and has been a test of the efficacy of great power resources under certain conditions, and thus a test of the efficacy of great powers themselves. This dissertation will address the following question: what are the policies and prevailing conditions that favor survival in foreign-imposed democratic regimes? The question is of extreme relevance to the recently concluded democratization project in Iraq and ongoing efforts in Afghanistan, as well as the controversial democratic experiment taking place under the Palestinian Authority. Why was democratization relatively swift and stable in Japan and Germany as compared to the contestation that has dogged the new regimes in Iraq and Afghanistan (Pei 2003, Porch 2003, Dawisha 2005, Bensahel 2006, Barak 2007)? The answer to this question will shed light on enduring puzzles in power politics, such as the fungibility of power, and why an occupier with a favorable and asymmetrical balance of forces can attain its goals in one point in space and time but not in another. 1 Significance Foreign-imposed regime change is a relatively rare phenomenon and forced democratization even rarer. However, these cases are important because of the attractive benefits of increased freedom, security and perhaps prosperity for both states in the event of success, the obviously high economic and security costs to the target state and region in the event of failure, and the economic and human cost of the endeavor for the intervening state, magnified by the uncertainty of success (Russett and Oneal 2001, Plumper and Martin 2003, Thompson 2004, Lo, Hashimoto and Reiter 2008). If imposed democratization has a sufficiently positive effect on freedom levels and regime longevity within a target polity while improving security condition for both parties, the risks and costs of such a policy may be acceptable. However, attempting such a policy without an informed model increases the probability that the costs will outweigh the benefits. It is possible that the intervention will have the opposite intended effect. Failure to democratize in and of itself may non-trivially delegitimize democracy as an alternative regime type of non-democratic societies in the future, either through adverse norm diffusion, historical memory, or modification of the perceived costs and benefits based on the information revealed by the case. Moreover, a failed intervention will reduce the security of both the initiator and the target by introducing new sources of instability into the international system (Rubin 2000, Murdoch and Sandler 2002, Kang and Meernik 2006, Coyne 2006). Any violation of sovereignty is typically an expensive undertaking, but interventions may lead to invasions, and invasions to occupations, adding dollars to the wrong side of the ledger and names to the eternal list of the fallen. 2 This work aims to address these questions and contribute to literatures in political science international relations, public policy and history. The majority of the (small but growing) literature on foreign imposed regime changes (FIRCs) has employed a largely quantitative methodology and a large-n sample, as noted above. While this is appropriate for a systematic understanding of the correlates between imposed regimes and stability or other outcomes, it is problematic at best. While large-n models can measure the relationship between independent and dependent variables, the underlying processes remain theoretical, unexamined, and/or untested. On the other hand, with a small-n process-tracing approach, we can derive a better understanding of these underlying processes. This particular research project will trace the process from intervention to democratization in multiple cases, exploring the factors that intervene to produce or prevent consolidation of the new regime. This approach aims to move beyond an existing literature that currently focuses on ?what? towards an understanding of ?why? and ?how.? Second, an in-depth and systematic exploration into democratic FIRCs could lead to the wiser implementation of related policies. The FIRC option is high risk, high cost, and high payoff. Many FIRCs involve a costly military invasion followed by a costly multiyear occupation, so maximizing successes and returns are key. Consider the unfortunate case of South Vietnam, wherein an inconsistent liberalization policy cost well over a million lives, resulted in no lasting increase in freedom in the target region and undermined existing democratic institutions south of the 17th Parallel. On the other hand, consider the cases of Germany and Japan: these former authoritarian and belligerent regimes, destroyed by fire bombings and two nuclear attacks, climbed to top the tier of 3 economic powerhouses as highly democratic and peaceful regimes. Consider also the high stakes of the ongoing democratization projects in Afghanistan, Iraq, Kosovo, Bosnia and Palestine. Understanding this topic is crucial to future planning and safe stewardship of the institutions that protect political freedom for millions living under fragile regimes. Third, the previous US presidential administration reserved the right to employ regime change as a tool in the War on Terror. In a democratic society, the government has a mandate to most appropriately apply its economic resources to extract tangible benefits, and to refrain from risking the lives of its citizens for non-trivial, non-capricious undertakings. It is the hopeful by-product of this research to make such policy less capricious and more rational?or at least to provide those in power and those in opposition with more reliable information?and to help refute or confirm the efficacy of a foreign policy that endorses regime change. Literature Review Introduction The following literature review surveys the broad inquiry into the drives of democratic consolidation and consolidation failure. The majority of the literature is comparative in nature, exploring the endogenous processes and characteristics that dictate whether democracies collapse, thrive or merely survive. The primary focus in this subset of the literature is on constitutions, electoral regimes, parliamentarianism vs. presidentialism, political economies, historical legacies, cultural patterns, generational 4 replacement, learning and the role of time itself. Less explored are the effects of interstate politics on democratization. This sub-canon is growing, but small. More importantly, the small body of work in this area fails to satisfy. It suffers from questionable methodological choices, contradictory findings and problems of external validity. I review this literature, its successes and failures, and conclude that there are approaches that are more suitable to the problem at hand. It is important to survey the broader consolidation literature and not simply the much smaller, less developed body of work addressing the role of interstate politics. For one, before we explore the role of foreign occupation in the consolidation of democracy we must specify what democratic consolidation is. This proves surprisingly difficult. As is often the case in political science (and social science in general), the concept suffers from a specification problem. I resolve this problem (for the purposes of this dissertation, not for the discipline as a whole, of course) at the outset. Second, this literature is essentially a survey of rival hypotheses. If one can perfectly specify a model of democratic consolidation in imposed democracies using solely the endogenous variables that apply to all democracies, then a model based on international-level conditions is superfluous. Third, to justify this project one must demonstrate that there are gaps in the literature. This review finds that, yes, there are lacunae, and a model of democratization that specifically incorporates foreign occupation can fill some of them. Fourth, in an occupation, the occupying forces actually have some influence on the supposedly endogenous variables that are so prominent in the literature. Depending on the level of involvement, the occupiers are often responsible for designing electoral procedures, balancing legislative and executive powers and authoring the constitution itself. The 5 occupiers control trade, regulate economic activity and authoritatively redistribute massive amounts of value within and across borders. The occupiers may even play a role in political education, thereby modifying cultural patters and mediating the effects of generational replacement. For each of these reasons, it is important that we review the broader literature on democratic consolidation: not just so we might discover new relationships, but also to explore how the presence and behavior of an occupying power might influence the relationships we already recognize. What is Democratic Consolidation? The term ?democratic consolidation? itself is contested. As the literature evolved in the wake of the Cold War, ?democratic consolidation? took on a variety of meanings as diverse as legitimation, value diffusion, neutralization of antidemocratic opponents, civilian control of the military, the elimination of authoritarian opposition, party building, stabilization of electoral rules, routinization of political behavior, decentralization, introduction of direct democracy, judicial reform, economic development and economic stabilization (Schedler 1998). In many cases these are examples of processes that favor consolidation. In others they are consolidation resultants. If we are to understand how foreign intervention may facilitate or retard democratic consolidation, we must first settle on a working definition. The source of the confusion is the latent nature of the consolidation process. We cannot directly observe consolidation?in fact we can only directly observe when consolidation fails. But the boundary of consolidation is a fuzzy one. (Schedler 2001a). That democracies may continue to survive due to exogenous factors unrelated to 6 consolidation complicates the issue (Svolik 2008).1 In the absence of direct, ex ante information, observers grasp for proxy variables. Huntington (1991) claims that we can pick out consolidated democracies by observing party turnover and diplomatic relations. In this scheme, a democracy is likely consolidated if the government shifts shifted twice from one party to another and if the state maintains positive relations with other democratic regimes. Svolik (2008) relies on a model he designed to predict reversals, and then fits that model to existing data. Here, consolidation is a continuous variable that is essentially the inverse of the probability of regime collapse. But there is still a lack of consensus as to what investigators should be looking for. Is consolidation evident when we see actors playing by the rules of a democratic regime? Are regimes consolidated when popular attitudes towards democratic institutions are favorable? Are democracies consolidated when the favorable institutions and overriding conditions are observed? Schedler (2001b) argues that the best approach is the one that takes into account the behavior of domestic actors: are they ?playing by the rules,? looking for ways around the rules that constrain them or actively looking to eliminate the rules and establish a new set of rules that is more favorable to them and less so to everyone else? According to Schedler, it is this behavior is the most causally adjacent to the phenomena of regime collapse. Yes, it is possible that, before actors start subverting the regime, their 1 The problem is akin to determining the structural integrity of a house from the outside alone. A house with a dilapidated exterior but a healthy frame that sits on bedrock could stand for decades. Conversely, a house with a beautiful fa?ade built on shifting soil an held up by a frame riddled with termites might blow over from the lightest breeze. We cannot tell from a snapshot which house is stronger or weaker; we can only observe whether the house stands or not. Instead, we must rely on more subtle indicators. Does it lean a certain way? How long do houses that lean so far stand before they fall? What is the typical life span of a house built in a similar style in a similar neighborhood and climate? Denied a more direct inspection we must rely on methods that are more superficial and actuarial in nature. 7 preferences might be reflected in societal attitudes towards democratic institutions. It is also likely that such actors would be more successful when institutions are too weak to marginalize, prevent or deter them. However, it is also true that weak institutions do not always remain weak, and that even weak institutions might persist if there is no threat to them from inside or out. It is also true that subversive behavior is risky (Lichbach 1998), and that the preferences motivating risky behavior tend to be latent prior to the moment of decision (Holt and Laury 2002). In other words, a survey of public attitudes might indicate positive affect for the regime only because those surveyed were not given sufficient incentive to demonstrate their dissatisfaction. Thus, it is best to rely on the observation of actual behavior. Observing costly anti-government behavior is a strong signal of the weakness of the regime precisely because of the signal?s cost. The shortcoming to this approach it is that destabilizing behavior might not become observable until the regime is already on the verge of collapse, but that is okay for the purpose of this project. We do not seek to construct an early warning system for democratic recidivism, but rather we seek objective ex post facto indicators such as delicacy and durability that might serve as dependent variables. The measure of consolidation in this dissertation will therefore take Schedler?s approach, defining it as the absence of behavior that seeks to undermine the regime. Inherent Instability Democratic consolidation is an interesting topic precisely because some democracies revert back to authoritarianism. Theoretical treatments of democratization and consolidation claim that democracies are inherently unstable to begin with. 8 Democratic systems must not bar the possibility that the winning coalition today might be the losing coalition sometime tomorrow, and thus democracies operate in?nay, depend on?an environment of some uncertainty. This uncertainty means that, given enough time, all members of the selectorate may be negatively affected by policies that redistribute economic resources and political power. The potential for redistribution creates an incentive for out groups to overthrow the existing order and replace it with one that is less uncertain and more favorable to them (Boix 2003). Instability is also a product of the relatively low costs of defecting from the winning coalition in democratic regimes (Bueno de Mesquita, et al. 2005). Winning coalitions in democratic regimes garner support by distributing public goods, since democratic selectorates are too large to buy off with private goods. Public goods are enjoyed by the entire selectorate, however, not just the winning coalition. As a result, pretenders to the throne can free ride off public goods while simultaneously attempting to position themselves as the winning coalition. The lack of barriers to peaceful competition results in an inherent instability in the composition of the winning coalition and, necessarily, a bit of uncertainty about future policy. This is unavoidable. As Przeworski (1986) writes, the ?process of establishing a democracy is the process of institutionalizing uncertainty, of subjecting all interests to uncertainty.? This uncertainty stems from the reality that, for a regime to be democratic, no coalition is guaranteed an opportunity to form a government. Furthermore, all regime transitions result in information deficits. O?Donnell and Schmitter (1986, Part IV, 6) characterize democratic transitions as naturally uncertain, during which ?the rules of the political game are not defined? in constant flux? [and] arduously contested.? Put another way, ?the emergence of uncertainty marks the beginning of regime change? 9 (Schedler 2001a, 2). Democracies produce a certain amount of uncertainty and new democracies produce even more. Diamond (1990) reveals additional sources of political instability in democracies. He points out that democratic institutions are essentially collections of paradoxes: democracies must permit political competition but constrain it in such a way that it plays by the rules; democracies must be responsive to preferences but not be so responsive that the winning coalition can permanently change the rules of the game; democracies must be constrained by the consent of the governed but still able to implement unpopular policies in times of crisis. Fortunately, these are not true paradoxes (they can be resolved) but rather opportunities for the new regime to satisfice. New democratic regimes inadvertently create some of this instability through their own behavior. Linz and Stepan (1978), with support from Haggard and Kaufmann (1995), find that new regimes try to solve too many problems at once, overburdening themselves and raising expectations to unreasonable levels. Some of this instability is evident in the revision of democratic institutions. Alexander (2001) points to the fact that institutional arrangements?such as federalism, parliamentary-presidential balance, electoral systems, and guarantee of negative and positive rights?remain in flux even in the longest-lived democracies. This is not to say that democratic states are mired in perpetual uncertainty with no capacity to self-stabilize. While Przeworksi is correct that democracies cannot guarantee the success of any given coalition, it is also correct that democracies are better at producing information about the distribution of preferences across society, better at producing the information about how political actors pursue power, and more efficient at distributing this information as it is produced. Alexander (2002) argues that there is just 10 as much variation in uncertainty within regime types as there is across. Democratic institutions produce information about how social, economic and political conflicts will be resolved, and in doing so encourage sub-state actors to converge on a smaller set of policy options. The role of uncertainty is especially relevant in the case of imposed democracies, wherein the presence of an outside force heavily distorts the rules of political competition. Occupation governments serve at the pleasure of the occupiers until they see cause to leave or unless the occupiers make the costs of the occupation unbearable. Occupiers choose the winning coalition and their continued presence is a threat to political competition of all sorts. Put simply, occupation governments are inherently antidemocratic. However, this could work in the favor of nascent democratic regimes. Given their preponderance, occupation governments can prevent redistributive activities that generate opposition to the regime. Or, alternatively, the occupation government may redistribute value in such a way that the distribution of value will be more favorable to future consolidation. An occupation government might be far more capable of withstanding retaliation by disinherited parties than an infant democratic government, and would therefore be the more suitable regime to undertake such reforms when necessary. Additionally, if new democracies falter by trying to solve too many problems, a better-resourced occupation government could lend its hand. Since the occupier?s power and presence produce strong incentives and disincentives for certain types of political behavior, the occupier is in a unique position to save unconsolidated regimes from the perils of their own uncertainty. 11 The Role of Time Scholars are split over whether new democracies are more or less stable than longer-lived democracies. Some hypothesize that new democracies will enjoy a ?honeymoon effect,? during which they will have more leeway for undertaking painful economic reforms and other redistributive policies. Haggard and Kaufman (1995) find that the capacity to sustain early challenges to the regime is a function of the ineptitude and illegitimacy of the previous regime. In other words, democratic regimes enjoy a grace period only when they replace autocratic regimes that presided over a period of low growth or contraction. Bernhard, Reenock and Nordstrom (2003) find no evidence of a honeymoon; quite the contrary, they find that new democracies are more susceptible to negative economic shocks. Linz and Stepan (1978, 1996) find that new democracies are generally more vulnerable when they are young. Svolik (2008) agrees, adding that the extant literature underestimates the likelihood of collapse in the first few years. Specifically, his models indicate that democracies of less than seven years? duration are more likely than not to regress. Moreover, the probability of regime collapse remains greater than one-in-ten until the fifty-second year of democratic rule (when controlling for development, growth and presidentialism). As far as imposed democracies are concerned, Enterline and Greig (2008b) find that they are just as fragile in their early years?30% fail by their tenth anniversary. Time is an important variable in the case of imposed democracies that are, by definition, new. More importantly, occupation governments may play the role of caretaker so that the democratic regime has enough time to consolidate. Occupations are not permanent and, when conducted by democracies, are very sensitive to their own time 12 effects. However, an occupation that enjoys support and resources can stick around. If democracies need time to grow, then a prudent occupation could shelter and nurture said democracy until it is strong enough to stand on its own. The Role of Institutions Research investigating the relationship between regime survival and institutional design is prominent in the literature on consolidation, and is particularly relevant to democratic imposition (where it is often the imposer who is building these institutions in the first place). Institutions are central to democracy, serving as a stable platform on which political actors can peacefully compete for the opportunity to form a winning coalition (Alexander 2001). Since losing coalitions have an incentive to overturn the system when the stakes are high, it is the role of institutions to credibly commit the winning coalition to open and competitive selection mechanisms (Collier 2009). Without credible constraints on the winning coalition, political contestants are driven to extremes to secure power or to protect themselves from the powerful. Democracies must set the rules for democratic political conduct, punish those who cheat, effectively check arbitrary leaders, replace arbitrary rules with rational ones, and devolve rulemaking authority to the general population (Moore, Jr. 1966). These rules originate on parchment as laws, regulations and constitutional provisions. While not a sufficient condition for fixing the behavior of political agents, legal documents produce information about the payoffs of supporting and opposing the regime (Carey 2000, Stepan and Skach 1993), thereby reducing uncertainty. 13 One of the ongoing debates regarding democratic institutions concerns the effect of presidentialism or parliamentarianism on democratic consolidation. More often than not, studies demonstrate a negative correlation between presidential regimes?where the chief executive is elected separate from the legislature and retains an independent base of power?and regime survival. Linz and Stepan find that parliamentary systems generally survive longer than presidential ones (1996). Huntington (1991) found that parliamentary democracy was more successful than presidential democracy during the third wave of democratization. Przeworski, et al. (1996) confirms these findings, as do Stepan and Skach (1993), who find that presidential regimes revert to authoritarianism thrice the rate of parliamentary regimes. Svolik?s (2008) findings suggest that presidential democracies almost never consolidate, or rather that the probability of democratic reversal in presidential regimes inevitably exceeds 10% even when controlling for development, growth and regime longevity. Presidential systems may be especially hazardous to first- time democracies, or democracies with no history of democratization prior to the last authoritarian regime (Hiroi and Omori 2009). Why would parliamentary systems be less prone to reversal? Haggard and Kaufman (1995) find that executives in presidential systems are less likely to cooperate and coordinate with the legislature?and vice versa? to resolve crises. Therefore, presidential systems fall prey to gridlock at the very moment when decisiveness is at a premium. If a democratic state faces a crisis and the government fails to act, the incentive to replace the regime with a more decisive, less deliberative government increases. Cohen (1997) confirms this hypothesis, finding that proportional democracies (most commonly found in parliamentary democracies) are 14 more successful in mediating civil conflict than majoritarian democracies (which tend to co-occur with presidential regimes). The efficacy of parliamentary democracy is largely dependent on the electoral system and the effect it has on party representation in the legislature. New democracies are most successful when the party distribution in the first elections reflects the ideological distribution of the electorate?but not too perfectly (O?Donnell and Schmitter 1986). If the distribution of preferences in the electorate is either exceptionally wide or bimodal, a faithful reflection of the true distribution of preferences will result in party fragmentation and polarization. Fragmentation is problematic as it endows minority parties with a veto over government formation and increases the minimum number of parties required for a coalition. Polarization thwarts government formation and policy making by depriving plurality parties of ideologically adjacent coalition partners. Since the probability of fragmentation increases with the total number of parties in the legislature, single-member, first-past-the-post districts, along with electoral thresholds for participation in the legislature, reduce the chances of fragmentation. This, in turn, lowers the probability of gridlock (Haggard and Kaufman 1995, Elgie 2008). Bandelj and Radu (2006) confirm that higher electoral thresholds correlate with democratic consolidation among post-communist democracies. The need to satisfice between deliberation and decisiveness highlights the role that parties play in democratic consolidation. Parties are responsible for aggregating constituent preferences and translating them into policy (Lai and Melkonian-Hoover 2005). They also facilitate the peaceful transfer of power?the costs of electoral failure are lower because the party can live to fight another day. Strong 15 party competition has been linked to reduced rates of democratic reversal and may solidify norms of peaceful political competition (Linz and Stepan 1996, Ibid). Due in part to the coordination problems faced by parliamentary democracies, there is a smaller but non-trivial collection of studies that find a negative correlation between parliamentarianism and regime survival. For instance, Hiroi and Omori (2009) find that parliamentary regimes are more likely to collapse in re-democratized states. However, pro-parliamentary findings vastly outnumber pro-presidential findings. Most contradictory to the hypothesis that presidential systems are more stable are studies indicating gridlock?the supposed scourge of parliaments?to be more debilitating in presidential systems (Elgie 2008, Stepan and Skach 1993). Furthermore, semi- presidential regimes (such as Weimar Germany) should be less stable than pure presidential regimes, since power sharing could introduce additional conflicts of interest between the executive and legislative branches. Przeworski, et al. (1996) tangentially confirms this hypothesis, finding that the combination of presidentialism and party fragmentation severely harms the prospect of consolidation in new democracies. On the other hand, Elgie (2008) finds that mixed systems are not significantly correlated to democratic reversal. If we broaden the definition of ?institutions? to include the military, Samuel Huntington (1991) is very clear about their ?designs.? Democracy necessitates the removal of the armed forces from policymaking. A system of government dependent on self-enforcing rules and egalitarian political competition cannot co-exist with elements that curtail an elected regime?s monopoly over the use of violence and retain a disproportionate capacity to influence outcomes. For democracy to succeed, it must 16 subordinate the military to civilian control. In pursuit of this aim, Huntington recommends that new democracies purge disloyal officers, clarify and enforce the chain of command, reduce ground forces to a reasonable size, increase compensation to remaining loyal elements, redeploy forces away from the capital and to the borders (or overseas), and develop a battery of active civilian supporters willing to take to the streets in the event of a military coup. This is compatible with Stepan?s (1986) assertion that ?externally monitored redemocratization? (such as the cases of West Germany and postwar Japan) is most successful when previous institutions, including the military, have been utterly reduced?paving the way for the imposers and the new civilian regime to rebuild institutions as they see fit. Finally, there is a vein of literature that asks whether stabilizing institutions? security provision, contract enforcement, social safety nets?is necessary prior to the implementation of elections and legislatures. The general consensus is that they are. Linz and Stepan (1996) go so far as to say that state consolidation is a necessary precondition of democratic consolidation. Enterline and Greig (2008a) find that rebuilding the military correlates negatively with political instability in imposed regimes once domestic force levels exceed 150,000 troops. M?ller and Skaaning (2011) reiterate these findings. They find that electoral rights, political liberties and rule of law are far more likely to increase once the new regime has established a monopoly on the use of force within its territory. As Linz (1997, 118) writes, ?No state, no Rechtstaat [rule of law], no democracy.? And yet, we must be careful not to credit variance in institutional design with too much empirical weight. First of all, the literature is not in total agreement on the role of federalism, presidentialism or party system. Bernhard, Reenock and Nordstrom (2003) 17 find no relationship between presidentialism, fractionalization and democratic survival. Haggard and Kaufman find that presidentialism and parliamentarianism can both paralyze crisis response in different ways. Boix (2003) finds that institutional design has little effect on consolidation in societies with high inequality or factor specificity. Diskin, Diskin and Hazan (2005) find no relationship between federalism, proportionality of the legislature, and democratic consolidation. Gasiorowski and Power (1998) find no evidence of a relationship between presidentialism, party systems, and consolidation when controlling for incomes, democratic neighborhoods, and inflation. Ulfelder and Lustik (2007) also reject the relationship between presidentialism and reversal when controlling for growth and development. Second of all, institutions may help powerful actors credibly commit to the pursuance of democratic politics in the future, but they are necessary components?not sufficient ones (Alexander 2001). Institutions may constrain the translation of political preferences into representation and policy, but they do not constrain the original preferences of the electorate (Alexander 2002). This is problematic, because the range of electoral preferences independently affects the probability of party fragmentation, polarization, and the capacity for disenfranchised actors to attack the new regime from the outside. Simply put, stable democracy requires not only the stabilization of the translation of preference to policy, but the moderation of those preferences, as well. The Political Economy of Consolidation The relationship between economics and democratization is complex, so much so that it is often difficult to tell which way the causal arrow is pointing. Nevertheless, 18 among the most common findings in the democratization literature is the positive relationship between democracy and incomes, and that economic contractions are hazardous to democratic regimes, especially new ones (Geddes 1999). Bernhard, Reenock and Nordstrom (2003) find that growing economies and rising incomes reduce the chance of a democratic reversal. Przeworski, et al. (1996) find that states are more likely to remain democratic if they are already affluent and production is on the rise. Ulfelder and Lustik (2007) confirm this specifically in the case of re-democratizing states. Przeworski and Limongi (1997) find that poor democracies are especially susceptible to reversals, and that regime collapse is hastened by shrinking incomes and sluggish growth. Arguably, the most important finding about the relationship between development and democracy is that, while high incomes do not predict the transition from autocracy to democracy, they do correlate negatively with transitions from democracy back to autocracy (Przeworski and Limongi 1997). Specifically, once average incomes surpass $4,000 per year (contemporary), the probability of democratic reversal is essentially nil. Diskin, Diskin and Hazan (2005) and Gasiorowski and Power (1998) confirm these findings. Haggard and Kaufman (1995) find that democracies are less vulnerable to recession in cases where the previous autocratic regime collapsed during an economic contraction. Enterline and Greig (2008a, 2008b) find that the poorest and most stagnant imposed democracies are the most likely to experience political instability and regime collapse. Svolik (2008) also finds that incomes correlate positively with consolidation, but more importantly reveals that unconsolidated democracies are harmed by recession while consolidated democracies weather recession just fine. Inflation and deflation also appear linked to democratic reversal. Przeworski, et al. (1996) finds that 19 moderate inflation (between 6% and 30%) is most conducive to the survival of democratic regimes. Gasiorowski and Power (1998) find that high inflation early in the life of a democratic regime is especially hazardous, while Haggard and Kaufman (1995) find that the stabilization of prices is a necessary precondition for economic growth and thus, consolidation. Why are growth and development good for democracy? There is no shortage of plausible causal explanations. Haggard and Kaufman (1995) suggest that growth produces value that the winning coalition can use to compensate losers, thus depriving out-groups of a grievance that they could use to rally a challenge against the new regime, as well as simply lessening the net economic incentives for rebellion. Mousseau (2000) submits that market prosperity fosters values that are specifically conducive to democracy, including full faith and credit, independent decision making, and peaceful negotiation and compromise. Since party identification is weak in the first elections of a new democratic regime (O?Donnell and Schmitter 1986), prosperity and poverty may establish the primary incentive structure for choosing between parties, or choosing to vote at all, in the early years of the regime. On the other hand, development and growth may only be proxies for other causal processes. Incomes and production are positively correlated with the modernization of the political economy?the replacement of ancient feudal modes of production with modern, commercial economies. This is the process?combined with a balance between aristocracy and monarchy and an independent trading class?that Barrington Moore (1966) credits for the rise of democracy in the United Kingdom, United States and France. Moore famously wrote, ?No bourgeois, no democracy,? (Moore, Jr. 1966, 418) 20 but that was not the only necessary condition he specified. Yes, there must be a capitalist class, but traders, bankers and manufacturers must be wealthy and powerful enough to marginalize landed interests such as the peasantry and landed aristocracy. The high levels of inequality in the feudal countryside raise the stakes of political competition, producing incentives for landowners to prevent democratization and for peasants to pursue harsh redistribution. If antiquated feudal production modes survive a democratic transition, they will likely seek to reverse said democratic transition, or at least retard the process of consolidation. Growth and development also co-vary with capital mobility and income equality, two processes which Boix (2003) considers essential for democratic transition and consolidation. Since democracies carry with them an inherent capacity for redistribution, high levels of inequality mean higher stakes for haves and have-nots. Those at the top of the wealth and income ladder face the possibility of a winning coalition comprised of the (far more numerous) lower classes, which could in turn use the instruments of state to relieve the upper classes of their wealth. In unequal societies, the incentives for the lower classes to pursue such a policy, as well as the absolute cost of such policies on wealth holders, are even higher. Higher risk of economic loss in unequal democracies translates into greater incentives to remove the existing regime. Greater equality, on the other hand, means lower stakes for rich elites and lesser incentives to challenge democratic processes?and less of an incentive for have-nots to pursue redistributive policies in the first place. Boix?s findings on the relationship between consolidation and income inequality confirm Przeworski, et al. (1996), who found that equalization of incomes correlates with democratic regime survival. 21 Boix finds that the same applies to capital specificity. In democracies with more open economies, capital tends to be more mobile (less specific) and harder to tax. Thus, the capacity for democratic governments to redistribute wealth is reduced, and the threat to wealth holders alleviated, independent of prior inequality. Therefore, capital mobility?which can be achieved by lowering trade and investment barriers and introducing convertible currency?has the same effect as egalitarian incomes in improving the chances of democratic consolidation.2 Boix also sees an interactive effect between economic variables and institutional ones. In asset specific political economies, presidential systems heighten the risks of redistribution and harm the prospects for consolidation. Federal systems, on the other hand, reduce the capacity for central governments to pursue redistribution. Boix?s claims about the benefits of openness aside, Bresser Pereira, Maravall and Przeworski (1993) caution against the dangers of sudden trade and capital liberalization in democracies. Haggard and Kaufman (1995) find that economic liberalization without rehabilitation of state capacity will lead to stagnation and anti-democratic challenges: ?wherever democratic governments followed neo-liberal tenets, the outcome has been stagnation, increased poverty, political discontent, and the debilitation of democracy.? Indeed, liberalization?and the volatility that accompanies exposure to the global market?may initially plunge a state into recession if it has to abandon corrective exchange and interest rate measures (Frieden 2007, Stiglitz 1999). 2 Boix (2003) is also careful to note that economic openness results in factor price equalization, which will depress the wages of laborers in labor-poor societies (such as the United States). This could exacerbate inequality and cancel out the benefits of reduced capital specificity. 22 Finally, scholars recommend caution when extrapolating economic findings in particular to new cases, as well as in their application in the policymaking. Haggard and Kaufman (1995) warn that it is difficult to differentiate ?good? economic performance from ?bad? without an appropriate reference point. For instance, a state with 5% GDP growth is a huge success if it is a western industrialized state, but a developing state with a similar rate of growth would be considered developmentally moribund. In addition, short-term growth may be funded by unsustainable policies (such as high inflation) that will undermine economic performance in the long run. All things considered, one must always remain vigilant of lagged effects. Growth, in particular, may not generate regime survival benefits until several years after the boom phase (Tang 2008). There are even some who claim that the relationship between recession and regime death is insignificant (Ulfelder and Lustik 2007, Bandelj and Radu 2006). In terms of imposed democracies, economic drivers are of special significance. It is up to the occupier how much they decide to manage the domestic affairs of the target state, including macroeconomic policy. If a state wishes to impose democracy, it will have to constantly consider the effect that occupation policy will have on inflation, growth and incomes. Moreover, the occupier may find it prudent to take a proactive approach and improve the economic prospects of the target state. The role that an occupation government would play in terms of the institutions of a new democracy is obvious. Occupiers have veto power over the occupied state?s choices, constitutional decisions included. The occupier could play a direct role in building the new regime?choosing the electoral system, determining the character of the legislature and the powers of the executive. Alternatively, the occupation government 23 could play an advisory and confirmatory role, allowing domestic actors their own government so long as it meets certain criteria. Or, the occupation may not involve itself in this process at all. The point is that the occupation has the choice to involve itself in this process or not, and thus retains control over the most important decisions related to the consolidation of the regime. The Roles of Culture, Learning and History Democratic regimes, being those that serve at the pleasure of the electorate, may be sensitive to public attitudes about the desirability of democracy itself. It is for this reason that cultural orientation and historical memory may play a role in successful consolidation. While attitudes themselves do not seem to be well correlated with collapse or survival (Fails and Pierce 2010, Gasiorowski and Power 1998, Przeworski and Limongi 1997), there is better evidence for the role of ethnic division. Since ethno- linguistic, religious and social cleavages are often (if inconsistently) linked to civil wars, and since civil wars have the capacity to overthrow regimes, a common hypothesis is that such cleavages are a threat to democracy. They may also interfere in policymaking and coalition formation by polarizing the electorate. Bandelj and Radu (2006) find that ethno- linguistic fractionalization predicts reversal in democratic post-communist states, while Enterline and Greig (2008a, 2008b) find that imposed democracies are more unstable and less likely to consolidate when fractionalization is high. Huntington (1991) offers the caveat that elections may serve only to emphasize religious or ethno-linguistic divisions. Presumably, conflict between these groups is more likely to concern indivisible goods? such as identity definition or the right to exist?which cannot be efficiently redistributed 24 by democratic institutions and which present high stakes regardless of the type of democracy in place. In contrast, Diskin, Diskin and Hazan (2005) find no link between fractionalization and democratic reversal. Several scholars claim to have uncovered a relationship between previous experience with democracy and the likelihood of success the second (or third, or nth) time around. Huntington claims this to be the case. Diskin, Diskin and Hazan (2005) find that previous democratization prevents democratic reversals later on, even when controlling for foreign intervention, economic contraction, and cultural cleavage. Like theories about the consolidating effects of democratic neighborhoods, there is a dearth of causal relationships that explain these findings (see next section). Unlike democratic neighborhoods, however, there seems to be no agreement on what direction the arrow points (if there is any relationship at all). This is not surprising. If a state has a history of previous democratization, it must also have a history of democratic collapse. Accordingly, Przeworski, et al. (1996) contradicts Diskin, Diskin and Hazan, finding a negative relationship between previous democratization (and overthrow) and future consolidation. Bernhard, Reenock and Nordstrom (2003) find that previous democratic regimes have no significant effect on future democratic survival. While occupying regimes have little control over a target society?s culture, and no control over its history, these variables are important in terms of establishing a base probability for consolidation. Foreign powers that arrive in states with a history or culture that is less conducive to democratic consolidation will have a harder time achieving their goals. They will need to expend more resources, and perhaps time, narrowing the 25 spectrum of political contestation so as to empower democrats and marginalize radicals. In order to empower a pro-democratic coalition, it helps if there is one to begin with. The Role of Interstate Politics Specifically germane to this project is a small but growing literature that explores the relationship between interstate political dynamics and successful consolidation, including the direct intervention of state actors to effect regime change. Imposed democracies are a special case, as the new regime was imposed by interests abroad and not by a domestic popular mandate. This raises questions about the legitimacy and stability of imposed democratic regimes. Diamond (1992) cautions that ?exporting? democracy?imposing institutions that popular back home but not necessarily appropriate in the target state?is rarely successful, and that states cannot impose a democracy where there is no preference for it on the ground. Whitehead (1986) finds that public declarations in support of democratic transition correlate poorly with actual transition and consolidation. Coyne?s (2008) work justifies skepticism about the efficacy of direct intervention for the purpose of democratization. While it is tempting to extrapolate from the successes of West Germany and Japan after WWII, it is questionable as to whether the conditions that made those successes possible will repeat themselves. Moreover, full-scale occupations are expensive and maintain questionable legitimacy among the electorate, a view espoused by Stepan (1986). Instead, Coyne favors a ?hands- off? approach that would minimize direct intervention and maximize the role of liberal economics, opening up the regime to foreign trade and investment in a way that would 26 generate pro-democracy attitudes among the target populations. Ulfelder?s (2008) work, which shows that membership in trade organizations reduces the chance of democratic reversal, supports Coyne?s position on free trade. On the other hand, Grimm (2008) finds that imposed democracies, specifically those following a war, are more likely to survive when state apparatuses are reconstructed. She offers that successful democratizers must ensure welfare, rebuild the state, and establish rule of law in order to give the populace in the occupied state an incentive to support the new regime. This implies that those who wish to impose democracy, particularly following wars and during occupations, should be active in rebuilding or replacing the institutions that they just helped destroy. Von Hippel (2000) agrees, claiming that re-establishing security, empowering civil society and strengthening democratic institutions are all necessary conditions for post-intervention democratization. Diamond (1992) suggests that foreign assistance is prudent in helping new regimes weather economic reform early in their existence. Stepan (1986), despite cautioning that analogous opportunities are unlikely to present themselves again, supports the notion that destruction and reconstruction of state institutions (as occurred in Japan and West Germany) is a prerequisite for the successful imposition of democracy. Meernik (1996) shows that absolute democracy levels do not increase in response to American military intervention, but he also finds that democracy levels do rise relative to a target state?s neighbors, especially when combined with the deployment of ground forces and a declaration of preference for democratization. Peceny?s (1999) work supports Meernik?s first point and adds that intervention combined with rhetorical support for free and fair elections predicts increased democracy. 27 Colaresi and Thompson (2003) find that external threats increase the likelihood of democratic reversals. This effect is exacerbated in young and economically open regimes. They offer the theory that external threats justify the suspension of political freedoms and open candidate recruitment, allowing those in power to abolish the regimes that might remove them. They also theorize that older regimes are less likely to resort to anti- democratic policies, as they are more likely to have developed a foreign policy establishment that can provide security efficiently. Walt (1996) would agree that younger democracies are more vulnerable to foreign threats, but would add that these regimes are actually more likely to generate new threats because they are prone to foreign policy miscalculations. In other words, democratization creates the very threats that compromise the regime. Diskin, Diskin and Hazan (2005) also find a relationship between foreign involvement and the likelihood that a democratic regime will fail. Edelstein disagrees (2004, 2008). He finds that external threats increase the likelihood that occupiers will fulfill the goals they set out, including democratization. Rounding out the findings regarding democracy and international dynamics, democratic neighborhoods seem to have an independent, positive effect on the democratization and democratic longevity (Gleditsch 2002, Gasiorowski and Power 1998, Przeworski, Alvarez, et al. 1996, Starr and Lindborg 2003). The converse may also be true: democracies in neighborhoods where other democracies are collapsing may be more prone to collapse themselves (Starr and Lindborg 2003). Bernhard, Reenock and Nordstrom (2003) challenge the neighborhood effect, finding no significant relationship between regional democracy and democratic consolidation when controlling for economic conditions and institutional design. While Pevehouse (2002a, 2002b) finds that 28 membership in international organizations aids the transition of autocracies to democracies, Ulfelder (2008) finds that membership predicts consolidation only in the case of organizations dedicated to free trade and human rights. Summation of Democratization Literature There is general consensus on a range of findings. Older democracies are less likely to revert to authoritarianism. Party fragmentation and polarization correlate well with democratic reversals. The persistence of military elements from the previous regime is a threat to continued democratization. New democracies are more likely to survive if they successfully re-establish the basic institutions of statehood, such as a monopoly over political violence, as well as the rule of law. Most democracies are capitalist, rich and egalitarian, but not all capitalist, wealthy or egalitarian states are democracies. The persistence of ancient modes of production in modern societies presents a serious roadblock to successful consolidation. Recession, inequality, excessive inflation and deflation indicate democratic reversal, while growth, moderate inflation and high, egalitarian incomes predict survival. Capital mobility and free trade predict survival, but exposure to international trade without the proper state institutions in place is a forecast for regime collapse. Other findings have majority, but contested, support. Younger democracies are more at risk for democratic reversal, although researchers dispute the susceptibility of brand new democratic regimes. Presidential democracies are more likely to revert than parliamentary democracies, although this relationship may be moderated by economic development, inequality, and democratic history. Federalism may or may not correlate 29 with regime survival. Some studies find that democracies are more successful the second time around, while other studies find just the opposite, and others still find no relationship. Foreign-imposed democracies may be more successful when the intervening state takes a hands-on approach in the target state?or they may be less so. Democracies in democratic neighborhoods are more likely to succeed, but we do not know why; it may be related to the fact that new democracies are less successful when external threats are more extreme. Certain compelling and persistent theories find little to no support in the recent literature. Mixed parliamentary-presidential regimes may or may not be more fragile than purely parliamentary or purely presidential regimes. Rapid growth does not appear to trigger democratic reversal. Negative affect towards democratic institutions does not tend to predict regime collapse, nor do ethno-linguistic or religious fractionalization. Of all the literatures presented here, the one with the biggest gap between research produced and real-world importance is that which deals with democratization and interstate politics. This is also the area in which the sample sizes are smallest. The three pieces in this review that most specifically address this question are Coyne (2008) and Enterline and Greig (2008a, 2008b). Considering the dearth of cases of foreign imposed democratization, perhaps it is not surprising that these are the three pieces with the most severe theoretical, methodological and evidentiary shortcomings. The pieces by Enterline and Greig produce very interesting results but do so by drawing from a questionable sample. In what appears to be an effort to expand the sample to the point at which a large-n examination would be possible, Enterline and Greig include cases that do not faithfully represent the concept of polity imposition. Their criteria lump together regime 30 change following military defeat with those that follow rhetorical calls by legislators for free elections (Enterline and Greig 2008a; p. 891). The sample also conflates autonomous postcolonial democratizers with foreign-imposed democratization attempts, so that the gradual, peaceful democratization of Canada?previously a compliant outpost of the British Empire?is part of the same population post-WWII Germany and Japan. Of the ninety-four cases in their sample, only seventeen involved a military victory by the imposer (p. 896). In only 27% of the event-years does the imposer deploy ground forces the target polity (p. 898). Moreover, their specification of instability is overbroad as well, including mundane disturbances such as protests and strikes (p. 895). The researchers do not seem to make any attempt to select out peaceful demonstrations that may not represent actual challenges to the regime, nor do they seem to consider the fact that strikes and protests are indicators of a healthy civil society, and thus may actually indicate consolidation rather than instability. Coyne?s treatment differs from Enterline and Greig?s in that it employs a small-n empirical narrative methodology, but it nevertheless suffers from a questionable research design in addition to serious historical inaccuracies. While Coyne?s sampling criteria are more faithful to what most would consider an attempt at foreign imposed regime change, there are some troubling inconsistencies. For instance, he counts the US interventions in Cambodia in 1970-73 and Somalia in 1993-95 as ?occupations? (p. 15) even though the former was an air war accompanied by a brief invasion, while the latter was primarily concerned with peacekeeping rather than regime imposition. On the other hand, he inexplicably excludes the occupation of the German Rhineland following the Great War, during which an American military government (alongside France and the United 31 Kingdom) controlled a third of the territory between the Rhine and the Franco-Belgian frontier from late 1918 through early 1923.3 Sample problems aside, the main shortcomings here concern the narrative itself. A narrative approach is supposed to yield empirical benefits by tracing causal processes, yet Coyne often fails to provide evidence that reveals a causal chain between independent and dependent variables. His methods are not incapable of rejecting rival hypotheses, but at no point does he share compelling evidence for his conclusion that free trade is actually generating norms or building incentive structures that condition political behavior. Finally, his historical data are suspect at best. In the cases of West Germany and Japan, he claims that the US and its partners avoided credibility and legitimacy issues since the state institutions in the occupied countries had largely survived the war. This position directly contradicts Stepan (1986), who cites the absolute dismantlement of the instruments of state by the Allies in West Germany and Japan as an explanation for the success of democratization in those cases. While in fact Coyne?s description may properly characterize the occupation of Japan, it ignores volumes of historical and political treatments about the occupation of Germany. In terms of the broader literature, this project can shine little light on those theories on which we as a discipline lack consensus. However, the role of the occupier is important to those relationships on which there is more agreement. As the preponderant power in the territory, the occupying force enjoys outsized influence over the construction of new institutions and the redistribution of public and private value. 3 Some democratization scholars truncate the Weimar experience from their sample due to the lack of data about economic performance prior to WWII, but this was not the case for Coyne, who included no fewer than ten cases that occurred prior to 1918. 32 Moreover, the occupation establishes the informational parameters of the regime by signaling to domestic actors who and who may not participate in formal political institutions and establishing incentives for domestic actors who work within those parameters. In doing so, the occupation can mitigate the negative aspects of uncertainty in new democratic regimes. In order for a democratic regime to successfully consolidate, occupying powers must behave in such a way as to narrow the acceptable spectrum of political contestation in favor of actors who prefer democracy as well as build institutions that continue to favor democrats and disfavor radicals long after the occupiers have left. Quo Vadis? The relative absence of literature on imposed democratization, and the problematic nature of the literature that does exist, presents us with an opportunity. The relevance of this area of inquiry to modern international relations and American foreign policy demands that curious scholars seize the opportunity to undertake a rigorous inquiry into the process of democratic consolidation and survival in imposed regimes. The cautionary examples provided above suggest that we should avoid embellishing the sample for the purpose of satisfying a methodological requirement. They also implore us to be careful and thorough in collecting and evaluating the data (not that this is something that a researcher should have to learn via cautionary tales) and explicit in describing causal processes. This is the type of inquiry that I aim to conduct here. In the next chapter, I shall describe a research design that proposes the use of small-n comparative case study and 33 process tracing to test the relationship between a few of the variables presented above and the likelihood of imposed democratic regime survival. This project will draw cases from a population defined by narrow criteria that is faithful to the idea of foreign imposed regime change. It will employ a large body of historical literature to triangulate events constituting a pathway to or away from regime collapse. It will certainly not artificially expand the sample for reasons of convenience or claim historical facts that are not supported by multiple secondary and/or primary sources. It is this research design and this project that I will unfold in the chapters ahead. The following chapter will present the research design for investigating the causes of instability and consolidation in imposed democracies. Chapters 3, 4, 5 and 6 will consist of historical case studies that test the hypotheses presented in Chapter 2. Chapter 7 will summarize the findings from the four cases and test hypotheses across cases. 34 Chapter 2: Research Design Introduction The previous chapter?s review explored the existing body of knowledge on democratic consolidation. The following chapter proposes a model that attempts to explain the success and failure of democratic consolidation in occupied regimes that incorporates and adds to the literature. Going forward, I argue that there is a common thread that ties together most of the theories about democratic consolidation. In essence, these theories all propose that consolidated democracies have successfully narrowed the spectrum of acceptable political contestation. The literature on democracy and instability explains the importance of allowing a certain amount of uncertainty as to permit the transfer of power between winning coalitions, but not so much as that transfer would result in a winning coalition intent on dismantling the regime. Uncertainty within certain boundaries is only possible once radical elements that might infiltrate the regime have been marginalized or eliminated?in other words, once the acceptable spectrum of political behavior is narrow and inclusive of moderates alone. The literature on democratic political economy focuses on marginalizing or preventing the formation of a constituency that has a stake in dismantling the regime. The acceptable political economy is one in which the distribution of resources and policies of redistribution do not encourage moderates to become radicals?a political economy that narrows the acceptable spectrum of contestation. The literature on institution focuses on building institutions that is responsive and inclusive enough to empower moderate, democratic 35 actors but not so responsive or inclusive that it allows antidemocratic elements to gain power through legal means; democratic institutions have to simultaneously represent moderates and marginalize radicals. Again, the story returns to the narrowing of the spectrum of political contestation. Therefore, a model that explains the role that occupiers play in the consolidation of democratic regimes must explain the role that the occupation plays in the narrowing of the spectrum of political contestation. Narrowing the Spectrum of Political Contestation Democratic regimes are insecure so long as there are organized elements within the territory that have the capacity and the will to revise or violate the rules set by democratic institutions. Additionally, democratic institutions are ineffective if the distribution of preferences is so wide that it cannot be reconciled by legislative compromise. Of the terms that scholars have conflated with democratic consolidation (according to Schedler 1998), many are synonymous with narrowing political contestation, including neutralization of antidemocratic opponents, civilian control of the military and the elimination of authoritarian opposition. When the distribution of preferences across the selectorate is narrow, outcomes are more predictable, radical elements are marginalized, and democracy stabilizes (Alexander 2002). Even when legislatures marginalize radical antidemocratic organizations, they do not necessarily narrow the range of actual preferences?only how they are expressed in government (O?Donnell and Schmitter 1986). This may prevent an internal coup, but it may also raise the stakes for those who find themselves disenfranchised by exclusionary electoral laws. 36 It will not hinder those who are prone to non-democratic channels to usurp the dominant regime. A broad range of political preferences is the source of some of the uncertainty in new democratic regimes that is described by Schedler (2001a) and O?Donnell and Schmitter. The rules are arduously contested in new regimes in part because some pretenders are unwilling to suffer what Przeworksi (1986) calls ?institutionalized uncertainty,? meaning that actors in democratic regimes cannot win a permanent spot in the winning coalition. Moreover, models that emphasize economic variables sometimes depend on constraining the range of political preferences. Constituencies in regimes with high inequality are likely to harbor a broader range of political preferences (especially when it comes to economic policy) than those in more equal democracies (all else equal). Just as democracies with growing economies may be more capable of buying out the losers of economic redistribution, they may also be more capable of buying out the losers of political redistribution (Haggard and Kaufman 1995). Democracy has a great potential for economic redistribution (Boix 2003), but it carries a near certain probability of political redistribution?this is the entire point of institutionalized uncertainty. Huntington (1991) and Stepan (1986) rely on a similar concept when they discuss the removal of unfriendly military elements in new democracies. Moore?s (1966) theory about reactionary landed classes retarding democratic transitions is also an argument about narrowing the spectrum of political competition. Polarization and fragmentation in highly proportional legislatures with low electoral thresholds (Haggard and Kaufman 1995, Elgie 2008, Przeworski, Alvarez, et al. 1996) are far more intractable when actual preferences in the selectorate are broadly distributed and/or bimodal. Theories that 37 prioritize the establishment of state institutions in new democracies (Stepan 1986, Enterline and Greig 2008a, M?ller and Skaaning 2011, Linz 1997, von Hippel 2000) highlight the importance of a new democratic regime?s ability to punish internal threats. Even Walt?s (1996) findings about regime change, uncertainty and war are compatible with this concept?foreign powers may have less information about a new regime?s intent when the range of actors who may capture the state is excessively broad. Occupations as Caretaker States Democratic regimes in sovereign states are solely responsible for constraining the spectrum of acceptable political contestation. However, in occupied states, there is another player who can play this role in imposed democracies: the imposer. Depending on the nature of the imposition, and therefore on the relationship between the occupying powers and the domestic government of the occupied state, occupation policy can have a strong effect on the distribution of political interests in the democratizing state. The authors of the new regime are responsible for designing the institutions that will establish the rules of the game, including relations between local and central governments and determining the type of legislative and executive systems. Formative elected governments have the responsibility of fashioning economic and foreign policy in order to promote growth, control inflation, and deter threats. But these are also responsibilities that the occupation can assume, if it so chooses. Moreover, the imposer has two advantages over the new regime: a preponderance of force to impose its will and the capacity to remain in power despite the preferences of domestic actors. It can do more 38 than set policy: it can forcefully rectify inequalities and unilaterally marginalize or eliminate undesirable actors. Now, how the imposer should use its predominance of capabilities in this manner is a matter of dispute? The following chapters will investigate this question, but the point remains: the occupation can fulfill the responsibilities of the new regime, and more. Knowing that an occupying force can take over the reins of state, temporarily, for a new democracy, one might ask what action an occupier can undertake to narrow the range of acceptable political contestation. The regular democratization literature provides some answers: encourage parliamentary regimes and impose a constraining electoral law; maintain capital mobility across borders through free trade agreements and enabling currency convertibility, but not before establishing rule of law and rebuilding infrastructure; ensure moderate growth in the money supply. Coyne (2008) investigated the impact of these processes, as did Enterline and Greig (2008a, 2008b). However, one variable that these scholars do not approach is the influence of interstate threat dynamics. External threat is relevant to democratic consolidation for a few reasons. The first is because of the nature of this type of democratization: when so much force is brought to bear, and when a military defeat is likely a recent development for the target state, one would expect threat levels and mutual suspicion to be unusually high. The second derives from the findings of Walt (1996), who sees a correlation between conflict escalation and regime change. The third derives from the findings of Colaresi and Thompson (2003), that external threats destabilize new democracies by justifying authoritarian security measures. The fourth derives from the findings of Edelstein (2004, 2008), that the presence of an external threat can actually predict the success of foreign occupations. 39 One of the reasons why Colaresi and Thompson presume that foreign threats can destabilize young democracies is that new regimes have not yet developed their own capacity to deter or forcibly respond to enemy provocation. But, is this not a role that an occupier can play as well? Can the occupation deter threats and respond to provocation in ways that preempt calls for authoritarian security measures among the target state?s selectorate? Enterline and Greig (2008a) hint at this possibility in a tangential manner by confirming an inverse relationship between military strength and political instability. Before moving forward with this theory, however, one must resolve a contradiction. Colaresi and Thompson find that foreign threats destabilize democracies, while Edelstein finds that foreign threats help occupiers satisfy their occupation goals. So, what occurs when the threat level is high, and the occupier?s goal is to impose a democratic regime? Which theory is correct in this case? The answer is both, but with a qualification regarding Edelstein?s theory. It is possible to rectify this contradiction by noting that Edelstein was specifically testing occupation outcome against threat perception, and by re-examining his citation of the occupations of South Korea and Japan. Edelstein found that the American occupation succeeded in Japan because the occupied population therein perceived a foreign threat from the Soviet Union, justifying the occupation. On the other hand, the South Korean population did not perceive the same threat, and?considering their desire to achieve independence after years of occupation by the Japanese?were unwilling to accept the violation of their sovereignty. This is puzzling. From an objective standpoint, the Soviet threat to South Korea was far greater than the Soviet threat to Japan. The war that erupted months after the American exit is confirmation of the true threat that the Soviets, and a 40 Soviet-sponsored North Korea, posed. North Korea had allied itself with the Soviet Union and, later, the People?s Republic of China, and the 38th parallel was an artificial boundary with no extraordinary geographical characteristics that could slow down an invasion of the South. On the other hand, Japan is separated by water from both the PRC and the Russian territories, and, even considering their surrender to American forces, had not faced an invasion of the main islands since repelling Kublai Khan in CE 1281 (Emanuel 2005). So why was the United States unable to justify its occupation of South Korea to the domestic political establishment, considering the considerable threat from the North? Two reasons: 1) the United States failed to demonstrate the same level of commitment to South Korean welfare as it did Japanese; 2) South Korea?s experience with occupation indicated that the occupation?s forces?not Stalin?s, Mao?s or Kim Il-Sung?s?were the more threatening presence in the region. Approaching the contrasts between South Korea and Japan in this manner resolves the contradiction between Edelstein and Colaresi and Thompson in two ways: 1) threat is a positive characteristic when the occupier demonstrates its capacity to defend against the threat, and a negative characteristic when it does not; 2) to demonstrate its value to the occupied state, the occupier must provide surplus?or marginal?value, to compensate for the threat, drain on resources and abrogation of autonomy that they themselves pose. Every occupation carries with it its own value. That value is a function of the threat posed by the occupier, the threat posed by foreign competitors, and the goods provided by the occupier. For the occupation to succeed?for domestic actors to see the occupation as less of a threat than the external actors?the occupier must provide enough value to exceed both the threat posed by 41 foreign competitors and the threat that they themselves create. They need to administer an occupation in which the ?marginal value of occupation? is positive. In the case of an occupation governing a new democratic regime, a positive value margin obviates the need for authoritarian security measures by compensating for the new regime?s underwhelming defense capacity. Marginal Value of Occupation and Modes of Political Contestation Providing marginal value through distributing public security goods or more narrowly distributing private economic goods empowers an occupation force to narrow the range of acceptable political contestation. Occupation-generated security produces a public good that can be passed on by the new regime to the selectorate, allowing the new regime to claim credit. Additionally, increasing security stems grievances that may arise in response to a perceived insecurity or threat. This in turn reduces domestic affect for organizations that may trade on such grievances, incentives for those who might choose to support such organizations (rhetorically, economically and/or physically). Moreover, increases in security delegitimize violence as a means of remedying such grievances, and reduces the possibility that replacement of R at io o f E le ct or at e Continuum of Political Preferences Outer Bounds: Radicals Center Column: Moderates Figure 2.1: Expected Effect of Marginal Occupation Value on Radical and Moderate Political Preferences Initial Distro Narrowed Distro 42 the new regime would yield internal or external security gains. Beyond the realm of security goods, the occupying forces find themselves in the unique role of authoritative redistributor; they have the capacity to redistribute and produce non-security goods? such as prosperity and personal security?thereby eliminating grievances and allowing the new regime to claim credit for improvements in the standard of living. What?s more, the occupying forces, with some finesse, may be able to redistribute said goods in such a way that specifically empowers pro-democracy factions and deprives radical groups. By pursuing policies that improve the standard of living specifically for pro- democratic factions, the occupiers narrow the spectrum of political contestation. Imagine that the distribution of preferences across the selectorate is two-tailed and bell-shaped: pro-democratic centrist, leftist and right-wing elements comprise the bell, while radical left- and right-wing elements constitute the long tails. In this metaphor, the first process grows the bell while the second process clips the tails. By narrowing the distribution of preferences, security provision narrows the range of possible electoral outcomes, reducing the probability of polarization, fragmentation, and the costs of political redistribution. It also reduces the possibility of political violence by marginalizing radical elements who may wish to contest the regime in this manner. A constraining factor here (as in the rest of the democratization canon) is uncertainty. Similar to the information premium during democratic transitions, information about actual security at any given moment is scarce (except in cases where the probability of surviving until the next sunrise equals zero, as in the case of total military defeat). This is why Edelstein bases his theory on perceptions of threat. Sub-state actors infer security based on prevailing conditions that indicate an increase or decrease 43 in the likelihood that their life, liberty or property will be threatened in the near future, whereas state actors calculate security based on the inverse probability of total military defeat. Accordingly, occupier behavior that credibly signals future security?such as a mutual security pact backed by a major force commitment?will have a positive influence on security perception. Conversely, actions that signal future insecurity?such as sponsoring external threats or directly engaging the occupied state in renewed hostilities?will have a negative effect. These signals cover a wide range of behavior, from public statements, to formal alliances, to direct military action. The more costly the signal, the more credibility it will hold in the minds of domestic actors (Fearon 1997). The same dynamics apply to the production and distribution of non-security goods? crime and domestic instability indicate future instability, prosperity signals future prosperity. In order to improve the marginal value of the occupation, and thus narrow the acceptable range of political contestation, occupiers must credibly signal their willingness and capacity to produce occupation goods. The remainder of the dissertation is dedicated to demonstrating how occupation commitments narrow the range of contestation, limit political instability, and improve the survival of new democratic regimes in occupied territories. 44 Specification of Terms Democracy This project will employ a minimalist conception of democracy as the important phenomenon under discussion. An executive and legislature (with lawmaking authority) constrained by popular elections will suffice. There are several measurements of democracy, one of the most common being the Polity IV rating system (Marshall and Jaggers 2009). For a more rigorous test, this project will employ a +5 Polity threshold to separate democracies from non-democracies. Regime Survival Regime survival is easy enough to pin down based on the criteria established above. If at any point during the case the democracy drops below the +5 Polity IV threshold the case will count as a failed consolidation. As Coyne (2008) notes, there is no consensus on how far out one should study a regime to determine consolidation; he looks at intervals of five, ten, fifteen and twenty years. Bernhard, Reenock and Nordstrom (2003) indicate that regimes consolidate their capacity to respond to economic shocks after the first three legislative elections. Svolik (2008) finds that the probability of collapse declines as early as the seventh year and that survival during the first two decades is ?crucial.? Enterline and Greig (2008a, 2008b) show that the historical rate for regime reversal passes even at about the twenty-year mark. There seems to be a consensus that forms around two decades as an outer bound for consolidation studies, so this project will examine regime democracy levels up to twenty years beyond the 45 founding of the polity?or when the ceases to be democratic, whichever comes first. Huntington?s (1991) emphasis on generational turnover is compatible with an outer bound of twenty years, which is a reasonable operationalization of the time span equal to a single generation. Consolidation and Instability Instability is a bit tougher to observe than survival. Instability is essentially evidence of behavior (or preparation for behavior) by domestic actors to replace the winning coalition in a manner not sanctioned by law or constitutional provision, or an effort to achieve electoral success with the goal of dissolving democratic regimes. At the extreme this would include civil wars, coups and attempted coups. Less extreme examples of stability include political violence directed at state actors, mass protests explicitly directed at the replacement of a democratic regime with another type, paramilitary violence and the raising of paramilitary organizations. In addition, this study will include electoral success of antidemocratic parties as an example of instability due to the possibility of so-called ?legal coups,? wherein radical forces come to power legally and use constitutional means to terminate democratic institutions. Consolidation is the most difficult of these terms to specify due to the inability to directly observe this process. There is a lack of consensus on what consolidation means, and Schedler (2001b) attempts to resolve this dilemma by recommending that we focus on the behavior of actors in the regime. Schedler claims that behavior of government actors is the most causally adjacent and observable indicator of consolidation. This is why instability is a phenomenon of interest?the actions described above are the most causally adjacent 46 conditions to regime collapse and thus the best indicators of the loss condition. Ultimately, consolidation is the opposite of instability and the inverse of the probability of survival. Occupation and Regime Imposition Instead of exploring all examples of foreign intervention and their effects on democracy, this project will narrow the range of cases under examination to occupations initiated by military intervention. This is a small-n project, so it is important to constrain the range of variables that might affect observed outcomes, making hypotheses easier to test. Occupations, broadly defined, occur when a foreign state abrogates the sovereignty of a target state by deploying armed forces on the target?s territory as a means of suspending that state?s right to conduct an independent foreign and domestic policy. This is a condition that would satisfy most, if not all, of the cases in Edelstein?s studies. Likewise, the population of cases in this dissertation derives from a narrower and more faithful definition of regime imposition than Enterline and Greig. Only new democratic regimes that emerge in occupied states or in anticipation of a future occupation count as imposed democratic regimes for the purposes of the dissertation. Optimal, Semi-optimal and Suboptimal Winning Coalitions If an example of instability is success among antidemocratic parties, then it is important to specify which parties meet the conditions of democratic and antidemocratic. Democratic parties are those that express a preference for open elections to powerful 47 representative legislative bodies via party platforms or public pronouncements. Together, these parties constitute the optimal winning coalition. This does not have to be an intentional coalition?it does not matter if these parties actually form a government together or not. Rather, a government can remain in the control of democratic parties while other democratic parties do not join the government or hold cabinet positions. These members of the ODC are the sub-state organizations, typically but not always political parties, which, if they did form a government, would represent the friendliest possible winning coalition (wherein all pro-democracy elements are in and all non- democracy elements are out). Furthermore, there are two classes of antidemocratic organizations. Organizations that remain publicly ambiguous about their support for democratic institutions constitute the semi-optimal coalition. Sub-state organizations that are publicly committed to the termination of democratic institutions constitute the suboptimal coalition. Marginal Value of Occupation Occupying forces are more successful in narrowing the spectrum of acceptable political contestation when the occupation they impose produces marginal value. This is difficult, as any occupation begins with a security deficit. Most of the time, the occupiers were recently enemies dedicated to the occupied state?s defeat. Moreover, the reality of the occupation produces three bits of information about the relationship between occupier and target: 1) the occupying states? aggregate capabilities are significantly greater than those of the target state; 2) the occupying states are necessarily the most geographically proximate states; and 3) there is a forward deployment of the occupiers? troops with the 48 capacity to inflict damage on the target state. These are all components of threat as determined by Walt (1987). The likely history of antagonism, the disparity in aggregate capabilities, geographical proximity and forward deployment almost guarantee that, at outset, the occupying forces are a greater threat to occupied state than any other state in the system. In order to overcome this deficit, the occupiers must provide a surplus of goods?security and otherwise?enough that it exceeds the costs of the inherent threat of the occupation itself. This excess value is the marginal value of occupation. There are three other terms related to the marginal security value of occupation: occupation relationship, threat environment and security configuration. The occupation relationship is the internal welfare effect of the occupation policy. The occupation relationship is positive when the occupying forces improve the welfare, standard of living and/or personal security of domestic political actors?what one might refer to as human security?and negative when it impairs domestic welfare. The threat environment is the degree to which the occupied territory is threatened by foreign states other than those that constitute the occupation force. When powerful, geographically proximate competitors antagonize the occupied state, the threat environment grows more hostile. The security configuration is the relationship between the occupiers and other enemies and allies of the occupied state. The security configuration is favorable when the occupiers confront enemies of the occupied state, neutral when it does not, and unfavorable when the occupiers are supportive of the occupied state?s enemies (historical or contemporary). A positive occupation relationship, hostile threat environment and favorable security configuration all contribute to the surplus of security provided by the occupation. These processes are more explicit in the section below. 49 The Marginal Value Model Threat Processes The following model outlines the mechanism by which the behavior of occupiers and of potential security competitors affects the perception of threat by domestic actors in the target state, and by extension how these dynamics affect the range of political preferences and contestation. This is determined by three relationships: 1) the occupation relationship, or the means by which the occupiers signal the provision of occupation goods to domestic political actors; 2) the threat environment, which is an objective measurement of the degree of hostility in the system; and 3) the security configuration, which is the relationship between the occupying states and states hostile to the occupied state. The occupation relationship, at its most simplistic, boils down to a single question: Does the occupation?s policy seem to be improving or reducing the welfare of domestic actors in the target state? The occupying state can improve the internal situation in the target state in a number of ways, such as targeting and eliminating violent internal threats and helping the state rebuild its police and external security forces. This relationship, however, may extend beyond traditional security. The direct provision of economic and personal security goods also improve this relationship. The reconstruction of infrastructure and industry, direct injections of capital, and policies that improve public health also fall under this rubric. Conversely, introducing violence into the territory (by, say, sponsoring insurgencies), preventing the new regime from establishing sufficient internal security, draining fixed and/or portfolio capital from the state, and any other 50 policies that reduce the personal safety of domestic actors constitute a negative occupation relationship. The capacity for the occupying state to provide external security goods in the first place is dependent on the status of foreign threats to the occupied state. Without a foreign threat to defend against, there is no way for the occupiers to provide surplus security, or security above the level that they withdraw from the occupied territory by their presence and any previous history of antagonism; the greater the threat by non-occupying states, the greater the capacity for occupiers to provide security goods. As always, the distribution of threat is in part dependent on capabilities and geographical security, but it is also dependent on offensive posture and demonstrations of intent towards the occupied state (Walt 1987). When geographically proximate states demonstrate hostile intent towards the occupied state, either through public pronouncements, conspicuous redeployment of forces, or actual military aggression, the threat environment grows more hostile. Conversely, if geographically proximate states demonstrate indifference towards the occupied state, or if they demonstrate willingness for peace and cooperation, then the hostility level decreases and the threat environment becomes less critical. Security configuration boils down to the capacity of the occupying states to establish themselves as the best possible allies to the occupied regime. The means by which the occupation signals the future provision of external security goods is by responding to those who pose a threat to the occupied state. By treating the enemies of the occupied state in an adversarial manner, the occupation sends a credible signal that it is committed to the occupied state?s national security. The costlier the signal the better: public pronouncements are marginally credible, formal alliances more credible, and 51 direct military engagement the most convincing (although expensive for obvious reasons). The least favorable security configurations are the resultants of A) a direct military confrontation between the occupation and the occupied state?s allies, B) an active military contribution by the occupation to the occupied state?s enemies, or C) both. A favorable security configuration increases the occupation?s potential marginal security. Information, uncertainty and time play important roles in the processes described above. At the outset, there is very little information about the political game in the new regime: about the payoff structure for political contestation, about the rewards for cooperating with new institutional rules and punishments for violating them, about the intent of the occupiers, and about how secure and prosperous the future regime will be. This uncertainty will not persist for long, however, as domestic and international actors begin to assert their interests. On the other hand, there is a wealth of information about the power relationship between the occupation and the occupied state, the overwhelming majority of which is negative. Both of these realities emphasize the importance of time, specifically the importance of producing information that is favorable to the consolidation of democracy as soon as possible. The occupiers signal?or produce information?about their intent through the occupation relationship and security relationship, but this information will be more valuable when this information is at a premium. Information is most valuable when uncertainty is highest, and uncertainty declines as actors assert their interests over time. Therefore, behavior early will have a greater impact on marginal security value than behavior late. Moreover, individuals do not respond to losses and gains equally in uncertain environments. Instead, cognitive psychology?of the utmost relevance here because the 52 updating of beliefs with new information is a learning process?indicates that individuals react more emphatically to losses than to gains (Kahneman and Tversky 1979, Levy 1996, 1997, 2002, Kahneman 2011). As a result, negative behavior of a certain degree harms the occupation?s security value more than positive behavior of an equal degree helps it. Additionally, negative behavior later in the occupation may cancel the effect of positive behavior early in the occupation, while positive behavior later in the occupation will not likely compensate for negative behavior early in the regime. It is impossible to know how much more of an impact negative signals convey versus positive signal, and it is impossible to know how much more of an impact early behavior will have versus later behavior. Nor is it possible to quantify the degree of any actions that may produce or reduce marginal security. What this tells us is that positive, early behavior is most likely to yield marginal security, that occupations should be more successful in narrowing the spectrum of political contestation when they begin producing surplus security in the first years of the occupation and in the first years of the new regime, and that democracies should be more likely to consolidate and survive when occupations start out this way. Variables Aside from the variables related to the existing literature on democratic consolidation?GDP growth, incomes, inflation, electoral laws and parliamentary- presidential balance?the independent variable here is the marginal security value of occupation. Marginal value cannot be directly observed, but it can be inferred through the observation of its constituent parts. Events in the historical record point to positive, negative, or stable trends in the occupation relationship, threat environment and security 53 configuration. The occupation produces marginal value when it contributes to the domestic wellbeing of the occupied state and convincingly pledges to defend it against foreign threats. The occupation fails to produce marginal value when it does not contribute to domestic wellbeing or when it aligns with the occupation state?s enemies. Everything else in the middle is a question of degree or just uncertain, specifically when the occupation relationship is positive, the threat environment lacks hostility, and the security configuration is unfavorable. Nevertheless, there are indicators that clarify the situation: alliances. Substantive security agreements between occupier and target, when entered into voluntarily, indicate that the occupying forces have credibly signaled their interest in preserving the security of the occupied state. This is an even more favorable indicator when the agreement involves material commitments. Parchment is cheap, but material transactions such as force contributions, basing agreements and coordinated military action are more expensive, and thus more convincing, indicators that the occupation produced marginal security goods. Considering this, informal security agreements that involve material compromises (such as training and arms production agreements) are better signals than formal alliances that do not (such as mutual security pacts that carry no additional requirements except for the joint declaration of war). Table 2.1: Effect of Component Factors on Marginal Occupation Value Occupation Relationship Threat Environment Security Configuration Internal Welfare External Security Marginal Occupation Value Positive Hostile Favorable Surplus Surplus Very Positive Positive Hostile Unfavorable Surplus Large Deficit Negative Positive Amiable Favorable Surplus Even Positive Positive Amiable Unfavorable Surplus Deficit Uncertain Negative Hostile Favorable Large Deficit Surplus Negative Negative Hostile Unfavorable Large Deficit Large Deficit Extremely Negative Negative Amiable Favorable Large Deficit Even Very Negative Negative Amiable Unfavorable Large Deficit Deficit Very Very Negative 54 Dependent variables include intra-case political instability and case-level survival or collapse. Political violence, paramilitary strength, economic support for antidemocratic organizations and electoral success by parties outside of the optimal winning coalition are all indicators of instability. Survival, as noted above, occurs when the regime in question closes out the twenty-year period of observation above +5 on the Polity IV scale. Hypothesis The model described above yields observable implications and testable predictions. The marginal value of occupation model yields predictions that operate both within each case and across cases. Generally, events indicating regime instability will be less common among cases in which the occupiers increase marginal occupation value through the provision of security and other goods than they will be among cases in which occupiers reduce the value of the occupation (H1). Additionally, events indicating regime instability with be less common within cases when the occupiers increase marginal occupation value than they will be when the occupiers reduce the value of the occupation (H2). Finally, democratic regimes will survive only in cases where the occupiers manage to provide economic aid and commit materially to defending the new regime from a foreign threat before the occupiers aid an enemy of the occupied state and/or before the occupied state forms an alliance with another, non-occupying state (H3). 55 Methodology This study will combine intra-case analysis with comparative case history in order to reveal the mechanisms that link explanatory variables to regime outcomes. The focus of the substantive chapters is the elucidation of processes that the hypotheses anticipate, as well as the role of those expected by control hypotheses. As McAdam, Tarrow and Tilly (2008) demonstrate, these processes can be uncovered through the methodological approach of historical process tracing.4 Process tracing is the method of testing hypotheses based on within-case and across-case variation. This approach requires the researcher to move above and below the level of analysis within which the key process operates (Bennett 2008). The limited sample of cases of imposed democracy demands the employment of counterfactual analysis of necessary-condition arguments (Ragin 1989, Hall 2003, Goertz and Levy 2007, Box-Steffensemeir, Brady and Collier 2008). This, perhaps in contrast with what the title of the method implies, does not mean constructing meticulous histories of events that did not occur. In fact, the counterfactual approach does as large-n statistical analysis does, attempting to ascertain how conditions would change if the data observed provided different values. The counterfactual approach conceptualizes historical outcomes as a chain of causal reactions. Each link in the chain is a point in time where, under different conditions, history may have taken a different course. Historians refer to these links as ?turning points,? while social scientists often refer to them as ?critical junctures? or ?windows of opportunity? (Goertz and Levy 2007). These periods establish points in the historical record where the researcher may test his theory against real 4 See also: Pierson (2000), George and Bennett (2001), Falleti (2006), Lichbach (2008), Falleti and Lynch (2009), Mahoney, Kimball and Koivu (2009). 56 conditions. They also allow the researcher to break down larger cases into several sub- cases (King, Keohane and Verba 1994). The working hypothesis and rival hypotheses, grounded in theory, provide explanations for what course history should or should not take following these junctures. Therefore, we can reject a hypothesis when the prevailing conditions during a critical juncture predict a counterfactual, or a history that did not occur. Identifying historical process chains is important for reasons other than counterfactual reasoning. The existing democratic consolidation literature varies in the degree by which the authors choose to undergird their predictions with theoretical claims, but a significant amount of the literature surveyed herein contains descriptions of the theoretical micro- or macro-processes that support the hypotheses that they test. For instance, Boix?s correlative prediction about the relationship between equality and democratic consolidation is based on a theoretical narrative in which holders of valuable goods employ antidemocratic tactics to impede confiscation. This means that if Boix is correct, we should not only observe that egalitarian democracies survive and that unequal democracies do not, but we should also find evidence of political actors who behave in ways that confirm Boix?s theory. We should see that political actors who are flush with immobile assets 1) express opposition to redistributive and confiscatory policies, 2) pool resources with similar actors in an effort to build coalitions to defeat the regime in response to the election of redistributive parties and the initiation of redistributive policies, 3) employ those resources to either take over the legislature or initiate a coup, 4) see these groups leading the effort to successfully bring down democratic regimes, 5) place themselves or someone who represents them in power, 6) eliminate democratic 57 institutions that may remove them from power, and 7) instituting policies that reverse previous episodes of redistribution and prevent future redistribution. Furthermore, the process should take place more or less in the same order as presented above, and this process should repeat itself in a preponderant ratio of cases of democratic reversal. Similarly, if Haggard and Kaufmann (1995) are correct about economic growth enabling political winners to buy out antidemocratic opponents, we should observe a causal chain in which the winning coalition manually transfers private goods to anti-democratic constituencies, followed by observations of those constituencies renouncing their preferences for regime displacement and/or a migration of supporters from antidemocratic organizations to democratic ones, in this order, in a preponderance of cases. If one does not observe these sequences, then one must consider the possibility that the hypothesis is A) wrong or B) right for the wrong reasons. The research in the following chapters will present findings in this way. The marginal security approach has its own theoretical narrative. Policy that seeks to improve the occupation relationship should result in an increase in positive attitudes towards the occupation forces, followed by these same constituencies renouncing the use of political violence and casting votes in favor of parties that constitute the optimal winning coalition. An increasingly hostile threat environment should encourage the constituencies most threatened to call for drastic measures to remedy their perceived insecurity, and it should encourage antidemocrats that hold legislative seats and cabinet positions to propose and support policies that curtail democratic freedoms as implied by Colaresi and Thompson?s theoretical narrative. When such threats arise, and if the occupation forces credibly commit themselves to deterring these threats, the same constituencies that called 58 for authoritarian solutions should withdraw their proposals, perhaps even migrating to the center. In cases of democratic reversal, we should see that the victorious parties had once rhetorically employed the perception of insecurity in programs and speeches. We should also observe that they blamed the occupation forces specifically for the insecurity they purported to experience. Finally, we should observe that antidemocratic political entrepreneurs commonly demonized pro-democracy elements in tandem with grievances directed toward the occupation, sometimes branding them as conspiring to further reduce the security of the state and its people for their own self-interest. We should see that politicians, parties, and other political organizations that employed such rhetoric won substantial electoral and, perhaps, economic and paramilitary support. The point is that behind every well-developed theory there is a narrative. Every narrative is a sequence of events in which one is largely dependent on the prior, and largely determines the subsequent, event. The order of these processes should be consistent across and throughout cases if the narrative is truly characteristic of the actual political process. In each case we should see a variety of processes that fit or do not fit the theoretical mold. Each juncture in which the narrative conforms or fails to conform is a test of the causal logic inherent in the theory and model. This presents the opportunity for rigorous qualitative research to derive a relatively large number of tests from a very small number of case studies. 59 Case Selection and Justification The total population of cases in which democratic regimes arose or were attempted during occupations includes the following country-years: Cuba 1898-1902, the Philippines 1898-1946, Haiti 1915-36, Germany 1918-30, Italy 1943-8, Austria 1945-55, Germany 1945-55, Japan 1945-52, South Korea 1945-9, Palestine 1967-present, Grenada 1983, Panama 1989, Timor-Leste 1999-2002, Afghanistan 2001-present and Iraq 2003- 11. This sample excludes cases that presume an attempt to impose democracy based on nothing more than the stated intent of policymakers; obvious application of state or coalition resources is a necessary precondition. Cases in which the intervening party may have considered democratization, but in which neither the occupying nor the occupied target state instituted a policy of democratization by writing a constitution, imposing elections, or at minimum sanctioning a representative form of government in the target polity, do not pass muster. Cases like Guatemala in the late 1950s or Nicaragua in the 1980s are not included because of this restriction since there was no overt commitment of economic, legal or military resources toward democratization. This particular project will focus on four cases: Germany 1945-55, Japan 1945-52 and South Korea 1945-9. These four cases maximize the amount of information available for study will minimizing differences in historical era, geography and culture. This case selection includes two successes (Germany [Federal Republic] and Japan) and two failures, and allows for the pairwise comparison of geographically and culturally similar cases with disparate outcomes (the two Germanys; Japan versus South Korea). 60 Table 2.2: Summary of Cases Territory Occupation Start* Occupation End* Regime Start** Regime End** Observed Here Cuba 1898 1902 n.a. n.a. Philippines 1898 1934 n.a. n.a. Haiti 1915 1934 n.a. n.a. Germany 1918 1929 1919 1934 Yes Italy 1943 1945 1948 Ongoing Japan 1945 1952 1952 Ongoing Yes Austria 1945 1955 1946 Ongoing Germany (West) 1945 1955 1949 Ongoing Yes Korea (South) 1945 1950 n.a. n.a. Yes Palestine 1967 Ongoing n.a. n.a. Grenada 1983 1983 1983 Ongoing Panama 1989 1989 1989 Ongoing Timor-Leste 1999 2002 2002 Ongoing Afghanistan 2001 Ongoing n.a. n.a. Iraq 2003 2010 n.a. n.a. *Occupation start and end dates drawn from Edelstein (2008) for cases through 1983. In cases of Panama, Afghanistan and Iraq the occupation begins when the state leaves the Polity IV database and ends when it re-enters the set (Marshall and Jaggers 2009). In Timor-Leste?s case, the occupation begins when UN coalition forces entered the territory and ends when it enters the Polity IV database. **The start of the regime is determined by the polity reaching +5 on the Polity IV scale, and ends when it drops down below that threshold. States that are still above +5 are considered ongoing, whereas cases wherein states never reached +5 are labeled ?n.a.? The exceptions are the cases of Palestine and Grenada. Since the Palestinian territory has not yet acceded to the community of states, it is also labeled ?n.a.? Grenada is not a large enough territory to qualify for the Polity IV set, so its regime start date occurs when it qualifies as ?free? in the Freedom House database (Freedom House 2011). Selecting on the Dependent Variable Social scientists are often urged to avoid selecting on the dependent variable, lest the research introduce bias into (or reduce the generalizability of) one?s findings (King, Keohane and Verba 1994). This caveat applies to research that seeks to test the average effect of interval variables on a population of cases. However, this is not the point of the project described here. Because this project uses historical analysis to test a model that relies on the logic of necessary conditions, the most appropriate methodological approach is qualitative, and the most appropriate method of choosing cases is dependent variable selection. This approach is supported by Goertz and Starr (2003), Mahoney and 61 Rueschemeyer (2003), Mahoney and Goertz (2006), Goertz (2008), and Seawright and Gerring (2008). The conditional logic approach defines positive outcomes as the coincidence of necessary or sufficient conditions. Therefore, representative cases are in fact determined by their values on the dependent?not independent?variable; qualitative research design based on dependent variable selection is not only acceptable but also desirable. Moreover, since historical explanations specifically explain why a particular case displays a particular value on the dependent variable, historical work demands dependent-variable selection. Organization of the Thesis The next four chapters will employ the research design presented above to investigate the determinants of survival, instability?and, by extension?consolidation in democratic regimes that arise under foreign occupations. Chapter 3 investigates the German Federal Republic, or West Germany, from 1945 through 1975. Chapter 4 investigates Japan from 1945 through 1972. Chapter 5 investigates the Deutsches Reich, also known as the Weimar Republic, from 1918 through the end of the democratic polity in 1933. Chapter 6 investigates the Republic Korea, or South Korea, from 1945 through 1950. Chapter 7 tests the inter-case hypotheses on the four cases and concludes by discussing the contributions of this research project and by stating plans for a future research agenda along these lines. 62 Chapter 3: West Germany, 1945-1975 Introduction The successful democratization of the Federal Republic of Germany (henceforth referred to as the FRG or West Germany) is among the most remarkable political turnarounds of the twentieth century. Emerging from the wreckage of WWII and born out of the Cold War, the FRG grew into a democracy as stable as any other. Pro-democracy parties that cooperated with the occupation thrived its confines and in sovereign West Germany.5 Anti-democratic parties faired poorly, never combining to equal ten percent of the popular vote in any federal election or winning a significant number of seats in the federal legislature6. There?s no shortage of theories explaining how the fatherland of National Socialism and Prussian militarism has grown into a stable federal democracy 5 The optimal domestic coalition includes the Christian Democratic Union, the most cooperative of the parties during this period, and their Bavarian affiliate?the Christian Social Union. The Free Democratic Party performed well as a bourgeois democratic party on the left of the CDU/CSU. The Zentrum and Bavarian Parties survived the Third Reich and emerged as pro-democracy parties whose constituencies were absorbed but the CDU and CSU. The Social Democrats (SPD) were pro-democracy as they were in the Weimar regime, although their support for the occupation was qualified. (Merkl 2002, Glees 2003) 6 Several authoritarian parties competed for seats in the federal legislature during the period under examination. On the right, the German Right Party (DRP) picked up the scraps of the DNVP while the Socialist Reich Party (SRP) claimed the mantle of the Nazi Party. The National Democratic Party was an extremist splinter group of the DRP. Both the DRP and NPD associated with the German Party (DP) which was not explicitly authoritarian but was not explicitly democratic, either. Other parties in this category included the Economic Reconstruction Union (WAV). On the left, the Communist Party of Germany (KDP) represented the old KPD, as did the later West German Communist Party (DKP). 63 (not to mention a bulwark of liberal capitalism, international cooperation and progressive ideology). The full allied occupation of Germany, comprised of the United States, United Kingdom, France and the Soviet Union, commenced on June 5, 1945 with the arrest and dissolution of the Nazi government under Admiral Doenitz (Plischke 1953, Glees 1999, Haftendorn 2006, MacDonald 2007). The four powers coordinated occupation policy for all of Germany through the Allied Control Council, which first met on July 30 of the same year and functioned until a Soviet walkout on March 20, 1948 (an incident that precipitated the Berlin Blockade; Willis 1962, Dunbabin 2008). Thereafter, the US, UK and France coordinated policy for the Trizone?the territory comprised of the three western occupation zones and that eventually became the Federal Republic. The three western powers presided over a civilian occupation of West Germany as the Allied High Commission from September 1949 until sovereignty was handed over to a West German government on May 5, 1955 (Plischke 1953, Hitchcock 1998, Haftendorn 2006, MacDonald 2007). Thus, the observation period in question stretches from the commencement of the occupation through twenty years after the return to sovereignty: 1949-1975. The following narrative will demonstrate that the US and UK generally acted as a cohesive unit, while the French delegates as a rule nominally resisted but eventually acceded to the policy goals of the Anglo-American bloc. Interest in democracy promotion was strongest among the British and American delegates, while French attitudes ranged from unopposed to enthusiastic. Here I shall summarize a number of the more prominent theories, test them against the history of the case, and assess their validity in light of the evidence. I shall 64 then test the working hypotheses of this project against the same history. The working hypotheses generally predict that domestic stability (the absence of riots, assassinations and insurgency) and electoral victories among the optimal winning coalition (described above) will be more prevalent when there is an obvious external threat, when the occupiers successfully demonstrate the capacity to protect the FRG and its people from said threat, and when the western occupiers successful distribute public goods such as personal security, freedom and prosperity in their zone. If we see this sort of success without the western occupiers contributing marginal security or promoting domestic welfare, then we should consider the working hypothesis rejected. Moreover, if rival explanations?such as economic growth?appear to contribute to democratization independent of the contributions of marginal security and welfare variables, then we cannot validate the working hypothesis. Therefore, I shall demonstrate that, in this case, domestic stability and ideal party success do indeed result from the constructive behavior of the occupying powers. Having done so, I will demonstrate that many of the competing explanations fail to pass muster when tested against the facts. Ultimately, I shall show that those hypotheses which history does not falsify are not incompatible, and in many cases complementary, to the theory that undergirds this dissertation: that democratic legitimacy in a young, occupied regime is dependent on the capacity of the occupying force to protect the new regime against foreign threats to security and domestic threats to democratic political competition. 65 Rival Hypotheses versus Case History Rival Hypothesis: Demilitarization One explanation for the durability of democratic institutions in West Germany that lends due credence to the role of the Allied occupation is Juan Linz?s (1990) ?debellicization? thesis, adapted from Stepan?s (1991) ?paths toward democratization.? In Linz?s terms, ??the total debellicization of the German state gave to the Allied Powers full control of political development.? (p.148). The Western Allies (along with the Soviet Union) utterly reduced the German war-making apparatus. Having done so, the occupiers faced no competition from militarized, reactionary political forces; the Western Allies were free to rebuild West Germany as they saw fit, just as the Soviet Union was free to reconstruct the East. This picture is in stark contrast to the state of Germany following WWI, when the war ceased before enemy boots set foot on German soil. Nearly all of the destruction resulting from the European theatre of the war took place in France, while Germany proper emerged relatively unscathed. As a result, institutions that maintained a monopoly on the use of force managed to survive the termination of the conflict and the imposition of the occupation. The Weimar occupation itself was also less extensive, exercising territorial control over the Rhineland but nowhere else. Demilitarization was part of the Allied plan for reconstructing Germany before the United States had formally entered WWII. At Newfoundland in August 1941, President Roosevelt and Prime Minister Churchill agreed that, in the event of the Third Reich?s defeat, Germany?s military apparatus would be dismantled (MacDonald 2007). This sentiment was reiterated at Teheran in 1943 (Willis 1962), at Quebec in 1944 (Ibid, 66 Killick 1997), at Yalta in February 1945 (MacDonald 2007) and at Potsdam in July 1945 (Dunbabin 2008, Glees 2003). It was implemented by the four-power Military Control Council that assumed power in June 1945 (Willis 1962, MacDonald 2007) and reinforced by the several Foreign Ministers Councils that shaped Allied policy thereafter (Hitchcock 1998). To reinforce this policy, the Allies transferred ownership of strategic resources, such as coal and steel, first to the French and then to an international body: the European Coal and Steel Community (Ibid, Krieger 1983, Glees 2003, Willis 1962, MacDonald 2007, Plischke 1953, Haftendorn 2006, Gillingham 1995, Adenauer 1984b). At the beginning of 1948, as the threat from the Soviet Union grew more apparent, UK Foreign Minister Bevin proposed the reconstruction of a German military that would serve under the auspices of the (ultimately failed) European Defence Community (Dunbabin 2008, Mee, Jr. 1984, Plischke 1953, Windsor 1969, Hitchcock 1998, Hillenbrand 1980, Banchoff 2002, Krieger 1983). Later that year, the US, Canada and European powers came together to sign the North Atlantic Treaty, creating a multilateral mutual defense organization that would supervise several German divisions (Willis 1962, Glees 2003, MacDonald 2007, Schmidt 1994). The democratic regime that emerged from the occupation incorporated these principles. Among the earliest of Chancellor Konrad Adenauer?s public pronouncements was a call for an end to the Franco-German rivalry (Banchoff 2002). The Basic Law itself incorporates the power to transfer sovereign authority over the armed forces to multilateral institutions (Parliamentary Council 1984). By 1954, the debate over demilitarization was a distant memory, and the US and NATO were discussing arming German forces with nuclear weapons (Schmidt 1994). By 1956, 67 NATO was complaining about West Germany?s failure to deliver the divisions it pledged to the alliance (Glees 2003). The Allies failed to ?debellicize? Weimar Germany in the same sense, though not for lack of trying. Demilitarization and disarmament were key pillars holding up the occupation policy of the US, UK and France following the first war. They were simply less thorough, less successful, and starting from a less advantageous position (with a smaller occupation zone following a war that caused less damage). But one should not pretend that the Great War left the German war apparatus unscathed. A large swath of the working-age male population lost their lives. The Allies did maintain a strong enough cordon on the production of materiel that the Weimar government had to pursue such aims in secret on foreign soil. The question here is not one of absolutes, but one of degree. How thoroughly must an occupier reduce its target?s capacity to make war (and generally monopolize political violence)? At which point must this process take place, and when should it be concluded? Most pressing for the purposes of this project, how?if at all?should the occupiers proceed to fill this vacuum? The debellicization thesis is compatible with the Marginal Value Model to a certain degree. It is important that the regime of the occupied state come to rely on its occupiers for security. If remnants of the war-making apparatus from the previous regime survive the new one, they could prevent the occupiers from increasing their occupation?s marginal security value. Even if the occupiers have succeeded in debellicizing the state, they must then rebuild an effective security apparatus that protects the citizenry but compromises neither the superiority of the occupation nor the durability of the new 68 democratic institutions. In other words, debellicization is only as important as rebellicization. The debellicization thesis, however, ignores this part of the equation. Rival Hypothesis: Political Re-education Two explanations rely on the concept of political education and generational turnover to explain the affinity of the postwar German population for democratic institutions. The first, put forth by Samuel Huntington (1991) attributes the gradual increase in preference for a democratic regime to institutional learning among those who were born and came of age under the post-Nazi state. The second, David Conradt (1980), attributes said value change to an intense campaign of political re-education led by the Allies and the West German government. This re-education took place through punishment of Nazi war criminals, banning of antidemocratic organizations, and nationalization?as well as moderate secularization?of the school system. As those in the new regime reached voting age, their institutional preferences reflected those that were inculcated with during youth. Surveys revealing an inverse relationship between age and indicators of civic engagement suggest that generational turnover did play a role. Those who experienced the Wirtschaftwunder (German economic miracle) during their formative responded more favorably than others. Conradt goes on to argue that political re-education was aided by the demise of the Wehrmacht and the Junker nobility during WWII.7 The paring of the protestant-dominated East, combined with the resettlement of millions of refugees, lessened the severity of ethnic, religious and geographic cleavages. 7 Surveys conducted under the military government indicate that only 15% of the Junker class survived World War II (MacDonald 2007). 69 Finally, Conradt offers that there was no ?credible alternative? in the earliest, roughest days of the Federal Republic to whom the disaffected could throw their support. That the Western Allies embarked on an abortive denazification campaign, including the prosecution of Nazis who committed crimes against humanity, is public record (Kvistad 2002, Willis 1962, Glees 2003, MacDonald 2007, Markovits and Reich 2002). Indeed, this policy correlates well with a general upward trend of civic engagement indicators and survey responses indicating affect for democratic institutions. The percentage of Germans who declared themselves to be interested in politics rose from 27% to 46% between 1952 and 1972 (Merkl 1980, Conradt 1980). Between 1951 and 1972, the ratio of respondents who felt that Germany was ?best off? under the Federal Republic rose from a dismal 2% to a staggering 81% (Conradt 1980). The percentage of Germans who said they discuss politics in their private lives increased from 50% in 1956 to 70% in 1972 (Ibid). The proportion of Germans who felt that parties represent the best interest of the people rose from 35% to 70% between 1959 and 1980 (Richter 2002). General approval of the Bundestag increased from 35% to 52% between 1951 and 1965, hitting 57% in 1955 (Loewenberg 1967). 70 See footnote for sources and explanation of survey questions and responses.8 Also important is the role of generational turnover. Huntington and Conradt were theorized about the impact of re-education and value acquisition over the course of a 8 (A) Voter turnout figures from Schweitzer, et al. (1984) (B) Percentage of respondents who answered affirmatively to the question ?Are you interested in politics?? Other responses included ?not especially? and ?no.? (Conradt 1980, Merkl 1980) (C) Percentage of respondents who responded affirmatively to the question ?Is democracy the best form of government? (Conradt 1980) (D) Percentage of respondents who chose ?Federal Republic? as the period when Germany was ?best off?? Other responses included the Third Reich, the Kaiserreich, the Weimar Republic and other. (Ibid) (E) Percentage of respondents who claimed to talk politics in social settings. (Ibid) (F) Percentage of respondents who claimed to be members of civic organizations. (Ibid) (G) Percentage of respondents who claimed to be active members of civic organizations. (Ibid) (H) Percentage of respondents who answered ?Excellent? or ?Basically Good? to the question ?What do you think of the Bundestag in Bonn as our representative assembly?? Other responses included ?Fair,? ?Poor,? ?No Opinion? and ?Indifferent.? (Loewenberg 1967) 0% 20% 40% 60% 80% 100% 19 48 19 49 19 50 19 51 19 52 19 53 19 54 19 55 19 56 19 57 19 58 19 59 19 60 19 61 19 62 19 63 19 64 19 65 19 66 19 67 19 68 19 69 19 70 19 71 19 72 19 73 Figure 3.1: Various Indicators of Civic Engagement in West Germany (A) Voter Turnout (B) Interested in Politics (C) Democracy Best System (D) Germany Best Off (E) Talk Politics (F) Membership (G) Active Membership (H) Approve of Bundestag 71 generation. But many new democracies?including Weimar Germany?simply did not have a generation?s worth of time to spare. As noted in the opening chapters, the average life expectancy of an imposed democracy is just over twenty years, and that sample includes the long-lived democracies of the FRG, Japan, Italy, and others that have survived since WWII. In other words, roughly half of imposed democracies did not survive long enough to benefit (or, for that matter, suffer) from the effects of political re- education on the values of a new generation in the new regime. The focus of this project, on the other hand, is the success of imposed democracies in their nascent years. Political re-education and value change may certainly play a role in the long-term success of democracies that emerge from occupations, but the theories operate on a time scale that extends beyond the temporal domain of the Marginal Value Model. Rival Hypothesis: Institutional Design Perhaps the most commonly cited contributor to stable democracy in the FRG is its constitution, known as the Basic Law (Grundgesetz). Given the history of earlier democratic regime in Germany, this is a natural starting point for any theory explaining the success of democracy in post-WWII Germany?especially if the aim is to compare and contrast the two republics. The Basic Law differed from the Weimar covenant in several ways, including but not limited to 1) a more federal government delineated explicitly by two articles assigning superior and concurrent jurisdictions for the central and Laender (state) governments, 2) a requirement for the election of a new majority government before dismissing the previous one (in other words, no more minority cabinets), 3) a requirement that political parties adhere to democratic principles, 4) a 72 hobbled presidency elected by the Bundestag and Laender governments (rather than by referendum), and 5) a federal constitutional court charged with reviewing the constitutionality of general legislation as well as the compliance of political parties with the democracy requirement. The Basic Law omitted the Weimar constitution?s Article 48, which granted the president sweeping, unchecked powers upon the declaration of a state of emergency by the Reichstag (Merkl 2002, Parliamentary Council 1984, Kvistad 2002, Bernhard 2001)9. In 1956, the FRG set itself further apart from the Weimar Republic by establishing a five percent vote threshold to win seats in the Bundestag (Merkl 2002, Niclauss 1982). Karlheinz Niclauss (1982) credits the federal character of the FRG for ?maintaining a balance of power in the political system,? particularly in that the opposition in the Bundestag was able to counterbalance the power of the presiding government by retaining control of the legislatures in the various Laender.10 Moreover, the rules on forming new cabinets and the weakened presidency curtailed the possibility of cabinet crises and authoritarian opportunism that characterized the Bundestag. The vote threshold made the formation of majority cabinets less cumbersome by limiting partisan competition to a few mass parties. Peter Merkl (1963) sees the establishment of a less direct, more mediated democratic system as contributing to the stability of the FRG 9 Article 48 of the Weimar Constitution permitted the banning of political parties that posed a threat to the state, but made no mention of democratic or anti-democratic platforms. Power to invoke and enforce this article was placed in the Reichstag and the presidency (and not the courts). During the Weimar regime, the government refused to use the power to ban right-wing authoritarian groups (Niclauss 1982). 10 It is my scholarly opinion, and nothing more, that students of the two regimes tend to overstate the federal nature of the Basic Law and understate the federal nature of the Weimar Constitution. Both were federal states, one certainly more federal than the other, but the degree of difference is questionable. This is a question begging further analysis. 73 (relative to the Weimar Republic). By eliminating plebiscites (that, under NSDAP control, had served as rubber stamps for authoritarian government) affording Land governments more power in the Bundesrat (council of states), and eliminating the directly elected presidency, the Basic Law ensured that the new regime would be run by elites, rather than a presumably more capricious mass constituency. Moreover, the voter threshold for representation ensured that only moderate parties would participate in legislation and execution of the laws, and that these parties would be better suited to uniting?rather than dividing?the body politic. Donald Kommers (1989) claims that the Basic Law ensured stability and legitimacy by striking a balance between Weimar-style majoritarianism and elite government constrained by the rule of law.11 He cites specifically the constraints of the Federal Constitutional Court in safeguarding the Basic Law and its government from majoritarian whims, but also its responsiveness in protecting the basic rights of the citizenry. One should not deny the importance of a well-written constitution or of well- written legislation enacted to protect said constitution. Almost certainly, these changes succeeded in accomplishing three key objectives: 1) moderating electoral competition under the new regime, 2) ensuring that parliamentary politics and legislative outputs remained faithful to republican principles and the rule of law, and 3) delaying the translation into policy of swift changes in public opinion. Since the Bundestag first convened, majority government has been the undisputed norm (Loewenberg 1967). As per the Basic Law, there was no opportunity for a chief executive to appoint a cabinet without majority support of the legislature, as President Hindenburg did in March of 1930 11 The author is careful to note that the ?ills of West German politics cannot be cured by constitutional law.? 74 (Lindenfeld 1999, Mommsen 1996, Snyder 1998, Kolb 1988). Governments in the Bundestag have successfully petitioned the Federal Constitutional Court to ban anti- democratic interest groups and parties. In 1951, the Court outlawed the neo-Nazi Socialist Reich Party (Merkl 2002, Glees 2003, Cromwell 1964). The Court banned the KDP (successor party to the Weimar-era KPD) in 1956 (Kvistad 2002, Merkl 2002), the student wing of the German Right Party (DRP) in 1960 (Cromwell 1964), and the League for Germanic Piety (a student group tied to General Ludendorff, early supporter of the Nazi party in the Weimar state) in 1961 (Ibid). In 1972, the government banned members of radical student groups from joining the civil service (Kvistad 2002). After the imposition of the 5% federal vote threshold for Bundestag inductees, the number of parties represented in the legislature dropped from thirteen in 1949 to six in 1953, four in 1957, and three from 1961-1972 (Schweitzer, et al. 1984, Loewenberg 1967, Heidenheimer 1958). From 1961 onward, those represented were exclusively of the Optimal Domestic Coalition (Ibid). In fact, from 1953-1972, the share of seats for parties in the ODC was 4.7% higher than their share of the vote, while the seat share of semi- optimal and suboptimal parties was 2.0% lower (Ibid). 75 Table 3.1: Translation of Votes to Seats in Federal Elections to the Bundestag Date Vote Share Seat Share Difference Optimal Semi Sub Optimal Semi Sub Optimal Semi Sub Sep '53 86.0% 9.1% 3.3% 91.7% 8.3% 0.0% 5.7% -0.8% -3.3% Sep '57 90.6% 8.0% 1.0% 96.5% 3.3% 0.0% 6.8% -4.7% -1.0% Sep '61 94.3% 4.7% 0.8% 100.0% 0.0% 0.0% 5.7% -4.7% -0.8% Sep '65 96.4% 1.3% 2.0% 100.0% 0.0% 0.0% 3.6% -1.3% -2.0% Sep '69 94.6% 0.0% 4.9% 100.0% 0.0% 0.0% 5.4% 0.0% -4.9% Nov '72 99.1% 0.0% 0.9% 100.0% 0.0% 0.0% 0.9% 0.0% -0.9% Mean 4.7% -1.9% -2.2% Source: Schweitzer, et al. 1984. The effects of the Basic Law and electoral law on general electoral and parliamentary politics are obvious. What is not obvious is how democracy under the Basic Law would have performed if the electorate had been more polarized, or had the ideological orientation of the citizens of West Germany been more radical. Had one of the many semi-optimal, constitutionally acceptable parties gained a significant constituency, the 5% threshold would not have been enough to keep them out of the Bundestag. Had one of these parties gained a plurality, building any coalition?let alone a purely pro-democratic one?may have proven significantly more difficult.12 Nor is it obvious whether the Basic Law had a significant effect on economic or paramilitary support for radical regimes. A better constitution could not have prevented financial institutions from secretly backing radical groups. The Basic Law could not have extinguished the Reichstag fire or kept the Brownshirts at bay. Without cooperation from those in the government, legal prohibitions on radical organizations would have been as effective as they were against the Stalhelm and the SA under the last democratic regime. 12 This was, in fact, the case following most elections under the Weimar regime. 76 The Basic Law explains a lot about democratic stability in Germany after WWII. What it does not explain is the relative absence of radical and/or violent political activity outside of the electoral realm. That is not to say that it is unimportant. Good regimes require good constitutions, and the Basic Law is an example of a constitution that successfully reinforced the Optimal Domestic Coalition. It is important to remember that this document was based on the input and subject to the approval of the occupiers who demanded a West German state constrained by a democratic, federal constitution (Plischke 1953, Glees 2003, Haftendorn 2006, Merkl 1963). Should an occupying force wish to ensure long-term democracy in its zone, it might endeavor to write, or at least encourage the authorship, of a constitution that limits electoral competition to pro- democratic constituencies. Rival Hypothesis: Delegitimation via Defeat One explanation for the absence of an authoritarian resurgence following the Second World War rests on the delegitimation of the Nazi regime following their complete defeat at the hands of Allied forces. Anne Sa?Adah (2006) presses this point fluently. She notes that ?the new regime must delegitimize its competition and construct its own legitimacy? (p. 310). In this case, the Nazi regime successfully delegitimized itself. An ideology built on conspicuous displays of power and rhetoric rooted in the historical inevitability of victory, the defeat of the Thousand Year Reich by armies fighting for decadent liberal democracy and treacherous bolshevism was thoroughly 77 invalidating.13 This is notable for two reasons. First, the ?culture of despair? (ibid) that Nazism thrived on seemed to be replicating itself in the post-WWII famine and economic desperation that preceded the Wirtschaftwunder (MacDonald 2007). Second, the victory of the Soviet Union brought the threat of international communism within the borders of the nation. It is not inconceivable that these two developments would have meant fertile soil in which a neo-Nazi movement, or some other authoritarian ideology, could thrive. Instead, Sa?Adah claims that the Christian Democratic majority and the Western Allied armies were more than able to channel any anti-communist sentiment into legal, pro- democratic channels. The author goes on to claim that the occupation gained support due to the fear of what success communism might enjoy in the absence of the Allied forces. It is more than plausible that defeat delegitimized the NSDAP and its heir apparent, the Socialist Reich Party. However, this does not explain the rejection by the electorate of the explicitly anti-democratic Communist Party (not outlawed until 1956) and West German Communist Party, nor the implicitly monarchist and nationalist, anti- democratic parties such as the German Right Party, the National Democratic Party, the Bund der Heimatlosen und Entrechteten (the Association of Expellees and Disenfranchised), the German Party, or the All German Bloc (Schweitzer, et al. 1984). Between 1949 and 1972, explicitly authoritarian parties polled no more than 7.5% in any Bundestag election. Borderline authoritarian parties totaled no more than 9.1%. Together, parties disloyal to democratic institutions never polled higher than the 14.4% in the first Bundestag election in 1949 (Ibid, Loewenberg 1967). Nor does this explain the relative absence of paramilitary activity in the FRG. There was no serious insurgent, terrorist or 13 The author cites Ralf Dahrendorf (1967) as the originator of this argument. 78 paramilitary threat in Germany until the rise of the left-wing student groups in the mid- 1960s through the early 1970s (Kvistad 2002, Merkl 2002, Glees 2003). Even these groups, including the Baader-Meinhoff Gang, were small compared to the militias, veterans groups and paramilitaries that operated with impunity during the Weimar regime (Cromwell 1964).14 Defeat may have successfully delegitimized the Nazi party, its supporters and its ideology. It may also have delegitimized the individuals and institutions that brought the 14 Hitler began to organize paramilitary groups, called Werewolves, in October 1944. Most of these groups dissolved quickly after the surrender. Before V-E Day, one Werewolf unit successfully assassinated the American-appointed mayor of Aachen in March of 1945. By July, General Clay reported no traces of Werewolf units in the American sector (MacDonald 2007). Aug '49 79.4% Sep '53 86.0% Sep '57 90.6% Sep '61 94.3% Sep '65 96.4% Sep '69 94.6% Nov '72 99.1% 6.9% 9.1% 8.0% 4.7% 1.3% 0.0% 0.0% 7.5% 3.3% 1.0% 0.8% 2.0% 4.9% 0.9% 0% 20% 40% 60% 80% 100% 19 48 19 49 19 50 19 51 19 52 19 53 19 54 19 55 19 56 19 57 19 58 19 59 19 60 19 61 19 62 19 63 19 64 19 65 19 66 19 67 19 68 19 69 19 70 19 71 19 72 19 73 19 74 Figure 3.2: Performance of the Optimal and Suboptimal Coalitions in Federal Elections to the Bundestag, 1949-1972 Optimal Semi-optimal Suboptimal Turnout 79 Nazi party to power. There is, however, no explanation as to why delegitimation would have quashed the prospects of right-wing, authoritarian parties that emerged following WWII and that claimed no connection to the previous regime. Nor should it have had this effect on the Communist Party. Indeed, the mechanics of the delegitimation thesis imply the opposite. The KPD was the archrival of the NSDAP, was the most viciously repressed and thoroughly persecuted political group at the start of the Third Reich. It, or any party claiming the mantle of proletarian revolution (such as the postwar KDP) should have benefited?rather than suffered?from the delegitimation of the NSDAP. It did not. The KDP mustered only 5.7% and 2.2% in the first two federal elections before it was banned in 1956 (Schweitzer, et al. 1984, Loewenberg 1967). The party reformed as the DKP in 1968, but never polled even 1% in a federal election (Kvistad 2002, Loewenberg 1967). Sa?Adah explains this away by claiming that ?the fear and revulsion the proximity of Soviet troops? made the KDP an unpalatable choice (p. 315). This explanation rings hollow. If anything, it should have activated communist support as it did in the east. Moreover, presence of British, American, or (even) French troops on German soil did not automatically harm the electoral performance of likeminded liberal, bourgeois parties. In summary, the author?s thesis is a potential explanation for the lack of success among neo- Nazi organizations in the FRG, but it doesn?t explain the lack of success among generally authoritarian political organizations. Rival Hypothesis: Geographic Division Building on Barrington Moore?s (1966) uneven development hypothesis?which ties the failure of democracy under the Weimar Constitution to uneven economic 80 development and the survival of reactionary political elites?Michael Bernhard (2001) claims that democracy thrived in the Federal Republic because the division of Germany had sequestered the nation?s most authoritarian and backward elements. As the argument goes, the FRG was established in the most modern part of Germany. The Iron Curtain cut off the Junker-dominated, agricultural East, obviating the issue of uneven development cited by Moore. The author uses district-level election data from the 1920?s test the hypothetical relationship between geography, economic development and attitudes towards democracy. In a related argument, Niclauss (1982) looks at the ban on the resurrection of a Prussian Land, as decreed by the Allied Control Council (the military government that presided in between the Third Reich and the Occupation Statute) in February 1947 (MacDonald 2007). The occupiers saw Prussia as the ?center of militarism and expansionism throughout German history,? and with two-thirds of the area under the Reich and three-fifths of the population, Prussia could have served as a major roadblock to the type of federalist democracy the Western Allies gradually came to favor (ibid). This argument is lacking. What Bernhard actually finds, instead of a relationship between economic development and democratization, is a relationship between region and partisan preferences. Certainly, the ?disloyal right? and ?disloyal left? performed better in the east in terms of percentage of votes for Reichstag elections?so much so that optimal parties continued to win a plurality (nearly a majority) of the national vote through 1932. Moreover, in many ways the east was less developed, more agrarian, and more ?backward.? Perhaps most importantly, the new republic didn?t include East Elbia, a region dominated by the Junker nobility. As I argued in the previous chapter, however, there are other explanations for the relationship between geography and support for 81 democracy. One such explanation is the conduct of that particular occupation and the threats encountered along the borderlands. Bernhard makes no effort to actually measure the level of development across German Laender or connect it to support for democratic coalitions. Moreover, by simply aggregating the Laender into ?East? and ?West,? the author ignores the antidemocratic tendencies of states that joined the FRG, most notably Saxony and Bavaria. While initial support for the Nationalists and National Socialists flowed from East Elbia and East Prussia, the NSDAP was founded and based in Bavaria, launched their first campaign against the government in Munich, and relied on that region for economic and paramilitary support (as did the DNVP). On the left, the Communists too launched their first coup in Munich. Moreover, the western regions?especially the Saar and Ruhr?were commonly subject to left-wing extremism in the earliest days of the Weimar state. These regions were not isolated as a result of the division of Germany (with the exception of the Ruhr until 1956). Nor were they breeding grounds for anti- democratic sentiment in the FRG. Yes, Bavaria voted down the Basic Law (Willis 1962, Rohrschneider 1999, Plischke 1953, Haftendorn 2006), but not out of opposition to the Bundestag or the generally democratic nature of the new constitution (Merkl 1963). Nor did Bavaria attempt to prevent the implementation of the Basic Law?the Bavarian Landtag explicitly accepted the new constitution as ?binding? upon ratification (Ibid). On the contrary, there are indicators that support for democratic and republican principles were strong in Bavaria early in the history of the FRG. The American- sponsored Union for Civil Liberties (Bund fuer Burgerrecht) found its strongest early support in Bavaria: in 1950, twenty-six of its forty local chapters were located in that particular Land; by 1952 the ratio was 33 of 60 (Rupieper 1997). In the Bundestag 82 election in 1949, the optimal domestic coalition dominated the vote as it did elsewhere in the FRG. Together, the SPD, the Christian Social Union (CSU), the Bavarian Party and the Free Democrats (FDP) tallied 81.3% on turnout of 81.1%. In 1953, that same coalition tallied 87.4%. The same applies to Prussia. As Niclauss (1982) correctly notes, the characterization of Prussia as a hotbed of militarism may have been outmoded even before Hitler came to power. Rather, Prussia was the last remaining bulwark of social democracy before Chancellor von Papen dissolved its SPD-led government. Pro- democracy sentiment remained strong in the Laender that once comprised Prussia in the new republic: Schleswig-Holstein, Lower Saxony, North Rhine-Westphalia, Rhineland- Palatinate, Hesse, and Baden-W?rttemberg (Alvarez-Rivera 2012). The history of Bavaria points towards a much deeper reform in German political preferences than can be explained simply by removing undesirable, anti-democratic elements. While one should not dispute that the division of Germany contributed to the development of democracy in the FRG, the evidence does not lend this fact the central role that Bernhard seems to believe. The optimal coalition was far more successful in all of Western Germany than it had been anywhere in the Weimar Republic, except perhaps following the first Reichstag election in 1920. Uneven development, a persistent Junker class, and an agrarian political economy might certainly have stood in the way of democratic institutions in an alternate reality in which Germany was both undivided and democratic, but their absence cannot explain the near absence of an anti-democratic nationalist or communist constituency in Bavaria or the rest of the FRG despite the 83 deprivation and starvation15 that reigned in the first years. Therefore, while Bernhard makes a compelling case for why a unified Germany (with an intact Junker class) might have been less democratic than the FRG, he fails to explain why democracy in the real- world FRG was stable. Summation The many hypotheses presented above each contribute something to our understanding of the success of democracy in the German Federal Republic, but they do not paint the whole picture. The demilitarization/debellicization thesis resonates well with a few key facts about the occupation and the subsequent government, specifically the absence of a well-armed domestic opposition that could have prevented or weakened the new regime. It is nonetheless an incomplete and imperfect explanation. It does not delineate the processes by which domestic actors in a less-than-debellicized state would undermine democratic process, and thus it does offer observable, falsifiable implications. Nor is it entirely compatible with the history of the FRG from 1950 onward, in which the debellicized state was thoroughly rebellicized, albeit in a multipartite sense that lacks any precedent in the history of international relations.16 The political re-education thesis is a compelling one for understanding the long-term consolidation of German democracy. 15 Giles MacDonald (2007) thoroughly details the substandard quality of life among German citizens under the military government, and before the voluminous American financial assistance that began to flow later in the decade. The average West Germans subsisted on a diet of at most 1000-1500 calories a day. General Clay at one point warned of a nutritional disaster, fearing that an impoverished Germany might serve as an incubator for communist sentiment. 16 From a practical standpoint, it is also worth noting that an explanation that relies in part on the success of belligerents prior to the cessation of hostilities might be less helpful for civilian policymakers in planning the occupation. 84 However, since it relies on the slow process of value change aided by generational turnover, it is not a particularly useful hypothesis for explaining the viability of the FRG in its earliest years.17 On the other hand, re-education (particularly denazification) and the institutional design hypothesis help explain why regular political intercourse in the FRG was the near-exclusive domain of pro-democracy forces. The forceful excision of the previous authoritarian regime from the new government, laws that excluded anti- democratic organizations and a constitution that ensured a more efficient parliament helped the FRG avoid many of the pitfalls of the Weimar Government. That said, these explanations fail to explain the relative absence of polarization in electoral politics or authoritarianism outside of the electoral domain. Moreover, the denazification and delegitimation theses explain only the success of preventing the resurgence of Nazism, not authoritarianism in general. Finally, the division thesis is appealing on the face but, for reasons described above, ultimately unsatisfying. Many of the lacunae left behind by these explanations can be filled by the Marginal Value Model. 17 Nor do scholars who lean on denazification as an explanatory variable duly appreciate that implementation was uneven across the zones (and almost completely disregarded by the French), a source of enmity among otherwise pro-occupation constituencies towards the occupation government, and hastily abandoned only four years after WWII (MacDonald 2007, Willis 1962, Glees 2003, Kvistad 2002). 85 Test of Working Hypotheses versus Case History Threat Environment Every play needs a villain, and in the stage play of the democratization of the Federal Republic is no different. The United States and United Kingdom turned in bravura performances as the champions of freedom and prosperity. France competently played the essential role of erstwhile-enemy-turned-cantankerous-skeptic who, ultimately, proves indispensible to the mission. Great as those performances may have been, they would have been entirely superfluous absent a villain. The antagonist moves the play forward by providing the heroes with a mission. Without the villain, the heroes are merely actors taking up space and the play is nothing but a waste of time and money. With a truly threatening villain, however, one that strikes fear into the hearts of the audience?in this case, the citizens of what would be West Germany?the mission becomes clear and the heroes gain purpose. The presence of a threatening villain is not a sufficient criterion for winning the audience?s hearts and minds, but it is a necessary one. The stage play of the democratization of the FGR had such a villain, one that justified the presence of the United States, the United Kingdom and France and proved the value of their mission. The role of villain was ably played by the Union of Soviet Socialist Republics. The Soviet Union generated the threat that the US, UK and France could deter, thus transforming Table 3.2: Influences on the Threat Environment in West Germany More Threatening ? Soviet ?fifth column? ? Soviets abrogate wartime agreements ? Berlin Blockade Less Threatening ? ?Peace Note? ? Renewal of unification talks 86 their occupation from a nuisance into a boon and defining the mission that would unite the occupiers with the occupied. That the play would unfold this way was not a fait accompli. The Soviets started out as partners in the occupation of Germany, and it was not clear early on that the French would be any more accommodating than the Soviets would prove to be. The conduct of the Soviets and the French might have led an occupation that was as catastrophic a failure as that which ruled Weimar Germany a generation earlier. There are three reasons that France?s initial desire to dismember Germany and the Soviet Union?s diktats and demontage did not derail West German democracy. First, the general autonomy that each occupier had in its zone ensured that the conduct of the Soviets and French would have little impact on Germans in the British and American zones. These zones together comprised about more than half of the land area and population (MacDonald 2007). Second, the French diplomatic and military positions were the weakest of the Four Powers, eventually requiring the delegation to toe the Anglo- American line. Third, the division of Germany not only relegated Soviet influence solely to the Soviet Zone, but also transformed a domestic threat to Germany into a foreign threat to a new Federal German Republic. In fact, the USSR demonstrated the threat it posed to a secure, autonomous German state before the occupation officially commenced In the waning days of WWII, the Soviet Union embarked upon a ?fifth column? strategy that pursued the infiltration of Eastern and Central European governments using local Communist parties as agents (Mark 2001). Because these parties answered directly to Moscow, these organizations posed a permanent threat to the target states? sovereignty and autonomy. This strategy first became apparent in Greece, where a communist 87 insurgency simmered between 1944 and 1948 (Dunbabin 2008, Mee, Jr. 1984). Communist parties later made limited gains in Romania, Yugoslavia, Italy, France and, eventually, Germany (Ibid). Upon taking control of their occupation zone, the USSR was the first in organizing ?anti-fascist? parties, initially legalizing both bourgeois (CDU, Liberal Party) and Marxist (SPD, KPD) organizations (Glees 2003, Schweitzer, et al. 1984, Rohrschneider 1999, Windsor 1969). But the era of the independent political party in the Soviet zone was short lived. As soon as June 1945, Stalin ordered his military government in Germany to press for the merger of all Marxist parties with the KPD (Dunbabin 2008). The SPD maintained its independence in the three Western zones, but in 1946 the SPD and KPD merged to form the SED, or Socialist Unity Party (Glees 2003, Dunbabin 2008, Merkl 2002). This merger took place under duress, as Soviet military intelligence had already begun arresting SPD members who opposed unification (Glees 2003). The experience in the Soviet zone mirrored that of the newly independent states of Eastern Europe.18 In August 1947, six hundred Christian Democrats were arrested in the Soviet Zone (MacDonald 2007), a prelude to the forcible removal from government of the CDU at the end of that year (Glees 2003). The establishment of a German Democratic Republic (DDR), answering to the USSR, following the Soviet arming of the Greek insurrection and Communist takeovers in Czechoslovakia and Hungary contradicted any beliefs that the Soviet Union was interested reinstating German sovereignty in anything but the most formal sense. However, it was not until the actual division that this ceased to be a threat by an occupation power and transformed into a threat from a foreign state, 18 Upon the signature of a Bulgarian peace treaty with the Allied forces, which included a condition to hold free and fair elections, the Soviet military government responded by hanging the Social Democratic opposition leader and proceeding with their takeover of the Bulgarian government via the local Communist Party (Dunbabin 2008). 88 thus opening a window for the Western Allies to increase the security value of their occupation. The Soviet threat arose not just from the Fifth Column but also from behavior that signaled an unwillingness to abide by its past commitments. In September 1945, the USSR unilaterally renounced their non-aggression pact with Turkey (Dunbabin 2008). Months later, the USSR reneged on its promise to end its wartime occupation of Iran.19 In September 1947, the Soviet Union declared the global schism between capitalism and communism irreconcilable by peaceful means (Haftendorn 2006).20 Most relevant, however, are the fissures that grew out of the joint military command in Germany leading to the Soviet abandonment of four-power control, the unification of the three western zones, the Berlin Blockade and subsequent airlift and, finally, the division of Germany. The quadripartite government of Germany collapsed with the sine die adjournment of the London Council of Foreign Ministers in December 1947. After pressing for the establishment of a unified German government, an Anglo-American gambit that was sure to meet stiff resistance from the USSR, the Soviet delegation walked out of the meeting without setting a future date for re-adjournment (Dunbabin 2008, Plischke 1953). The three Western powers proceeded to unite the US-UK Bizone with the French zone (Glees 2003, Plischke 1953). The USSR responded by leaving the Allied Control Council (Dunbabin 2008, Willis 1962). The introduction of a new currency to the Trizone? formalizing its economic unity?was the last straw. In an effort to reverse the policy of 19 They withdrew only after Iran appealed the US, UK and the newly established United Nations and offered to form a joint oil company with the Soviet Union (Ibid). 20 In response to Churchill?s Iron Curtain speech?it is worth noting that Churchill, along with several high-ranking government officials in the US and UK, needed no help from Moscow in painting the USSR as an implacable, expansionist threat. 89 the Western allies, Stalin pressed his biggest advantage. Authorities in the eastern zone first cut off traffic to the western sectors of Berlin in late March 1948 (Willis 1962, Mitchell 2005, MacDonald 2007, Dunbabin 2008) and imposed a full-scale blockade on June 24 (Glees 2003, Haftendorn 2006). The Berlin Blockade lasted through May 1949, took place during one of the coldest winters on record and brought West Berlin to the brink of famine (MacDonald 2007, Mitchell 2005). The Blockade itself was a clear signal that the Soviet Union was unconcerned with the security and welfare of the German citizenry. The Blockade and division of Germany cemented the Soviet role as foreign threat, turning the threat environment critical and thereby raising the marginal value of the occupation. Following the failure of the Soviet hardline stance to prevent the unification of the three western zones, the Soviet Union employed a more peaceful (but perhaps more disingenuous) tone. In November 1950, the USSR offered to restart quadripartite talks with the Western Allies while the DDR made a similar overture to discuss unification with the FRG (Glees 2003, Hitchcock 1998). Adenauer spurned the initial invitation, citing a preference for western integration (Dunbabin 2008). Stalin made a broader overture on March 10, 1952, in his famous ?Peace Note? (Windsor 1969, Hanrieder 1980, MacDonald 2007). The note offered support for a unified German state with free elections on the pretense that the new state builds only a small defensive army, that all occupiers withdraw their troops, and that Germany would be prohibited from joining any foreign alliances (Hyland 1980, Hitchcock 1998).21 The FRG and Western allies remained skeptical of this new offer, and rightly so. When the Four Powers resumed 21 The FRG was already in the process of joining the European Defence Community and was on its way to full partnership in NATO. 90 quadripartite talks at Berlin in January 1954, the USSR?s formal offers were far less conciliatory (Haftendorn 2006, Dunbabin 2008, Hitchcock 1998, Schmidt 1994). Stalin (who died before the conference) and Khrushchev aimed to prevent the continued integration of the FRG into the Western economic and security bloc (Schmidt 1994). Had their overtures been credible, they may have succeeded in convincing the Federal Republic and its citizenry that the Soviet threat was overblown, thus reducing the security value of the western occupation. Instead, the shadow of the Berlin Blockade, the history of the military government?s oppression of democratic partisanship in the eastern zone, the repression of democratic politics on the east side of the Iron Curtain?including the ruthless crushing of a workers? revolt in East Berlin (Dunbabin 2008, Merkl 1980, Rohrschneider 1999, Haftendorn 2006, Glees 2003, Reifenberg 1980)?and the USSR?s refusal to concede UN monitoring undermined the credibility of the peace overture. Accordingly, a 1952 public opinion poll revealed that 65% of those surveyed in the FRG considered the Soviet Union a threat to their security. Following the failed talks in 1954, the number remained relatively stationary at 63% (Merkl 1980). And so it was that the Soviet threat remained strong through the end of the occupation, providing the Western Allies ample opportunity to demonstrate the positive marginal value of the occupation. Had the occupation not ended in 1955, Khrushchev?s renewed threat to isolate West Berlin (Government of the USSR 1984, Mitchell 2005, Dunbabin 2008, Haftendorn 2006, Hyland 1980, Windsor 1969) would only have increased the occupation?s marginal value, as would the erection of the Berlin Wall (Catudal 1980, German Democratic Republic 1984, Post, Jr. 1997, North Atlantic Council 1984). In the years following, the USSR assumed a policy of d?tente with the 91 West. This time, their advances were far more successful, helping to usher in the era of Ostpolitik. By then, however, the occupation had long since ended. About the same time, the government in Bonn marked an important step on the path to consolidation when Willy Brandt?s Social Democrats completed Germany?s first peaceful, democratic transfer of power since the rise of Adolf Hitler. Signaling Security Value While the protagonists in our play eventually proved to be true champions of freedom, security and prosperity, in the opening act this was less than assured. Germans would not have viewed the earliest Allied postwar plans as an enhancement of their security. The original position of the United States was to pastoralize Germany, dismantle its factories, neutralize its war-making capability and split it into five separate states. Through 1946, the UK, USSR and France would all endorse some version of this proposal (MacDonald 2007, Killick 1997, Willis 1962, Mitchell 2005, Glees 2003).22 In truth, there was never a consensus behind this proposal, known as the Morgenthau Plan23 and, eventually, the US came around to the British point of view that the best approach was one of reform rather than relegation. The French reluctantly followed suit. By the time of the Yalta conference (in February 1945) FDR was distancing himself from Morgenthau and his Carthaginian peace. Instead, unconditional surrender for Germany 22 German fears of a Carthaginian peace may have been stoked by ill treatment of civilians by Allied forces as they entered Germany and forced Berlin?s surrender. While the Soviets were the worst perpetrators of these crimes, the US, UK and French forces were not innocent theft, arson, rape and murder in the first months of 1945 through V-E day (MacDonald 2007). 23 After the US Treasury Secretary who originated it. 92 would mean a federal German state and an assuredly temporary occupation dedicated to the five D?s: demilitarization, disarmament, democratization, decentralization and factory dismantlement or demontage (Willis 1962, MacDonald 2007, Pollock 1944, Mitchell 2005). The US, UK and USSR formalized this proposal at Potsdam in August 1945 (Glees 2003). They reiterated the Quebec Conference demand for retrocession of territories seized after 1938 and recognition of a Polish frontier along the Oder and Neisse rivers. This proposal would restore the Polish state and provide a buffer for the Soviet Union, but it also required the forced relocation of millions of German citizens (MacDonald 2007). Nevertheless, the occupation plan as implemented carried a far higher marginal security value than the one that Morgenthau proposed, that FDR once endorsed, and that France would eventually and reluctantly abandon. Still, for want of an external threat there is little security value to enhance. This changed once the Soviets took a more aggressive posture. Prime Minister Churchill?s Iron Curtain speech of March 1946, which reflected a growing wariness with Soviet dominance in Eastern Europe and all but accused the Soviet Union of expansionist ambitions (Gaddis 2000), accentuated the nascent East-West divide. Truman doubled down on this line in March 1947 when he pledged $400 million (contemporary) in financial assistance to Greece?which was in the middle of a heated left-wing insurgency? and Turkey?where the USSR was pressing their influence to obtain free passage to the Table 3.3: Indicators of Occupation Security Value in West German Occupation Positive ? US expenditures in Germany, Europe ? Operation Vittles (Berlin Airlift) ? German integration in NATO ? European Coal and Steel Community Negative ? Morgenthau Plan ? Factory dismantlement ? Control of Saar, Ruhr 93 Black Sea (Mee, Jr. 1984).24 Even though these declarations did not directly involve Germany, they clarified the threat environment and signaled that the US would not eschew European entanglements as it had after WWI. Instead, the United States was going to assume the responsibilities that the British could no longer afford and actively balance against the Soviet threat.25 For Germans in the western zones, this raised the possibility that the western occupiers might someday be western allies. The security value of a plausible alliance served to mitigate the threat posed by the occupying powers, increasing the likelihood that the occupation would yield marginal value. The Western powers? response to the Berlin Blockade transformed a loosely beneficial relationship into a de facto alliance. Despite fears that the Berlin crisis could lead to all-out war, the Western Allies launched a massive campaign to supply West Berlin from the air (Mitchell 2005, MacDonald 2007). The Berlin Airlift?officially known as Operation Vittles?delivered two million tons of fuel and an indeterminate volume of non-fuel supplies over three hundred thousand flights during the Blockade?s eleven months. Thirty-nine British and thirty-one American servicemen would give their lives to the operation (Glees 2003). In support of the operation, the US deployed several B-29 bombers to the region (Mitchell 2005). The deployment of the same bombers that delivered nuclear weapons to Hiroshima and Nagasaki was a clear signal that the United States was willing to retaliate with a massive response in the event of a Soviet invasion. 24 Truman?s proclamation, that ??it must be the policy of the United States to support free peoples who are resisting attempted subjugation by armed minorities or by outside pressures? (Truman 1947) was the relevant signal to both the USSR and occupied Germany. The monetary pledge in and of itself is what defined the pledge as costly and, therefore, credible. 25 Concrete action along these lines reinforced Secretary of State Byrnes? speech in Stuttgart, where he promised that the US would stay so long as the USSR remained (Dunbabin 2008). 94 No single act signaled the positive marginal value of occupation better than the massive effort by three western occupation powers to save West Berlin from starvation and political subjugation. That the operation was costly in terms of resources spent and security gambled made the commitment all the more credible. Finally, the US, UK and France promoted this de facto alliance to de jure status by building an unprecedented network of trans-Atlantic multilateral organizations and, eventually, by welcoming the German Federal Republic as a full member. This process began in December 1947, when UK Foreign Minister Bevin began to contemplate a multilateral security pact among Atlantic democracies (Dunbabin 2008). Bevin then invited France and the Benelux states to form a mutual security pact that would eventually include the US. This new alliance would comprise the North Atlantic Treaty Organization (Willis 1962, Glees 2003, MacDonald 2007, Mitchell 2005, Mee, Jr. 1984, Dunbabin 2008, Kennan 1967, Killick 1997, Glees 2003, Hitchcock 1998). The NATO signatories later agreed to the construction of a West German army on condition that it would be under Allied control, satisfying the French demand for a forward position in Central Europe (Schmidt 1994, Plischke 1953, Hitchcock 1998). In October 1954, NATO invited the FRG to join as a full member (Dunbabin 2008, Glees 2003, Haftendorn 2006, Hitchcock 1998) and agreed to countenance a Bundeswehr (federal army) of up to one half million men (Kelleher 1980). Upon the conclusion of the occupation in May 1955, the FRG took its place alongside the other member nations (Banchoff 2002). The construction of NATO by the occupying states and the offer of a German role within the organization represented the strongest of signals that the occupiers are committed to West German security and protection against the Soviet threat, thus indicating the value of the 95 occupation. West German assent to these terms confirms their recognition of the threat in the East and their belief that salvation lie to the West. Alongside NATO, and with US approval, France attempted to organize a purely European multilateral force. French Premier Plevin reversed his government?s opposition to German rearmament when he invited the FRG to contribute a small contingent under French supervision (Dunbabin 2008, Willis 1962, Haftendorn 2006, Hillenbrand 1980, Schwartz 1986, Banchoff 2002, Krieger 1983, Glees 2003, Hitchcock 1998). While the EDC ultimately failed ratification in the National Assembly, it signifies the reversal of Allied attitudes to German militarization. Between 1947 and 1955, Western policy on Germany shifted from total demilitarization to the rearmament of a German state under a multilateral framework to protect against the common threat of the Soviet Union (Schmidt 1994, Schwartz 1986). The narrative serves as both signal and indicator: the alliances themselves signaled a lasting commitment by the occupiers on behalf of the FRG; the Federal Republic?s willingness to join said alliances indicates that Germany now saw its occupiers not as enemies, but as allies. Over the course of the military government and legal occupation, the Western Allies increased the value of the occupation by halting many of the security-diminishing aspects of the military administration. For instance, while the US, UK and France were originally required to ship 10% of their dismantlement proceeds to the Soviet zone, the US ceased these transfers in May 1946 after it became clear that the Soviets were not using their proceeds to ship food and supplies to the West as stipulated (Glees 2003, Dunbabin 2008, Hitchcock 1998, Hitchcock 1998, MacDonald 2007). France issued its 96 final demontage orders in October 1947 (Willis 1962).26 In June 1947, the UK began to roll back its denazification policy, which was proving to be unpopular with the majority of Germans, disproportionately unfair to Germans whose roles in the Nazi party were inconsequential and a bureaucratic nightmare (Glees 2003). The US would follow suit in May 1948 (MacDonald 2007, Willis 1962).27 There is one area, however, in which the Western Allies continued to undermine the value of their own occupation. Allied designs on the Ruhr sent a clearly negative security signal to Germany. The permanent loss of German coal and steel threatened to derail Germany?s economic rehabilitation and neuter their capacity to establish even a defensive military. In January 1946, France proposed that the Ruhr be placed under international control (Willis 1962).28 In February, France proposed the wholesale annexation of the Saar (Ibid). Both policies gained early support from the US and UK (Dunbabin 2008, MacDonald 2007, Ibid). With support from the Western allies, France set about economically isolating the Saar from the rest of Germany. In December 1946, France surrounded the region with a customs barrier (Glees 2003, MacDonald 2007). Between June and November of the following year, they introduced a new currency to the region?the Saar Mark?and declared it legal tender (Willis 1962, Glees 2003). During that time, Saarlanders had expressed overwhelming approval (87% in an October plebiscite) for economic union with France. They ratified the new arrangement in their constitution that December (Willis 1962). 26 Although actual dismantlement proceeded until the signing of the occupation statute. 27 The French never fully embraced denazification. In their view, German aggression was not a Nazi problem but a German problem (Willis 1962). 28 That same month, De Gaulle called for a permanent occupation of the Rhineland, and a Germany reconstituted as a confederation rather than a unified state (MacDonald 2007). 97 In threatening to rob Germany of strategic resources, the occupiers were diminishing the value of their occupation. Allied policy shifted abruptly in April 1948, however, when the France agreed not to export Saar coal for profit. They also agreed to halt any factory dismantlement in the Saar and to count all production and dismantlement towards future reparations payments (Ibid). The catalyst for this change was the USSR?s threat to cut off surface freight to West Berlin, which they did in June (MacDonald 2007).29 As for the Ruhr, the Allies did not roll back plans for internationalization. Instead, in May 1950, the French offered to place the Ruhr under the control of a transnational organization with the new German Federal Republic acting as a full partner (Glees 2003, Plischke 1953). Said authority would come to be known as the European Coal and Steel Community (Hitchcock 1998, Haftendorn 2006, Gillingham 1995).30 The 1948 shift in German coal and steel policy notwithstanding, France was reluctant to yield control of the Saar, continuing to publicly favor economic and political fusion (Hitchcock 1998). In one of his first public statements as chancellor, Adenauer slammed France?s plans for the region. (Willis 1962, Glees 2003). In November 1950, the Bundestag called on Saarlanders to boycott a plebiscite on integration with France, the results of which heavily favored autonomy from both Germany and France under a framework of European governance (Hitchcock 1998). In January 1952, France converted their High Commissioner in the Saar into a Saarland Ambassador, essentially 29 Ironically, prior to the Blockade, the USSR was the sole opponent to the cession of the Ruhr and Saar (Willis 1962). 30 Negotiations on the new arrangement began in June (Hitchcock 1998). When negotiations stalled, US High Commander John McCloy inserted himself in the process to force an agreement (Willis 1962). The treaty was initialed in the spring of 1951 (Hitchcock 1998, Haftendorn 2006, Glees 2003, Schwartz 1986), ratified by the Bundestag in January 1952 (Plischke 1953, Willis 1962), and took effect that June (Gillingham 1995). 98 declaring the territory sovereign (Haftendorn 2006). As late as January 1954, France held firm on their policy when four power talks resumed in Berlin (Hitchcock 1998). It wasn?t until January 1957 that the Saar Basin returned to West German control, nearly two years after the end of the occupation (Willis 1962). France?s policy on German coal and steel was no small grievance among the West German electorate and, in a vacuum, may have seriously compromised the security value of the occupation. Politics, of course, do not take place in a vacuum; the Saar dispute was but one negative in a sea of positive undertakings by the Western Allies that enhanced the security of the German Federal Republic. Facilitating Regime Consolidation Just as the Soviet Union became the clearest threat to a new German regime in the West, and just as the Western Allies were ramping up their commitment to protect the new regime and its citizens from said threat, they also moved to make the occupation more palatable and shored up the prospects of the optimal domestic coalition. Early on, they licensed partisan political activity but excluded right-wing authoritarian parties. They worked directly to remove right-wing authoritarianism from the occupied territory through a policy of denazification and trial of war criminals. They embarked on a political re-education campaign to safeguard the republican character of the new regime. They worked with the government of the FRG to eliminate authoritarian organizations on Table 3.4: Evidence of Facilitating Regime Consolidation in West Germany Constructive ? Denazification ? Frankfurt Documents ? Petersberg Agreements ? Economic reconstruction Destructive ? n/a 99 the left and right. Finally, the US, UK and France agreed on a new relationship committing themselves to the reconstruction of western Germany, the protection of democratic political intercourse, and the end to factory dismantlement and unconditional demilitarization. At a London meeting of the western Foreign Ministers/Secretaries of State, the delegations agreed on the fusion of the Trizone and its basis for the formation a new German state. Between February and June 1948, the delegates proclaimed that such a state must conform to the principles of federal democracy and separation of powers (Plischke 1953, Dunbabin 2008, MacDonald 2007). In late June, the three western military governors formally invited the ministers-president of the various Laender to promulgate a constitution (Plischke 1953). The documents they issued, known later as the Frankfurt Documents, ordered that the constitution be federal, democratic and spell out specific protections for civil rights (Glees 2003, Kruger 2002, Ibid). Over the next year, representatives of the Laender hashed out a charter along these lines, what would be known as the Basic Law (MacDonald 2007, Merritt 1976, Ibid, Haftendorn 2006, Merkl 1963). They ratified the Basic Law in May 1949 (Plischke 1953, Kvistad 2002, Willis 1962, Glees 2003, Haftendorn 2006, Schweitzer, et al. 1984). In August 1950, the western military commandants approved a constitution espousing identical values in West Berlin (Schweitzer, et al. 1984, West Berlin Commandants 1984). This is a key example of how a foreign occupier can directly intervene in the domestic politics of an occupied state to build (or rebuild) institutions that narrow the acceptable range of political contestation. The occupiers, as preponderant powers in the territory, are in a unique position to decide who benefits from the occupation?and who does not. 100 Another key example of direct intervention in the domestic politics of the state, denazification may have played an important role in excluding National Socialism from regular politics in the new regime. It was a two-pronged attack: the occupation would remove Nazis from positions of political importance and prosecute the worst perpetrators for war crimes. They were also excluded from economic production?coal and steel concerns once owned by Nazis or Nazi collaborators were confiscated, the owners arrested (Glees 2003). By the conclusion of denazification in 1949, over 53,000 public officials were stripped of their position and banned from future participation due to past ties to the Nazi Party (Kvistad 2002).31 The wholesale arrest and prosecution of Nazi war criminals began in June 1945, when the British military government issued a warrant under the jurisdiction of the Nuremburg Code.32 Later, the Four Powers signed the London Accords, an agreement on the prosecution of war criminals as well as a new class of plaintiff: those who committed ?crimes against humanity? (MacDonald 2007). Military policy arrested 100,000 Nazis and collaborators through the end of 1945. Ten thousand of those arrested were civilians in the financial sector, while 25,000 were linked to paramilitary groups (Ibid). Through 1946, the Allies arrested 250,000 former collaborators classified as ?dangerous? (Glees 2003). Ultimately, they tried only 8,000 cases. About 10% of these resulted in death sentences, half of which were commuted (MacDonald 2007). By June 1947, the British were already drawing down their denazification efforts, refusing to arrest any additional war criminals after September 1948 (Glees 2003). The US had made the same decision in May (Willis 1962, 31 The FRG would later reinstate about one thousand former Nazis (Ibid). 32 This itself was a synthesis of the 1907 Hague Convention on Land Warfare and the Geneva Convention of 1928 pertaining to the treatment of prisoners of war (MacDonald 2007). 101 MacDonald 2007). Even though the occupiers ultimately abandoned denazification, the policy defined the acceptable range of political contestation in the earliest years of the occupation and the new republic, marginalizing radicals and empowering moderates. This ensured that any the benefits the occupation conferred would be enjoyed most by stakeholders in the new regime. Even as the official policy of denazification concluded, the Western Allies and occupied West Germany continued to prevent the return of authoritarianism. In January 1949, the US military government sponsored a German civil liberties union, the purpose of which was to address citizen concerns of residual authoritarian sentiment in the civil service, sponsor conferences discussing equal opportunity and human rights, produce educational films and lobby for the creation of the Federal Constitutional Court (Rupieper 1997). These organizations flourished, growing to forty chapters in September 1950 and sixty in April 1952. The majority of local chapters were based in Bavaria, birthplace of National Socialism (Ibid). At Petersberg from September through November 1949, the High Commissioners of the three occupying states met with Adenauer and pledged to cooperate on the eradication of Nazism (Plischke 1953). While still under the legislative control of the occupying powers, the ruling coalition of Christian and Free Democrats continued the policy of marginalizing (left- and right-wing) antidemocratic groups by banning their members from the civil service (Kvistad 2002, Merkl 2002). Invoking Articles 21 and 43 of the Basic Law, the government would successfully petition to ban the neo-Nazi Socialist Reich Party in 1952 and the German Communist Party in 1956 (Merkl 2002, Parliamentary Council 1984, Glees 2003, Cromwell 1964). And despite the FRG?s general abandonment of the occupation?s denazification policy, they continued to 102 hunt, arrest, prosecute and imprison those who participated in the Holocaust (Markovits and Reich 2002). The Petersberg Agreements that bound the Federal Republic to the continued eradication of authoritarian politics were part of a larger agreement defining the nature of the occupation. The joint declaration reaffirmed the Western Allies? commitment to the economic redevelopment of Germany, to the resettlement of WWII expellees, and (in ambiguous language) declared security concerns in Eastern Europe a major determinant of security in the Federal Republic (Adenauer 1984a). The Agreements included an Allied pledge to ease demilitarization restrictions, to invite the FRG into international organizations including the International Authority of the Ruhr, to cease dismantlement of factories, and to permit the re-establishment of consular relations with foreign powers (Hitchcock 1998, Plischke 1953, Haftendorn 2006). Here, we have an example of the occupation explicitly tying security commitments to democratic reforms, as well as an example of the West German government accepting the bargain. The actions of the Western Allies boosted the political power of the optimal domestic coalition through multiple channels. They limited competition from antidemocratic challengers on both sides of the political spectrum. They then directly empowered the optimal domestic coalition handing the new ruling coalition of Christian and Free Democrats a major victory on occupation policy (see below for more on this). The shift in policy from a punitive occupation to a reconstructive one shored up support for pro-democratic coalitions and safeguarded the democratic nature of the new regime in the interim while specific affect towards democratic institutions grew among the West German population. In doing so, the occupiers ensured that the regime that benefitted 103 from the goods produced by the occupation would be a democratic regime more or less of the occupier?s choosing. The benefits specifically flowed to pro-democratic constituencies and away from radicals, making the marginal occupation value positive not just for West Germany in aggregate but for the optimal winning coalition in specific. Domestic Outcomes West Germany was an immediate and prolonged success of imposed democratization. At the point it debuts as a sovereign state in the Polity IV data set, it rates as a 10/10 (most democratic) and remained there until unification (Marshall and Jaggers 2009). The prolonged success of pro-democracy parties in West Germany and the general lack of domestic instability reflect the occupation?s positive marginal security and its direct efforts to boost moderates and marginalize radicals. In the first elections within the three western zones, optimal parties were extremely successful. Citywide elections in Berlin prior to the blockade showed strong support for the CDU (~22%) and SPD (~40%), together overshadowing the communist Socialist Unity Party (~20%; Glees 2003, MacDonald 2007, Rohrschneider 1999). That same month (October 1946) the CDU won landslides across town and city elections in the UK zone, polling on average 48.5%. The SPD added 31.3%, the FDP 2.7% and the Zentrum added 5.5% to the total of the optimal domestic coalition. The KDP?which would see its membership slowly evaporate before its 1956 ban?won only 7.4% (Glees 2003). In Landtag (state legislature) elections in the UK zone the following April, it was the SPD who outpolled the CDU (36.8% to 32.2%), nevertheless returning a sound majority for the optimal coalition (Ibid). Optimal parties likewise dominated the first elections to the Bundestag, 104 with the CDU/CSU, SPD, FDP and Zentrum totaling 79.4% of the vote. The KDP and the German Right Party (DRP)?both openly disloyal to the new regime?scraped together only 7.5% of the popular vote in the first election (Loewenberg 1967, Schweitzer, et al. 1984). Marginally disloyal parties such as the Economic Reconstruction Union (WAV) and the German Party (DP) managed a similarly paltry 6.9% (Ibid). The Socialist Reich Party (SRP), considered the true heir to the Nazi Party, fared poorly in the first federal elections, winning only one seat in the Bundestag and adding an independent that switched parties after the fact. However, the SRP performed won enough seats in the Landtagen in Lower Saxony and Bremen to alarm moderates in the Bundestag (Glees 2003, Cromwell 1964).33 The results prompted the Bundestag to initiate proceedings to have the Federal Constitutional Court ban the SRP (Cromwell 1964).34 The initial success and positive electoral trends among the optimal winning coalition conform to the expectations of the Marginal Value Model?the increasing marginal value of occupation yielded increasing electoral and parliamentary victories for the optimal winning coalition. The CDP and FDP would go on to form the Bundestag?s first government, in a coalition with the CSU, the CDP?s affiliate in Bavaria, and the semi-optimal German Party (DP) (Schweitzer, et al. 1984, Loewenberg 1967). This 33 The SRP won 11% of the vote (16 seats) in Lower Saxony and 7.7% (8 seats) in the Bremen Land elections of 1951 (Lee 1997). The SRP was not only disloyal in their opposition to democratic institutions (they did publicly deny the legitimacy of the new regime and considered Admiral Doenitz the last rightful German leader). But the most damning example of their loyalty was fealty to the USSR. Despite their decidedly right- wing orientation, the Soviet Union sponsored the SRP in an effort to foment anti-Western sentiment. Oddly, the USSR refused to sponsor the KDP?which was both communist and more successful in electoral politics?labeling them ?ineffectual? (Lee 1997). 34 One member of the FDP, Dr. Werner Naumann, had plans to attract the SRP diaspora following their 1952 ban. A former aide to Josef Goebbels, he was arrested by British authorities in January 1953 (Glees 2003). 105 partnership would dominate German government through October 1957, and again from November 1961 through December 1966, relegating the SPD to the role of loyal opposition for seventeen years (Loewenberg 1967). Optimal parties?specifically the CDU?consolidated their power in the election of 1953 and in the subsequent elections following the return of sovereignty. The optimal domestic coalition increased their share of the popular vote to 86% in September 1953 and 90.6% in September 1957. Those victories translated into 91.7% and 96.5% of the seats in the Bundestag, respectively (Loewenberg 1967, Schweitzer, et al. 1984). The Electoral Law of 1956 succeeded in pruning the number of parties with seats from thirteen to three, entirely at the expense of semi- and suboptimal parties. From 1961-1972, no party with an explicit or implicit aversion to democratic institutions won a single seat in the Bundestag (Ibid). In terms of the cabinet, optimal parties almost exclusively held ministerial positions. The few exceptions were the semi-optimal Bund der Heimatlosen und Entrechteten (BHE, or the Union of Refugees and Expellees) between October 1953 and October 1960 and the DP between September 1949 and December 1966 (Ibid). Cabinet positions held by semi- optimal parties were few and typically inconsequential, however, and the chancellorship remained with the CDU through December 1969 (Ibid). With trivial exceptions, optimal parties maintained exclusive control over West German parliamentary politics. Support for the occupation was, of course, never unanimous among the West German people or even parties closest to the occupation (the CDU, FDP and, to a lesser extent, the SPD). Both Adenauer and his counterparts in the SPD hit the Western Allies hard for supporting France?s goal to annex the Saar (Glees 2003). The CDU opposed the Oder-Neisse frontier that the Four Powers had agreed upon at Potsdam. The SPD, which 106 opposed rearmament and accession to both the EDC and the Council of Europe, was relatively more interested in developing positive relations with East Germany and the Soviet Union (Glees 2003). They attacked the CDU following the Petersberg Agreements, accusing Adenauer of being in the pocket of the Allies (Haftendorn 2006).35 That said, their opposition to the CDU stemmed more from differences in economic policy and not so much from outright opposition to the occupiers or their policies (Glees 2003). The SPD, already losing members, would abandon the lingering Marxism in their platform and take a somewhat more western orientation on security and economic policy at their 1959 party conference in Bad Godesberg (Merkl 2002, Kvistad 2002, Ibid). Whether or not their moderate opposition towards liberalism and European integration was the true cause of their disappointing electoral performance in the 1950s, the changes they made to their platform suggest that they believed this to be the case. Moreover, the CDU more than any other party toed the occupation line and was never seriously punished for doing so at the ballot box. The CDU-CSU-FDP coalition that ruled from the birth of the republic through December 1966 was steadfastly in favor of European integration, international economic liberalization, and Franco-German cooperation (Glees 2003, Haftendorn 2006, Willis 1962). When the official occupation line was one of disarmament, the CDU made the case for disarmament. When the official 35 This charge borders on the incredible, considering that the Allies and Adenauer shared a rather acrimonious relationship from the start. The future Chancellor had once been barred from local politics after opposing British economic policy in their zone. Adenauer himself had once accused the British of working too closely with the Social Democrats. The UK enabled the SPD to broadcast their propaganda efforts, while the SPD successfully lobbied for a proportional legislature in contrast to the CDU?s desire for single-party districts. Considering their history and the number of concessions Adenauer won from the Western Allies, this reads far more like kneejerk partisan reaction than a true elucidation of the party?s opinions regarding the relationship between the occupiers and the occupied (Glees 2003). 107 line turned to rearmament and military integration with the western powers, the CDU made the case for rearmament and integration. In doing so, Adenauer?s government managed to secure important concessions from the Western Allies, not the least of which was sovereign recognition. In federal elections of September 1953, the CDU mustered 45.2% of the popular vote?14% more than in 1949?while the SDP vote slipped slightly to 28.8% (Schweitzer, et al. 1984, Loewenberg 1967). Four years later, the CDU would capture the first-ever outright majority in a free German parliamentary election (Ibid). The early relative success of the party that was most compatible with the goals of the occupying powers indicates acceptance, if not enthusiasm, for the occupiers? vision of West Germany. During the occupation, political extremism was rare but not unheard of. Notwithstanding the assassination of the US-appointed mayor of Aachen by a Nazi ?Werewolf? militiaman prior to Germany?s surrender, there was virtually no residual paramilitary presence loyal to the old regime (MacDonald 2007). Most of the extra- parliamentary activity under the military government consisted of protests and the occasional riot in protest of poor living conditions and changes to agricultural policy. In March 1948, labor leaders held a daylong general strike to protest the lack of food (Ibid). In the more agrarian south, protests against the deflationary monetary policy?which suppressed returns for agricultural goods and increased the real cost of mortgages?grew into full-scale riots. Unlike agrarian unrest in the Weimar regime, however, the government was responsive and eased up on inflation controls (Esposito 1995). Arguably, the most telling?and certainly most conspicuous?demonstration of extra- parliamentary political power was not in protest against the Western Allies, nor did it 108 even take place in the western zones. In September 1948, 300,000 residents of the Eastern Zone took to the streets in front of the Reichstag building to protest the blockade. This was no trivial act, as the Soviet military police had already been cracking down on pro-western political groups. Indeed, approximately two hundred were injured when the police fired into the crowd, and seven were sentenced to twenty-five years hard labor in the USSR (MacDonald 2007). The protest made quite clear the opinions of the German constituency regarding the occupation policies of the Western Allies and the USSR. Otherwise, extra-parliamentary politics were generally peaceful until the rise of radical student groups in the mid-1960s, several years after the occupation had ended (Cromwell 1964, Merkl 2002). Election data, political demonstration and party politics aside, the single best indicator of the marginal security value of occupation may be the FRG?s foreign policy under occupation control. By 1950, Adenauer?s government pledged to support NATO and the EDC, and also sought a separate mutual security pact with the United States, United Kingdom, and France (Plischke 1953). Not six years after their predecessor state unconditionally surrendered, the new government was willing to enter into no fewer than three mutual security pacts with their erstwhile mortal enemies. More specifically, the new regime asked for and received multiple guarantees of security from the occupiers without significant opprobrium from those who freely elected them despite that many (most?) of them had sworn in the name of their country and leader to lay waste to those same foreign powers barely a decade earlier; to the conquerors that cremated Dresden, Hamburg, Pforzheim and nearly 100,000 fellow countrymen; to the first foreign armies to wage war on German soil since the Battle of Spicheren in 1870, a battle fought in the 109 opening stages of the war that gave birth to their nation. Is there any better evidence of the occupation?s security value that the constituency of the occupied state did not regard this arrangement as a betrayal by the new regime? I submit to you that there is not, and that the success of the optimal domestic coalition in the years following?as well as the democratic institutions that said coalition helped build?is largely the result of this fact. Summation This case meets all the conditions for a successful democratization. The Soviet Union presented a clear threat that became a clear foreign threat following the division of Germany. The Western Allies credibly committed to protect the FRG from this threat via the Airlift, the various security institutions they promulgated, and by the permanent commitment of US and UK military to the Continent. The occupying states improved their security value to West Germany by minimizing the threat that their more detrimental policies may have posed. They also undertook specific policies to protect and empower democratic politics in the new regime, including the supervision of a constitution that enabled democratic organizations to protect their own interests. The pro- democratic parties that emerged from the occupation supported occupation policy, and these parties grew ever stronger in the Bundestag. In the earliest days of a new democracy, the Western Allies conducted an occupation that protected and promoted democratic interests and institutions, demonstrating to a new generation of German citizens that a democratic German state could be stable, effective, efficient, and strong. 110 Reconciliation of Competing Hypotheses The case seems to support H1: there were fewer events indicating instability when occupation policy turned more palatable and when the Western Allies brought West Germany into their security perimeter. What little political instability existed seemed to occur in the earliest years of the occupation, before the Allies ended their policies of demontage and fully renounced the desire to pastoralize Germany. There is no clear-cut case in which the Allies aligned with any clear and present foreign threats. While the case does lend empirical support to the Marginal Value Model, it also resonates with a few of the implications of the rival hypotheses. Thorough defeat and demilitarization reduced the likelihood that authoritarian groups could challenge the occupation or its preferred domestic coalition via extra-parliamentary channels. Conquest and political re-education delegitimized the previous regime and eliminated any appeal Nazism would have among future generations. The design of the Basic Law and the Electoral Law of 1956 marginalized fringe movements and empowered the ruling coalition to take action against threats to the new system. The geography of the occupation ensured that political actors who opposed democratization were excluded from the new regime. Does this mean that the Marginal Value Model fails to provide analytical leverage? No, because many of the more compelling bits and pieces of the rival hypotheses can in fact be subsumed and improved upon by the Marginal Value Model. Demilitarization contributes to the goal of increasing the marginal security value of occupation. What this thesis ignores, however, is the importance of a policy of remilitarization that contributes to the territorial integrity of the state and the security of the citizens within. Democracies succeed in occupied territories when the occupation 111 increases the security of its citizens and reinforces the political strength of the optimal domestic coalition. Therefore, political re-education and institutional design, when undertaken with the goal of protecting and reinforcing pro-democratic constituencies and coalitions, is a necessary condition for early democratic consolidation. In these cases, it critical that the occupation force is active in the institutional redesign, and perhaps even takes a more active role in the education system. If better institutional design and political re-education improve the likelihood of democratic consolidation, then an occupying force interested in building a stable democracy should involve themselves in these questions. Finally, an occupation force may, under certain conditions, improve the chances of democratization by excising territories that may prove detrimental to its aims. This appears to be the case for the Federal Republic, but not for the reasons offered in the political geography thesis. The value of the division was less related to the political economy of the eastern zone than it was to the exclusion of an occupier that had no interest in implementing a democratic state. By forming a separate polity, the western powers increased the marginal security value of occupation in two distinct ways. First, they removed from the occupation a state whose policies were materially detrimental to the domestic constituency. Second, this allowed the US, UK and France to position themselves as the primary line of defense against this threat. This is important, as excision may not always be the answer in other cases. If the primary concern was less about security and more about political culture, the act of division may itself be perceived as threatening. In the case of Germany, this means that the exclusion of the East would not have been propitious if it had not also been an instrument of Soviet intimidation and a threat in its own right to the security of those in the Western zones and, especially, to 112 those in West Berlin. Division is only favorable to democratization if it improves the security of the democratic regime. Economic Drivers As in the case of all studies of democratization, one should not ignore the importance of economic development and growth. The Wirtschaftwunder?which commenced sometime between 1949 and the early 1950s?put the Federal Republic?s new regime in an advantageous position. While the optimal coalition may have benefited from a positive marginal security value of occupation, one might also surmise that they benefitted substantially from the massive economic growth that the Federal Republic enjoyed through the 1950s and 1960s. After all, Germany came out of the war in an unusually strong economic position, at least in terms of capital stocks. Estimates indicate that the war erased 13% of investment capital in (the eventual territory of) West Germany. This exceeded France?s losses (8%) but fell short of the 25% loss suffered by the USSR. By some estimates, German productive capacity was as high in 1945 as it was in 1940 (Reichlin 1995). Coal production jumped fivefold between the surrender and the end of 1945 (Glees 2003). By 1947, German roads had made a full recovery (Willis 1962). But positive indicators in the commanding heights of industry did not instantly translate to a better standard of living or even aggregate growth. War-driven inflation was somewhat to blame?the Reichsmark lost 80% of its value against the US dollar between 1938 and 1945 (Glees 2003). The RM continued to lose value after the war?dropping another 50% in 1946 for a total trop of 90% from 1938 levels?largely due to the Soviet 113 Union?s printing of currency to bankroll occupation costs (MacDonald 2007). Cheap money did nothing to boost employment, which topped out at 76% that year (Killick 1997), nor industrial production, which equaled 40% of 1936 levels and 28% of 1938 levels (Wolf 1995, Ibid). Assuming a liquid global economy and a more productive German industry, cheap prices would have boosted demand for German products. That notwithstanding, even if Germany was producing at a higher level in 1946-7, the RM was hardly convertible and gold reserves across Western Europe were fleeing the continent to cover capital account imbalances with the United States (Hitchcock 1998, Lindlar and Holtfrerich 2002). Perhaps the most pressing danger was the destruction of Germany?s agricultural capacity and the logistical, infrastructural and distributional issues that kept food from reaching the kitchen table. As Germany demobilized, grain production remained stagnant (Killick 1997). The caloric intake of the average German citizen hovered between 1,000 and 1,500 kcal/day during 1946 (MacDonald 2007). The situation deteriorated enough for General Clay to warn of a pending ?nutritional disaster,? that might not only generate a famine but also fuel anti-occupation sentiment and political extremism (MacDonald 2007). That famine never arrived. Instead, the years of 1947 and 1948 mark the economic turning point for West Germany. Employment in 1947 jumped to 85% and caloric intake rose to 1,800 kcal/day (Killick 1997). Employment soared to 95% in 1948, indicating that the German labor glut had largely been exhausted. Inflation stabilized after 1948, as did Germany?s capital account balance (Ibid). Industrial production increased to 51% of 1938 levels?an increase of 82% since 1946 (Wolf 1995). The savings rate averaged a stunning 19% during this period (Eichengreen 1995), a fact that?s 114 less surprising when one considers that consumer markets had yet to emerge and that the central bank had reigned in inflation. The stellar employment statistics of 1948 turned out to be an anomaly, as unemployment hovered around 8-10% from 1949 to 1952 (Killick 1997). Nevertheless, 1949 marked the start of the West German Export boom, and thus the commencement of the West German economic miracle. Exports doubled that year, and would grow 23.5% year-over-year through 1957 (Mee, Jr. 1984, Lindlar and Holtfrerich 2002). GDP would grow at a 4.2% clip through the 1990s. Coal and steel production grew by leaps and bounds (Ibid). Industrial production reached 72% of 1938 levels, a 157% increase since the end of the war. By 1950, industrial production had essentially returned to prewar levels. Five years later, production was at 167% of prewar levels (Wolf 1995). As the economy boomed, Germans saved more and more, with the savings rate averaging 27% from 1950-1962 (Eichengreen 1995). That this recovery took place without the country going into debt, and without the high rate of savings putting a damper on aggregate demand, would appear miraculous at first glance. But perhaps the ?miracle? label is misleading. If one truly considers the economic boom a miracle, then one need not ascribe agency to the improvement in economic indicators. If one considers economic growth as purely an independent variable, them one might assume that the successful democratization of the FRG is unrelated to non-economic variables. It is a tempting view to take, as it is a parsimonious explanation with empirical and theoretical support. This approach is wrong. 115 In reality, the economic welfare of any polity is dependent on good policy, and in these cases growth and living standards require substantial inputs just to restore the economy to its prewar state. In the case of West Germany, the occupation could have short-circuited the Wirtschaftwunder in any number of ways?and it almost did. The attitude of the Western Allies (specifically the United States) on German economic matters passed through three phases: 1) deindustrialization of the German state; 2) reconstruction of the western zones as part of a broader plan to help rebuild Western Europe; and 3) integration of the Federal Republic into regional and global trading blocs. Had the occupation continued to pursue the first option, there may have been no economic miracle. The stage was set when the US forced the UK to abandon the imperial preference system as a condition of Lend-Lease. This paved the way for Commonwealth states to participate in the bilateral agreements that would form GATT and reduce transatlantic tariffs by one half. (Irwin 1995, Frieden 2007). Since open borders are trivial without stable, fungible and flexible currency, the US and UK organized a fixed-rate monetary regime decoupled from specie and built the International Monetary fund to ensure convertibility. Since openness and convertibility are trivial without products to buy and sell, they seeded the International Bank for Reconstruction and Development. The delegates at Bretton Woods intentionally reimagined the global economy that was dynamic but not volatile?one less likely to swamp fragile democracies like the Weimar Republic (Frieden 2007). As impressive as this wave of economic liberalization was, these institutions were of little value to occupied Germany pending an increase their production capacity, itself endangered by the policy of dismantlement. Reparations aside, by the end of 1946, the US and UK were 116 already convinced of the value of a revived, liberal democratic Germany. In an April 1947 paper, then-delegate to the UN General Assembly John Foster Dulles (1947) called for an end to industrial reparations (already underway), combined with monetary policy reform to revalue the German mark and direct transfers of dollar-denominated assets. Dulles? recommendations were indicative of future policy. Foreign assistance ramped up substantially in 1947 via the UN Relief and Recovery Agency, government aid and grants, foreign loans and credits totaling $10 billion. As for monetary policy, the Western Allies reformed the German currency by phasing out the Reichsmark for the new Deutschmark, deflating the money supply by 93.6%, and by shepherding the fusion of the Land banks into a central ?Bank of the German Laender? (Willis 1962, Glees 2003, MacDonald 2007, Dunbabin 2008, Merkl 1963, Mee, Jr. 1984, Haftendorn 2006, Hillenbrand 1980, Merritt 1976). Substantial as these measures were, they simply were not substantial enough to balance the region?s capital accounts. Belligerents had exhausted much of their gold and dollar reserves financing the war and spent what was left on American goods or wartime debt service (Killick 1997). The IMF existed to solve balance of payments issues, but the imbalance was simply too large for the Bretton Woods institutions to counter. Between 1947 and 1953, the IMF loaned $753 million?just over $100 million per year?to European borrowers (Eichengreen 1995). Meanwhile, in 1947 alone, Western European central banks exhausted $2.5 billion worth of gold reserves amounting to one-third of the total supply (Hitchcock 1998). This trend, clearly unsustainable, prompted American policymakers to propose a far more valuable package for Europe. Enter Secretary Marshall. 117 Fearing that economic destitution could fuel communist sentiment on the free side of the Iron Curtain, Secretary of State Marshall proposed what would become a $12-13 billion package known as the European Recovery Plan, or simply the Marshall Plan (Glees 2003, Mee, Jr. 1984, Willis 1962, Haftendorn 2006, Killick 1997, Dunbabin 2008, Windsor 1969, Reichlin 1995). The Marshall Plan would flood Western Europe with surplus military goods (useful for agricultural and industrial production), credits for infrastructural development,36 and debt forgiveness credits. In exchange, signatories promised to balance their budgets, rein in inflation and make their currencies convertible (Hitchcock 1998).37 Despite the large sum, there were some in the US administration that feared it would be too little too late. President Truman successfully requested over $600 million in aid to France, Italy and Austria before the ERP disbursements came through (Hitchcock 1998, Killick 1997, Willis 1962). The UK and US ramped up their expenditures in Germany as well, spending $1 billion annually through 1950 (Lindlar and Holtfrerich 2002). These credits helped balance the current accounts of the Western European economies. Aid to the Trizone/FRG helped transform a $1 billion current accounts deficit in 1948 into a $144 million surplus in 1951 (Killick 1997). The aid also prevented states from abandoning their recent commitment to lower tariffs and currency convertibility as they did after WWI. As the Trizone became the German Federal Republic and transitioned from a military-controlled territory to an occupied, semi-sovereign state, the Western Allies 36 Originally the province of the IBRD, the World Bank?like the IMF?lacked the resources necessary to make a significant impact. 37 States behind the Iron Curtain were not excluded, but the Soviet Union rejected the ERP and discouraged those in its sphere of influence from joining (Killick 1997, Dunbabin 2008). 118 aimed to make the FRG a full member of the new global economic institutions that they themselves had crafted. During a four-week span in October and November 1949, the occupation powers invited the FRG to join the GATT, the Organization for European Economic Cooperation (the board which oversaw the distribution of Marshall Aid) and the ill-fated International Trade Organization (Glees 2003, Haftendorn 2006, Plischke 1953). The FRG lowered its average tariff barrier to 15%, down from 48% in 1931 and lower than it had been since 1925 (Irwin 1995). In 1950, with Western Europe opening up and capital becoming more liquid, West Germany?s trade volume expanded by 75%; it doubled the following year. (Mee, Jr. 1984). Not too long after the occupation ended, the Federal Republic and France joined Italy, Belgium, Holland and Luxembourg in signing the Treaty of Rome, the substance of which founded the European Economic Community?an idea that grew out of discussions between Adenauer and the French Government in 1950 (Glees 2003, Willis 1962, Mitchell 2005, Lindlar and Holtfrerich 2002, Haftendorn 2006). It is clear: the West German Economic Miracle is less an act of God, and more a conscious design by the occupation forces to rebuild the regional economy in a way that would specifically protect West German (and Western European) democracy. By doing so, the occupiers reduced the possibility that an economic tsunami of the style that weakened the Weimar Regime in 1923?and crushed it in 1933?would occur. Moreover, these policies did more than boost indicators of growth and standard of living. They also signaled a credible commitment by the Western Allies to the welfare of occupied West Germany, while directly improving the lot of the most valuable constituencies in a fledgling democracy: capitalists and middle class workers. During the 119 Weimar Regime, trade openness had trended in the opposite direction. States worldwide were raising tariff barriers for the first time since the mid-nineteenth century, and the Allies themselves raised a punitive tariff wall around Germany. Trade barriers, along with reparations, hobbled Germany?s capacity to balance its current accounts and forced the government to meet its fiscal requirements by taking on debt. This prompted inflation and, eventually, the hyperinflation of 1923-1924, which alienated owners of capital from the democratic regime, decimated the ranks of the working middle class, and empowered the reactionary landed constituencies. In contrast, the post-WWII occupation ensured a stable monetary base upon which working households and capital concerns could rebuild. New trading regimes ensured that barriers would remain minimal. Capital infusions encouraged governments to keep these barriers low. These reforms helped the West German economy graduate from surviving to thriving, and in doing so reinforced the constituencies most favorable to the consolidation of democracy in the German Federal Republic. Summation of Findings and Conclusion This case is confirmatory of the hypothesis. The model expects democracy to succeed in occupied territories where and when the occupiers establish a positive marginal value of occupation. The conditions for this were favorable. The Soviet Union established itself as a threat to the Trizone/FRG in various ways. They communicated these threats implicitly, through political oppression of moderate parties in their zone and in the Eastern European states they had infiltrated, and explicitly via direct demands to 120 the Western occupiers regarding the economic and political unification of occupied zones. They established threat credibility via the yearlong Berlin Blockade. Soviet capability to seriously threaten the survival and prosperity of those in the western zones was never in question?the Soviets demobilized their war machine at a slower pace than the western powers (Dunbabin 2008, Mee, Jr. 1984) and retained motorized and armored divisions in the eastern zone and in their Eastern European satellites (Adenauer 1984c, Hitchcock 1998), to say nothing of their atomic capability after 1949.38 The Western Allies, including the United States, United Kingdom and France, positioned themselves as credible allies against a Soviet threat. After rolling back the most threatening aspects of the occupation itself (and thus reducing the likelihood that the occupiers would continue to be enemies rather than become allies), the Western Allies built a network of mutual security agreements to protect Western Europe against the Soviet threat, eventually elevating the FRG to full partnership in NATO. During the Berlin Airlift, the US and UK proved their resolve. By halting and reversing the demobilization process, ramping up the number of American divisions on the Continent and around the world, the US and its allies signaled ample capability. The occupiers directly enhanced the political position of pro-democracy political elements in the FRG via their suppression of anti- democratic politics during under the military government, and by their involvement in the promulgation of the Basic Law. The correlation is obvious?the optimal domestic coalition has performed exceptionally well in every federal election since the ratification of the Basic Law, 38 In fact, the Soviet Union, despite their first successful nuclear test in 1949 (Haftendorn 2006, Dunbabin 2008), probably had no deliverable nuclear weapons as late as 1953 (Merkl 2002). That said, most citizens of the Federal Republic had no way of knowing how underdeveloped the Soviet nuclear capability was. 121 winning both popular sentiment and formal power in the Bundestag. Radical, violent anti- democratic political activity was sporadic throughout the period of the occupation and afterward. But there are also signs of a direct causal link between the behavior of the occupying forces and support for the optimal domestic coalition. Political parties and affiliated organizations that explicitly renounced the occupation and the government that supported it faired poorly in elections and never mustered a significant economic or extra-parliamentary challenge to the ruling coalition. The Christian Democratic Union (along with their Bavarian affiliate, the Christian Social Union) was, for all intents and purposes, the preferred party of the occupation powers. Their success during the occupation and a full decade thereafter indicates popular support for their relationship with the occupiers. The Social Democrats? opposition to the CDU?s western orientation correlated with a decline in their domestic support, and their electoral failures prompted an excision of planks from their party platform that were incompatible with the western security and economic order. The pro-Western demonstrations in Berlin during the Blockade/Airlift indicated a strong sentiment in favor of the Western occupation style and its concomitant adhesion to democratic institutions and free association. The hundreds of thousands of refugees that fled East Berlin for the west prior to the erection of the Berlin Wall voted with their feet for the western system. That the Basic Law specifically allowed for the transfer of sovereign powers regarding security to international organizations, and that they did transfer those powers to the western interstate security system, is itself a confirmation of the perceived link in Germany between democracy and the security of their state, both of which depended on the enlightened self-interest the occupying powers. 122 Finally, while there are a handful of theses that explain one or two of the differences between the failure of Weimar Germany and the success of the German Federal Republic, no single explanation provides as much empirical leverage as the Marginal Value Model. The Marginal Value Model explains the relationship between threat, commitment, democracy and economics, and subsumes the competing theories presented in this chapter. It does so while presenting hypotheses about democratic outcomes in occupied territories that rely on a priori and not post hoc observations (and nevertheless performs well with object post hoc observations as well). The case of the German Federal Republic is a strong test of the model, one that the model passes. 123 Chapter 4: Japan, 1945-1972 Introduction Japan acquainted itself with democracy in 1947, courtesy of the same country the Japanese had ?suddenly and deliberately attacked? barely five years prior. The occupation itself began on September 2, 1945 and ended on April 28, 1952 (Wells 1948, Braibanti 1950, Morley 1965, Williams 1968, Passin 1990, Dower 1999, Tsuzuki 2000, Moore and Robinson 2002). The relevant observation period for this case thus extends from 1945 to 1972. While technically the Yalta Agreements called for four power (US, UK, Nationalist China and the Soviet Union) control of Japan, the United States administered the only local occupation authority and thus retained total operational control under the essentially unilateral leadership of General Douglas MacArthur (Tsuzuki 2000, Dunbabin 2008). Cohesion, therefore, was not an issue. As noted in the following sections, democracy promotion in Japan became official?albeit classified? policy in 1946 and remained the steadfast, public position of the United States from 1947 through the end of the occupation. As with the Federal Republic of Germany, Japan grew into a stable democracy in which pro-democracy parties (in this case, one pro-democracy party in particular) dominate the political landscape.39 While marginal but non-trivial anti-democratic 39 The optimal winning coalition in this case includes a cluster of center-right and centrist parties, including the Japanese Progressive Party (JPP) that evolved into the Democratic Party (DP), and the Japanese Liberal Party (JLP) that, in 1955, merged with the Democratic Party to form the Liberal Democratic Party (LDP). Each of these parties were/are typical center and center-right parties, often with indistinguishable, pro- 124 elements competed with the optimal winning coalition throughout the observation period, they were never a serious threat to seize the reins of government.40 I will begin by summarizing a few of the Japan-specific explanations for the successful democratization that this case study documents. He will continue to test their validity against the historical record, following this with a test of the marginal security model against the same data. A successful test would show that the electoral success of parties in the optimal winning coalition and domestic stability follow the provision of security and welfare goods by the US occupation government. It would also show that anti-democratic parties perform best when the occupier fails to provide such goods, and that this failure should be followed by riots, assassinations and other political violence. It should also demonstrate that positive democracy platforms lacking a redistribution program or an ideology rooted in class terms (Borton 1948b, N. G. Kim 1997, Crespo 1995, Morley 1967, Tsuzuki 2000). A left-of-center party, the Cooperative Party (CP), evolved into the Reform Party (RP. The RP later merged into the as well LDP (H 1949, Richardson 1997, Morley 1967). One left wing party consistently operates within the criteria of the optimal winning coalition: the Democratic-Socialist Party (DSP), a right-wing splinter from the Japanese Socialist Party (JSP), was the closest thing Japan had to a traditional social democratic faction (Morley 1967, Destler, et al. 1976). In certain years, when not dominated by its left wing, the JSP qualifies as a pro-democracy, loyal party. This is true during the elections of 1947, ?49, ?58, ?60 and ?63. At other times, the JSP?s line is largely indistinguishable from the Japanese Communist Party (Auer 1990, McNelly 1965, Borton 1948b, Dionisopoulos 1957, Morley 1967). 40 The set of semi-optimal parties in Japan during the observation is small, often including only the Japanese Communist Party (JCP). Unlike other Communist Parties in other cases, this party retained a Popular Front strategy that prescribed coalition building with other left-wing parties. They were not committed to violent overthrow of the government, but they were not committed to the preservation of democracy, either. Generally, the JCP opposed occupation policy, contributing to its denunciation and purgation by MacArthur in 1950 (Dower 1999; Caprio and Sugita 2007). This study treats the JSP as a semi-optimal party during the years in which the extreme left wing is dominant?the elections of 1952, ?53, ?55 and ?67 (Auer 1990, McNelly 1965, Borton 1948b, Dionisopoulos 1957, Morley 1967). The set of sub-optimal political parties is null, as is the set of right-wing semi-optimal parties. There are no examples of publicly anti-democratic, violently extreme, ultranationalist or racialist parties in Japan between 1946 and 1966 that are of any consequence in electoral or parliamentary politics. 125 trends among variables relevant to rival hypotheses are causally subsequent?not prior to or independent of positive behavior on the part of the occupier. Having done so, I will demonstrate that the marginal security model provides the best explanation for the consolidation of Japanese democracy. Rival Hypotheses versus Case History Rival Hypothesis: Delegitimation and Demilitarization The delegitimation hypothesis is as applicable to the Japanese case as it is to the West German. Sonja Grimm (2008) finds that external democratizers are more likely to succeed if the target state is not capable of dictating the terms of peace. Clear-cut failure makes elites on the ground open to reform. That the Japanese surrendered without conditions (Moore and Robinson 2002) indicates that they were indeed unable to dictate the terms of peace, and the elites on the ground did very much seem open to reform. And, just as in the case of West Germany, Japan?s abject defeat allowed the United States to totally demilitarize Japan, a necessary step according to Linz?s (1990) and Stepan?s (1991) debellicization hypotheses. Nowhere is this more obvious that in the (in)famous Article IX of the imposed Japanese constitution, in which the new regime permanently renounces the use of force even in the case of self-defense.41 In fewer than two years, the 41 The English translation of Article 9, Chapter II of the Constitution of Japan reads, ?Aspiring sincerely to an international peace based on justice and order, the Japanese people forever renounce war as a sovereign right of the nation and the threat or use of force as means of settling international disputes. In order to accomplish the aim of the preceding paragraph, land, sea, and air forces, as well as other war potential, will never 126 occupiers caused Japan to disband its military, dispersed all paramilitaries,42 and imposed upon the country a constitution that renounced the use of force as a means of international politics (see more about military dismantlement in the ?Threat Environment? section). Among conquered territories that would regain their sovereignty, there are few if any instances of demilitarization equivalent to this one on the historical record. Just as in the case of West Germany, however, this theory ignores the importance of succeeding demilitarization with remilitarization. Sometime between late 1947 and early 1948, occupation policy reversed course from demilitarization and early withdrawal to re-armament and long-term commitment (Eiji 2002, Caprio and Sugita 2007). Japan was no longer a belligerent; they became a potential ally in a developing Cold War. Along with this new image of Japan came a push for rearmament, which to some Japanese was a relief and to others was especially hypocritical (considering that the ink on the new constitution had barely dried). Either way, Japan did heed this call, building a ?self-defense? force that served as both domestic and foreign auxiliary, replete with land, air and sea forces (Tsuzuki 2000). This rearmament would continue long after the occupation ended (Morley 1967). Aiding Japan?s internal balancing efforts was a long- term American commitment. The US retained troops on Japanese territory long after the occupation formally ended, committed their naval forces to the region, and essentially rebuilt Japan?s marine forces (N. G. Kim 1997). The only differences between the be maintained. The right of belligerency of the state will not be recognized.? (Government Printing Bureau 1946) 42 A GHQ order in January 1946 to disband all paramilitaries directly targeted one organization, the Guard Force, a wing of the WWII-era Imperial Guard charged with defending the main islands to the last man. They were not a significant force in postwar Japanese politics (Tsuzuki 2000). 127 German and Japanese examples is that Japan?s demilitarization was more thorough and its remilitarization more swift. Dismantling Japan?s war-making capability and then leaving it to its own devices in the time and place this new regime happened to occupy would have been in no way propitious for democracy. Left exposed to the realpolitik of the Soviet Union and China, without American military assistance, Japan would have had to choose between diverting crucial resources to rebuilding its own self-defense capability at a time when such resources were in short supply, or risk subordination to the Soviet Bloc. Either choice would have put a serious strain on the new democratic regime. An increase in threat without reliable foreign support could have re-invigorated the once-purged militarists and ultranationalists.43 Diversion of economic resources may have delayed Japan?s economic recovery and tested the people?s patience with the capacity of the new regime to deliver goods both public and private. To rebuild Japan as a democracy, the United States had to both tear down and rebuild Japan?s defense capability. American assistance in these matters constituted a costly and credible commitment to the state?s welfare and distributed public security good that the new regime could effectively redistribute. The remilitarization of Japan is an example of the occupation?s production of marginal security. 43 As per the findings of Colaresi and Thompson (2003); see Chapter 1. 128 Source: James W Morley. Japan and Korea: America's Allies in the Pacific. 1965. Walker and Company, New York. Rival Hypothesis: Retention of Institutions Seemingly in direct contradiction to Linz?s demilitarization case is Coyne?s (2008) hypothesis regarding the retention of institutions from the previous regime. Unlike Germany, Coyne is correct that the United States largely retained Japanese institutions during the occupation period. Whenever possible, SCAP governed through existing Japanese institutions rather than attempt to build new ones (Caprio and Sugita 2007). The United States continued to recognize the government that had prosecuted the war and made no attempt to remove it wholesale. The structure of the SCAP constitution mimicked that of the Meiji charter, only gutted and reconstituted with American style republican-democratic organs (Dower 1999). There were benefits to this approach. Particularly, GHQ could immediately begin its state building operation rather than wait ?0 ?100 ?200 ?300 0 100 200 300 19 49 19 50 19 51 19 52 19 53 19 54 19 55 19 56 19 57 19 58 19 59 19 60 19 61 19 62 19 63 Billi on s Th ou sa nd s Figure 4.1: Expansion of the Japanese Armed Forces Maximum Personnel (Left) Spending (Right) 129 until it established a monopoly over the use of force in a power vacuum (Barnet 1983). Coyne argues that retaining Japanese institutions solved the credible commitment problem that imposed regimes typically face by relying on sources of legitimacy that the citizenry already respected. In this case, the main problem with Coyne?s claim is not a mischaracterization of the occupation experience, but rather a lack of elucidation about the processes upon which the hypothesis depends. It is not exactly clear why retention of these institutions should solve the credible commitment problem inherent in occupying a foreign state and imposing a new regime. Presumably, Japanese people would be more likely to obey Japanese orders than they would American orders. This claim is simplistic and implausible for two reasons. First, it assumes that the Japanese people would respect Japanese institutions with full knowledge that said Japanese institutions were hijacked by American power, making American decisions for the benefit of Americans. Second, it may explain why it solves the credible commitment dilemma as concerns the general population, but it does not explain why it does so for elites. In order for any occupation to be successful without resort to all-out repression, the occupying authorities must reign in those classes of actors with the greatest power potential. The ?people? are one of these classes, and securing their assent is indeed helpful. Securing their assent does not guarantee that the predominant holders of economic and physical power?colloquially, guns and money?will be on the side of the occupiers. Moreover, ruling through Japanese institutions does not guarantee their support. By definition, an occupation will have supplanted the privilege that the previous wealthy and powerful classes enjoyed. Also by definition, imposed democratization cannot leave those previous elites intact; to 130 do so would be to preserve the last regime. This is exactly the folly that USSR feared MacArthur was committing, especially in preserving the Chrysanthemum Throne. Soviet High Command was so worried about MacArthur?s unwillingness to level utterly the institutions of the old regime that they compared his behavior to that of George and Clemenceau at Versailles (Barnet 1983). While these fears proved unwarranted, they raise an important point: in the case of imposed democratization, retaining old institutions may (though not necessarily) entail the retention of autocratic institutions.44 It is obvious that retaining institutions of autocratic character would pose a threat to the consolidation of the new regime and, depending on the institution and its nature, negate the regime change all together. Even if the new institutions solve the credibility dilemma, it matters not if they do so by leaving elites with autocratic preferences in place. This is a dilemma of its own: if the choice of institution through which the occupation chooses to govern has little positive effect on the attitudes and preferences of elites, then one cannot assume that it solves any credible commitment dilemma when it comes to them. Since one cannot assume that ruling through Japanese institutions solves any credible commitment dilemmas among the mass population or among the elite classes, then one should question whether this is an aspect of the occupation that had any effect whatsoever. And while I do not find it acceptable to rely on this explanation as a causal feature of the success of the Japanese occupation-cum-democratization, there are kernels of truth 44 MacArthur avoided the mistakes that the Soviets warned against by retaining the Emperor?s office only in name, by retaining only institutions that were democratic in character?such as the diet, by ensuring that those institutions were made more democratic (by expanding suffrage and subordinating the cabinet to it), and by abolishing formal and informal institutions that were antithetical to democracy, such as the peerage, the Home Ministry State Shinto, and absentee landlordism (Government Printing Bureau 1946; Ray 1947; H 1949; Ulmer 1957; Dower 1999; Moore and Robinson 2002). 131 within. Had the United States decided to break down and rebuild anew the institutions of state, as they had in Germany, the total outlay in terms of money, manpower and time might have been much greater. Perhaps the effort required would have been even greater than in Germany, where the United States shared the burden with three other states, and where a long tradition of federalism meant that the devolution of power to local authorities was not an alien concept. If the retention of the Diet did not alleviate credibility costs, it certainly reduced coordination costs. In essence, the only modifications MacArthur made in this area were the suffrage rules under which representatives were elected, the powers the legislature retained, and its power to choose the cabinet and head of government.45 The persistence of the Diet meant that it was one less institution that SCAP had to build from scratch. That the MacArthur Constitution passed as an ?amendment? to the Meiji Constitution is indicative of the lower coordination costs of retaining Japanese institutions (Dower 1999). Finally, American policymakers were convinced that retaining Emperor Hirohito was necessary for ensuring the legitimacy of the new regime (Ibid). This is an untestable supposition, but it is not one worth ignoring. Rival Hypothesis: Prewar Modernization If the retention of Japanese institutions cannot itself explain the success of the occupation, perhaps it was the case that the institutions themselves were already favorable for a democratic overhaul. Related to Coyne?s hypothesis is the proposition that 45 Moore and Robinson 2002; Dower 1999; these changes are, of course, everything that matters. 132 Japan had concluded much of its political, societal and economic modernization before WWII. Therefore, by the time that the United States imposed a democratic regime, Japan was ?ready? for the new type of government. Pempel (1992) places the origins of Japanese democracy in the Meiji Constitution and the rapid modernization of the Tokugawa Era. This claim resonates with Seymour Lipset?s (1959) seminal work, crediting rapid consolidation of Japanese democracy to the nation?s unusually high level (for its income) of education. Lipset argued that an educated populace is more likely to understand the importance of tolerance, eschew extremism, and make rational choices. This was the reason, according to the author, that Japanese democracy consolidated despite (then) trailing the rest of the free world in economic development. The claim that states can be more ready for democracy if they have developed certain formal and informal institutions is persistent and fraught, and it dovetails with many of the consolidation hypotheses from the first chapter. Economically developed democracies are more successful consolidators. Democracies that rule a state with a prior history of democratization are more successful consolidators. Socially egalitarian democracies are more successful consolidators. Each of these branches of the consolidation literature supports the idea that Japan may have been a successful democratizer because it was a developed state with a history of constitutional government. These explanations, however, cannot stand independently. The rise of the Sh?wa military government that ruled Japan in the 1930s and took the empire to war seemed in no way hampered by either the modernity of the society or the existence of the Meiji Constitution. Modernization in and of itself forged paths towards communism and fascism just as it did towards bourgeois democracy, as Barrington Moore (1966) has 133 demonstrated. And while making rational decisions is perhaps an important characteristic of an informed electorate, it is not clear that individual voters must act rationally in order to produce rational electoral outcomes. Two originators of modern democratic theory, Jean-Jacques Rousseau and James Madison, are very clear that individual rationality is not a prerequisite for positive democratic outcomes.46 Nor is rationality always a stumbling block for anti-government extremists.47 These indicators of modernity are important correlates of democratic consolidation, but they do not alone explain success in the case of foreign-imposed democratization. They cannot explain why Japan took the same road as West Germany rather than that of Weimar Germany. To understand why, one needs to look further and deeper, into the origins and prosecution of the occupation that imposed a democratic regime on Japan in the first place. 46 ?The latent causes of faction are thus sown in the nature of man? So strong is this propensity of mankind to fall into mutual animosities, that where no substantial occasion presents itself, the most frivolous and fanciful distinctions have been sufficient to kindle their unfriendly passions and excite their most violent conflicts? A republic, by which I mean a government in which the scheme of representation takes place, opens a different prospect, and promises the cure for which we are seeking.? (Publius 1787); ?The citizen gives his consent to all the laws, including those which are passed in spite of his opposition? The constant will of all the members of the State is the general will; by virtue of it they are citizens and free? When therefore the opinion that is contrary to my own prevails, this proves neither more nor less that I was mistaken, and that what I thought to be the general will was not so?all the qualities of the general will reside in the majority: when they cease to do so, whatever side a man may take, liberty is no longer possible.? (Rousseau 2001, p.277-8). In the quotes above, both authors imply that the structure of representative government can reign in the distortions of disparate opinions. Liberty and the public interest can survive even when individuals are acting passionately instead of rationally so long as majority rule is both preserved and mediated by appropriate institutions. 47 Lichbach (1998) holds that the group dynamics of collective dissent may overcome the collective action dilemma that otherwise should prevent the rational actor from rebelling against the state. Thus, given the right conditions, the rational actor may choose rebellion over consent. 134 Test of Working Hypotheses versus Case History Threat Environment Foreign threats create an opportunity for occupiers to produce marginal value by distributing security goods, thus making consolidated, imposed democracy a possibility. But until 1950, neither the USSR nor the Nationalist or Communist regimes in China posed an existential threat to Japan. Even with the advent of war on the mainland, none of the belligerents communicated any malicious intent toward Japan specifically. Besides, Japan had since abandoned its imperial ambitions, including its possessions on the Korean Peninsula, now embroiled in an internationalized civil war. Furthermore, no foreign power directly targeted Japan in the same way that France, Belgium and Poland antagonized Weimar Germany or that the Soviet Union pressured the Federal Republic. There is no Japanese counterpart in the historical record to the Berlin Blockade. Taken individually and without context, these conditions indicate a placid threat environment as compares to the two German cases. Context, however, is important. The historical and contemporary background of the Japanese case indicates that, while not extreme, the level of threat in this case is non- trivial. Dating back to the mid-1800s, Japan?s primary concern in terms of overseas territories was northeast Asia, particularly Manchuria, Siberia and Korea. Japan?s interest in this region was unsurprising due to its geographical proximity, inevitable due to the Table 4.1: Influences on the Threat Environment in Japan More Threatening ? Historical relationship with Russia ? Korean War Less Threatening ? Archipelagic geography 135 accession of Russia to great power status and the colonization of East Asia by Westerners, and irresistible due to the region?s comparative wealth of strategic resources. Following wars with China and Imperial Russia, Japan brought Korea into its sphere of influence and eventually subjugated the state as a colonial possession (Cha 1999). Japan would later do the same to Manchuria. Moreover, Russo-Japanese neutrality during (most of) WWII reoriented Japan towards Southeast Asian and Pacific expansion. From the late nineteenth century onward, the structure of interstate politics in the region?as well as their relationship with the US?significantly impacted Japanese domestic politics. The Meiji Restoration of 1868 and the Sh?wa regime of the 1930s were, in part, attempts to modernize the organs of state to address and exploit the changes in Asian-Pacific affairs. Isolationism was no longer a reasonable foreign policy; modern warships and Western interest in the Asia-Pacific raised the possibility of a foreign power turning Japan into the next Philippines or East Indies. Geography dictates that the gravest, most immediate threats to Japanese sovereignty would emanate from the northeast mainland. Naturally then, upon Japan?s imperial expansion, this region would also prove to be the most enticing target. Just as the security of Britain largely depends on maintaining a balance of power on the Continent, the security of Japan largely depends on who controls Korea, Manchuria and Siberia, as well as the Chinese and Russian littorals. This is the context in which we must infer Japan?s perception of threat during the American occupation. Japan neutralized Korea in 1910, Manchuria in 1934, and in 1941 signed a neutrality agreement with the USSR, but this security buffer collapsed in a matter of weeks when the Soviet Union re-entered the Pacific theatre (Tsuzuki 2000). The loss of Manchuria (which would return to China as per the Potsdam Agreement), the 136 abrupt end to the USSR?s assurance of non-intervention, and the Soviet occupation of Korea north of the 38th Parallel left Japan exposed to northeast Asia once again.48 This time, however, Japan lacked the option of either offensive or defensive rearmament, as per the American occupation?s disarmament policy (see more below). Perhaps this exposure would have been less stark, if only slightly, had the Soviet Union responded positively to MacArthur?s calls for an early peace treaty in March 1947 (G. F. Kennan 1967, Borton 1948b, Kim 1997), but the USSR rejected the proposal (Ibid, Borton 1948a). The situation would only grow more foreboding thereafter. That Japan, or at least the cabinet, perceived the Soviet Union as the primary threat to national security is a matter of public record. On September 13, 1947, Foreign Minister Ashida presented a memorandum to General Eichelberger, commander of Eighth Army, requesting a long- term American military presence on Japanese soil. The premise behind the communiqu? was a fear that the USSR would threaten Japanese interests should US-Soviet relations deteriorate coupled with a lack of confidence that the United Nations could effectively protect Japan if such were the case (N. G. Kim 1997).49 In other words, despite the lack of a formal alliance binding Japan and the US, the absence of any obvious Soviet designs on Japan, and the fact that the United States was responsible for most of the losses the Japanese sustained during WWII, high-ranking government officials on both sides had already presupposed a security partnership. 48 While Japan was aware that they would have to relinquish their dominions on the mainland when they agreed to the terms of surrender put forth at Potsdam, they managed to hold onto these territories up until the Soviet Union re-entered the Pacific theatre on August 9, 1945 (Spector 2005). 49 Prime Minister Yoshida adopted this line and repeated Ashida?s position publicly in 1949 and 1951 (Ibid). 137 Ultimately, the Cold War did come to the Pacific Rim. It is difficult to ascertain exactly when this occurred, but odds are this transition occurred sometime between August 29, 1949 and June 25, 1950. On August 29, the Soviet Union detonated its first atomic bomb, further weakening Japan?s geographic defensive advantage (Haftendorn 2006, Dunbabin 2008). The US once proved it feasible to incapacitate Japan without attempting a treacherous and unprecedented invasion of the Main Islands by bombing Hiroshima and Nagasaki, and now the Soviet Union might press the same advantage. Two months later, Mao Zedong and the People?s Liberation Army?who honed their skills fighting the Japanese during WWII?expelled Chiang Kai-shek?s Guomindang forces and assumed control over Mainland China (Morley 1967, K.-K. Kim 1995). They soon established favorable relations with the Soviet Union and the Democratic People?s Republic of Korea, which itself had established relations with the USSR in March 1949 (Morley 1967). In February 1950, the PRC and USSR signed a treaty of military alliance (K.-K. Kim 1995). On June 25, 1950, the Cold War in Asia turned hot as the DPRK invaded the Republic of Korea (Morley 1967, Tsuzuki 2000, N. G. Kim 1997, Moore and Robinson 2002, Guthrie-Shimizu 2007). The Korean War eventually grew into an internationalized civil war with US forces fighting for South Korea under the banner of the United Nations, with Soviet materiel and training supplementing the efforts of the North (N. G. Kim 1997, Dower 1999, Moore and Robinson 2002, Morley 1967, Tsuzuki 2000) and, on November 26, 1950, with thousands of Chinese Communist volunteers launching an offensive across the Yalu River (N. G. Kim 1997, Eiji 2002). While geographic circumstance and decades of contestation in the region gave reason for wariness in the late 1940s, the sweeping 138 victories of the People?s Liberation Army in mainland China and the DPRK in the first months of the Korean War, and the nuclearization of politics upon the USSR?s successful test of a nuclear weapon, made it clear that Japan could no longer rely on the sanctity of its island fortress. The war, which endured beyond the conclusion of the American occupation, clarified the threat environment in two ways. First, it represented the crystallization of security relationships in the region, with an autocratic, communist bloc forming on the mainland among the People?s Republic of China and the USSR in support of North Korea, directly opposing a democratic, capitalist bloc revolving around the United States in support of the South. Second, the introduction of war to a peninsula commonly described as a ?dagger aimed at the heart of Japan? (Dower 1999) represented the first serious external instability in the region since the conclusion of WWII. The intensity of this threat is magnified by the American success in demilitarizing Japan early in the occupation. It is this situation that allowed the United States occupation force to produce marginal security for Japan. While SCAP had abandoned the policy of demilitarizing Japan in 1949, five years of enforcing this policy?not to mention the decimation of the Imperial military during the War?left Japan with a limited capability for self-defense (Morley 1967). Until the new regime could rebuild its self-defense capability, Japan was faced with a choice: align with the United States, or go it alone and risk being swamped by an increasingly belligerent Sino-Soviet bloc. Japanese demand for an external balancer was high, and the United States established itself as the most logical, most willing and most generous supplier. 139 Signaling Security Value While the opportunity to produce marginal occupation value by distributing security goods presented itself in the form of a growing threat on the mainland, the occupiers did not take immediate advantage. An early mistake by Truman handed territories to the Soviet Union that Japan had acquired peacefully. To make matters worse, for two years it seemed that the US occupation policy in Japan revolved around its own three Ds: democratize, demilitarize, and depart. These early mistakes were not disastrous, however; policy during this era was more na?ve than injurious, and MacArthur & Co. made up for their deficiencies by improving Japanese security in other ways. Most importantly, as the threat environment deteriorated, US policymakers signaled credibly by completing a policy about-face and enhancing their commitment to Japan. As the environment grew more threatening, the US invested more in Japanese security. By the time the Korean War broke out, the US had successfully demonstrated their long-term commitment to the security of Japan, thus producing marginal value and aiding in the consolidation of the new democratic regime. The breadth of the US commitment to Japan specifically, and to northeast Asia generally, was initially unclear. At Potsdam in July 1945, the US, UK, China and Soviet Union signaled that the occupation would be temporary, ending as soon as Japan relinquished its empire, reformed domestic politics (this did not necessarily mean democratization, though the terms required a government in accordance with the ?will of Table 4.2: Indicators of Occupation Security Value in Japan Positive ? Evasion of reparation demands ? Remilitarization ? Inclusion in US security perimeter ? US-Japan Security Treaty Negative ? Surrender of the Kuriles 140 the people?), liberalized trade and renounced militarism (Borton 1948a, Dower 1999, Williams 1968, Wells 1948). The language of the surrender terms makes no mention of any long-term commitment to Japanese security, and the conditions for ending the occupation specifically imply that the US would not make any such commitment. That the four parties to the Potsdam Declaration renounced any intentions for a permanent occupation, for placing Japan under vassalage, and made no mention of dividing or pastoralizing the country were certainly positive features. But as it stood, Potsdam declared that Japan would be stripped of its military and the security buffer provided by its inland colonial possessions, and then left to the whims of the powers in the North Pacific (perhaps under the nominal protection of the United Nations). Furthermore, despite the Potsdam provision requiring Japan to return only territories acquired through aggression, Truman agreed to hand over the Kuriles (a chain of islands north of Hokkaido) over to the Soviet Union (Gallicchio 1991). Japan did not expect that the US would cede the Kuriles, which had been acquired by treaty with Imperial Russia, nor did the US officer corps or the Soviet Union (Gallicchio 1991). Nevertheless, the Soviet Union would take possession of the Kuriles, and the US refused to correct their mistake and support the retrocession of the territory until 1956 (Ibid). On the other hand, the US worked hard to prevent other former belligerents, including the Soviet Union and China as well as England, the Netherlands, and newly sovereign states such as India and the Philippines, from asserting their influence in Japan. The Soviet Union pressed for an occupation zone in Hokkaido, while Great Britain and her dominions pushed for a multilateral council in Tokyo (Levi 1946). However, following the surrender, the United States was the sole belligerent with a military 141 presence in Japan. By virtue of their advantageous strategic position, Washington enjoyed de facto veto power over occupation policy. They exercised this veto to keep Soviet boots off of Japanese soil (Dunbabin 2008) and to significantly enervate demands for multilateral control. Instead of a quadripartite occupation authority in Tokyo?as proposed by Molotov and implemented in Germany?Washington countered by creating the Far East Advisory Commission (FEAC), a body with no formal authority over policy (Borton 1966). When the Soviets refused to join, the US ditched the FEAC proposal and secured an agreement on the formation of the Far East Commission (FEC). The function of the FEC, comprising of the US, UK, USSR, China, France, Netherlands, Canada, Australia, New Zealand, India, and the Philippines, was to review and approve (or veto) the decisions of the Supreme Commander (Ibid, Borton 1948a, Tsuzuki 2000, Levi 1946, Wells 1948). Their actual power to effect policy was limited, however, due to the aforementioned American monopoly on military force over the four main islands and Okinawa, as well as a formal veto the US enjoyed on FEC decisions (Borton 1966).50 Moreover, the terms of surrender?which were approved by the Soviet Union and China as well as Japan and the US?placed full executive authority in the person of the Supreme Commander of Allied Powers (Borton 1948a, Moore and Robinson 2002, Wells 1948).51 US General Douglas MacArthur, who occupied the office between October 1945 50 In one example, the US employed this veto to block a resolution calling for the submission of the SCAP-imposed Japanese constitution for review to the FEC. The FEC later adopted a resolution calling for a constitution embodying the same principles that SCAP?s constitution had already authored and that the Diet had already agreed to consider (Ibid). Clearly an effort to avoid the appearance of irrelevance, this transparent move only serves to underline the FEC?s impotence. 51 Emperor Hirohito reinforced this arrangement in Imperial Ordinance 311. Issued on July 15, 1946, the ordinance directed his subjects to regard SCAP directives as equivalent 142 and April 1951 (Tsuzuki 2000), had no intention of relinquishing this authority and was generally successful in having his way. Along these lines, the US dragged its heels on issuing reparations payments to former belligerents, instead directing its energies to rebuilding the Japanese economy (Reday 1949).52 At the very least, this prevented the mainland powers from taking a retributive course of action (as the USSR did in Eastern Europe) and from subordinating Japan?s wellbeing to their own?a likely outcome, considering the decades of antagonism in this security triangle. The value of this strategy would become even more obvious as the Cold War developed, especially to the nascent democratic elements under the new, imposed, democratic regime. It was not until 1948 that the United States signaled a long-term commitment to Japanese security. How long the United States would remain in Japan, and continue to protect Japanese territorial integrity, remained an open question. In accordance with Potsdam, MacArthur called for a peace treaty and subsequent withdrawal once his forces had successfully democratized and demilitarized the state (Kennan 1967, Borton 1948a, to Japanese law, even when the Emperor or Diet did not formally back the directives with their own authority (Braibanti 1948). 52 Comparative US reparation policy toward German and Japan after WWII presents an interesting study. The US officially pursued reparations in both cases, but the US was far less committed to reparation deliveries from Japan than from West Germany. This is directly related to the contrast between the free hand that the US effectively enjoyed in Japan and the necessity of coordinating German occupation policy with France and the Soviet Union, stats that were fervently pro-reparation and -dismantlement. The Potsdam Declaration, which dictated the terms of surrender, stipulated reparations including factory dismantlement, but the US contingent was wary about this policy as they were in Germany. The difference in this case was that the members of the Far East Commission, the international body in charge of the Japanese occupation (as stipulated in a 12-way agreement among former belligerents including the USSR and China), had no mechanism for effecting policy on the ground. The FEC was unable to obtain reparations beyond machinery and abandoned works; rarely if ever were active factories dismantled (Reday 1949, H 1949, J. B. Cohen 1948, Passin 1990). Nor could the FEC stop the US from unilaterally abandoning reparations deliveries in 1949-50 (Kennan 1967, Reday 1949, Morley 1967). 143 Borton 1948b, N. G. Kim 1997). In Washington, attitudes towards the long-term strategic value of Japan were beginning to change. George Kennan (1967) cautioned against an early exit, as leaving Japan before it was economically rehabilitated and remilitarized would leave it susceptible to bullying by the Soviet Union and, perhaps, even a communist takeover (Marxist elements were still very active in Japan at this time). In late 1948, the US reversed the policy of demilitarization and began encouraging Japan to rebuild its capacity for self-defense (Caprio and Sugita 2007). Kennan recommended a long-term troop presence on the islands (G. F. Kennan 1967), while planners in Washington began calling for the reconstruction of a Japanese paramilitary force that would maintain order domestically and deter threats internationally (Williams 1968, Barnet 1983). The strongest American commitments to Japanese security were delivered during the Korean War. Indeed, American involvement in the Korean War itself was the strongest possible signal that the US considered the region, which includes Japan, a strategic priority. This, however, represented a total reversal on American policy towards Korea. Prior to the conflict, the priority was extrication from the costly occupation of Korea as quickly as possible (N. G. Kim 1997). The US fulfilled this policy goal upon their departure from Korea in mid-1949 (N. G. Kim 1997, Morley 1967, Matray 1985, McGlothlen 1989, Oh 2002, Oliver 1999). Secretary of State Acheson went so far as to declare the Korean peninsula beyond the American security perimeter (Morley 1967, N. G. Kim 1997, McGlothlen 1989). This message, while unfavorable to South Korea, was favorable to Japan: the US may not be interested in Korean security, but by excluding Korea Acheson implicitly included Japan (Morley 1967, N. G. Kim 1997). The UN- 144 sponsored, US-led ?police action? represented the first material commitment by the occupiers to countering the expansion of Japan?s traditional security competitors in the region, and the strongest proof of the value of the American occupation at the time. The US ramped up its contribution to Japanese security as the Korean War dragged on. In September 1951, negotiations concluded regarding the end of the occupation of Japan and the status of the post-occupation relationship (Morley 1967, Eiji 2002). By the terms of the Treaty of San Francisco and the US-Japan Security Treaty, the US retained the right to intervene to protect Japan from both external and external threats to the new regime. Also in September 1951, the US reinforced its commitment to the greater Asia-Pacific region when it concluded the ANZUS mutual security pact with Australia and New Zealand (Eiji 2002). Domestic elements in Japan offered only divided and conditional support to the treaties. The joint responsibilities of the mutual security pact were asymmetrical, requiring Japan to accept American assistance without an iron clad commitment to come to Japan?s defense in the event of a crisis. Dulles? command to consult with, rather than dictate to, Japan notwithstanding, the final treaties reflected the leverage that the Americans had by virtue of their predominant force and Japan?s desire for both independence and protection (Destler, et al. 1976, Aruga 2001). Nevertheless, Prime Minister Yoshida?s fear of a quick American departure (once Korean truce talks commenced in July 1951) belies the cabinet?s support for a long-term American commitment of some type (N. G. Kim 1997).53 While the earliest signals by the United States and its allies were of ambiguous nature, they were not dramatically positive or negative. Without perfect foresight about 53 Yoshida had publicly called for a post-occupation US presence in Japan as early as May 1949 (Ibid). 145 how the Cold War would unfold?or that it would occur at all?the initial absence of a long-term security commitment to Japan was more na?ve than malicious. The surrender of the Kuriles was perplexing and a sore point in US-Japan relations for over a decade. That said, since postwar Japan did not lack for warm water ports and had little use for them, their cession was probably more beneficial to the Soviets than it was detrimental to the Japanese. The US would make up for that by serving as the primary guarantor of Japanese maritime security, indicating that the Kurile cession was more of a nuisance to the US than to Japan. Most importantly, as the environment grew more threatening, the US responded by ramping up its commitment. By late 1948 it was clear that the United States held a significant stake in the survival and independence of Japan, and in mid-1950 the US resolved to affirm that commitment with money, prestige and blood. By assuming these costs, the United States credibly signaled its future commitment to the welfare of Japan, thereby producing marginal occupation value. Facilitating Regime Consolidation In the earliest years, when the American commitment to Japan?s security was anything but concrete, MacArthur and General Headquarters directed their efforts to improving domestic welfare and marginalizing antidemocratic coalitions. The purge of ultranationalists and militarists from government helped narrow the range of acceptable political contestation, while the end of controls on political expression and the introduction of full adult suffrage helped to grow the center. It also helped to grow the far left, necessitating another purge that marginalized another possible challenge to liberal democratic government. The United States poured billions of dollars into Japan during 146 the occupation period, although not soon enough for the millions of Japanese who suffered from malnutrition during the first year of the occupation. Supreme Command tried the highest-ranking officials in the old regime for war crimes, further thinning out the ranks of those who might challenge the new government. Crucially, one actor left out of these proceedings was Emperor Hirohito. It was in this context that MacArthur and Supreme Command imposed a representative democratic constitution that survives, un- amended, as the basic law of Japan to this day. Among many parallels that the post-WWII German and Japanese cases share is a direct effort on the part of the occupying powers to purge the elements of the previous regime. On September 10, 1945, the US military government imposed censorship of all Japanese media (Dower 1999, Caprio and Sugita 2007). The purpose of this directive was to ensure that disruptive elements?this meant mostly right-wing extremists from the old regime but, as the occupation continued, increasingly included left-wing elements?could not regenerate domestic support for the previous regime. The next day, the occupation authorities issued the first warrants for the arrest of war criminals, which it would follow up with the formal establishment of the International Military Tribunal for the Far East on January 19, 1946 (Dower 1999, Tsuzuki 2000, Morley 1967). These trials ran from May 1946 through November 1948 and resulted in over 4,000 convictions and 984 executions (Ibid, Moore and Robinson 2002, Morley 1967). On October 22, SCAP prohibited the Table 4.3: Evidence of Facilitating Regime Consolidation in Japan Constructive ? Preservation of the Emperor ? Militarist purge ? Expansion of suffrage ? Imposed constitution ? Land redistribution ? Direct aid Destructive ? Early failures to manage inflation, starvation ? De-purge 147 promotion of militarism and ultra-nationalism in textbooks, and furthermore ordered the dismissal of teachers with similar views (Nozaki 2007). On December 15, the occupiers banned the elements of the Shinto religion that promoted Japanese racial superiority, claims for eventual world domination, and those that proclaimed the emperor?s divinity (Nozaki 2007, Moore and Robinson 2002, Dower 1999, Ray 1947). In January 1946, President Truman issued the ?Purge Directives,? disqualifying over 200,000 officials from participation in the cabinet, Diet or civil service due to their views or their connections to the previous military government (Dower 1999, Tsuzuki 2000, Moore and Robinson 2002, Morley 1967). That said, the purge of militarists and ultranationalists in Japan was not quite as zealous as the Denazification policy in Germany. In contrast, MacArthur?s men were far more meticulous in crafting the institutions that would define the new regime. The occupation authorities opened up the realm of political expression and competition with the ?civil liberties? directive of October 4, 1945, which re-legalized political parties, freed 3,000 (mostly left-wing) political prisoners, and repealed all laws restricting freedom of thought, religion, assembly, speech and criticism of the government (Tsuzuki 2000, Dower 1999, Moore and Robinson 2002). A week later, MacArthur met with the Japanese cabinet to discuss his plans to reform the whole structure of Japanese Government. (Auer 1990, Dower 1999, Moore and Robinson 2002). Initially, SCAP left it to the Japanese to redesign the Meiji Constitution in according with the Potsdam Declaration provision for self-government. However, by February 1946, MacArthur had lost faith in the ability of the existing Japanese ministers to promulgate a document that would represent a true break with the past (Borton 1966, Dower 1999). Between 148 February 2 and 11, a secret SCAP committee led by General Whitney wrote the document that would become the draft of Japan?s postwar constitution (Gibney 1996, Dower 1999, Moore and Robinson 2002). The final Japanese version, which was promulgated in October and went into effect on May 3, 1947, differed in translation very little from the draft authored by the Whitney committee (Ibid, Borton 1948b, Auer 1990, Wells 1948, Morley 1967, Williams 1968, Nozaki 2007, Okudaira 1990). Meanwhile, the first postwar elections took place in April 1946?with blacklisted politicians purged from the ballot and suffrage opened to all adults without regard to gender, class or property ownership (Borton 1948b, Dower 1999, Moore and Robinson 2002). By expanding suffrage while blacklisting known antidemocratic elements, MacArthur and the occupation officials directly manipulated the domestic political environment to privilege and empower pro-democratic constituencies. While the occupation significantly reformed the terms of acceptable political contestation in the new regime, one institution was left intact, if only symbolically. Emperor Hirohito retained the Chrysanthemum Throne, his title as head of state, and, not least, his life. As early as 1942, there was a faction within the US postwar panning community that favored retaining Hirohito in some capacity (Borton 1966). In mid-1944, the State Department thought he might be useful as an instrument of the occupation authorities (Ibid). The OSS felt that the Emperor would provide a moderate panacea to the ultra-nationalism that pervaded the Japanese government at the time (Tsuzuki 2000). MacArthur (and his advisors) felt that a forced abdication would result in a violent reaction that would complicate American postwar aims (Dower 1999). During the diplomatic back-and-forth leading up to surrender, Japan requested that the US agree to 149 the preservation of the Emperor. The American response was either non-committal or implicitly affirmative, depending upon one?s interpretation (Moore and Robinson 2002). In November, the Joint Chiefs of Staff barred MacArthur from removing Hirohito, although MacArthur required very little prodding in this matter (Borton 1966). MacArthur and the joint chiefs successfully fended off calls to depose, try, imprison and execute the Emperor from actors American and otherwise (Moore and Robinson 2002, Caprio and Sugita 2007). When MacArthur?s officers wrote the constitution that Japan would later ratify, they ensured that the office of the Emperor would survive under the new regime, if in a powerless form (Moore and Robinson 2002, Dower 1999, Borton 1966, Government Printing Bureau 1946). Economic reform under the auspices of the occupation authorities served to root out those domestic elements that would be most incompatible with a democratic society. In December 1945, SCAP first ordered the reorganization of land ownership on a more equal basis (H 1949). These reforms effectively eliminated the landlord class. Between 1946 and 1956, owner-operators increased from 36% to 70% of the farming population, and landless tenants dropped to 4% (Eyre 1956).54 In August 1946, SCAP ordered the dissolution of the zaibatsu, the enormous monopolies that dominated Japanese industrial production and fueled the war effort (H 1949). Most of these conglomerates (the notable exceptions being Mitsui and Mitsubishi) were broken up by January 1947 (Caprio and 54 These reforms left the previous landlord class in charge of half of the forested acreage, however. Since timber was a necessary resource for the rural population, landlords did retain some power over the new tenant-owners, if through market channels instead of political ones (Ibid). 150 Sugita 2007).55 These reforms helped neutralize potential enemies to Japanese democracy?specifically landed interests and monopoly capitalists. In support of these reforms was direct aid, which fed the people during the depths of the postwar period, re- industrialized the economy, legitimated the new regime and built the Japanese middle class. Following a year of food shortages and mass starvation, in August 1946 MacArthur began pleading with Truman for additional aid to Japan in order to preserve the progress they had made thus far (Fuchs 2007). The United States was already shipping half a million tons of grain and 142,000 tons of canned goods to Japan to alleviate the food crisis (Aruga 2001). That aid, along with military spending, essentially financed Japan?s trade imbalance through 1962 (Hunsberger 1997). If dollar infusions from the United States were removed from the balance of payments between September 1945 and December 1962, Japan would have run a current accounts deficit of $7.5 billion. Counting only goods and services, Japan would have run a capital accounts deficit in every year excepting 1958 and 1959. Instead, GARIOA (Government Aid and Relief in Occupied Areas), Korean War procurement and post-Korea military expenditures injected $8.6 billion into the Japanese economy. Specific expenditures in support of American efforts in Korea totaled at least $2.2 billion (Dower 1999, Barnet 1983).56 As a result, Japan ran a current accounts surplus of greater than $1 billion during the period, with the annual balance of payments in the black every year excepting 1945-6, 1953-4, 1956-7 and 1961 (Hunsberger 1997). The outbreak of war directly benefited the average 55 Most of the factories themselves were allowed to remain intact; those remaining on the dismantlement list were returned to Japan at the end of the occupation (N. G. Kim 1997, Reday 1949, Passin 1990, J. B. Cohen 1948, Morley 1967). 56 This is one reason why Prime Minister Yoshida referred to the outbreak of the Korean War as a ?Gift of the gods? (N. G. Kim 1997). 151 Japanese worker; the US would employ 290,000 Japanese in clerical and auxiliary positions in support of the police action (N. G. Kim 1997). And while the Korean War was responsible for a large share of the cash influx, military expenditures largely prior to the outbreak of hostilities overcame what would have been a $1.4 billion deficit between September 1945 and December 1950 (Hunsberger 1997). GARIOA aid totaled almost $1 billion through mid-1948 (N. G. Kim 1997, J. B. Cohen 1948) and $2 billion through 1949 (Dower 1999). To say that the United States financed Japan?s ?economic miracle? would be an accurate but incomplete statement. One could argue without hyperbole that the United States essentially underwrote the Japanese economy during first decades of the new regime. Even though the occupation?s original ideological concern was right-wing extremism, before long Washington and SCAP reoriented policy towards reigning in the very active left wing of the Japanese political spectrum. Left-wing political activity began soon after MacArthur ordered the release of political prisoners in late 1945, and first became disruptive in the ?Food May Day? protests of May 1946. MacArthur?s response was a ban on certain types of political assembly (Maki 1947). He specifically enjoined against a general strike planned for February 1, 1947, which (as a testament to the authority which MacArthur had acquired in Japan) subsequently failed to materialize (Ibid, Dower 1999, Tsuzuki 2000, Morley 1967, Caprio and Sugita 2007). By the fall of 1947, the majority of publications under prior censorship by Civil Censorship Detachment were leftist at this point, signaling a shift in their concern from extreme right-wing elements to the extreme (and sometimes less extreme) left (Dower 1999). In the summer of 1948, MacArthur withdrew the right to strike from labor unions and began 152 purging left-wing politicians, civil servants and news professionals (Ibid, Caprio and Sugita 2007). Between mid-1949 and the outbreak of the Korean War, SCAP blacklisted 11,000 union activists, resulting in their dismissal (Dower 1999). They did so despite the order New Years Eve 1949, in which the occupation authorities formally (if not factually) relinquished their jurisdiction over civilian activities (Park 1949). SCAP?s campaign against the far left plateaued just before the Korean War; MacArthur ordered the purge of the leadership of the Japanese Communist Party in May, ordered its dissolution in June, and purged all JCP members from government, industry and the media on June 6 and 7 (Caprio and Sugita 2007, Dower 1999, Moore and Robinson 2002, Tsuzuki 2000, Morley 1967).57 The day after North Korean forces crossed the 38th Parallel, SCAP closed hundreds of left-wing newspapers and blacklisted 700 journalists (Dower 1999). The ideological reverse course was complete by July 1950, when MacArthur de-purged 10,900 personnel who were originally banned from public service by the anti-Showa purges of the early days of the occupation (Morley 1967, Eiji 2002, Tsuzuki 2000, Quigley 1951). The politico-economic reverse course began in late 1947, as a growing contingent in the Army, Joint Chiefs of Staff, National Security Council and State Department came to oppose the anti-Zaibatsu policy, the disarmament doctrine, pro-labor ordinances, and 57 In fact, Japanese authorities never followed through on MacArthur?s order to permanently dissolve the Japanese Communist Party (McNelly 1965). They did not need to. The purges drove officials into hiding and censorship prevented the JCP from communicating with its constituency or drumming up electoral support. Because the JCP had committed to an electoral rather than revolutionary strategy, the party received no support from Moscow or Beijing (and in fact resulted in their denunciation by the Comintern and Mao Zedong; Dower 1999). Lacking access to electoral channels, financial support from abroad, and having eschewed violent revolution as a path to power, the party could not have survived even if MacArthur had not specifically mandated their dissolution. 153 any other policies that conceivably hindered the reconfiguration of Japan as a powerful and prosperous Cold War ally. It in the fall of 1947 that Army Undersecretary Draper and special advisor George Kennan launched dual campaigns to effect a ?reverse course? in occupation policy (Barnet 1983, Tsuzuki 2000, G. F. Kennan 1967). Over the next two years, SCAP halted demilitarization and the de-monopolization of Japan (Caprio and Sugita 2007, Barnhart 2007, Wilkins 1982, Eiji 2002, G. F. Kennan 1967). By late fall 1948, official US policy on Japan called for a rebuilding of a dual police/self-defense force and drastic anti-inflation, pro-growth measures including a balanced budget (Williams 1968, Barnet 1983, Eiji 2002). In April 1949, SCAP imposed the ?Dodge Line,? named after special advisor and banker Joseph Dodge. The Dodge Line established a fixed (and artificially undervalued) yen-to-dollar exchange rate, required a balanced budget, and drastically cut payrolls for nationalized industries (Tsuzuki 2000, Dower 1999). A month later, the US announced its intention to halt reparations delivery, much to the chagrin of FEC members (such as China and the Philippines) set to receive them (Reday 1949, Morley 1967). Finally, the Korean War provided the strongest impetus for remilitarization and the most substantial levels of American economic assistance. In July 1950, MacArthur ordered the Japanese to recruit a 75,000-man National Police Reserve, to serve as a domestic police force and as a military reserve (Article IX?s ban on militarized self- defense notwithstanding; Auer 1990, Eiji 2002, Bey and Kataoka 1985, Moore and Robinson 2002). One of SCAP?s final acts was to lift the prohibition on arms production in Japan, in March 1952 (Morley 1967). The Japanese acceded to American demands for remilitarization as a condition for the conclusion of a peace treaty, and would continue to 154 expand their commitment to national defense after the occupation (N. G. Kim 1997, Morley 1967). In the course of enforcing the UN mandate, the United States made use of 1,212 Japanese military installations and injected $2.2 billion into the Japanese economy due to ?special procurement,? or the purchase of Japanese goods and services in support of the war effort (Barnet 1983, Morley 1967). On the domestic front, the American forces reinforced the prospects of the optimal winning coalition by marginalizing political extremists, shoring up the prospects of the center, and economically strengthening the optimal winning coalition?s main constituency. By eliminating anti-democratic challengers, and by enriching pro- democratic constituencies, the occupiers improved the chances that the new regime would successfully consolidate. The occupation was successful in narrowing the spectrum of acceptable political contestation. Moreover, this support?economic support especially?represented an investment in the future of Japan: yet another costly and credible signal that the United States was committed to the long-term future of a democratic Japanese state. Domestic Outcomes As with West Germany, postwar Japan rates a 10/10 on the Polity IV scale in every year following the return to sovereignty (Marshall and Jaggers 2009). The electoral history of the Japanese occupation begins in November 1945, when political parties return to the scene following SCAP?s ?freedom of expression? ordinance (Tsuzuki 2000, 155 Morley 1967).58 They and their candidates competed in the first postwar elections on April 10, 1946, prior to the drafting of a new constitution but after SCAP had purged the militarists and ultranationalists from the voter rolls and mandated universal adult suffrage (Borton 1948b, Dower 1999, Caprio and Sugita 2007, Moore and Robinson 2002, Wells 1948). While the victorious parties in this election were the same associations that had backed the Emperor in the previous regime, they had sloughed off their anti-democratic elements and pledged to support MacArthur?s vision for a more open Japanese political system. None was publicly or privately committed to a resurrection of an autocratic regime. Among those who appeared on the ballot, 95% had never before held office (Dower 1999). Turnover was a staggering 80% (Tsuzuki 2000). With only small gains for the far left (if there was one?the JCP is rather moderate as far as Communist parties go) one should consider the April 1946 elections a victory for the optimal winning coalition. The elections placed Yoshida (Japanese Liberal Party) in the premiership, where he helped shepherd through the constitution proposed by SCAP (Morley 1967, Dower 1999, Moore and Robinson 2002, Tsuzuki 2000, H 1949). The April 1947 election, the first under the new constitution, was another win for the optimal winning coalition. The three most successful parties in the second postwar elections were the Democratic Party, the Japanese Liberal Party, and the Japanese Socialist Party (Morley 1967, Abe, Shind? and Kawato 1994, Richardson 1997). The platforms of the Liberals and Democrats were largely indistinguishable, and would eventually merge to form the dominant Liberal Democratic Party. These were right-of- 58 Prior to the April 1946 elections, the Emperor established a caretaker government, placing the cabinet under Prince Higashikuni, and later Prince Shidehara (Tsuzuki 2000, Morley 1967, Moore and Robinson 2002; Dower 1999; Caprio and Sugita 2007). 156 center, bourgeois democratic parties. (Tsuzuki 2000, H 1949, Borton 1948b). The Japanese Socialist Party would vacillate between its radical and moderate leftist phases throughout the occupation and thereafter, but at this point was largely a proletarian democratic party. They renounced violent revolution and adopted an electoral strategy (as, in fact, had the Japanese Communist Party; McNelly 1965, Borton 1948b; Dower 1999). The left-of-center Cooperative Party, which also drew votes, was a minor left-of- center party (H 1949). The strong showing of the JSP allowed it to form its first and only cabinet (Dower 1999, Borton 1948b, Tsuzuki 2000, Caprio and Sugita 2007). This cabinet lasted little over a year, and fell in May 1947 after falling out of favor with GHQ (as they became less enchanted with Japan?s left-wing) and the electorate (after failing to stem inflation and solve other general economic problems; Morley 1967, Borton 1948b, H 1949). Ashida (Democratic Party) formed a new cabinet that would itself fail, replaced by Yoshida?s second government (Morley 1967, Moore and Robinson 2002, H 1949, Williams 1968). The conservatives consolidated their power over the next two elections. Despite a strong showing (9.7%) for the Japanese Communist Party in January 1949, the conservatives absorbed a great deal of the constituency that had previously supported the JCP and JSP. The Democratic-Liberals took home 43.9% of the vote in that election and a full majority (59.5%) in October 1952, the final election before the end of the occupation (Morley 1967, Abe, Shind? and Kawato 1994, Richardson 1997). These elections ushered in the third, fourth and fifth Yoshida cabinets (Morley 1967). This pattern, where the conservative parties would win a combined majority in each election and dominated the cabinet, continued nearly unabated through the end of the observation 157 period (Ibid, Tsuzuki 2000, Dionisopoulos 1957). The exception is January 1967, when the Liberal Democrats won ?only? 49% of the popular vote (Morley 1967, Abe, Shind? and Kawato 1994, Richardson 1997). One interesting trend was the continued support the Japanese Socialist Party received in each of these elections, even during years when the moderate wing split off and the JSP looked rather similar to the JSP.59 Throughout the period, the JSP and the LDP essentially split the electorate to establish a two party system, but the JSP was always the second party (Mendel 1966, Auer 1990, Crespo 1995, Okudaira 1990, McNelly 1965, Dionisopoulos 1957). The electoral history of the observation period is one in which semi-optimal parties would sometimes challenge for a double-digit share of the electorate, but one in which the optimal winning coalition dominated an a truly suboptimal threat was essentially non-existent. 59 The radical wing dominated the JSP during the elections of 1952, 1953, 1955 and 1967 (Morley 1967, McNelly 1965, Tsuzuki 2000). In these years the study counts the Socialists as semi-optimal rather than part of the optimal winning coalition. 158 Sources: James W Morley. Japan and Korea: America's Allies in the Pacific. 1965. Walker and Company, New York; Hitoshi Abe, Muneyuki Shind? and Sadafumi Kawato (translated by James W White). The Government and Politics of Japan. 1994. University of Tokyo. Tokyo; Bradley Richardson. Japanese Democracy: Power, Coordination, and Performance. 1997. Yale University. New Haven, CT, USA. The occupation was not without its civil unrest. The first year or so was characterized by an insufficient food supply and several accounts of starvation. Likewise, there were several protests and the occasional violent confrontation stemming from grievances along these lines. The first instance of political action against the food shortage occurred in October 9, 1945, when fifteen women comprising a group known as the ?Housewives Association? petitioned for an increase in the rice ration (Dower 1999). The demonstrations escalated through May 1946. On May Day, 500,000 gathered in Tokyo (and another three quarters million across Japan) in protest of the small food rations, without incident (Ibid, Fuchs 2007). A smaller group demanded entrance to the Apr '47 85.1% Jan '49 78.5% Oct '52 78.4% Apr '53 80.2% Feb '55 78.1% May '58 90.7% Nov '60 94.0% Nov '63 91.1% Jan '67 61.0% Apr '47 3.7% Jan '49 9.7% Oct '52 12.2% Apr '53 15.0% Feb '55 17.3% May '58 2.6% Nov '60 2.9% Nov '63 4.0% Jan '67 26.0% 0% 20% 40% 60% 80% 100% 19 47 19 48 19 49 19 50 19 51 19 52 19 53 19 54 19 55 19 56 19 57 19 58 19 59 19 60 19 61 19 62 19 63 19 64 19 65 19 66 19 67 19 68 Figure 4.2: Performance of the Optimal and Suboptimal Coalitions in National Elections to the Lower House of the Japanese Diet, 1946-1967 Optimal Semi-optimal Suboptimal 159 imperial palace on the 12th to direction petition the Emperor for more rice (Ibid). The most remarkable of the protest occurred on May 19, known as the ?Food May Day.? A quarter million demonstrated in front of the Imperial Palace in protest of the rationing system, posing enough of a threat that MacArthur was forced to intervene by force. The event was significant enough to convince the Japanese government to reform the rice collection system resulting in a significant increase in yield (Ibid). As significant as these events were, the food protests abated as the food supply improved. However, the vast majority of disturbances during the occupation arose from dissatisfaction from Japan?s newly enfranchised left wing. In December 1945, SCAP allowed the formation of trade unions. A side effect of the introduction of workplace democracy was an increase in leftist radicalism (Dower 1999). On April 7, 1945, the first clash between leftists and the authorities occurred in Tokyo. Fifty thousand demonstrators attempted to gain access to Prime Minister Shidehara?s residence in order to issue a list of demands. Local police fired into the crowd hoping to disperse them, resulting in several injuries and only pushing the crowd further toward the residence. American troops intervened, settled the disturbance, and let a delegation of thirteen issue their demands to the premier (Ibid). Just over a month later, a peaceful demonstration among students and university professors called for the removal of ?war criminals? from higher education and offered their assistance for doing so (Ibid). These demonstrations were but a prelude to what was supposed to be the mother of all demonstrations on February 1, 1947. Throughout January, union after union pledged to join part in a general strike on the First of February in response to both food shortages and high unemployment. By the end of the month, unions comprising three million workers were 160 set to strike throughout Japan. It was only MacArthur?s intervention on January 31 that prevented the strike from occurring (Ibid, Maki 1947, Tsuzuki 2000, Morley 1967, Caprio and Sugita 2007). The failed general strike would not be the last confrontation between the occupation authorities and the Japanese left. The more violent confrontation took longer to develop, and was an indirect result of the Dodge Line economic reforms. Dodge?s call for liberalization resulted in a balanced budget and massive public sector layoffs, including 126,000 workers of the national railway in July 1949 (Tsuzuki 2000). The ensuing dispute between the railway workers, the Japanese government and the American occupation authorities resulted in episodes of sabotage resulting in several deaths by runaway train accidents. The president of the railway disappeared for a time, his body eventually found mangled and beaten. While the deaths were originally attributed to agitation by Communists, there is some evidence that the sabotage was actually the result of American agents looking to discredit both the JCP and the workers (Ibid). This intrigue foreshadowed a more violent turn in anti-government activity and government response. A confrontation between JCP partisans and American GIs on May 30, 1950 left four Americans injured and eight Communists arrested (Dower 1999). On May Day 1952, police repelled protesters demonstrating in protest of rearmament in Tokyo with tear gas and pistols. Twenty-two protesters sustained injuries, two died. Eight hundred policemen and four American GIs sustained injuries (Ibid, Tsuzuki 2000). It was not until the near-end of the occupation, with the end of the purge and the beginning of the de-purge, that right-wing extremism became a problem (Morley 1967). The late 1950s and 1960s were characterized by a rise in right-wing extremism, including 161 the advent of right-wing youth organizations and terrorist groups responsible for the occasional assassination attempt (Ibid, Tsuzuki 2000). This is not unexpected when one takes the de-purge into account. MacArthur essentially lifts the constraints on political contestation as concerns the right wing, and the Marginal Value Model expects that this would produce characteristics of an imperfectly consolidated regime. This type of anti- democratic activity was not a common characteristic of the occupation or the years immediately following, nor did the far right ever seem to pose a serious threat to the stability of democratic politics in the new Japan either during the occupation or after. This may, however, explain why Japan?s consolidated democracy was exceptionally conservative (compared to Germany?s) and why there was so little party turnover (an indicator of an unstable democracy according to Huntington 1991).60 60 There are some who offer that Japan?s centralized, conservative style of government does not qualify as a democracy at all. Bradley Richardson (1997) is among them. That the Liberal Democrats held the government between 1955 and 1993 (Crespo 1995) supports those who claim that Japan was more a single-party state than a competitive democracy. This conflicts with Japan?s polity score (+10) from the year the state regained sovereignty through the present day (Marshall and Jaggers 2009). The data indicate that Marshall and Jaggers are in fact correct on this matter, though there are irregularities. The LDP?s ratification strategy for the 1960 US-Japan security treaty employed tactics that were constitutionally questionable (Destler, et al. 1976; Barnet 1983), while the 1950 electoral law allowed the LDP to gerrymader ridiculously disproportionate and favorable electoral districts (Hata 1990). Others, like S. Sidney Ulmer (1957) may cite the lack of democratization at the local level as evidence that Japanese democracy is not ?deep? and, therefore, unconsolidated. Regardless, these irregularities do not indicate that Japan was not a democracy by modern standards; in even the most advanced and consolidated democracies there is broad distribution of states across proportionality, legislative behavior and centralization variables. Moreover, the fact that there was no party turnover in postwar Japan is not in and of itself proof that party turnover is restricted due to the actions of the ruling party or non-democratic institutions; one must consider the possibility that the LDP secretariat was exceptionally adept at identifying and wooing the median voter. By the Polity standard that this thesis relies on, executive and legislative recruitment remain open, indicating that Japan is indeed a democracy. 162 Summation The Polity IV data hold that Japan emerged from the occupation a democratic state, a status it holds at the present (Marshall and Jaggers 2009). Along with (West) Germany, Japan is among the most successful examples of democratization and liberalization (but not, in this case, social progressivism) of our day. Also like West Germany, the Japan case meets the marginal security model?s conditions for successful democratization. The advent of the Cold War in Asia and the very ?hot? war in Korea re- established the threat environment that had driven Japan to modernization, militarization and empire at the turn of the century. The United States, essentially the sole occupying force, convincingly established itself as Japan?s closest ally despite the mutual hostility they shared during WWII. A threat arose, and the United States rose to face it. The Supreme Commander, General Douglas MacArthur, and General Headquarters imposed a democratic constitution, then took many of the necessary steps to neutralize the new regimes enemies, both real and potential. The winning coalition that emerged was ultimately supportive of the occupation policy and the conditions of the peace. Reconciliation of Competing Hypotheses The events of the case support H1, although perhaps the relationship is not as strong as it is in the case of West Germany. Civil unrest is more common prior to the Korean War and in the earliest days of the occupation when the nation was on the brink of famine. The food protests subside once the US intervenes to increase food production 163 and when Washington increases economic aid, and the left-wing unrest subsided after the US aligned against Japan?s traditional enemies (Russia, China) in Korea. The model itself has trouble explaining the rise of left-wing challenges in the late 1940s, other than the fact that the occupation authorities were initially less concerned about addressing anti- government activity from that side of the political spectrum. Moreover, the electoral indicators do not conform as well as the extra-parliamentary indicators. While the strong performance of the Socialists in 1947 indicates a negative reaction against the occupation?s shortcomings vis-?-vis food economics, unemployment and its continued presence in the absence of a strong, explicit foreign threat, the Japanese Socialist Party still counts as a pro-democracy faction at that point in time, if only barely. The Communist Party?s unique success in the 1949 election directly contradicts H1, as the US occupation seemed to be producing marginal security by that point. Perhaps it took the very real threat of the Korean War, rather than the vague threat of historic tensions between Japan and the mainland Asian states, in order for the Japanese to perceive the marginal security value that the occupation was producing. Perhaps also, the persistence of the semi-optimal coalition (sometimes the JCP alone, sometimes the JCP and JSP) results from the fact that the threat from the mainland was implicit, based on historical memory, rather than explicit, based on direct threats and signals of hostile intent. As noted above, the case does resonate with some competing hypotheses. However, the marginal security model remains stronger in tests against the historical record while subsuming some competing explanations. As in the case of West Germany, the demilitarization and delegitimation hypotheses explain part of the case but not all of it. The occupation remilitarized Japan almost as quickly as it demilitarized Japan, and in 164 self-help, anarchic world, it is necessary to take both sides of the coin into account. This is as far as I will discuss this thesis, as its shortcomings are identical to those in the previous chapter. Coyne?s hypothesis on the retention of Japanese institutions is difficult to test due to a lack of explicit observable implications that derive from a well-wrought model. However, one should allow that the retention of certain institutions did solve a coordination dilemma, if not necessarily the credibility dilemma. Furthermore, it is implausible that retaining institutions can solve the credible commitment dilemma if the occupation does not responsibly employ these institutions to improve the perceived welfare of the state. As is always the case in democratic politics, the existence and persistence of institutions matter less than the way in which powerful actors utilize and operate within said institutions to realize their preferences. The prewar modernization hypothesis is a far more convincing explanation about how Japan became a developed state than about how Japan became a democratic state. The path of modernization can lead to either democracy or autocracy, though in most cases we must accept that the path towards democracy begins with modernization. Other variables must intervene to decide the fate of modernized states, and the marginal security model olds that threat environment, security configuration, and occupation relationship are the variables that do so. Economic Drivers An analysis of the case of Japanese democratization is not complete without an analysis of the economic situation. Compared to both German cases, however, the situation is far more clear-cut. Democracy arose before Japan?s ?economic miracle? and 165 appears to have stabilized quickly. What?s more, said ?miracle? is even more obviously a result of American policy in the region than in the other cases. The first years of the occupation were plagued with inflation, unemployment and underwhelming growth. The tide turned with an influx in aid, with the Korean War, and with the eventual success of organic economic reconstruction. Inflation was a persistent dilemma through 1949. During the course of the war, the yen-to-dollar exchange rate increased from ?3.4 to ?13.6, an annualized rate of just over 30%?high but not nearly as high as what was to come. The rate more than doubled by the end of the year, doubled again by September 1946, once again by June 1947, and once more by September 1948 until the Dodge Line set the exchange rate at ?360 to $1 in June 1949 (T. Cohen 1987). In an attempt to contain inflation, the American authorities established a command economy with controlled prices, production quotas and rationing. However, SCAP increased these prices over time in order to keep pace with the black market that inevitably developed alongside the official economy. Official prices doubled between the onset of the occupation and the conclusion of 1945, increased a whopping 539% in 1946, 256% in 1947 and 127% in 1948 (Dower 1999). Official prices were so artificially low at first that, despite an inflation rate that exceeded that of the yen-to-dollar exchange rate, they still remained one half as high as market prices by 1949 (Ibid). Understandably, underground trade flourished throughout Japan during this period, and landowners were at first slow to fulfill the mandated rice quotas which supplied the subsistence rations (Ibid, Fuchs 2007). 166 Source: Theodore Cohen. Remaking Japan: The American Occupation as New Deal. 1987. The Free Press. New York SCAP employed a stick-and-carrot approach to prevent the diversion of staples to the black market. Sensing that market prices were too quickly outpacing the official prices, the military government doubled the payment to farmers for rice and other staples in November 1946 (Dower 1999). Regardless, collections fulfilled only 60-85% of the quota, depending on the prefecture. In addition to wartime devastation, unusually inclement weather, and the lures of the black market, suspicion of the military authorities? rationing program encouraged hoarding (Fuchs 2007). The official adult staple ration in the 1946 harvest year equaled 1,000-1,100 kcal/day, half the recommended amount during a time when many (if not most) had no other source of food 0% 100% 200% 300% 400% ?0 ?100 ?200 ?300 ?400 Jan '4 5 Ju l ' 45 Jan '4 6 Ju l ' 46 Jan '4 7 Ju l ' 47 Jan '4 8 Ju l ' 48 Jan '4 9 Ju l ' 49 Jan '5 0 Ju l ' 50 Figure 4.3: Inflation in Occupied Japan: Yen to Dollar Exchange Rate, 1945- 1950 $1 buys (left) Annualized Inflation Rate (Right) 167 (Ibid, Dower 1999). In reality, there were months that the true ration was well below these levels due to problems in the distribution system (Ibid). Food imports from abroad, many of which had come from Japan?s mainland holdings in Asia?especially Korea? had dwindled to barely a third of their wartime peak (Fuchs 2007). MacArthur in January 1946 warned Washington that Japan was on the edge of famine, and that such a disaster would decimate the lower and middle classes, engender political subversion, and perhaps even spark an uprising (Ibid). Aware that the carrot was not producing the intended effect, SCAP turned to the stick. In February 1946, SCAP announced that they would seize undelivered rice, with the farmers compensated at the official price. That same year, SCAP ordered the Japanese government to pass the land reform law, making arable land that had been the property of absentee landlords available to production (Tsuzuki 2000, H 1949, Fuchs 2007). These reforms, and a bumper crop (Fuchs 2007), improved the situation in 1947. The government, at the behest of GHQ, introduced incentives for delivering quotas early (Ibid). Meanwhile, GHQ threatened that, without a production increased, there would be no additional aid from the US forthcoming (Ibid). In the 1947 harvest year, collections equaled 107% of the quota (Fuchs 2007). The situation would improve in 1948 and thereafter (Braibanti 1948, Dower 1999). By 1948, Japan was emerging from the shadow of impending famine, but by no means was it a kernel of the economic powerhouse it would eventually become. The imposition of the Dodge Line was harsh medicine. While it did contain inflation and the government spending that was partly driving it, it did so at a substantial cost. In 1947, the government employed fully 25% of the work force (Barnet 1983). Over 168 400,000 workers lost their jobs when the government cut public sector employment in 1949 (Tsuzuki 2000). And despite Dodge?s professed goal of driving exports to the United States, Japan?s dollar reserves actually declined through 1950 (Barnhart 2007). A global recession prevented Japanese producers from taking advantage of its artificially high exchange rate, and as a result bankruptcies increased and inventories went unsold (Barnet 1983). The situation was such that US News and World Report predicted Japan to be on the verge of economic collapse as late as April 1950 (Dower 1999), and local mainstream publications considered the Dodge Line an unmitigated disaster (Barnet 1983). Finally, inflation in the yen-to-dollar exchange rate was trending steadily downward over the course of the occupation (T. Cohen 1987), so one wonders whether the new policy was necessary for reigning in inflation in the first place. Regardless of the virtues of the Dodge Line, Japan?s recovery began in 1950 due to the increase in demand resulting from the onset of war on the Korean Peninsula. It the war that helped drive the 80% rise in the Japanese stock market between June 1950 and December 1951 (Dower 1999), that boosted steel production by 38% and tripled exports (Ibid), that returned industrial production to prewar levels (N. G. Kim 1997, Eiji 2002, Morley 1967) and that turned Japan?s balance of payments positive for the first time under the occupation (N. G. Kim 1997). American dollars flowed into Japan through traditional market channels as well as through special procurement and direct assistance. In May 1950, the Japanese government passed the Law Concerning Foreign Investments, allowing US foreign direct investment into the country so long as American firms pair with Japanese concerns (Wilkins 1982). US FDI was a small contributor during the period of observation, however, barely totaling $100 million (Ibid). 169 In conclusion, one cannot rely purely on economic hypotheses to explain the consolidation of democracy in Japan. For starters, one should not get carried away with these numbers. Despite the war boom, real per-capita income remained below prewar levels through the end of the occupation, and total trade remained about 50% of 1934 levels in 1952 (Morley 1967). As noted above, Japan would have run a near-perennial trade deficit for over a decade post-occupation were it not for the influx of American money. It was not until 1955, the first year that GNP surpassed prewar levels, when the boom would truly begin (Tsuzuki 2000).61 Additionally, the optimal winning coalition received strong domestic support even during the poorest years of postwar Japan. Finally, Japan?s growth largely depended upon pragmatic expenditures and purchases by the occupying forces. Economic drivers simply do not correlate well enough with outcomes to serve as a viable explanation, let alone an independent one, while it is impossible to separate Japan?s postwar boom from the reality of Japan?s postwar occupation. The marginal security model, on the other hand, fits better from a correlative as well as causative perspective. Summation of Findings and Conclusion This case is confirmatory of the hypothesis, although not as convincingly as the case of the Federal Republic of Germany. The historical record indicates that the United States, during the occupation of Japan, was able to produce marginal security. The threat environment was primarily implicit rather than explicit?no powers in the region directly attacked Japan or signaled a desire to compromise Japan?s already compromised 61 And it was not until 1965 that real incomes caught up with the trajectory implied by the prewar growth rate (Passin 1990). 170 territorial integrity. Rather, the biggest threats to Japan were the same threats that preoccupied Japan prior to WWII and for the last three generations: the sleeping giants of China and Russia. What changed following WWII was Japan?s vulnerable position? without an army, a navy, or an empire?the names of the People?s Republic and the Soviet Union, and the alliance they forged at the dawn of the 1950s. When Soviet-backed North Korea, and later the People?s Republic of China, invaded the Korean peninsula, the implied threat became that much greater. After half a century of imposing its will on the Korean Peninsula in an attempt to expand its influence onto the mainland, Japan?s mainland security competitors were on the verge of controlling the ?dagger? aimed at its heart. The United States, initially reluctant to commit to Japan?s future (or to take on any of the responsibilities of a global superpower), transformed from enemy to ally by rebuilding Japan?s self-defense capability and deploying its own forces to the Pacific Rim, long term. These actions, in addition to economic assistance and military procurement, credibly signaled their commitment, thus producing marginal security. The production of marginal security, plus the state-building work of General MacArthur and the office of the Supreme Commander, served to narrow the acceptable range of political contestation. It was under these circumstances that the new Japanese regime appeared to consolidate. 171 Chapter 5: Weimar Germany, 1919-1933 Introduction The case of the Weimar Germany is an unfortunate one, a failure. The military occupations of the German Rhineland under the United States, United Kingdom and France began in November and December 1918 (Wright 1919, Dawson 1933, Cornebise 1982, Williamson 1991). The military governments remained in place until the Inter- allied Rhineland High Commission (IARHC)?with delegates from the US, UK, France and Belgium?took control on January 10, 1920 (Williams 1991, Pawley 2007). In addition to the direct occupation of the Rhineland, the occupying powers retained control of foreign policy for the whole of Germany and reserved the right to post troops in initially unoccupied areas should Germany fail to meet the conditions of the Treaty of Versailles (Ibid). France and Belgium expanded their occupation to the Ruhr on January 11, 1923, prompting the exit of the US delegation and marking the end of the US occupation of the Rhineland (Ibid, Lauterbach 1944, Nelson 1970, Cornebise 1981, Mommsen 1996, O?Riordan 2005). The UK, France and Belgium continued to exercise legal control over the German armed forces (Reichswehr) through January 31, 1927 and direct control over the Rhenish territories through June 30, 1930, when the occupation came to a close (Kolb 1988, Williamson 1991, Mommsen 1996, Webster 2006, Pawley 2007). The period of observation thus begins in 1918, but ends in 1933 when the Enabling Act permanently dissolved the Reichstag (Nicholls 1991, Halbrook 2000, Snyder 1998). 172 Cohesion was weak among the occupying forces. As the historical record will show, the occupying powers generally retained broad authority in their zones (as compared to the way in which the allies coordinated policy in Germany following WWII). Moreover, the IARHC split into two blocs: the Franco-Belgian bloc, which favored a more punitive approach to the occupation, and the Anglo-American bloc, which was less interested in wringing concessions out of Germany at Versailles and was generally reluctant to cite Germany for treaty violations and follow through on punishments. Of the four occupiers, the US was keener on democracy promotion than the other states, followed by the UK. France and Belgium appear generally uninterested in the type of German regime so long as it was weak and incapable of expansion. The pair of economic crises that brought serious hardship to the German people heavily dominates the Weimar case. Therefore, to support the hypothesis, the historical record must indicate that the behavior of the occupiers significantly contributed to domestic unrest, the dwindling share of the optimal winning coalition and the success of extreme, anti-democratic parties in a way that is independent or causally prior to economic variables. 62 62 The winning coalition includes the Social Democratic Party (SPD), the German Democratic Party (DDP), and the Zentrum for most of its existence in this case. Parliamentary government had been a longstanding part of the SPD program, while the DDP was formed specifically by pro-democracy elements of both the wartime Progressives and DVP. The Zentrum?s platform emphasized the importance of democratic politics as a basic criterion of Christian government. The Communists (KPD), Independent Socialists (USPD), Social Democrats (SPD), Democrats (DDP), Catholic Center (Zentrum), and Liberals (DVP) were all opposed to the Kaiser?s regime by the time the war ended, but the sub-optimal KPD and semi-optimal USPD and the (post- WWI iteration of the) DVP all rejected representative democracy and so fall outside the optimal winning coalition. The sub-optimal Nationalists (DNVP) vehemently opposed democracy and suffered greatly from the loss of the abolition of special privileges for the Junker nobility, a key demographic of Nationalist support. The Nazis (NSDAP, and the 173 True, the Great Depression ultimately brought down the Weimar Republic, but by 1933, democracy in Germany was a collapsing edifice, mortally weakened but still standing; the global economic crisis yielded only the final push that brought down a once-promising structure. Below, I shall test the working hypotheses against the case history, as well as test the competing hypothesis. I will then demonstrate how the marginal welfare model is a superior approach to understanding the case by demonstrating how the competing hypotheses are unsupported or are subsumed by the Marginal Value Model. This case will reinforce the premise that nascent democracies in occupied territories require an occupation that protects the new regime from foreign threats and reinforces the optimal winning coalition. Rival Hypotheses versus Case History Rival Hypothesis: Institutional Design It would be overly comprehensive to summarize the exhaustive body of literature blaming the Weimar Republic?s institutional design for the fall of the regime, so I will produce a sample of that literature here. Three pillars of the Weimar regime sustain the lion?s share of criticism: the astonishingly open Weimar party system, which established a very low barrier to entry for political parties in the Reichstag; direct democracy, in the form of the direct election of the reichpresident and in the referendum system; and Article 48, which conferred emergency powers on the reichpresident in the event of precursor DAP) were not a major force in German politics at the time, winning their first Reichstag delegates in 1924 (Kaes, Jay and Dimendberg 1994). 174 extreme civil strife (German National Assembly 1919, Kaes, Jay and Dimendberg 1994). Put together, these institutions amplified each other?s failures. The indecisiveness of parliamentary government in Weimar Germany in moments of crisis is what prompted the votes for emergency government in the first place. The direct election, and lengthy term, for the reichpresident resulted in the translation of an overactive populism into a presidency controlled by a leader without acceptable democratic credentials. The most obvious flaw of the Weimar Constitution was the provision granting emergency powers, which is the legal construct Adolf Hitler eventually used to prosecute his ?legal coup.? While the clause was supposed to apply only to widespread civil strife, the Reichstag expanded the operational definition of civil strife with each invocation (Lindseth 2004). While it is self-evident that a clause granting broad executive powers can lead to a constitutional crisis and the collapse of a democratic regime, it is by no means certain that it will. The Weimar Constitution was not the first republican constitution with an emergency clause, and it will not be the last. Moreover, democratic regimes without emergency clauses often force leaders to assume powers not granted, precipitating a constitutional crisis of another color (see Lincoln?s suspension of Habeas Corpus in the American Civil War). Emergency authority leads to a regime collapse only once the regime is captured by anti-democratic forces, at which point we could already assume the regime is in trouble. Elements of direct democracy in the Weimar Constitution are a common target for criticism of the founding document. Combined with the seven-year presidential term, and the emergency powers the reichpresident may come to enjoy, the directly elected presidency can result in severe political instability. As Linz (1990) notes, the direct 175 election of presidents crystalizes winning coalitions an extended period of time (again, in this case, seven years) with no hope of realignment or reconfiguration. This can lead to a crisis of legitimacy, should a crisis lead to a shift in the preferences of the electorate, especially if the president unwilling to seek a domestic consensus. While this is not exactly the story of the Weimar, the presidential system did elevate Paul von Hindenburg?a popular figure with clear disdain for the constitution under which he ruled?to a position in which he used his emergency authorities to pave the way for Adolf Hitler. As luck would have it, the second presidential election under the Weimar Constitution occurred in 1932, during the depths of the Great Depression, translating the full force of populist furor into the most powerful office in the land. It is impossible to ignore the role that this institution played in Weimar Germany?s failure to consolidate. Finally, the ease at which political parties could gain access to the Reichstag, and the resultant parliamentary fractionalization and gridlock, receives a fair amount of blame for the ineffectiveness of the legislative body (Grumm 1958, Kuechler 1992). Gridlock in the Reichstag led to minority (unelected) governments in 1922, 1926 and then from 1930 through Hitler?s seizure of power in 1933 (Kolb 1988). In one notable instance, the Reichstag?s indecisiveness in dealing with the Great Inflation and the Great Depression drove contemporary thinker Carl Schmitt?s (1988) criticism of the regime and later influenced his allegiance to the National Socialists. As the theory goes, the Reichstag?s ineffectiveness led to a sense of disenfranchisement among the German populace? especially among the lower-class agrarian population and the professional middle class (Childers 1976, Bendix 1953, Lipset 1963). Without an outlet, these constituencies 176 instead channeled their frustration into the populist movements that swept Adolf Hitler into power (Berman 1997). Rival Hypothesis: Demilitarization The demilitarization hypothesis (J. J. Linz 1990, Stepan 1991, Grimm 2008) presented in the previous chapters applies to the Weimar case as much as it applies to the post-WWI Japan and German cases. Whereas the occupying dismantled the war making capabilities of both Japan and Germany after WWII, the Allied and Associated Forces (AAF, including the US, UK, France and Belgium) failed to do so in Germany after WWI?and not for lack of trying. Disarmament was a major component of occupation policy. Britain?s Prime Minister Lloyd George initially suggested disarmament at Versailles as an alternative to the occupation of the Rhineland, but diplomatic maneuvering on the part of French Prime Minister Clemenceau led to a peace that included both disarmament and occupation. The disarmament clauses required Germany to destroy offensive weaponry and maintain an army of no greater than 100,000, a ban on conscription, with soldiers serving twelve-year contracts. (Willamson 1991, Diehl 1977). The occupiers would monitor Germany?s progress on disarmament, which was perpetually satisfactory to the UK and unsatisfactory to France (Willamson 1991). Ultimately, the demand for unilateral disarmament encouraged Germany to seek cooperation with the only external security partner willing to help rearm the Weimar Republic: Russia (Gatzke 1958, Kolb, The Weimar Republic 1988, Nicholls 1991, Mommsen 1996). In 1927, the Inter-Allied Military Control Council (or IAMCC, the organization in charge of disarmament and monitoring) transferred monitoring 177 responsibilities to the equally clueless League of Nations and subsequently dissolved (Kolb 1988, Webster 2006). Germany had already been secretly rearming for years. In addition to the AAF?s failure to disarm Germany, they also turned a blind eye to the composition of the Weimar Republic?s military establishment. When domestic security began to collapse at the end of the Kaiserreich and the beginning of the Republic, the new republican government turned to the military establishment of the old regime in order to maintain order (Diehl 1977, Mommsen 1996). There was no Weimar equivalent to denazification. Instead, AAF implicitly sanctioned the inclusion of the old regime?s military apparatus in the new government. This is the flip side of the process that Linz, Stepan and Grimm claim played such an important role in the consolidation of Japanese and German democracy after 1945. Rival Hypothesis: Prewar Modernization (or lack thereof) Among the most cited explanations for the failure of democratization in Weimar Germany is Barrington Moore?s (1966) thesis on uneven development and agrarian interests. When industrial interests must compromise with agricultural interests in building a modern, industrial state, they marginalize the banking and trading classes that typically push for democratic institutions. This leaves two possibilities: fascism and communism. When the peasantry is not substantially organized and mobilized, the result is a fascist state. Moore specifically cites this process as that which undermined pro- democracy forces in the Weimar Republic, and employs Nazi Germany as one of the three archetypes resulting from the processes he describes. 178 There is some evidence to support this hypothesis. As noted in the chapter on the post-WWII Federal Republic, parties outside the optimal winning coalition performed substantially better in rural regions than in urban ones (Bernhard 2001, Childers 1976). However, as I also noted, this ignores antidemocratic tendencies in the more industrialized laender of Saxony and Bavaria, and in the Saar and Ruhr. Uneven development, a persistent Junker class, and an agrarian political economy might certainly have stood in the way of democratic institutions in an alternate reality in which Germany was both undivided and democratic. But they explain only one instance of anti- democratic tendencies in one region of Germany, and not the persistent anti-democratic challenges the Weimar regime faced among eastern and western?as well as urban and rural?constituencies. This does not mean we can reject the hypothesis outright. Moore?s theory explains very well why Germany transitioned to Fascism in 1933. The problem is that it fails to explain the several other indicators that the Weimar regime had failed to consolidate dating back as early as 1922. Summation The hypotheses above are all plausible, despite shortcomings here and there. Individually, however, they leave major gaps in our understanding of the collapse of Weimar democracy. The failure to demilitarize Germany does seem to explain why the optimal winning coalition was unable to monopolize political violence, but it does not explain the rise and fall of the electoral and parliamentary fortunes of the optimal winning coalition. It also fails to account for the fact that Hitler was able to seize power in a legal rather than extra-parliamentary manner. Conversely, Moore?s hypothesis does a 179 very good job in explaining why Hitler was able to seize power, but does a very poor job of explaining the near-success of left wing and other less extreme, yet anti-democratic political movements in the early years of the Republic. The institutional design hypotheses all present coherent explanations for why perceived disenfranchisement of the German population resulted in a constitutional crisis that the government was unprepared to handle. In other words, it explains why a failure to narrow the spectrum of acceptable political contestation resulted in the collapse of parliamentary democracy. However, it is not particularly convincing in explaining the widespread political violence and the rise of paramilitary politics before the Great Inflation and Great Depression. Immoderate political forces were rampant in Germany before the ineptitude of the Reichstag became obvious, weakening the hypothesis that parliamentary gridlock was the primary cause of political extremism and virulent populism. I shall demonstrate that there is a conspicuous relationship between the threat dynamics and occupation management techniques of the Weimar case and the subsequent domestic outcomes. The Marginal Value Model provides a coherent and comprehensive explanation for the rise and fall of the Weimar Republic. Test of Working Hypotheses versus Case History Threat Environment In the case of Weimar Germany, it is difficult to find a player who sufficiently fits the role of villain. There is nothing on the order of the threat that the USSR posed to 180 Germany and Japan after WWII, no powerful state threatening to destabilize the region in the way that the PRC threatened the Pacific Rim. However, there was still good reason for the first German democracy to fear its neighbors. First, the Treaty of Versailles re- established the Polish state (that was absorbed by its neighbors in 1795) on the German border. The rebirth of Poland was not threatening in and of itself. That said, the insurgency on the German-Polish border that raged through the early years of the occupation, their push to acquire territories that were majority German, and the support and/or sponsorship that Poland and the insurgents received from other states was a threat to the personal and territorial security of the Weimar Republic, not to mention a source of grievance to a society that (rightfully or no) was exceptionally aggrieved. While the potential to establish marginal value by providing security to Weimar Germany was not as clear-cut as in other cases, the opportunity was there. In November 1919, hoping to gain territory before proceedings at Versailles cemented Poland?s new borders, Polish republican Wojciech Korfanty organized an abortive coup aimed at separating Upper Silesia from Germany by force (Willamson 1991). Polish-German violence characterized the threat dynamic through 1921. The most violent of these uprisings, the Third Korfanty Revolt/Silesian Uprising, began on May 2, 1920, during which Polish irregulars overran the eastern two-thirds of Upper Silesia and insurgents killed three hundred Germans and wounded 1,500. The German Freikorps paramilitary to regrouped during a ceasefire and launch a successful and decisive attack Table 5.1: Influences on the Threat Environment in Weimar Germany More Threatening ? Silesian Uprisings ? Polish annexation of Danzig corridor Less Threatening ? Russo-German military production agreements 181 at Annaberg on May 23. Only after German forces gained the upper hand did the conflict cease. Polish forces retreated between June 10 and July 21, ending the rash of violence in the German-Polish hinterlands (Ibid, Dawson 1933). As the uprising receded, so did the AAF?s best opportunity to produce security goods and thereby produce marginal occupation value. However, this does not mean Germany was finally secure. Internal threats in the form of separatist movements in the Rhine grew in the early 1920s (see next section), raising the possibility additional territorial cessions. Germany still shared a relatively indefensible border with an unfriendly neighbor, and they still had limited means to defend themselves. Since the supply of goods for internal balancing had dwindled under the disarmament regime, but demand for security had not declined (or at least not as swiftly), Germany looked for a partner who would help them externally balance against foreign threats. These constraints encouraged Germany to reach out to Russia in secret and in violation of the Versailles prohibitions on offensive weapon production and foreign security pacts. Evidence of Russo-German security cooperation dates back as early as August 1920, when Defense Minister Seeckt and Foreign Minister (and future Prime Minister) Gustav Stresemann authorized the issuance of low-level communiqu?s to the Soviet armed forces. In May 1921, the Berlin and Moscow signed a commercial treaty, joining forces out of mutual need to alleviate their economic deprivation (Gatzke 1958). The nations also shared a mutual interest in constraining and suppressing Poland. The Russo- German answer to the Polish question was the Treaty of Rapallo, signed on April 16, 1922. The public version of the treaty established most favored nation status, secured Germany from Soviet claims under Versailles, and secured the Soviet Union from 182 German claims over property seized by the Communist regime. Privately, Rapallo contained the first agreements pertaining to military assistance (Nicholls 1991, Kolb 1988, Mommsen 1996).63 Following Rapallo, German pilots and tank commanders began secret training in Russian tanks and planes?an unequivocal violation of the Versailles Treaty (Gatzke 1958, Kolb 1988, Nicholls 1991, Mommsen 1996). From the middle-1920s onward, a tacit agreement between Seekct and Stresemann enabled the falsification military budgets to prevent discovery of their violations by Inter-Allied Military Control Commission (IAMCC) officials. While it appeared smaller on paper, the actual Reichswehr budget increased by 75% during the period 1924-1928. Once the responsibility for weapons monitoring was transferred to the League of Nations, German rearmament continued without direct administration by the allies. Towards the end of the decade, Germany entered into industrial arms production agreements with the Soviet Union, in which Germany would foot the bill for factory construction, while the two states would share the fruits of its production. The 1930 Rheinmetall agreement, a plan to fund construction of a munitions factory in the Soviet Union, was the last of these (Nicholls 1991, Webster 2006, Gatzke 1958). Germany?s partnership with Russia is evidence of Germany?s desire to remilitarize despite the threat of punishment by the AAF?punishments that might include expansion or extension of the Rhineland occupation?indicate a sense of insecurity. Germany?s actions tell us that 63 Notably, these secret proceedings?including the decision to announce the treaty while a German delegation was seeking concessions in London on disarmament?was made by the Reichswehr leadership without cabinet approval; even the Reichswehr operated as a free agent in this regime. Later in this thesis I indicate the dominance of pre-Weimar officers in the armed forces, illustrating the theories espoused by Huntington (1991), Linz (1990), Stepan (1991) and Sa?Adah (2006) about the fragility of regimes in which the pre-republican military apparatus remains in place. 183 there was demand for security goods were the AAF to produce some. Their alliance with Russia indicates one of the following: 1) the AAF refused to produce these goods; 2) the AAF failed to produce these goods in the quantities that Moscow offered; 3) the AAF pursued policies that were themselves threatening, thereby undermining the credibility of any security commitment from the occupiers. The outside threat environment over the entire observation period is perhaps the least critical of the four cases in the dissertation. And yet, it is clear that the environment was threatening enough for Germany to seek outside assistance. Part of this demand for external balancing was surely manufactured by the occupation and Versailles: the territorial cessions, the reconstruction of Poland and the neutering of Germany?s security apparatus that drove this demand were the work of the same states that occupied the Rhineland. However, the demand for outside security assurances was there, and the Allied and Associated Forces were in the best position to fulfill that demand. Would the AAF be security providers or security deniers? Signaling Security Value Instead of satisfying Germany?s desire to balance externally and, in doing so, producing marginal occupation value, the AAF focused on permanently weakening the Weimar Republic so that it could never again project its power. While Poland nipped at Table 5.2: Indicators of Occupation Security Value in Weimar Germany Positive ? Locarno Pact Negative ? Closed-door negotiations at Versailles ? French support for Polish insurgents ? French support for annexations of predominantly German territory ? French support for Rhenish separatism 184 Germany?s borders, France?which retained administrative authority in Upper Silesia? stood by and allowed it to happen (Williamson 1991). While Rhenish separatists threatened to pull away from the Fatherland, France not so quietly urged them on. The occupiers attempted to dismantle the German war machine, but they did so in a way that left elements of the old regime in control of the means of violence, that did not include a path to remilitarization (as the Allies offered to the FGR via NATO, or as the US would present to Japan in the form of a large defensive reserve) and that only temporarily hampered Germany?s power projection capability. When the AAF eventually offered a mutual security pact at Locarno, France had already proved themselves a better enemy than a friend and Germany had found help elsewhere. The rational starting position of any occupied state is that the greatest of all threats are the occupiers themselves. This remains the case until the occupiers prove otherwise. They did not. This is not how the occupation had to proceed. Early on, there was reason for Germany to expect an equitable peace. Between late 1917 and early 1918, the US and the UK began to express their support for a postwar peace along the lines of Wilson?s Fourteen Points. In response to these signals, Chancellor Prince Max of Baden initiated truce negotiations with Wilson and led the Reichstag in a vote to begin Germany?s transition to democracy. (Wright 1919, Willamson 1991, Mommsen 1996, Kolb 1988, Dawson 1933, Pawley 2007, Nicholls 1991, Snyder 1998).64 Germany hoped to conclude a peace based on Wilson?s Fourteen Points, despite a lack of demonstrated commitment 64 Nominally, the new regime was a constitutional monarchy, but with the imminent abdication of the Kaiser and the absence of any formal role for the monarch, these reforms established a de facto parliamentary republic (Mommsen 1996). 185 to the American plan from the UK and France. (Dawson 1933, Nelson 1975, Pawley 2007, Willamson 1991). When peace talks opened at Versailles on January 18, 1919, closed-door negotiations excluding the German delegation directly contradicted Wilson?s plea for ?open covenants openly arrived at? (Wilson 1918). The negotiations followed a very familiar pattern throughout: the French delegation would press harsh demands that the UK and US could not abide, receiving Belgian support. The French would then offer a counterproposal that was still beyond what the Anglo-American bloc was willing to implement. However, President Wilson, anxious to write his plan for a League of Nations into the Treaty of Versailles, would trade support for Clemenceau?s demands in exchange for his League. Wilson?s capitulation isolated Lloyd George, leaving no choice but to support the French position (Schiff 1930, Dawson 1933, Kolb 1988, Willamson 1991, Pawley 2007). This pattern first manifested itself in several key negotiations: French territorial claims that stemmed from secret prewar treaties, a temporary occupation, disarmament, reparations, et cetera? (Dawson 1933, Nelson 1975, Willamson 1991, O'Riordan 2005, Pawley 2007). Many of these negotiations resulted in clear negative security consequences for Germany, especially considering the issue of territorial cessions. In the wake of the Armistice, the victors began to draw a new map of Europe, ostensibly based on recognition of national self-determination. The Germans would soon discover that the Allies did not intend that this principle apply to them. The Versailles Treaty established a Plebiscite Commission that selected the regions for possible redistribution to existing states or the formation of new polities. The agreement also mandated the use of 186 plebiscites to determine the relevant ethno-geographic distributions, but the AAF?s interpretation of self-determination was more than a bit disingenuous. The AAF prevented German-speaking individuals from registering for the plebiscites and ignored others.65 In Upper Silesia, French administrators granted local Poles premature access to German funds in local banks on the assumption that the vote would favor Poland. Despite the imposition of eligibility requirements by Polish and French Commissioners that excluded the 70,000 Germans who had settled in Upper Silesia since 1904, 59.6% of eligible voters elected to remain part of Germany. (Willamson 1991, Dawson 1933). The plebiscite in Allenstein would declare 98% for Germany, while plebiscites in what would become the Polish Maritime Corridor declared between 60-80% for Germany. It did not matter: the Commission handed the territories over to Belgium and Poland anyway. It had become very clear that the postwar peace would not be governed by the Fourteen Points, but rather by realpolitik and Clemenceau?s (historically understandable yet ultimately misguided) efforts to permanently incapacitate Germany (Ibid). The AAF?specifically France?succeeded in making itself into Germany?s enemy vis-?-vis Poland. On March 12, 1919, the Commission on Polish Affairs recommended that Upper Silesia and the port city of Danzig be transferred to Poland.66 Instead, France and the UK decided that Danzig would be a League of Nations? administered free city within the Polish customs union, and that that the fate of Upper Silesia be decided by plebiscite (Willamson 1991). During the Silesian Uprisings, France pledged its support to Poland via a treaty of mutual assistance. While the pact ostensibly 65 In Eupen-Malmedy, a territory that was 80% German in 1910, only 271 voters registered as Germans (Dawson 1933). 66 Clemenceau supported this, but Lloyd George threatened to pull out of the peace talks if the French president pursued this course (Williamson 1991). 187 united the two states against Russia, it also created a bloc that flanked and largely isolated Germany. On September 15, 1922, France further aided Poland by recognizing the Polish Navy?s navigation rights to the Baltic Sea and at Danzig, essentially transforming the ?Free City? into a Polish port-of-call. The two parties amended this agreement on May 12, 1923, with France sending advisers to build a Polish air force, along with aircraft, automatic weapons, a loan of 300 million Francs to build their military capability, and finally by training Polish officers in French war colleges (Dawson 1933, Kolb 1988). Ultimately, Poland?s qualified legal jurisdiction and geographic dominance in the region turned Danzig into a de facto ward of Poland. France very clearly aligned itself with German?s strongest contemporary foe?an unquestionable example of an unfavorable security configuration. France pushed for the erosion of German borders in the west as well as east. As early as November 17, 1918, Marshal Foch was authoring plans to turn the Rhineland into a buffer state (or states). On December 12, Plebiscite Commissioner Tirard suggested to Paris that they incorporate the Rhineland into their economic sphere, thereby gaining economic and political leverage over the territory (Willamson 1991, Pawley 2007). Beginning in 1919, despite officially resigning their initial demand to annex the Rhineland, the French embarked on a campaign to sponsor Rhenish separatist groups in their occupation zone. Commissioner Rault, who controlled the Saar Coal Basin, tried to coax the region into independence and into French hands by commissioning a Saar flag and coat of arms, introducing the franc as legal tender, financially supporting the construction of French-language schools, Gallicizing the existing German ones, and by openly requesting a post-occupation plebiscite that might favor separation from the Reich 188 (O'Riordan 2005). France later attempted to introduce a new currency to the region, the Rhenish Notenbank, in order to isolate the region economically and ensure its dependence on French demand for raw materials (Willamson 1991, Pawley 2007).67 The French redoubled their efforts to separate the Rhineland in June 1923, as General Mangin conspired with separatists in Guelph to form an independent Lower Saxony. The movements turned violent when, in Dusseldorf on September 30, 10,000- 15,000 separatist demonstrators clashed with German police. Twelve separatists and five police died, while 400 were wounded. Three weeks later, separatists declared an independent Rhenish government in Aachen, occupying municipal buildings with the implicit cooperation of the French. Following this declaration, separatists launched coups in Bonn, Mainz, and Trier (O'Riordan 2005). The French turned their attention to separatist movements in the Palatinate in early November. France gave up on their goal to separate the Rhineland only after the British intervened in 1923 (Willamson 1991, Pawley 2007). Between 1920 and 1924 the French and Belgian contingents made very clear that they were not interested in producing security for Germany. The security relationship turned more favorable for Germany in mid-1925. In June, the AAF and Germany renegotiated the terms of disarmament and reparations, leading to the Locarno Conference and resultant Locarno Pact (Willamson 1991, Pawley 2007). At Locarno, the Germans scored their greatest diplomatic victory since the conclusion of the Great War. In exchange for recognition of the Franco-German and Belgian-German borders (including Alsace-Lorraine) a permanent DMZ along the Rhine, a non-aggression pact with France and the United Kingdom, and peaceful arbitration 67 The gambit failed only when the British Committee on Clearing Banks refused to participate (Ibid). 189 agreements between Germany and France, Belgium, Poland and Czechoslovakia, the Allies agreed to pave the way for Germany?s induction into the League of Nations and evacuate the northern zone by the beginning of 1926. By January 30, 1926, the Allies evacuated Koln and the northern zone, returning the first section of occupied territory to German control (Kolb 1988, Snyder 1998). On July 29, 1926, Chancellor Stresemann was able to convince Britain?s Chamberlain to appeal to French Prime Minister Briand for a reduction in force levels in Germany. The French were only willing to concede small reductions at first, but when the chancellery transferred to Marx in the following months, he jumped on an opportunity to negotiate for the early withdrawal of all Allied forces. Efforts in February 1927 to achieve a one-third to one-half reduction in forces, and a play for a full withdrawal submitted in July 1928 were unsuccessful (Willamson 1991), but future governments continued to push for withdrawal. British public opinion, perhaps swayed by the German experience during the Ruhr and inflationary crises, played a key role in precipitating these changes. Even though French fears regarding a rearmed Germany remained typically strong, the British were facing domestic pressure to further reduce their commitment to the enforce Versailles (Willamson 1991, Snyder 1998). For the first time it was the Franco-Belgians?not the Brits?who were diplomatically isolated. The change in occupation policy, especially the security pact that resulted from Locarno, represented a shift in the marginal security value of the occupation. Finally, the occupying powers recognized Germany?s need for external security and took steps to provide it. However, the Locarno Pact did not take place until after the simmering of tensions between Germany and Poland, and after Germany had found a way to rearm in 190 secret. In other words, the demand for external security had largely subsided, while Germany had already found its supply (from Russia). This is not to say the offer was not welcome; Stresemann pushed hard for this arrangement. But the degree to which this agreement could produce marginal value is questionable, considering that the deficit the AAF had created during the first six years of the occupation drove the Weimar Republic to look for help elsewhere. Facilitating Regime Consolidation Occupation policy in the Rhineland and other AAF administered territories ranged from the inconsistent to the outright retaliatory. The inconsistency stems partially from the fact that the AAF had no postwar plan until September 1918, and the largely improvised occupation policy resulted in cooperation and coordination issues (Mommsen 1996). The occupation relationship in the case of Weimar Germany was probably the most destructive of the four cases in this dissertation. The ?War Guilt? clause of the Treaty of Versailles, which required Germany to accept full responsibility for the Great War, humiliated Germany and discredited the coalition that took Germany out of the war. Reparations sucked reserve capital out of the new republic while a punitive trade regime and regional autarky prevented capital from flowing back in, bankrupting the bourgeois and white-collar base of the optimal winning coalition. The AAF frequently, if unintentionally, empowered those outside the winning coalition while punishing those Table 5.3: Evidence of Facilitating Regime Consolidation in Weimar Germany Constructive ? Locarno Pact ? Dawes Plan Destructive ? ?War Guilt? Clause ? Reparations ? Ruhr occupation 191 within. While putting Germany in a weaker position internationally, the AAF crafted a domestic occupation policy that directly weakened the optimal winning coalition, reducing the value of the occupation to key democratic constituencies. The AAF rarely intervened to quell insurgencies on either the left or right that threatened the future of the Weimar regime, but they readily punished the new democratic regime when it attempted to preserve itself. The year 1920 saw both a radical leftist uprising in the Ruhr and the right-wing Kapp Putsch. The AAF ignored Kapp regime, leaving the German government to handle the situation itself. The occupation forces similarly twiddled their thumbs during the Ruhr uprising, although they considered occupying the demilitarized zone to quell violence. The AAF nevertheless remained inactive, and the uprising continued until Germany deployed Reichswehr regulars to pacify the region. This, however, was a direct violation of the Versailles Treaty, one that French forces punished by expanding the occupation to include Frankfurt, Homburg and three other cities while the British arrested Reichswehr troops that entered their zone (Willamson 1991, O'Riordan 2005, Pawley 2007). This was nothing if not a signal to those in charge that they lacked a mandate to protect the regime, and nothing if not a signal to anti-democratic constituencies that the occupation would stand aside while insurgents endeavored to destroy it. The more infamous occupation of the Ruhr occurred years later, on the pretense of Germany?s failure to pay reparations. Reparations ran counter to Wilson?s overtures, Prime Minister Lloyd George?s better judgment, and the advice of American economic advisors. Nevertheless, Wilson was willing to trade reparations for the League, especially if reparations would help the UK and France pay their war debt. The UK and US agreed 192 to reparations in April 1919 (Nelson 1975, Kolb 1988). The reparations issue was a thorn in the sides of the first Weimar governments. The AAF threatened sanctions against Germany if it did not accede to the reparations demands. Failing to heed these threats, on March 7, 1922, the Allies occupied Dusseldorf and two other German cities and imposed a 50% levy on German exports. On March 19, Germany nominally acceded to AAF demands (Kolb 1988, Nicholls 1991, Lauterbach 1944, O'Riordan 2005, Pawley 2007) but this did not modify their financial situation. On January 10, 1923, following several failed attempts to reach a compromise, the Reparations Commission declared Germany in default. The next day Franco-Belgian forces seized the Ruhr. Outraged, American forces in Germany withdrew in protest by the end of the month (unfortunately relieving the Weimar Republic of one its more amiable occupiers). While disapproving of the venture, the British felt the need to preserve what unity and credibility the AAF retained (Kolb 1988, Nicholls 1991, Willamson 1991, Mommsen 1996, Pawley 2007, Cornebise 1983). Eventually, Germany and the AAF were able to reach an agreement on reparations, and on other occupation issues, at Locarno and by way of the Dawes Plan. The Dawes plan restructured the German banking system to stabilize the Reichsmark and rescheduled reparations debt. On the heels of success at Locarno, Germany called for a more favorable revision of the Dawes plan. In February of 1929, as Germany was already beginning its slide into the Great Depression, the Allies reopened the issue of reparations at the Paris Conference. Germany caught a huge break on May 31, 1929, when the Labor Party took back the House of Commons on a pledge to end the occupation. Germany took advantage of the situation by linking reparations talks to disarmament discussions as they continued at The Hague on August 5-31. The product of these conferences, the first draft 193 of the Young Plan, was accepted in late August of that year. As per the Young Plan, all parties agreed to modify the reparations schedule to slowly increase the per annum payment from 1.7 billion to 2.5 billion gold marks through 1966, and thereafter reducing the payment rate until reparations were fulfilled in 1988 (Nicholls 1991). The occupation relationship was undeniably negative between Versailles and the end of the Ruhr occupation, but it took a term for the better at Locarno. The War Guilt clause did nothing but empower those who wanted to cast the pro-democratic government as traitors. Uneven treatment by the AAF further aided extremists against the optimal winning coalition. Reparations and the Ruhr Occupation further radicalized the population. The occupation relationship did not remain permanently negative?it improved significantly after Locarno and continued to do so until the departure of foreign troops in 1930?but early damage is lasting damage. As the numbers show, the optimal domestic coalition never entirely recovered from the poor relations of the occupation?s early years, leaving the Weimar Regime especially vulnerable when the Great Depression arrived. Domestic Outcomes The history of the optimal winning coalition in Weimar German breaks down conveniently into four different periods. The first period, preceding the signature of the Treaty of Versailles, is characterized by electoral successes for the optimal winning coalition but tainted by escalating political violence and the rise of paramilitaries on both sides of the spectrum. The second period, between Versailles and the Locarno agreement, 194 indicates a severe retrenchment where paramilitaries multiply, domestic security markedly deteriorates and the optimal winning coalition suffers at the polls. The third period, between Locarno and ending just before the Great Depression, witnessed a partial resurgence by the optimal winning coalition, a dramatic decline in political violence but a substantial increase in paramilitary enlistment. The fourth and final period, beginning just before the Great Depression and ending in Hitler?s seizure of power, saw the defeat of the optimal winning coalition, renewed political violence and the sustained dominance of paramilitary politics. Many of these trends confirm to what we would expect based on the hypotheses of the Marginal Value Model. The period immediately preceding the November 11, 1918 Armistice, leading up to the elections of January 19, 1919, characterized a phase of great promise for democracy in Germany. A republican coalition almost entirely comprised of enthusiastically democratic parties brought an end to four years of deprivation and, in so doing, won widespread electoral support. These early victories, however, were tainted by the onset of domestic strife. Moreover, anti-democratic factions maintained control over the domestic and foreign security apparatuses. In the first elections in Weimar Germany, the optimal winning coalition of Social Democrats (SPD), Democrats (DDP), and the Catholic Center (or Zentrum) won 76.1% of the popular vote and 78.3% of mandates to the National Assembly (Kolb 1988, Kaes, Jay and Dimendberg 1994, Hunt 1964). 195 Figure 5.2: Succession of Governments and Chancellors in Weimar Germany, 1918-1933 Sources: Eberhard Kolb (translated by P.S. Falla). The Weimar Republic. London. Unwin Hyman: 1988; A.J. Nicholls. Weimar and the Rise of Hitler. London. Macmillan: 1991. Jan '19 76.1% Jun '20 43.6% Dec '24 45.9% May '28 34.7% 28.3% 19.2% 13.1% 38.1% 18.3% 30.7% 29.4% 16.4%10.3% 17.2% 34.1% 29.9% Sep '30 42.2% Mar '33 64.4% 0% 20% 40% 60% 80% 100% 19 18 19 19 19 20 19 21 19 22 19 23 19 24 19 25 19 26 19 27 19 28 19 29 19 30 19 31 19 32 19 33 Figure 5.1: Performance of the Optimal and Suboptimal Coalitions in Federal Elections to the Reichstag, 1918-1933 Optimal Semi-optimal Suboptimal Turnout 196 The success of the optimal winning coalition is largely the result of a favorable shift in threat dynamics and an absence of information regarding the strategic setting. Leading up to this point, the German people had little knowledge of what effect regime change and occupation would have on their personal security. The only information they have about the occupation-constituent relationship is that they were no longer at war with the AAF. As far as the any political actor in the new republic was aware, Germany and the AAF had entered into negotiations on presumably equal terms to establish a lasting, Wilsonian peace. The greatest threat to the average German during this period was domestic unrest emanating from the left and the right, as communist revolutionaries faced off against the right-wing Freikorps and associated paramilitaries. Temporarily, however, the Freikorps were allied with the government against the radical socialists. German domestic instability was so obvious even outside of the Reich that Lloyd George, along with then Munitions Minister Winston Churchill, considered arming German forces against the radical left. Their concern only grew as economic deprivation spread throughout the Reich. Radical USPD partisans and Communists (KPD) were the primary source of political violence early on, spurring a Christmas revolt that exposed the SPD?s tenuous hold on power. The SPD attempted to employ soldiers returning from the Western Front to maintain order, but many of these war-weary individuals disobeyed commands and/or broke rank (Willamson 1991). Into this security vacuum stepped private or semi-private paramilitaries, including the Freikorps volunteers, civil guards (Einwohnerwehr), militarized veterans associations such as the Stalhelm, and volunteer auxiliaries (Zeitfreiwilligenverbande; Diehl 1977, 197 Snyder 1998). These forces grew quickly, fed by a glut of demobilized servicemen. Volunteer auxiliary forces numbered 15,000 by February 1920, while the Freikorps grew into a 200,000-strong corps by that summer. Many paramilitaries honed their skills during the revolts of late December 1919 and early January 1920 (Diehl 1977). Since the Freikorps were loosely aligned with the optimal winning coalition, a vote for the SPD, DDP or Zentrum represented a vote for stability, anti-Bolshevism, and peace. As a result, the optimal coalition enjoyed a great deal of support from an historically anti-republican, anti-Communist, but also pro-peace peasantry who had borne the brunt of costs of the Great War.68 But the strength of the optimal coalition was not distributed evenly across political economic institutions. The constituencies that supported the German People?s Party (DVP) and the German Nationalists (DNVP) controlled a disproportionate share of landed wealth and capital (Heberle 1943, Van Riel and Schram 1993). Moreover, just as with the military under the Kaiserreich, the paramilitaries were by-and-large nationalist and anti-republican. Regardless, these forces were the first short-term policing options for a newly empowered winning coalition of Socialist, Nationalist and Catholic Democrats. Ideological considerations were not paramount during this period of widespread instability (Nicholls 1991), but this measure of expediency placed the legitimate use of force in the hands of a suboptimal coalition. Support for the SPD evaporated following the signature of the Treaty of Versailles. Only the SPD favored signature of the Versailles Treaty outright, while the Zentrum and DDP favored a signature with reservations (a symbolic distinction at best). 68 For instance, Bavaria suffered a relatively high number of casualties due to the fact that a majority (65%) of the population was peasantry, generally less educated and far more likely to produce infantrymen than officers (Heberle 1943). 198 (Schiff 1930, Dawson 1933, Third Army 1943, Kolb 1988, Nicholls 1991, Willamson 1991, Diehl 1977, Pawley 2007).69 The first elections following the Treaty of Versailles are notable for the string of political defeats for the optimal winning coalition in general, and the SPD in particular. Support for the SPD-DDP-Zentrum coalition plummeted from a sizeable majority to 43.6% of the popular vote (44.7% of Reichstag mandates; Kolb 1988). For the first time since the abdication of the Kaiser, the SPD was excluded from the cabinet, replaced by the semi-optimal German People?s Party, or DVP (Eyck 1970). Left-wing insurgency and right-wing assassinations (of Rosa Luxemburg, Karl Liebnicht and Kurt Eisner, leader of the November 1918 Bavarian independence movement) characterized the extra-parliamentary political climate. In early March 1919, a failed leftist coup precipitated the clash of 15,000 demonstrators and 42,000 Freikorps and Civil Guard units. A month later Communists seized M?nchen, initiating a standoff during which KPT forces executed hostages. The Freikrops response on May 2 resulted in the slaying of 600 suspected agitators. Early the next year, a left-wing demonstration on the steps of the Reichstag devolves into a riot, leaving 42 dead. In July 1919 Chief Papst, leader of the Einwohnerwehr (civil guards) launched an aborted putsch against Berlin. On March 12, 1920, journalist Wolfgang Kapp was more successful, seizing control of Berlin and forcing the government to retreat to Stuttgart. The coup would cost Defense Minister Noske and Reichswehr Chief Reinhardt their positions. Inauspiciously, both positions were filled by right wing anti-republicans Gessler and Seeckt. This marked the consolidation of control over the official security apparatus by actors outside of the 69 The United States, due to opposition from unilateralist forces in the Senate, ratified a separate peace with the Germans on October 19, officially concluding hostilities between the United States and the Central Powers. (Dawson 1933, Cornebise 1983, Kolb 1988). 199 optimal domestic coalition, where it would remain until the collapse of the republic (Diehl 1977, Cornebise 1983, Kolb 1988, Nicholls 1991, Willamson 1991, Mommsen 1996, Snyder 1998, Halbrook 2000, O'Riordan 2005, Pawley 2007). Domestic unrest tracked the growth of paramilitary forces. The Freikorps numbered 200,000-400,000 by the end of 1919, and as its power increased began acting more as a free agent than an instrument of the state. In March-April 1919 the Prussian Interior Ministry ordered the formation of class-agnostic, non-partisan ?republican? civil guards in an attempt to head off a right-wing monopoly over the use of force. The March 6, 1919 Reichswehr Law failed to spur the assimilation of paramilitary units. Now illegal, the Freikorps and other paramilitaries went underground, while the Stalhelm, a private military force disguised as a veterans association, exploited its particular loophole. In fact, the reputation of Versailles was very profitable to veterans groups: the Stalhelm openly recruited those who blamed the Treaty?and those who concluded its terms?for their hardships. Openly defiant, they declared rearmament and Lebensraum as their main aspirations. Similarly, the Bayerische (Bavarian) Einwohnerwehr grew from 100,000 before Versailles to a quarter million thereafter (Willamson 1991). At a strategy session of Bavarian Guard leadership, delegates characterized their organization as the ?vanguard of the Fatherland? whose goal was to protect young Germans against capitalist decadence, communist immorality, and democratic corruption. Such rhetoric presaged future Nationalist Socialist (NSDAP) propaganda. Coincidentally, the SA (the Nazi Party?s paramilitary wing) began recruiting members in May 1921. (Diehl 1977, Nicholls 1991, Mommsen 1996, Berman 1997). Overall, the disarmament mandate did not 200 significantly reduce German militarism; rather it shifted control over the use of violence from the central government to private actors. By September 1923, unrest in Saxony and rumblings of a right-wing coup originating from Bavaria prompted Chancellor Stresemann and Reichswehr Chief M?ller to declare a state of emergency and grant absolute authority to Seeckt. Saxony?s President Zeigner, a leftist leader of a leftist state, called for the formation of left-wing working class volunteer auxiliaries as a defense against the increasingly strong anti-republican, anti-socialist paramilitaries. While the KPD responded to Zeigner?s call and set about forming paramilitaries that they called ?Proletarian Hundreds,? the SPD refused to cooperate in this measure out of fear that Zeigner was attempting to initiate a left-wing coup. Without the support of the moderate Social Democrats, the call to arms only succeeded in empowering extreme left-wing anti-government forces in Saxony. The KPD took advantage of their newfound strength to launch statewide communist revolt. Seekct quickly deposed the Saxon cabinet and appointed a minister from the DVP to establish a new government in the state. Simultaneously, Seeckt faced pressure from his allies in the business community and the former Junker nobility to lead his own anti-labor, anti- socialist, authoritarian coup. While Seeckt continued to profess loyalty to the constitution, privately he referred to the Weimar Constitution as ?contrary to [his] political thinking? (Hunt 1964, Nelson 1975, Mommsen 1996). The passionate but fractious nature of Bavarian ultra-nationalism precipitated one of the more peculiar episodes in the history of the Weimar Republic. As Seeckt carefully weighed his options, Munich witnessed the rise of the Nationalist Socialists (NSDAP) and its SA paramilitary. NSDAP leader Adolf Hitler, with the cooperation of war hero 201 Erich von Ludendorff, outflanked the DNVP and Seeckt with a plan to seize Munich and then march to Berlin, determined to take control of what they believed was an impending revolution. Nationalists close to Seeckt assumed total political control over the Bavarian division of the Reichswehr, presumably in preparation for the final Putsch, but Hitler moved first (Diehl 1977, Mommsen 1996). On November 8, Hitler, Ludendorff, and 600 SA troops launched a putsch in a Munich Beer Hall, surrounding the hall during an event hosted by Munich public officials. The attempted coup quickly and dramatically collapsed when Seeckt, more disdainful of Hitler than the Constitution, ordered his troops to decisively end the Nazi demonstration. After the resulting clash, 16 NSDAP and 3 police laid dead. Hitler escaped, but was later found and sentenced to five years in prison for treason. He used his trial as a platform for his ideals, and was released within months (Kolb 1988, Nicholls 1991, Mommsen 1996). Justifiably worried that the security forces were becoming overrun by middle- and upper-class anti-republican elements, the SPD urged working-class members to join the Civil Guards. The few who tried were typically turned away. The problem was not a lack of interest, but the consolidation of right-wing nationalist influence in the institutional structure of security (Diehl 1977). After 1923, moderate republican parties reversed their position on the use of paramilitaries. While strategically this was an appropriate and belated move on behalf of the parties that constituted the optimal domestic coalition, it also serves as an indication that the German state had conceded its monopoly over political violence, which itself is an indicator that the range of acceptable political contestation had not been narrowed. 202 The failure of one extra-parliamentary movement was a serious one for the optimal domestic coalition. In September 1923, the end of passive resistance against the Ruhr occupation spelled the end for Chancellor Stresemann, a member of the republican wing of the DVP. The perceived failure of passive resistance, coupled with the dire threat of violent political competition from the right and extreme left, led to the collapse of the first Stresemann cabinet and the continued shift of the cabinet towards the anti- democratic right-wing side of the ideological scale (A. Cornebise 1982, Kolb 1988, Willamson 1991, Mommsen 1996, O'Riordan 2005, Pawley 2007). The optimal winning coalition would not go down without a fight, however. In February 1924, the SPD, DDP and Zentrum caught on to the benefits of organizing a loyal paramilitary force. The ideal parties united to form the Reichsbanner, a paramilitary force pledged to defend the republic against radical elements. Membership grew to over 3 million by early 1925, but the Reichsbanner never became the broad-based pro- republican coalition the SPD hoped it would be. The private corps never enjoyed strong support within the Zentrum, where right-wing Catholics opposed the idea of a police force under the control of left-wing elements. Moreover, the paramilitary suffered a major public relations blow on July 23, 1927 when its most visible member, Chancellor Marx, resigned in order to placate the far-right DNVP members of his ruling coalition (Diehl 1977, Nicholls 1991). Meanwhile, the years 1925-1927 were relatively calm in terms of anti-government violence. During this period, the death of President Ebert on February 28, 1925 triggered the first-ever German presidential election. The parties of the optimal coalition nominated former Chancellor Marx of the Zentrum. The right and far-right nominated Paul von 203 Hindenburg, noted anti-republican and originator of the ?Stab in the Back? myth that blamed socialist agitation for Germany?s military collapse (Kaes, Jay and Dimendberg 1994). In two rounds of elections, Hindenburg defeated Marx (Diehl 1977, Mommsen 1996). The Hindenburg victory funneled the frustration of the whole German constituency into a single powerful office, one with broad emergency power that would have a very important role to play when the economy collapsed years later. In the meantime, however, center and center-left groups were very successful in the May 1928 election: the optimal coalition improved their popular share of the Reichstag vote to 46.8% and the SPD returned to the ruling coalition for the first time since 1923 (Eyck 1970, Kolb, The Weimar Republic 1988). At the same time, the optimal coalition finally began to compete with the right and far right in terms of paramilitary politics, building by far the largest semi-private army during the course of the Weimar Republic. Still, between the end of the war and the beginning of the Great Depression, 4-5 million white- collar workers became blue-collar workers. These ?d?class?s,? politically disillusioned and economically disenfranchised, seriously weakened the constituency of the optimal winning coalition in economic and numerical terms (Lauterbach 1944, O'Riordan 2005). The 1928 elections marked the last hurrah for the optimal winning coalition, as the Republic suffered two blows on the individual level in December 1928 and October 1929. First, former Chancellor Marx retired from the Zentrum, leaving the seat to a Catholic priest who was more concerned about forwarding Catholic interests and less about republicanism. Second, Foreign Minister Gustav Stresemann, once a staunch defender of republicanism in a moderate authoritarian party, died (Nicholls 1991). After this point, the Zentrum essentially ceased to exist as a member of the optimal winning 204 coalition. These events, combined with the decline of the DDP, and the radicalization of the electorate at the beginning of the Great Depression, left the SPD as the lone protector of the republic. The coalition cabinet that saw the return of the SPD to government and the chancellery would not survive the next election cycle. Resulting from a split over the Young Plan (the US-led deal on reparations), plans to rebuild the German navy, and agrarian aid to eastern Germany, the SPD resigned from the cabinet for the last time. A minority parliament under Chancellor Br?ning, appointed by President Hindenburg, marked the first of several minority governments that would govern until Hitler?s ascension to power. The first Depression-era election in 1930 set the stage for the debut of the revamped NSDAP, the party that Hitler had transformed in the wake of the failed Munich Putsch. In what can only be considered one of the most unfortunate political comeback stories of all time, the NSDAP?s share of seats in the Reichstag grew from 2.4% in May 1928, to 18.5% in September 1930, to 37.8% in July 1932, trailing only the SPD and outgaining every other radical nationalist party (those to the right of the DVP) combined (Stachura 1993, Mommsen 1996, Snyder 1998, Kolb 1988). Hitler attempted to parlay his recent success into a victory in the president. The second presidential elections of the Weimar Republic took place over two rounds in March and April in the depths of the Great Depression. In contrast to the 1925 poll, the SPD, Zentrum, BVP and DVP all supported Hindenburg, fearful of the possibility of a Hitler presidency. The far right split their vote between the DNVP?s D?sterberg and the NSDAP?s Hitler, with Hitler finishing second. Indicative of how much even the far-right feared the Nazi candidate, D?sterberg threw his support to Hindenburg instead of Hitler. Hindenburg won, a small victory for 205 the only ideal parties remaining. Of course, Hindenburg certainly was no friend of the republic, and at 84 had very little fight left in him for a political system he disdained in the first place. Despite the fear-based vote that elected Hindenburg, his reelection as president on April 10, 1932 did nothing to stem the rising tide of ultra-nationalism. The NSDAP now had the support of, or control of, the largest agrarian and corporate civic associations, including the Reichslandbund (German Farmers? Association) and various German industrial firms (Nicholls 1991, Berman 1997, Snyder 1998). At this point, landslide victories on the state and local level had become so commonplace that in the special election on November 6, the loss of four percentage points encouraged Hitler that continuing with a purely electoral policy was too risky. Early in 1933, banking magnate and NSDAP financial contributor Baron von Schroeder arranged a meeting between the recently deposed Chancellor Papen and Hitler, with the goal of gaining political support for a Hitler Chancellery. Papen subsequently met with President Hindenburg, who was unsatisfied with the performance of then-Chancellor Scheicher, the man who had previously convinced Hindenburg to appoint Papen as the head of a minority Reichstag, as well as the man who subsequently convinced Hindenburg to depose him. Papen and Hindenburg were wrongly operating under the assumption that bringing Hitler into a minority government in coalition with the DNVP would allow the Nationalists to control the Nazi party and moderate their leader?s influence (Snyder, Encyclopedia of the Third Reich 1998). Hindenburg appointed Hitler to the chancellery on January 30, 1933. The events that followed, including the Reichstag Fire and the March 5 election, in which the NSDAP won 44.5% of all seats, provided Hitler with the legitimacy he required to 206 complete his revolution. The election presented an opportunity for the NSDAP and DNVP to form the first majority government coalition since the fall of the Great Coalition. Instead, Hitler persuaded the Reichstag to pass the Enabling Law, making permanent the various states of emergency that previous chancellors had declared in order to govern without a majority cabinet, and thus allowing Hitler to rule permanently by decree. By the time he made his move, Hitler had the Reichstag delegates of the KPD arrested as part of the investigation of the Reichstag Fire, while the Catholics Center was too busy protecting their own to challenge the SA, the center and moderate right had ceased to exist, and the SPD only held 18.5% of the seats in the Reichstag?enough for second best, but not enough to unilaterally block the two-thirds quorum or the two-thirds vote required to pass an extra-constitutional measure. On March 23, 1933, as the SPD casted the only dissenting votes against the Enabling Law, the passage of which completed Hitler?s ?legal coup? (Nicholls 1991, Snyder 1998, D. F. Lindenfeld 1999, Halbrook 2000). In the three elections following the advent of the Great Depression, the optimal coalition saw its share decline from 28.3% in 1930 to 19.2% in 1933. Not a single majority formed a cabinet during the period, with the appointed chancellery passing from the right-centrist Br?ning to anti-republicans Papen, Schleicher and Hitler. Even though the SPD enjoyed a massive advantage in manpower with the Reichsbanner, the SPD never did mobilize this force to prevent Hitler?s victory (Nicholls 1991). Moreover, the Presidential elections of 1932 indicated how far to the right the individual parties had shifted. In 1926, the republicans supported Marx against Hindenburg, the latter of whom 207 was sponsored by the BVP, DVP and DNVP. In 1932, the SPD, Zentrum, BVP and DVP all supported Hindenburg (Kolb 1988). The radicalization of the once-moderate Weimar Republic was primarily due to the decline of the once-strong optimal domestic coalition, now weak electorally, financially and militarily. So weakened, pro-republican coalitions were less prepared to repel radical threats as they had during the Great Inflation, or as their counterparts in the US, UK or France were able to during the Great Depression. Ultimately, the ultranationalist forces that tried to take the government by force in 1923 seized power through perfectly legal means one decade hence. They were empowered largely by the actions of the occupying forces. The AAF weakened Germany without offering a credible security guarantee. By doing so, they undermined the position of the parties that brought an end to the war and allowed the occupation to occur. The optimal winning coalition was further discredited by the punitive peace they signed at Versailles. All the while, occupation policy impoverished the white collar and bourgeois elements that could have bolstered the winning coalition, and the occupiers looked the other way as radical elements outflanked the government on the left and right. The damage was complete by the time the occupation reversed course. The winning coalition recovered, though never to the levels they enjoyed before Versailles, and not to such a point where they could reinforce democratic institutions to weather the oncoming tsunami of a global economic collapse. The Great Depression was the final wave that eradicated the rotting structure of the Weimar Constitution, but the structure collapsed primarily due to the rot that set in soon after the edifice was dedicated?in the early days of uncertainty that followed World War I. 208 Summation The case of Weimar Germany is a very clear example of a failed democratic consolidation under an occupied regime. The first Federal Republic of Germany was a marginal democracy following the end of the Kaiserreich and until the rise of Hitler, scoring a 6 on the Polity IV scale between 1919 and 1932. With the rise of Adolf Hitler, however, Germany slides into autocracy (score: -9) and never recovers until the resurrection of a unified, democratic German state in 1990 (Marshall and Jaggers 2009). Moreover, frequent instances of domestic political violence and the dominance of domestic security by paramilitary organizations paint a picture of a regime that failed to establish a monopoly over the use of force. The early plunge of the prospects of the optimal winning coalition in the first elections following the Treaty of Versailles, and their slow decline thereafter, reinforce the fact that the regime never successfully consolidated. Beginning with the Treaty of Versailles, continuing during the separatist violence in the west, the insurgency in the east, during the battle over reparations, and concluding in the occupation of the Ruhr, the occupation forces established themselves as enemies?not allies?of the occupied state. Under such conditions, it was not possible for an optimal winning coalition?responsible for concluding the Armistice and originally in power directly as a result of Germany?s exit from the war?to succeed. 209 Reconciliation of Competing Hypotheses As in the previous cases, the findings support H1. At Locarno, the AAF pledged a mutual security pact with Germany. More importantly, Locarno marked the end of the most punitive aspects of the Versailles peace, including French sponsorship of Rhenish separatists and Polish insurgents, unreasonable demands on reparations, and the occupation of the Ruhr. During this period, we do see some decline in events indicating regime instability. The period between Versailles and Locarno was witness to some of the worst paramilitary violence and major losses for the optimal winning coalition at the polls. On the other hand, the pre-Versailles period saw the biggest win by the optimal winning coalition: the SPD, DDP and Zentrum comprised a 76% majority. There is still significant political violence during this period, and the regime lost its hold on power after the onset of the Great Depression. But the Great Depression, the return of domestic political violence and the swift collapse of the optimal winning coalition all occurred after the occupation had ended. Following Locarno, the optimal winning coalition saw a small boost in its electoral fortunes and won significant victories in the Reichstag? taking back the chancellery from the DVP and holding onto it through March of 1930. There was in fact a home for regime consolidation following the improvements in the occupation relationship and alliance configurations following Locarno through the end of the occupation. Therefore, this case fails to reject the null hypothesis and lends support for H1. Furthermore, events indicating regime instability were indeed more common while the powers detracted from the domestic wellbeing of occupied Germany. The optimal winning coalition absorbed its harshest electoral blow between Versailles and the 210 end of the Ruhr occupation. The 76% majority collapsed into a 44% plurality in 1920 and briefly fell behind the suboptimal coalition in the pre-Locarno 1924 election. The semi- optimal DVP took its first seats in the cabinet during this time, and the chancellery fell to Stresemann (himself a moderate, but the representative of the immoderate DVP). This period also witnessed the worst domestic insecurity, including several putsch attempts and an explosive increase in the rate of paramilitary recruitment. Finally, events indicating regime instability?paramilitary expansion, coup attempts, assassinations and poor showings at the polls by the OWC?were most common when the AAF aligned itself with the enemies of the Weimar state, including Rhenish separatists and Polish insurgents. The case study does not directly reject any of the competing hypotheses, but they do not explain anything the Marginal Value Model cannot. The Reichstag was indeed deadlocked and incapable of responding to either the Great Depression or its precursor, the Great Inflation?until the AAF stepped in. This, however, highlights the importance of positive moves by the occupying forces to resolve domestic crises that threaten democratic regimes. While the Dawes plan may have been politically infeasible before 1924, it was not technically infeasible. Far from it, in fact?the US, UK and France would have been far more capable of monitoring the Reichsbank than it was in monitoring disarmament (if only because markets are very bad at keeping secrets). Moreover, while Article 48 was the instrument by which Hitler conquered German democracy, it was wielded by a reichpresident who rose to power directly as a result of his rhetorical attacks on the occupying forces and the treaty they presented to a 211 government comprised solely of parties from the optimal winning coalition.70 While the Germans did rearm in secret, indicating a failure of disarmament policy, that they did so indicates a demand for external security that the allies themselves could have fulfilled. Instead, the young Union of Soviet Socialist Republics supplied these goods. There is no better indicator that the AAF failed to establish itself as an ally than the fact that Germany?whose fear of a Bolshevik revolution within its borders persisted long enough for Hitler to exploit it?had to look to Soviet Russia for help. Finally, the new regime may have ultimately failed due to the residual Junker class, but it was on weak footing long before the unholy alliance of farmers associations and urban industrialists succeeded in putting Hitler in charge. The Marginal Value Model subsumes these competing political hypotheses, and it is also compatible with competing economic explanations. Economic Drivers The dire ideological and security threats to German Democracy provided a backdrop to the collapse of the German monetary system and the subsequent Great Inflation. From January 1919 through November 1923, the exchange rate of the Reichsmark shrunk by a factor of 500 billion. Germany officially entered into a period of hyperinflation (a minimum 50% devaluation per month; Cagan 1956) in January 1922 and did not emerge from it until late 1923. While Reichsbank?s purely nominal independence from the central government was the primary cause, it occurred in the context of the cession of Germany?s most productive territories, a global decrease in trade 70 To paraphrase Dr. Moore, ?no ?stab-in-the-back,? no fascism.? 212 openness (and thus the ability of the state to balance its capital accounts by means other than domestic production) and the imposition of reparations. The Great Inflation eradicated the value of deposits and debt owned, hobbling the financial sector. Millions in the managerial class, petit bourgeoisie and civil service were cast down into the ranks of the working class and the unemployed. While unemployment among trade unionists had plunged from 6.6% to 1% during moderate inflation that followed the war, trade union unemployment shot up to 28.2% during the period of hyperinflation as incentives for investment collapsed. While currency speculators and mortgage holders profited, inflation crippled most of the economy (Lauterbach 1944, Hill, Butler and Lorenzen 1977, Berman 1997, Voth 1995, Voth 2003). It was not until August 1923 that the central bank took significant steps to ameliorate the currency crisis, freezing the government?s line of credit. On October 15, the government founded a new central independent bank, the Rentenbank, which would issue new currency to replace the Reichsmark. On November 15, the Rentenmark officially supplanted the Reichsmark and began trading at a rate of one trillion to the US dollar. The Rentenbank established its independence immediately, refusing to issue a loan to the German government the following month. Other difficult but necessary decisions, such as the dismissal of large numbers of public employees, helped Germany balance its budget by March 1924 (Webb 1986, Willamson 1991, Mommsen 1996, O'Riordan 2005). Unfortunately, the process of currency stabilization both exacerbated and crystallized the political consequences of the Great Inflation. The new exchange rate locked in the losses of the middle class and most former members of the bourgeoisie. This weakened the economic strength of the optimal coalition (Hill, Butler and Lorenzen 213 1977) and gave credence to nationalist, anti-capitalist and anti-democratic rhetoric blaming a cabal of republicans and bankers for bankrupting Germany. In summary, the Great Inflation weakened the economic strength of pro-democracy forces and improved the rhetorical appeal of nationalism, all of which informed the broader constituency of the negative security value of the occupation and those who had caused it to occur. During the Ruhr occupation, the US and UK agreed to restructure wartime debt, fixing UK liabilities at $4.6 billion.71 Chancellor Stresemann and Konrad Adenauer, then senior mayor of Koln, successfully appealed (on national security grounds) for talks that would eventually lead to the Dawes Plan. US Treasury Secretary Dawes devised a plan to balance the budget, cut payments to 50% the initial rate, issued a loan to cover German currency, balanced fiscal and monetary policy through austerity measures and bank independence, to discount occupation costs from reparation liabilities, and ensured that the Ruhr occupation would wind down as soon as Germany passed the legislation implementing the agreements. The Dawes Plan took effect on September 1, 1924 (Kolb 1988, Nicholls 1991, Willamson 1991, Snyder 1998, Pawley 2007). The positive effects of currency stabilization and fiscal sanity were quickly apparent. A balanced budget, stable exchange and real interest rates drove a stock market boom in Germany and a decline in unemployment. It would not be too much of a stretch for the casual observer to believe the German republic had finally turned the corner. On the contrary, improvements in basic economic indicators masked fundamental weakness in the German economy. The number of people on unemployment insurance through 1928 hovered between 9-18%?an improvement but still a far cry from post-WWI levels. 71 The US and France would not reach a similar agreement until April 29, 1926, when the two parties agreed to fix liabilities at $4 billion (Kolb 1988). 214 Assets backing the currency declined from 85% to 65% through 1927 as the central bank vacillated between policies meant to bolster securities and those meant to prevent capital flight (ultimately fulfilling neither goal). The illusive nature of the recovery revealed itself in 1928. While economic growth among most industrial democracies peaked around the time of the Wall Street Crash in October 1929, Germany?s slide into the Great Depression began in late 1928 (Voth 1995, Voth 2003). Persistent unemployment, asset liquidation and durable declines in the standard of living meant that the middle class was unable to recover before the Great Depression set in. The advent of the Great Depression multiplied the gains that the NSDAP began to register in 1928. In the last years of the republic (1928-1932), unemployment rose from 9% to 33%, agricultural prices collapsed by 50%, and DANAT Bank, a preeminent banking institution in Germany and Austria, collapsed. At the same time, gridlock in the Reichstag and the rapid rise in unemployment led Berlin to severely cut unemployment insurance, the percentage of unemployed receiving benefits plummeting from 83.2% to 37.8%. The rising unemployment that drove so many disenfranchised voters to the NSDAP rolls began to subside just after the Nazi seizure of power in January 1933, further legitimizing the authoritarian regime (Kolb 1988, Stachura 1993, Van Riel and Schram 1993, Nicholls 1991, Stachura 1993, Nicholls 1991, Stogbauer, 2001, Voth, Hans-Joachim 2003, Stogbauer and Komlos 2004). The Great Inflation may have permanently weakened the economic strength of the optimal coalition?s prime middle-class constituency. We cannot, however, blame this effect solely on exogenous economic trends; the AAF was not entirely innocent in this disaster. The occupying coalition seized territories responsible for a disproportionately 215 large share of Germany?s industrial capacity, and also a small share of the Reich?s labor force (Lauterbach 1944, Dawson 1933). In an open economy, this might have been remedied through trade flows and internationalized production, but international trade was shrinking rather than growing as industrialized European states transitioned to an autarkic political economy (Frieden 2007). During the lowest point of the inflationary crisis, France and Belgium seized the Ruhr, another major source of German industrial production. All the while, the occupying coalition refused to restructure their debt obligations and thus their reparation demands, providing assistance to Germany only after the domestic security environment became so dire that American interests (who themselves had withdrawn from the occupying coalition) intervened, primarily to ensure payment of Anglo-French war debt. Finally, while the Dawes Plan stabilized the situation, the incumbent reflationary policy locked in the losses of the German middle class along with the gains of the more radical Junker landowners. There is no doubt that the Great Inflation was an economic disaster. However, it was a man-made disaster that largely stemmed from the policies imposed by the occupying forces, notably reparations, punitive tariffs and the seizure of productive lands. This did not go unnoticed among the electorate?nationalists and ultranationalists considered the occupation, the Great Inflation and the ?stab in the back? as intimately related issues (von Hindenburg 1994, Mommsen 1996, Nicholls 1991), and used the strife as artillery against moderate pro-democracy parties. Perhaps if the occupying coalition was unwilling to produce national security, they could at least have won support for the optimal winning coalition by producing economic and human security. The life raft arrived, but only after a large segment of the pro-democracy constituency drowned. 216 In this environment it would have been impossible for a foreign occupier to successfully narrow the spectrum of acceptable political contestation. Just the opposite, in fact: the AAF essentially lopped the top off of the bell curve. Summation of Findings and Conclusion The advent of democracy in Germany was far from hopeless. In the first two windows discussed here, the optimal coalition was strong electorally, albeit weak financially and disorganized militarily. From 1918 through 1925, the devastation of war, the decline of German territorial and military resources, and the Polish threat created the political space in which the occupying coalition could have consolidated the strength of the pro-fulfillment, and thus pro-democracy, elements. Unfortunately for German democrats, France and Belgium used their position of strength to further reduce the security of the German people, attempting to gain additional territorial cessions while aiding the Poles in their insurgency in Silesia and East Prussia. The other members of the occupying coalition, the United States and United Kingdom, were accommodating, both towards Germany but also towards France and Belgium. By the time relations between the Allies and Germany began to improve, the opportunity of the Polish threat had subsided, and the information produced by the Allied forces and their treatment of Germany had established detrimental beliefs among German constituencies regarding the value of supporting pro-fulfillment, pro-democratic coalitions. With the decline in the Polish threat, and with the assistance of the Soviet Union in secretly re-arming Germany, the Allies forfeited their chance to serve as allies to a 217 democratic Reich. Instead, the negative reputation of the AAF and the uneven economic recovery ensured that while improving relations would yield some support for the optimal domestic coalition, this support was insufficient to help democratic interests regroup before the next disaster, far greater in magnitude. Aided by these politico-economic factors, along with serendipitous historical coincidences (such as the postponing of presidential elections from 1930-1932), anti-democratic factions leveraged the radicalization of German politics into a takeover of her democratic institutions. The lesson to take from this history is that no single event is solely responsible for the rise of Hitler and the fall of the Weimar Republic. Rather, the confluence of the Great Depression, the failure of the Ruhr Occupation, the Great Inflation, the harsh and inconsistent enforcement and promulgation of the Versailles peace, and the institutional choices made early on in the German revolution brought Hitler to power. In other words, while the Great Depression was the straw that broke the camel?s back, it was the conduct of the AAF that conferred the majority of the burden. The security configuration during the occupation period neither helped nor hindered the occupiers and the democratic coalition from an objective standpoint. What it did was create an opportunity for the AAF to serve as an ally to Germany against rising powers in Eastern Europe. By exploiting these opportunities, the AAF could have consolidated enduring political power within the optimal domestic coalition. Instead, by disregarding them the AAF permanently established itself as an enemy to the optimal coalition. The elimination of military employment exacerbated the labor economy and fueled the rise of extra-parliamentary violence emanating from paramilitary politics. The 218 resultant decline in security led Germany to seek a protector. Ideally, this would have been the AAF. Instead, the Allies pushed Germany into the arms of the Soviet Union. In other words, weakening Germany was not poor strategy in and of itself. Indeed, by heeding Churchill?s advice, the AAF could have built a weak and willing Germany that was dependent outside military assistance (like post-WWII Germany and Japan). However, instead of aiding the Germans against a Polish-sponsored insurgency that disproportionately harmed Bavarian and Silesian agrarian interests, the French and Belgians often intervened on behalf of the insurgents while American and English forces stood mute. Germany found it had two enemies, the original French rival and now Poland. The Bavarians found they had three: the French, the Poles, and the left-center government that continued to cooperate with them. As a result, the Germans turned to the Soviet Union for security, while the Bavarians threw their lot in with the Freikorps, the Stalhelm and, later, the SA. The AAF could have taken leadership over the German economic system early in the process. Conversely, they could have completely avoided any interference in German economic affairs. Instead, the AAF dictated its economic demands to the new government?the military and diplomatic officials overriding their own economic advisors?and left the new government to find a way to meet those demands on their own, with no assistance and hardly any guidance until the situation had reached the level of political, economic and humanitarian crisis. The result was constant fluctuation in interest and exchange rates, reducing the informational content of market transactions and thus the supply of information about returns on future investment. By failing to clarify 219 the strategic setting for investment, the AAF failed to establish a favorable environment for the economic betterment of the optimal domestic coalition. 220 Chapter 6: South Korea, 1945-1950 Introduction The case of the attempted democratization of the American occupation zone in Korea is the most unqualified failure depicted in this dissertation. While the Weimar Republic may have collapsed more dramatically and with the worst of repercussions, the first German republic at least enjoyed a full decade of democracy. The US and USSR jointly occupied the Korea following WWII, with the US military government taking control on September 8, 1945 (Liem 1949, Park 2002). Despite half-hearted talks on establishing a uniform policy with an aim to restoring a sovereign, united Korea, neither the Soviets nor the Americans successfully exercised authority on each other?s side of the 38th Parallel. Thus, in the case of South Korea, the US military government remained the only relevant occupying power until the occupation ended on August 15, 1948 (Croissant 2001). During the occupation, the official policy of the United States was to establish a democratic government in Seoul, regardless of whether or not said government would attain control over the entirety of the peninsula. Regardless, the military government often seemed to subordinate democracy promotion to repressing left-wing radicals (often at the expense of the center-left) and expediting the process of withdrawal. The first Republic of Korea was stillborn?a republic in name only. South Korea?s first government was replete with the trappings of democracy but lacking in some of the most basic democratic characteristics, including open recruitment, respect for civil rights and adherence to the rule of law. Thus, the period of observation extends from 221 1945 only through 1950?the first year that data sets such as Polity IV consider South Korea a sovereign state. I shall present evidence demonstrating that, during this period, the case of the occupation of South Korea confirms the intra-case working hypotheses put forward in Chapter 2. I shall also explain how competing hypotheses and endogenous economic variables fail to explain the whole of the case. Rival Hypotheses versus Case History Rival Hypothesis: Leadership Failure Channing Liem (1949) predicted the failure of the American occupation of Korea before it actually concluded. Citing a combination of negligence and inconsistency on the part of the occupation authorities and Washington, Liem anticipated the monopolization of South Korean politics by Syngman Rhee and the deterioration of the democratic institutions that the AMG imposed on the territory. Liem correctly cites the lack of planning that went into the Korean occupation before the end of WWII. Unlike the substantial attention that the US paid to postwar planning in Germany and Japan, Korea was an afterthought at best. This lack of planning left US policymakers woefully undereducated about the preferences, politics and complex socioeconomic situation of the society that they were about to engage. Truman, the Departments of War (later Defense) and State, MacArthur and Hodge all consistently underestimated the amount of time and energy required to stabilize the zone economically and politically. Despite American claims that the Koreans would get to choose their own leader, Liem accuses Hodge of 222 foisting Syngman Rhee?an expatriate who rode out the Japanese occupation in Washington, DC, and who had been disavowed by the Korean government-in-exile (Cumings 1981)?on the Korean people. In doing so, Hodge failed to lend his support to moderates that would have acted as better custodians of nascent democratic institutions. Bruce Cumings (1981) makes a related claim, blaming Hodge?s single-minded campaign to eliminate the threat of Soviet communist infiltration in the American occupation zone. The fear of communist interference in AMG occupation party drove the military government to isolate, marginalize and eliminate left-leaning radicals. However, it also led Hodge and the occupation authorities were seemingly incapable of separating the radicals from the left-leaning moderates?members of the optimal winning coalition who associated with Communists during WWII or supported causes as innocuous as labor unionization and the redistribution of former Japanese colonial possessions. This led Hodge to back nationalists Rhee and Kim Ku?a radical nationalist who was implicated in more than one assassination attempt and embarked upon a rather hopeless coup against the AMG. According to Cumings, Hodge allowed Rhee and Kim free reign over the most powerful organs of state, including the police. They used the police to further marginalize the liberal left while permitting right-wing extremists free reign. According to both Liem and Cumings, Hodge realized his mistake only too late. By the time it was clear that Rhee was not the right choice, he had already gained control of the police and strong-armed the moderate forces that might challenge him. Even then, Hodge refused to cut his losses and retract his support. Essentially, Liem and Cumings accuse Hodge, the AMG and Washington of failing to respect the preferences of the people of South Korea. 223 The weakness in this line of argument is the assumption that there was no overwhelming preference for Rhee and other nationalists in Korea. This is a weakness only because we lack hard data on electoral preferences in pre-sovereign South Korea, and the available data do not provide as much insight as they do in other cases. In occupied South Korea, partisan politics revolved around personalities rather than platforms and programs (Croissant 2001). Moreover, independent candidates with ambiguous ideological positions won large shares of the vote (Ibid). Finally, the vast majority of this data is for elections to the National Assembly, only the final of a series of attempts at a South Korean government that formed long after the nationalists allegedly began to employ the organs of political violence to skew electoral outcomes. Cumings does cite the apparent popularity of the Korean People?s Republic, a left-leaning coalition of communists and liberals that formed between the end of WWII and the arrival of the AMG, as evidence that the nationalists lacked popular support. However, Cumings himself provides the best study on the KPR, and his research agenda precludes the prudent researcher from relying solely on his findings. Moreover, since the outgoing Japanese colonial regime was responsible for organizing the precursor to the KPR (the Committee for Preparation of Korean Independence, or CPKI), there is some question as to the degree of voluntary support for the KPR as well. The hypothesis that the US was unprepared for their experience in Korea does find some support in the historical record. The AMG did in fact frequently underestimate the amount of time and energy it would take to build a stable government in their occupation zone, making no fewer than four attempts to build a provisional government. Moreover, Hodge and the AMG frequently overestimated the support of the Korean right 224 for occupation policy as well as the disloyalty of the Korean left (Park 2002, Oh 2002a). Most damningly, Hodge admittedly overestimated Syngman Rhee?s interest in a democratic Korean government and assumed a confluence of interests when none existed (Cumings 1981). When Hodge realized this mistake, he refused to withdraw his support (Ibid). Such miscalculations belie an information asymmetry that implies an absence of forethought. However, we should not place so much emphasis on the importance of pre- occupation planning. Neither of the initial plans for the American occupations of Germany and Japan closely resembled ultimate occupation policy. Rival Hypothesis: Prewar Modernization (or lack thereof) In addition to supporting the wrong people, Cumings charges the AMG with failing to implement socioeconomic reforms that would have put South Korea on a more favorable course to democracy. The AMG in Korea never attempted a comprehensive reform of the political economy in its occupation zone. Contrast this to Japan, where MacArthur & Co. implemented land reform and stabilized the Japanese yen, or to Germany, where the US and UK directly injected capital in to the region while curbing inflation. In Korea, there were no direct capital injections for reconstruction efforts, no currency stabilization schemes and no land reform until the occupation was nearly over (Farley 1950; only Japanese colonial holdings, not land owned by absentee landlords, fell under the aegis of the plan). There was an attempt to open the market for agricultural goods, but this succeeded only in fueling inflation without a subsequent increase in production. That inflation, especially in the price of rice, allowed landed interests to shore up their own positions in the new regime. Combined with the favoritism the AMG 225 demonstrated towards right-moderates and right-wing extremists, elections for the various provisional governments disproportionately represented the interests of landlords (Matray 1985, Oliver 1999, Oh 2002a) at the expense of professionals and the industrial working class. Landed interests and right-wing bureaucrats, fearing the possibility of a peasant-based, communist revolution akin to the one occurring simultaneously in China, chose authoritarianism over democracy. As noted in the opening chapter, agrarian political economies (with their low per capita incomes) and inequality do tend to undermine democratic consolidation. Democratic institutions pose a redistributive threat to owners of land-based wealth factors. The fear is that less privileged interests will use legislative authority to confiscate land or to levy extortive taxes on property, raw materials and commodities (Boix 2003). This threat could lead land owners to push for constitutional safeguards on private property, contracts and taxation. In a low-information environment when institutions are young and protections afforded by constitutional safeguards are uncertain, this fear also creates an incentive to block democratic reforms altogether. Such fears are not unfounded. Koreans did call for the redistribution of both Japanese colonial holdings and fallow farmland deeded to absentee landlords. Just across the East Sea, Hodge?s boss? General MacArthur?authorized a comprehensive land reform that successfully redistributed holdings from absentee Japanese landlords to tenant farmers. It is perfectly understandable in this circumstance that the wealthiest South Koreans would choose a mild authoritarianism over a democratic regime, especially if that regime had the backing of a powerful occupation government. 226 Neither of Liem?s or Cumings? explanations fully explains the failure of democracy in this case, but they are supportive of the hypotheses of the Marginal Value Model. States that do not plan enough before they occupy another state are plausibly less likely to make the necessary commitment to domestic pro-democracy elements and to the security of the occupied state in general, if only because the lack of planning indicates a lack of interest in the region itself. The fears that Liem shared as a contemporary and the failures that Cumings saw in retrospect both support the theory occupiers must narrow the range of acceptable political contestation in order to foster and protect democracy in the long run. The initial strength of the optimal winning coalition becomes irrelevant when the occupying force uses its power to marginalize them and empower others. By looking at the case through the Marginal Value Model, we see below what effects the conditions cited in the rival hypotheses had on the prospects for stable democracy, as well as others that played a larger, if not more obvious, role. Test of Working Hypotheses versus Case History Threat Environment Despite Edelstein?s (2004, 2008) characterization?that there was no significant foreign threat during the US occupation of South Korea?there is substantial reason to view the threat environment as critical. By some measures, this may have been the most threatening of the four cases in this dissertation. The external threat to South Korea existed on three dimensions: historical, geopolitical and contemporary. From an historical 227 standpoint, the Korean peninsula was a long- standing target of great power expansion. Imperial Russia and Japan competed over the territory until Japan established its superiority following the Russo-Japanese War and the Treaty of Portsmouth. The reason for this competition was geopolitical: Korea was a source of arable land (desired by a resource- poor Japan) and seafaring warm-water ports (particularly valuable to a hemmed-in and frequently ice-bound Russian Empire). This was no less true one-half century later. What changed was the forward position of military forces arranged behind the porous boundary of the 38th parallel, at first the property of the USSR and later that of a Soviet-trained and ?armed North Korean army. Finally, both the Soviet Union and the provisional government in the North deliberately and repeatedly provoked their counterparts south of the 38th. The next few paragraphs will demonstrate that, from any objective standpoint, the occupied society that would become the Republic of Korea faced a dire existential threat that yielded an exceptional opportunity for the United States to produce marginal occupation value. History indicates that the two primary threats to Korean sovereignty prior to WWII were Imperial Russia and Japan, with a heavy emphasis on Japan. In the late nineteenth century, both Russia and Japan maintained spheres of influence on the Korean Peninsula. In 1896, they mutually recognized each other?s interest in the territory, agreeing that the Soviets would control the north and the Japanese the south, separated Table 6.1: Influences on the Threat Environment in South Korea More Threatening ? Historical Russian interest in Korean Peninsula ? Geopolitical vulnerability ? Direct Soviet control of government in North ? Division and mutual non- recognition of governments in Pyongyang and Seoul ? USSR-PRC-DPRK Alliance Less Threatening ? n/a 228 by?notably?the 38th Parallel (Fisher 1954). This arrangement was nullified by Japan?s 1905 victory in the Russo-Japanese War, after which the two belligerents agreed in the Treaty of Portsmouth?brokered by the United States?to place the entirety of the Peninsula under Japanese dominion (L and G 1946, Krishnan 1984). Japan formally annexed the territory by 1910, with American approval (L and G 1946, Hart-Landsberg 1998, Cha 1999). Korean life under Japanese rule turned for the worst after Japan?s invasion of China in 1937. Starting that year, through the end of WWII, Japan forcibly immigrated over 700,000 Koreans to assist in the war effort (Caprio 2007). Most were drafted into labor, but after 1943 the Japanese used them as soldiers (Hart-Landsberg 1998). Meanwhile, Japan systematically attempted to cleanse Korea of its cultural identity, banning all independent organizations, the use of the Korean language in all instruction, publications and public meetings, and forcing Koreans to adopt Japanese names (Ibid). These policies continued until the end of the war, and they firmly established Japan as the greatest threat to Korean security and autonomy. The conclusion of WWII and the maneuvers of two victorious superpowers during an eighteen-day period in mid-August 1945 largely determined the geopolitics of the Korean occupation. On August 8, the Soviet Union fulfilled the obligation they took on at Yalta and entered the war against Japan (Hart-Landsberg 1998, Stueck 2002, van Ree 1989, N. G. Kim 1997). The US and USSR agreed on the 38th Parallel as the appropriate line of division on the 16th (Cumings 1981, Stueck 2002, van Ree 1989) On the 25th and 26th, Soviet troops took up positions along the 38th Parallel (van Ree 1989). In sum, one half of the peninsula was once again under the control of a foreign power that 229 once (albeit, under a different regime) had expansionist aspirations in the area. Moreover, the boundary of occupation was an imaginary line of latitude that correlated with the pre- colonial division of Korea into spheres of influence but provided no strategic barrier to invasion in either direction. And why should it? Korea was to be reunited and, in time, granted independence, if the Americans and Soviets were to be believed. A Russian presence on the Peninsula, this time en force, may very well have raised some alarms. Indeed, Syngman Rhee pushed the US to recognize the Korean government-in-exile?to which he could no longer claim any direct ties but nevertheless presumed to represent in Washington (Cumings 1981)?as the legitimate government of all Korea. Rhee pushed Washington to embrace Korea as a bulwark against Russian expansionism (Oliver 1999). The plea for a positive security relationship from the man who would run South Korea is a powerful indicator than an opportunity to produce security goods and increase the occupation?s marginal value existed and that the US was aware of it. In essence, with the defeat of Japan in WWII, Korea?s most recent oppressor had been heeled. However, it was replaced in part by another threat. In a relative sense, this new threat was chronologically distant and lacked the horrifying record of Japanese colonialism. In an absolute and hypothetical sense, this new threat wanted something (strategically valuable ports) that Korea had and, now more powerful than either Imperial Russia in 1904 or Japan in 1937, had the capability to take it.72 The Soviet Union now loomed large over a Korea that would be relatively weak even if it remained united. The territory is of course a peninsula, hemmed in by water and, at the time, a distracted but 72 This is not to say that the USSR expressed a specific interest in taking over Korea, but rather that Korea had reason to be concerned: they possessed something of military value but not the resources to keep others from taking it. 230 unstable Republic of China. The Soviet Union also shared a sliver of border with what is now North Korea, while Korean territorial waters are easily accessible from the port city of Vladivostok. A state that emerged south of the occupation line would be especially weak: the southern reaches were agrarian and unindustrialized, lacking the capital to build weapons and the natural resources to power them.73 Moreover, the 38th Parallel itself provides no natural barrier. All told, the national capabilities of either a united or divided Korea paled in comparison to the industrial capacity and manpower that any middling, great or ?super? power would bring to bear in the region, even one that was devastated in a recent war and aligned primarily towards the west. From an objective standpoint, the threat environment was critical. The objectively critical threat environment edged towards an explicitly critical state as both regions developed their own provisional governments. Each occupier sought to create a government in its own image. While the US occupation authorities made several attempts to establish a nationalist-leaning parliamentary democracy in their zone (more on this below), the USSR quickly established a ?Provisional People?s Committee? in the north, populated by the parties compatible with Soviet occupation policy and led by a hand-picked communist freedom fighter by the name of Kim Il Sung (van Ree 1989, Washburn 1947). The establishment of the People?s Committee in February 1946 coincided with a purge of bourgeois parties and politicians in the Soviet zone (Ibid). Almost immediately, AMG-backed reconnaissance claimed that Soviet-led military maneuvers north of the 38th signaled an imminent invasion (Cumings 1981). It was a 73 The Japanese had oriented industrial development of their domain in the north, where the geography was favorable for hydroelectric power and, thus, cheap energy (Matray 1985). 231 false alarm, but it grew out of a fear that the thinly disguised Soviet puppet state in Pyongyang posed a material threat to the security to the South as well as the likelihood of a peaceable reunification of Korea.74 The eventual establishment of a provisional government south of the 38th Parallel only crystallized the division, as did the failure of the first US-Soviet Joint Commission and the resultant diplomatic volleys and recriminations.75 These failures increased the likelihood that one or both provisional governments would attempt to unify the peninsula by force, thereby amplifying the already precarious threat environment. Soviet-American negotiations over the fate of the territory dragged on through 1947 without resolution (Oh 2002a, 2002b, Matray 1985, Stueck 2002). During that time, the Soviets reduced their troop commitment from 40,000 to 10,000 (van Ree 1989). This reduction in Soviet troops in 1946 only preceded the construction of a 100,000-strong volunteer army in the north by the end of 1947, supplied and trained by the USSR (McCune 1947a), easily outnumbering its equivalent in the south. Tensions increased as it became clear that the US and Soviet Union would be unable to reach an agreement on unification and independence. In May 1948, following elections to South Korea?s first representative assembly?but before the end of the US occupation?and following months of threats to do so, the Soviet occupation authorities in the north (not yet officially the DPRK) cut transmission of power from their hydroelectric plants to the 74 Just as the establishment of a more-or-less pro-American government in the South would give those north of the 38th cause to worry about the prospects of unification. 75 Following the adjournment of the Joint Commission, the USSR closed its consulate in Seoul and blocked the establishment of an American consulate in Pyongyang (van Ree 1989). That fall, in the UN, the Soviets (who still retained over 10,000 troops in the north) characterized the US presence in the south as a threat to world peace (Oliver 1999). 232 south (Morley 1965, Dobbs 1985, Matray 1985). The Soviets offered to restore power only if the US recognized the Provisional People?s Committee (Matray 1985). This represented the first direct provocation, and educated anyone who was not already aware about the relative military and economic disadvantage of the emerging Republic of Korea. After the USSR removed the last of its occupation forces from the north, they left behind a few hundred military advisors (van Ree 1989, Morley 1965). More importantly, the USSR maintained total control over North Korean trade?rendering the new state a de facto suzerainty of the Soviet Union (Ibid)?while commencing a crash course of modernization and mechanization of the North Korean People?s Army (K.-K. Kim 1995). The first border skirmishes between DPRK and ROK troops began not long after (Hart- Landsberg 1998). Such was the situation that South Korea faced at the end of the American occupation. In the north was a separate state, backed by a superpower with a forward position in the region and good reason to covet the geopolitical advantages that the Korean Peninsula conferred. The Soviet-backed DPRK boasted a substantial, rapidly modernizing army that was separated from the South only by an imaginary line on a map?and that was beginning to press its advantage. The threat to South Korea and the provisional regime that preceded it had roots in turn-of-the-century great power competition, the distribution of geographic defenses and material capabilities in 1945, and the signals produced by the Soviet Union and the northern provisional government during the timeframe of the case. At any given point after 1945 there was a very real and very obvious threat to the southern regime, and this threat grew critical by 1949. The critical nature of the threat is confirmed the invasion of 233 North Korean troops the following year. Considering the geographic and military disadvantages that South Korea faced, the threat environment in this case is all but certainly the most critical of all four cases in this dissertation. In no other case was the possibility of interstate conflict so great as it was on the Korean Peninsula in 1949, and in no other case would the outbreak of war so likely result in the destruction of the regime and its inchoate institutions?absent the intervention of a powerful foreign state. This threat environment was favorable to the production of marginal occupation value, an opportunity the US was well positioned to exploit. The United States maintained its own forward positions in the region?not just in Korea but also in Japan?and enjoyed naval superiority in the corresponding littorals and open seas. Now boasting the world?s most potent military, and emerging relatively unscathed from the ravages of the Second World War, there is little doubt that the US?if so motivated?could have produced marginal occupation value by offering a credible commitment to defend the territorial integrity and existential prospects of a nascent Republic of Korea. Signaling Security Value It was into this threat environment that the United States initiated the military occupation of Korea. On September 7, 1945, General MacArthur?in his capacity as Allied commander in the Pacific?issued Proclamation #1 formally establishing the American Military Government (AMG) in Korea south of the 38th Parallel (Hart- Landsberg 1998). General Hodge and the 24th Corps of Tenth Army arrived at Inchon on September 8 (Oh 2002a, Sook 2002, Park 2002, Hart-Landsberg 1998, N. G. Kim 1997, Liem 1949). The threat environment present in this case provided an excellent 234 opportunity for the AMG to signal a credible defensive commitment. In doing so, the occupation could have produced marginal security and created a political space in which moderate democratic coalitions could survive and thrive. A state in a threat environment as precarious as Korea?s in 1945 cannot rely on internal balancing alone to ensure its own security. Potential external balancers would have to be capable enough to deter the local threat, demonstrate a credible commitment to Korean security, and communicate their resolve to intervene if necessary. The only state in the international system that could possibly have fulfilled all three criteria at that time was the United States. Objectively, the threat environment in 1945 provided the opportunity for the United States?and only the United States?to produce marginal security. However, the US was not successful in doing so, either as a result of a failure to intentionally protect South Korean interests or an indifference to South Korean sovereignty and territorial integrity. In terms of convincing domestic elements in Korea that they came bearing gifts of security and prosperity, the US military and government had already dug themselves quite a hole before they had a chance to climb out. Prior to WWII, the United States had publicly and explicitly approved of Japan?s dominion over Korea. While mediating negotiations for the Treaty of Portsmouth (which concluded the Russo-Japanese War), the US and Japan agreed to recognize each other?s influence over Table 6.2: Indicators of Occupation Security Value in South Korea Positive ? US defeat of Japan ? Signals of aid for Korea (unrealized) ? Modest remilitarization Negative ? US support for Japanese annexation ? US demands for Koreans to obey Japanese administrators ? Retention of colonial-era regulations and personnel ? Moscow Declaration ? US-Japan security cooperation ? Undersized US aid to Korea ? Exclusion of Korea from US security perimeter 235 the Philippines and Korea, respectively (Cha 1999). When Japan officially occupied Korea in 1910, the United States was the first state to recognize the annexation (Caprio 2007). After Woodrow Wilson called for self-determination in his Fourteen Points speech, Syngman Rhee, then with the Korean Provisional Government in Shanghai, called on the United States to liberate Korea from Japan and steward Korea?s transition to independence. Washington refused to heed Rhee?s entreaties, reluctant to disturb the balance of naval power in the Pacific or queer the deal that preserved American influence in the Philippines (Choi 2002). The US relationship with Korea grew no closer as Washington resumed its cozy relationship with Tokyo. By 1947, MacArthur was set on turning his once-mortal enemy into an American ally in the nascent Cold War. In doing so, the US very publicly broadcast its new commitment to the security and prosperity of Korea?s historic enemy across the East Sea. The United States? about face on Japan (covered in detail in Chapter 4) included efforts to remilitarize the state and culminated in formal security pact in September 1951 (Tsuzuki 2000, Destler, Sato, et al. 1976, Morley 1965, Eiji 2000). When considering what sort of signal this broadcast to Korea, note that by 1947, Korean repatriates from Japan?vast numbers of whom served as the indentured auxiliary to the Japanese war effort?represented 15% of the total South Korean population (Cumings 1981). The memory of Japanese dominion was fresh and the wounds had yet to heal (Cha 1999), and this was the country with which the United States had decided to be friends. In order to produce marginal security, the US not only needed to overcome the inherent threat that any occupying power posed to the occupied territory, but also make up for the role it played as accessory to Japan?s imperialist crimes and the new role it 236 played as Japan?s chief benefactor. This was by no means an insurmountable obstacle. One cannot overlook the fact that the occupiers were not so much conquerors as the force that had vanquished the Japanese colonial overlords, an advantage the US did not enjoy in its occupations of Germany or Japan. But it was an obstacle nonetheless. And if the US was sensitive to how their relationship with Japan would play in Korea, the actions of the occupying forces indicate otherwise. Initially, the ranking officer in the American Military Government (AMG), General John R. Hodge, planned to rely extensively on the Japanese colonial forces that remained in Korea (Park 2002). Before his arrival, he ordered the dropping of leaflets instructing Korean citizens to obey their Japanese overlords; privately he commanded his troops to treat Korea as enemy territory (Hart- Landsberg 1998). He quickly backtracked when Washington objected to the employment of Japanese, but left it up to the Japanese themselves to nominate potential replacements from the local population (Hart-Landsberg 1998). Instead of liberating South Korea after four decades of subjugation, Hodge left many Korean institutions in the status quo ante bellum. The AMG initially ruled out any plans to redistribute Japanese colonial properties to Koreans, although they changed their position later (McCune 1947b). When the US reopened the Japanese-administered Korean schools that closed on V-J Day, they did so with Japanese administrators still at the helm (Armstrong 2003). On November 2, the AMG issued Ordinance 21, explicitly retaining all laws and ordinances that predated the Japanese surrender (Park 2002). These laws included those that the Japanese used to prohibit political opposition, including the Act Prohibiting Political Meetings of 1910 and others that prohibited free political expression (Cumings 1981). Perhaps most importantly, the US kept the Japanese colonial 237 police establishment?known as the National Police Force?and filled it with former collaborators (Han 1974). As late as December 1946, five hundred Japanese troops (not civilians) remained on the Korean peninsula, not as prisoners but as an auxiliary force in aid to the AMG (Caprio 2007). Incidentally, the AMG ignored the recommendation of the Left-Right Coalition Committee (an early attempt at a provisional government) that all former Japanese and collaborators be purged from government positions (Park 2002). Slowly but surely, the Hodge and the AMG were eroding Korean confidence that the United States would be an agent of liberation rather than continued oppression. The American plan for Korea was a temporary occupation that would help Korea transition from colony to sovereign state?a period of ?tutelage.? Such plans did not sit well with Koreans who hoped for immediate independence. The US had been planning on a period of trusteeship for Korea as early as the Cairo Conference in late-1943. The Cairo Declaration promised independence for Korea, but ?in due course? (Choi 2002, Cumings 1981, L and G 1946, Borton 1966). The Soviet Union signed onto this plan at Yalta in February 1945 (Choi 2002). Korean supporters of trusteeship were few and far between, Korean Democratic Party leader Song Chin-u being one of the lonely advocates of tutelage prior to the Moscow Conference (Ibid).76 Opponents to trusteeship occupied a diversity of positions on the ideological spectrum, including Yo Un-hyong on the left wing and right-wing nationalists Syngman Rhee and Kim Ku (Ibid). Domestic opposition notwithstanding, the US and USSR (and UK and nationalist China) settled on a maximum five-year period of trusteeship (Ibid, Cumings 1981, van 76 Song reversed his position on December 29, 1945, one day before an agent working under anti-trusteeship extremist Kim Ku murdered him for his prior support of trusteeship (Ibid). 238 Ree 1989, Han 1974, N. G. Kim 1997, Morley 1965, Choi 2002). The decision drew strong and swift rebuke. The Korean Democratic Party labeled trusteeship a Communist plot to divide Korea (Cumings 1981), despite the fact that the Korean Communists also initially opposed trusteeship (Choi 2002). The Korean People?s Party joined the chorus (Ibid). Moderate Song Chin-u, who initially favored trusteeship, reversed his position (just prior to his assassination by anti-trusteeship radicals; Ibid). Syngman Rhee publicly broke with Hodge for the first time over the issue (Oliver, Transition and Continuity in Korean-American Relations in the Postwar Period 1999). The Korean Communist Party would eventually reverse its position (under pressure from its counterpart in the North and the Soviet Union) but the rest of the Korean political scene remained strongly opposed (Choi 2002, Cumings 1981, Morley 1965). Facing massive domestic opposition, Hodge did his best to spin trusteeship favorably?perhaps to the point of insubordination. Publicly, Hodge claimed that the US wanted to leave as soon as possible and blamed the Soviet Union for the trusteeship provision of the Moscow Declaration, and US Secretary of State Byrnes clarified that trusteeship would end as soon as the US and Soviet Union reached agreement on an exit strategy (Cumings 1981, Sook 2002). Hoping to corral unstable domestic groups and constituencies, Hodge began to secretly rally opposition to the Moscow Declaration (Cumings 1981).77 77 Cumings? (1981) characterization of the episode depicts Hodge stirring anti-trusteeship sentiment in an attempt to challenge official policy. I find this assertion questionable. Even a cursory review of the historical record indicates that anti-trusteeship elements in South Korea required no outside assistance to arouse their constituencies. The prior declarations of Yo, Rhee, Kim Ku and others?most importantly their riotous response to the Moscow Declaration?and the near absence of any pro-trusteeship sentiment should have, would have, made this patently clear all but the most oblivious American officer. One could just assume that General Hodge was in fact the most oblivious American officer, but this is a lazy assumption. The sound assumption, and the one that yields a 239 Along with a quick return of sovereignty, Koreans desired unification. Washington delivered neither. Following the Moscow Communiqu?, the two occupiers formed the US-USSR Joint Commission in order to reach a compromise on the Korean question (Choi 2002, Morley 1965, Oh 2002a). The first attempt to reach an agreement was a failure, and the commission dissolved in a matter of months. When MacArthur offered suggestions on how best to pull out of Korea in January 1947, he and officials back in the US already assumed that there would be no cooperation from the Soviet Union (Matray 1985). By the time a second US-USSR Joint Commission convened in May 1947 (Lee 2006, Wells 1948), the US was set on an independent track (Morley 1965). Part of this new strategy involved submitting the Korean question to the United Nations (Matray 1985). Knowing that the USSR would likely oppose UN intervention in their zone, policymakers in Washington were confident that this would force a two-state solution (Ibid). In the summer and autumn of 1947, the US took the necessary steps to request a UN resolution of the Korean matter, including a nationwide election for a provisional government and joint troop withdrawal (Ibid, Oh 2002a, Oliver 1999, Hart- Landsberg 1998, Lee 2006, Stueck 2002, Morley 1965). more rigorous test, is that Hodge was in fact aware of the situation and retained an ulterior motive for his rabble rousing. While Hodge opposed trusteeship, he could not publicly state so without drawing charges of insubordination from Washington. Nor would his verbal opposition seem credible to Koreans without some sort of collateral to back up his claims. Hodge?s private support for anti-trusteeship activities was the collateral he needed, the costly signal that might retain the cooperation of Rhee, Kim Kyu-sik and the other ?respectable? (read: conservative) factions. Doing so privately, he avoided opprobrium from Washington and?insofar as it mattered?Moscow. This is a far more plausible explanation of Hodge?s behavior than one that has him stoking a fire that blazed furiously before him, and one that bears the added benefit of compatibility with the underpinnings of the Working Model. 240 In October, the US delegation moved to form the United Nations Temporary Commission on Korea (UNTCOK), a motion that was formally approved by the General Assembly on November 11, 1947 (Hart-Landsberg 1998, Oliver 1999). The motion also called for elections no later than March 31, 1948, a provision that pleased Rhee and other nationalists (Oliver 1999). Eight states? delegations comprised UNTCOK: China, Canada, Australia, the Philippines, France, India, El Salvador and Syria (Hart-Landsberg 1998). The Soviet delegation understandably protested, claiming that such a motion violated the UN Charter articles pertaining to joint consultation and postwar settlements (Articles 32 and 107; Hart-Landesberg 1998). When the USSR signaled that they would deny UNTCOK access to the north, the US successfully set about convincing the UN to follow through on the motion below the 38th Parallel (Oh 2002a). The UNTCOK delegation arrived at Seoul in January 1948 (Lee 2006, Hart- Landsberg 1998, Oliver 1999). Opinion in the South was split. Rhee supported the idea of separate elections in the South (as he would, since he felt he had the support to win them), while Kim Ku denounced the latest version of foreign intervention in Korean affairs (Oliver 1999). In February, the UN backed the American delegation?s position and ordered UNTCOK to proceed with elections wherever they were capable of holding them (Hart-Landsberg 1998, Lee 2006). Nevertheless, the plan went forward, leading to elections in the South and a new republic. The Soviets responded by establishing their own government in Pyongyang. The American delegation won one more victory when the UN recognized the government in Seoul as the legitimate government of all Korea (Hart-Landsberg 1998, N. G. Kim 1997). Ostensibly, the occupation authorities succeeded in establishing a government that was anti-Communist and democratic, but the 241 division of Korea created two issues. The first was that the southern state was inherently insecure without American assistance. The second was that it failed to resolve a primary grievance held by Korean citizens opposed to the occupation. It is possible that the Soviet Union sensed dissatisfaction among those Koreans in the South with the American Military Government. While civil unrest continued and the US continued to implement unpopular ordinances, the Soviet Union appeared to be devolving authority to local elements (Park 2002). North of the 38th Parallel, the USSR retained the People?s Committees established by Yo Un-hyong and the CPKI before the official start of the occupation; superficially, the North appeared calm. 78 This, and the uproar over trusteeship, may have encouraged the AMG to begin devolving power to local authorities in March 1946 (Ibid). This process included a demobilization of several key military posts, the handing over of administrative duties to Korean civilians, and the general attempt to build a provisional government populated by Koreans (Ibid, Cumings 1981). In an alternate history, these could have been the first signs of an about-face on American occupation policy?much like the reverse course that MacArthur implemented in Japan. In this history, however, Korea was hardly ever a priority for Washington. For instance, Hodge and the AMG operated without instructions from Washington well into October 1945 (Park 2002). Foreign aid to Korea in the year ending 1945 totaled a paltry $4.9 million (Morley 1965). Aid levels rose to $49.5 million in 1946, $175 million in 78 We now know this to be an illusion, that the local Korean authorities were Soviet puppets, and that the calm was less indicative of public sentiment and more indicative of the occupiers? ruthlessness. As Edelstein (2004, 2008) asserts, the Soviet occupation of North Korea may be the only modern example of a state successfully realizing its occupation goals solely via repression. 242 1947, $180 million in 1948 and $116.5 million in 1949 (Ibid), but the cost of the occupation was nothing but a source of consternation to the War Department and, later, the Department of Defense. As early as January 1947, General MacArthur and the War Department were citing cost overruns resulting from occupation management as a rationale for evacuating the peninsula as soon as possible (Matray 1985). Plans to evacuate Korea ahead of schedule did not dissuade the State Department from devising a plan for social and economic reform in Korea, not dissimilar from its plans for Germany and Japan. The plan?which called for the stabilization and rehabilitation of the industrial and agricultural sectors?was never fully implemented due to lack of support from Congress and the War Department, and the already-dilapidated state of the economy in South Korea (Park 2002). Yet for a while, there were signs that policymakers stateside might elevate Korea to the level of priority accorded Japan and Western Europe in General. In February 1947, then Undersecretary of State Dean Acheson called upon Secretary Marshall to request $600 million in reconstruction aid for Korea, to be distributed over a three-year period (McGlothlen 1989, Matray 1985, Park 2002). Acheson also called for a high commissioner to replace Hodge and the AMG, and argued that US forces should remain in Korea until the US could guarantee the territorial integrity of a democratic Korean state (Ibid). This represents the first example of an American official seriously calling for a long-term commitment to a democratic Korea. If there were constituencies in the South who still looked to the United States as a guarantor of security and prosperity, the spring of 1947 would have been their most optimistic season. On March 12, in his famous ?Doctrine? speech, President Harry S. Truman (1947) pledged to "assist free peoples to work out their own destinies." In May, Acheson 243 spoke publicly about the benefits of rebuilding postwar Asian economies (McGlothlen 1989). Later that month, he presented to the Senate Foreign Relations Committee a three- year Korean aid plan totaling $540 million (Ibid, Matray 1985 N.G. Kim 1995; opposition from the War Department had already caused Acheson to pare the request down by $60 million). That request would remain the most promising signal from the United States, as it was soon erased by congressional politics and lack of interest. The aid bill never did reach a vote on the floor of the Senate (McGlothlen 1989). There was some talk of approving a one-year, $215 million appropriation, but Truman opposed this on the grounds that it would divert from funding for European reconstruction (Matray 1985). In late June, Senate Foreign Relations Chair Vandenberg decided to refuse all grants-in-aid requests for Korea (Ibid). The following month, MacArthur abandoned plans to fund Korean reconstruction with Japanese reparations (Park 2002). This ensured that all money sent to Korea during the occupation period would be directed towards basic occupation maintenance and preparations for exit. The Joint Chiefs saw this as a necessity, since defense appropriations had been cut down to $13.1 billion from their wartime high of $81.6 billion; they saw a quick exit from Korea as a means of saving funds that were dwindling quickly (McGlothlen 1989). Despite a few small victories on Acheson?s parts, those who remained uncommitted to Korea in the long run succeeded in imposing their policy stateside. That same August, George Kennan declared Korea strategically unimportant and advised the JCS on a quick exit, a position they already held (McGlothlen 1989, N. G. Kim 1997). On the issue of economic assistance, Kennan called for disbursements through the Economic Cooperation administration small enough 244 to impose a minimal drain on the US economy and to avoid the implication of any commitment to South Korea (McGlothlen 1989). One last opportunity to rekindle Washington?s interest in Korea presented itself with Acheson?s reappointment as Secretary of State in January 1949 (McGlothlen 1989). Truman and Acheson hashed out a policy that would continue the military withdrawal but increase direct military and economic assistance (Ibid). Of course, this was after the occupation formally ended, once the US had already tipped their cards to domestic elements in the ROK. Besides, the promises were fleeting; Congress continued to reject any large financial outlays to South Korea until the start of the Korean War (Matray 1985). The occupation?s effect on marginal security in this case was not entirely negative. Despite the initial wishes of MacArthur and those in Washington, Hodge did succeed in remilitarizing the South and rending the new state dependent on American security. By mid-1948, the AMG had raised a 50,000 strong constabulary from nothing, an army in all but name so as not to violate official policy or raise suspicions about US intentions in Korea (Matray 1985, Park 2002). In February 1948, Truman approved a plan that would pull American soldiers off of the Peninsula by the end of the year (Matray 1985). Washington agreed to delay their exit, however, leaving troops in the South to aid the new government in corralling insurgents (McGlothlen 1989). By the time the US left the constabulary had grown to 65,000, constituting the entirety of the South Korean armed forces (McGlothlen 1989). The small force left Rhee, who once called for the immediate departure of occupying forces, begging the US to keep its forces on the peninsula (Matray 1985). In short, the US succeeded in convincing domestic elements in 245 Korea that they relied on American forces for their security, only to tell them that they were out of luck. By 1950, even Acheson had given up on Korea when he declared the region beyond the ?US security perimeter? (McGlothlen 1989). The Secretary of State effectively told the world?and most fatefully, Moscow, Beijing and Pyongyang?that the US was not willing to risk American lives in the defense of Korean ones. In reality, he was only affirming publicly what Kennan, the National Security Council and the Joint Chiefs of Staff had settled on privately. The few positive signs that the US might commit itself long-term to the security of a Korean state were overwhelmed by American indifference. The evidence indicates that American actors, outside of perhaps General MacArthur and his staff, failed to see the threat posed by Kim, Mao and Stalin.79 While the security configuration during and immediately after the occupation of South Korea was not as destructive as it was during the occupation of the Rhineland a quarter century earlier, it was hardly constructive. Undoubtedly, the US was opposed to Soviet expansionism and the new communist regime in Beijing. Truman and Acheson saw ?opportunities? to contain the Soviet Union in several regions: Central Europe, Indochina, Japan, the Philippines, Iran and Latin America. Acheson?s statement of January 1950, however, indicates that the US was not (yet) willing to extend the containment doctrine to the Korean Peninsula?or, for that matter, the Taiwan Strait. The US refused to commit fully to backing the Kuomintang during the civil war in Mainland China, and in January 1950 refused military support to 79 There is in fact evidence that MacArthur and Rhee had leading edge intelligence suggesting an imminent invasion by North Korea, but they failed to pass it along to Washington (Stone 1952, Cumings 1983). 246 the exiled nationalists on the island of Formosa?operatives with whom Syngman Rhee and his bureaucracy had maintained close ties (Buhite 1978). Generally, the US made it clear that it was not getting into the business of defending small, precarious territories in East Asia against Communist aggression whether it derived from Beijing or Pyongyang. In October 1946, General Hodge lamented that there were not enough US forces to repel an invasion from the North (Matray 1985). What was perhaps hyperbole then was now the undeniable truth. The haste with which the US had made its exit from South Korea only emphasized a lack of interest in maintaining a forward position. Three days prior to the invasion that set off the Korean War, US administrators of the Korean Military Advisory Council?a small corps of just under 500 American military officers stationed in South Korea to help modernize the newly sovereign country?s armed forces?recommended a 50% reduction in forces (N. G. Kim 1997). The swift?and initially very successful?North Korean invasion might have confirmed fears that the US was not committed to South Korean security. Just the same, US diplomatic maneuvering in the UN, MacArthur?s historic landing at Inchon, the UN counteroffensive, and the multi-year involvement in the Korean War (Morley 1965) might have changed some minds. Rhee saw things differently, however. Following the Chinese intervention, Truman (and later Eisenhower) set their sights on an armistice (Ibid). Rhee?s frame of reference, and that of his key constituency, was unification. A peace that seemed a best-case scenario to the US and the UN coalition was a strategic failure for pro-unification South Koreans. Rhee demonstrated his commitment to unification at a high cost by actively trying to sabotage peace talks between Washington, 247 Beijing and Pyongyang (McWilliams and Piotrowski 2009). But Rhee?s meddling aside, American escalation in Korea beginning in the summer of 1950 reveals the greatest irony of the occupation. It turns out, the United States was concerned with and willing to defend South Korean sovereignty and territorial integrity?they just failed to share this crucial bit of information with everyone else! Indeed, Truman and his cabinet failed recognize their own latent preferences regarding the Republic of Korea before it was very nearly driven into the sea. Facilitating Domestic Consolidation The narrative of the AMG?s interaction with domestic elements in the southern occupation zone is one of multiple halting attempts to establish a provisional government that was both moderate and democratic. The landscape was not an easy one to navigate? when AMG arrived there were already three different bodies politic competing to establish themselves as the legitimate government of a new Korean state. Singling out the optimal winning coalition was not easy. The political spectrum was replete with communists masquerading as liberals, liberals posing as communists and radical nationalists pretending to be moderates. In order to empower the appropriate coalition, Hodge needed to sift through these muddied partisan identities and identify moderate democrats who would cooperate with the military government. Ultimately, General Hodge chose to work through an anti-communist, nominally democratic coalition led by Syngman Rhee. General Hodge chose poorly. 248 When American occupation forces first reached Korea, they encountered Korean organizations already claiming to constitute a legitimate government. The Japanese established an interim regime to keep the peace in the interregnum, and the Korean left took advantage of this opportunity to establish a toehold in postwar Korean domestic politics. The Japanese placed Yo Un-hyong, a left-of- center nationalist, in charge of the new government. Yo formed a coalition with communists and progressive nationalists under the banner of the Committee for the Preparation of Korean Independence or CPKI (Hart- Landsberg 1998, Han 1974, Park 2002). While the original raison d??tre for the CPKI was peacekeeping and preventing reprisals against Japanese colonial officials, Yo and his compatriots set about organizing people?s committees across the peninsula that served as local governments (Han 1974, Cumings 1981). To this end, the CPKI renamed itself the Korean People?s Republic on September 6, 1945, declaring itself to be the legitimate government of Korea just before the Americans arrived (Ibid, Oh 2002a, Sook 2002, Hart-Landsberg 1998, C.-P. Park 2002, Oliver, Transition and Continuity in Korean- American Relations in the Postwar Period 1999). Another competitor for the rightful government of Korea was the Korean Provisional Government, founded in Shanghai in 1919 (Han 1974). In 1940, they formed a militant wing known as the Liberation Army, which fought alongside Chiang Kai-shek in China during WWII, killing both Japanese Table 6.3: Evidence of Facilitating Regime Consolidation in South Korea Constructive ? Attempts to build representative South Korean provisional governments Destructive ? US support for anti-Communist, anti-democratic factions ? US support for Rhee ? US refusal to punish right-wing radicals ? Initial refusal to redistribute Japanese, fallow Korean properties ? Inability to control inflation ? Voting procedures skewed towards anti-democratic constituencies 249 invaders and Chinese Communists (Ibid). The two sides had very little in common other than joint preferences for independence and unification and a mutual opposition to trusteeship (Choi 2002). While the KPG was in exile at the time the occupation began, nationalists in Korea formed the ?Preparatory Committee for Welcoming the Return of the Korean Provisional Government? (Cumings 1981). Truman prohibited the occupation authorities from endorsing any pre-existing coalitions (Oh 2002a). The AMG?s response to competing authority centers was swift and resolute. On September 7, 1945, General MacArthur simultaneously ordered the creation of the American Military Government in Korea and declared pre-existing government organs illegitimate. (Han 1974). General Hodge reinforced the position on October 10 when he declared the AMG to be the only government south of the 38th Parallel (Krishnan 1984). Later, in December, Hodge declared the KPR an ?unlawful? organization (Cumings 1981, N. G. Kim 1997, Hart-Landsberg 1998). Despite Truman?s orders, the AMG showed clear preference for right-wing factions over left-wing. As early as September 21, the AMG allowed Song Chin-u, head of the Korean Democratic Party (which the Committee for welcoming the KPG reconstituted itself as after the AMG?s declaration of its illegitimacy) to denounce the Korean People?s Republic over official radio (Cumings 1981). The AMG and Washington considered the KDP to be a ?respectable? organization. Accordingly, Hodge relied on them to counter radical leftist threats to the occupation (Park 2002). One of the more useful members of the KDP was Cho Py?ng-ok, whom the AMG placed in control of the National Police Force (Cumings 1981). Moreover, when Hodge attempted to build a South Korean Army (without the knowledge or consent of Washington), he recruited right-wing paramilitary groups to 250 swell its ranks while eschewing left-wing paramilitaries (Ibid). In what was likely an honest attempt to narrow the spectrum of political contestation, Hodge?s policies in fact empowered radicals of one feather at the expense of radicals of another, all the while marginalizing moderates who got in the way. The beneficiary of the American state-building policy was Syngman Rhee. Rhee had once represented the KPR, but was dismissed due to improprieties such as embezzlement and over-stepping his authority decades before the occupation began (Ibid). Nevertheless, he continued to claim to represent the KPR, making connections and building support in Washington, DC, during the 1920s, 30s and 40s (Ibid). When he returned to Korea on October 16, 1945, he did so on a military plane chartered by General MacArthur (Hart-Landsberg 1998, Cumings 1981, Oliver 1999). When it came time to appoint a provisional government independent of the KPR or KPG, Hodge appointed Syngman Rhee chairman (Choi 2002). Rhee?s stated aims were plain enough. Via his American spokesman in an article in the New York Times, Rhee declared several goals: disband the AMG; establish a separate South Korean government with its own arms and a barrier against the North; Soviet withdrawal from the North; secure admission for the new state in the UN; establish a South Korean currency with convertibility; and rehabilitate the economy (Oliver 1999). He remained steadfastly opposed to trusteeship. When the KDP reversed their position and endorsed trusteeship?ostensibly to gain access to the AMG and the interim government?Rhee denounced them as traitors (Han 1974). While Rhee often accepted support from various political parties, he preferred to operate outside the party system whenever possible (Ibid). 251 Hodge?s preferential treatment of Korean nationalists extended to social and economic policy. Perhaps out of fear of communist infiltration, the AMG?s ?right-to- work? Ordinance 19 denied labor unions the right to intervene in labor disputes and presaged the ?Reverse Course? in Japan?s occupation policy by over two years (Ibid). Naturally, this sapped the power of the pro-labor left, both radical and moderate. Later, Ordinance 34 banned strikes entirely and required mediation of labor disputes via a five- member panel (to which the AMG nominated three officials from the KDP; Ibid, Hart- Landsberg 1998). In contrast to MacArthur?s policies in Japan and Soviet occupation policy in their zone in Korea, Hodge?s AMG postponed the redistribution of lands?both former Japanese colonial holdings and fallow land held by absentee Korean landlords? until the South Korean government was about to take over (Farley 1950). Also in contrast to MacArthur?s policies in Japan, Hodge attempted to impose open-market reforms almost immediately (Cumings 1981). These policies sat well with Rhee and both Kims, as the marginalization of organized labor undermined political opponents who relied on that constituency (especially the Korean Communist Party and Yo Un-hyong?s moderate- left coalition). Anti-labor and anti-redistribution policies placated the landlords and professional classes that supported the KDP, KNP and Rhee. However, in empowering the conservative landed classes and marginalizing moderate workers, these policies weakened the natural constituency of the optimal winning coalition. These are appropriate tactics for improving the prospects of the optimal winning coalition, so long as they applied to radicals independent of their position on the ideological spectrum and so long as there is no collateral damage. However, Hodge?s shock therapy and zeal for purging communist elements harmed left-wingers both inside 252 and outside the optimal winning coalition. At the same time, Hodge?s policies improved the prospects of anti-democratic nationalists such as Rhee and Kim Ku. Hodge?s suspicion of communist infiltration ran deep. Long before the ?Red Scare? gripped the United States, Hodge privately questioned the motives of moderate, pro-democracy groups south of the 38th Parallel, lumped them in with the radical leftists and branded them enemies (Cumings 1981). The commanding US general in South Korea routinely attributed just about every early policy failure to communist infiltration despite a lack of corroborating evidence (and a wealth of evidence pointing towards interference from Rhee, Kim Ku and other violent and meddlesome extremists). In one instance, Hodge claimed definitive evidence that pro-trusteeship politician Song Chin-u was a victim of a left-wing terror attack. There was no evidence to back this claim, and in fact it was right- wing extremist Kim Ku who confessed to orchestrating Song?s assassination (Ibid). Kim Ku?s punishment amounted to a stern opprobrium and loss of access to the American Military Government (Ibid). Hodge?s views in this area remained steadfast, even after May 1946, when he realized that Rhee and Kim Ku were unsuitable partners for building the type of Korean (or South Korean, if necessary) state that Washington demanded (Cumings 1981). When the Left-Right Coalition Committee came under attack from both sides of the ideological spectrum, Hodge viewed the right-wing opponents as blameless while arresting moderate left-wing organizations (such as Yo Un-hyong?s Democratic National Front, a coalition of left-liberals) and politicians on the charge of fostering dissent (Ibid). It seems in this case that Hodge?s anti-communist zeal got the better of him. While he could have 253 employed this motivation to strengthen the center, instead?either intentionally or unintentionally?his fear of infiltration led him to pick the wrong side. In a private letter dated February 2, 1948, Hodge seemed to recognize his folly, if belatedly. He blamed Rhee for obstructing US policy for over a year (in truth he had been doing so from the very start) and for fomenting mistrust of the AMG and its policies (Oliver 1999). Unfortunately, it was too late to reverse course. The AMG had succeeded in neutralizing any opponents to Rhee?s contingent, radical or moderate. He had the landlords, the support of the people, and the police?whom Hodge had all but handed over to the would-be president of the Republic of Korea. The spectrum of political contestation had indeed narrowed. Unfortunately for Korean democrats, it narrowed in such a way as to exclude the optimal winning coalition. During the period of occupation, there is a clear divide between policymakers in the State Department and the officers of the Army, Navy and what would eventually become the Department of Defense. State committed itself to fostering democracy in Korea, calling for the AMG to build a coalition of moderates who shared the values expressed in the late President Roosevelt?s ?Four Freedoms? speech (Oh 2002b). Furthermore, on February 28, 1946, State instructed the AMG to sever its ties with the groups under the control of Rhee, Kim Ku, and those infiltrated by the Soviet Union, and specifically called for a draft of progressive elements to form a legitimate, provisional government (Cumings 1981). In March, attempting to walk back some of the earlier occupation policies and paint a better picture of the AMG, Hodge asserted that the United States was committed to enabling freedom of expression in its zone (Ibid). The US took several opportunities to establish a provisional government?with Korean participation? 254 in the South, in hope of establishing a legitimate state organ that could pave the road to an American exit. The first of these efforts was the Representative Democratic Council (RDC) of South Korea (Choi 2002). Hodge appointed Syngman Rhee to preside as chair over a presumed coalition of left- and right-wing moderates (Ibid, Oh 2002a), but allowing Rhee to appoint its entire membership led to an unbalanced, extremist selectorate (Choi 2002). This should have been no surprise to Hodge, as Rhee had been publicly critical of his opposition (via the radio time that Hodge himself had offered; Cumings 1981) and had made it clear that he opposed party politics and preferred a unified front of likeminded anti-communists to lead Korea into independence (Han 1974). With Rhee as chair and several extreme right-wing politicians on the council, however, the left-wing moderates refused to join. The rightists, on their part, simply stonewalled the AMG and used the RDC as a platform to further oppose trusteeship (Park 2002, McCune 1947b). In response of the failure of the RDC, officials in Washington pledged only to work with ?truly democratic? parties and organizations (Ibid, Choi 2002). Hodge took up this cause in April, attempting to build a Left-Right Coalition Committee that included only moderates, specifically Kim Kyu-sik and Yo Un-hyong (Cumings 1981, Sook 2002, Oh 2002b, C.-P. Park 2002, Borton 1948). It was also about this time that the AMG began aiding the NPF in its investigation and eventual arrest of Korean Communist Party members (Park 2002). In June, Hodge was proud to announce a partnership with the Left- Right Coalition under Yo Un-hong (brother of Yo Un-hyong, who refused to join) of the newly formed Social Democratic Party, although the coalition was not as broad as the AMG or State Department would have hoped (Cumings 1981). Any optimism was 255 unwarranted, however, as there was virtually no cooperation between the left and right wings of the Coalition Committee. The Democratic National Front?a coalition of far-left and liberal-left organizations including the KCP and KPP?demanded land reforms, the resurrection of people?s committees and punishment for former Japanese collaborators, all of which the right wing refused (Ibid). Syngman Rhee refused to acknowledge the Coalition Committee, which became a far more relevant organization for coordinating right-wing political activities (Ibid, Oh 2002b). Unable to build a large enough coalition under the banner of the Left-Right Coalition, the AMG proposed a new provisional government known as the South Korean Interim Legislative Assembly (SKILA). Unlike the previous attempts at a provisional government, this legislature would be half-elected (indirectly, and only by taxpayers and landlords) by Korean citizens and half-appointed by the AMG (Cumings 1981, Sook 2002, Oliver 1999). Elections notwithstanding, SKILA was illegitimate in the eyes of many Koreans and contrary to the interests of others. Kim Kyu-sik originally declared some of the elections to be rigged. Following a re-vote, Kim would serve as the body?s president (Sook 2002). Rigged or not, the elections placed only right-wing officials in the forty-five elected seats. To remedy this imbalance, Hodge appointed mostly moderates and leftists to the forty-five appointed seats. In response to this move, Rhee?s faction of twenty-seven right-wing representatives boycotted the interim legislature (Oh 2002a, Matray 1985). But in many ways, it appears that the AMG was never so interested in building a democracy in Korea, despite the avowed policy of Hodge, of policymakers in Washington, or the claims of the right-wing politicians with whom they aligned 256 themselves. Unlike in Japan, the AMG resisted large-scale land reform until early 1948, leaving the landlord class?and a power base for those opposed to democratic and liberalizing reforms?intact (Farley 1950). There are many instances that bring their intent into question. First is the aforementioned retention of laws from the colonial era that granted the AMG and domestic organs such as the NPF wide berth in suppressing political activities it deemed subversive (Cumings 1981). Another instance is the issuance of Ordinance 55, which required parties to open their finances and membership rolls to the government while allowing local entities to dissolve parties that, in their eyes, failed to comply (Ibid). Laughably, General Lerche called the ordinance "usual...for regulating political parties in democratic countries" (Ibid). Ordinance 72, issued in May 1946 just as Hodge undertook his attempt at building a moderate coalition, virtually outlawed all dissent against the AMG (Ibid, Park 2002). When the coalition faced criticism from left- and right-wing parties and politicians who refused to join, Hodge?s response was to issue arrest warrants (pursuant to Ordinance 72) to left-wing dissenters (but not those on the right, such as Kim Ku and Rhee; Cumings 1981). In February 1947, Hodge did warn that right-wing parties that dissented against the AMG might be dissolved, but there is no evidence that this actually occurred (Matray 1985). By December 1947, the number of political prisoners topped 21,000?4,000 more than those who served time for political crimes under the Japanese (Hart-Landsberg 1998). The number soared to 30,000 before the withdrawal of the AMG despite the repeal of the oppressive, colonial era laws restricting political expression in April 1948 (Ibid, Cumings 1981). Whether due to negligence, indifference or naivet?, the occupation policy of the US military government in Korea empowered a coalition that paid only lip service to 257 representative democracy. They narrowed political competition in such a way that excluded radicals on the left but also pro-democracy elements in the middle, handing the territory to a coalition of nationalists organized around the personality of Sygnman Rhee rather than a coherent policy platform. Despite the failure to ensure a truly democratic government in the American occupation zone, Washington and the AMG pushed ahead with the establishment of a sovereign South Korean state. The final preparations began in March 1948, when candidates began to register for elections to the Constituent Assembly (later, the National Assembly; Han 1974). The first parliamentary elections took place on May 10 in a semi-competitive environment (Ibid, Oh 2002b, Hart-Landsberg 1998, Park 2002, Morley 1965, Stueck 2002, Croissant 2001). The Assembly convened in July and promulgated a constitution two weeks in, claiming it to be the basic law of all Korea (Hart-Landsberg 1998, Oliver 1999, Matray 1985, Morley 1965). On August 15, the National Assembly elected to the presidency Syngman Rhee, who then declared the independence of the Republic of Korea (Oh 2002a, Lee 2006, Hart-Landsberg 1998, Morley 1965, Han 1974, Croissant 2001). Between September 1948 and June 1949, US forces undertook a phased withdrawal from the peninsula (Buhite 1978). When American troops left the peninsula in June 1949, de facto and de jure sovereignty returned to a Korean state for the first time since 1905 (Oh 2002a, N. G. Kim 1997, Oliver 1999). They would not be gone for long. Domestic Outcomes The historical record of the American occupation of Korea indicates that emergent regime was neither democratic nor consolidated, despite the efforts of the AMG 258 to establish a representative government based on competitive elections. Syngman Rhee?s government politicized the national police force and ignored constitutional restraints on power before, during and after the Korean War. On the Polity IV scale, South Korea?s government never exceeded -3 before the First Republic was dissolved in the 1960 coup that deposed Rhee (Marshall and Jaggers 2009). Moreover, South Korea was beset with multiple rebellions and uprisings that began during the occupation period. The case of the American occupation of Korea was a failure from a democratic standpoint. Election data is not particularly useful in this case for three reasons. First, the elections themselves were, at best, semi-competitive. Second, the parties were more personalist than programmatic, and therefore say less about the ideological spectrum of those who supported them. Third, independent candidates swamped party candidates in the popular vote in three of the four elections during the first republic (Croissant 2001). These are candidates about whom we have basically no information regarding their positions and politics, and therefore the data tells us little about the preferences of voters. All is not lost, however; there are a few things we can learn about South Korea based on electoral data. One can tell that the right and far right were successful in establishing control over the government due to the total absence of far-left participation in electoral activities, and a near-absence of participation among moderate-left and centrist parties (Croissant 2001). Even if one were to consider these elections competitive, one would see that parties tied to candidates that commonly espoused violence and opposed democratic procedure won the lion?s share of votes that did not fall to independents. Moreover, the dominance of independent candidates as a group confirms the finding that party-based democracy never fully developed in South Korea. The good 259 fortunes of independent candidates would not subside until Rhee?s Liberal Party and the opposition Democratic Party divided the electorate in the late 1950s (Ibid). The electoral system itself was heavily weighted in favor of the ruling coalition, with the Liberal Party averaging 55.1% of the seats in the National Assembly following the 1956 and 1958 elections, despite averaging only 39.5% of the popular vote. (Ibid) These data all corroborate the findings of the Polity project?s evaluation of South Korea?s ?First Republic? as undemocratic. The one issue that may have forged consensus in occupation-era Korea did so only in the narrowest sense, and not in a way that benefitted the occupation. While essentially all parties and organizations, save the KCP, came out in opposition of trusteeship, the right wing of the Representative Democratic Council (the earliest attempt at an AMG-directed interim regime) chose to continue opposing trusteeship but not the actual implementation of an interim government (Park 2002). The moderate left opposed both (Ibid), putting them on the opposite side of the AMG. The largest party of the moderate left, the Korean People?s Party, allied with the KCP, left-leaning nationalists and numerous workers groups to form the Democratic Nationalist Front as a counter to the political heft of the right (Choi 2002). In order to salvage a moderate coalition, the AMG urged Yo Un-hyong to leave the KPP and form a new Social Democratic Party that would cooperate with the right and the AMG, but Yo refused (Cumings 1981). The trusteeship debacle served more to divide the domestic political scene than to unite it, and not along lines that were favorable for the consolidation of a democratic regime. The initial reaction against trusteeship erupted into a violent crescendo that resulted in an assassination and an attempted coup. Song Chin-u, a member of the Korean 260 Democratic Party who maintained a positive relationship with the American Military government, had sought the cooperation of ardent trustee opponent Kim Ku. Han Hyon- u?a supporter of Kim Ku?murdered Song hours after that meeting (McCune 1947b, Cumings 1981, Choi 2002). This was only Kim?s first strike against the government, which culminated in a weak and easily repulsed coup against the Korean Provisional and American Military Governments (Cumings 1981, L and G 1946, Sook 2002). Kim Ku was seeking an alliance between the Korean Democratic Party and the Korean Provisional Government to establish a united front in favor of independence, and the Moscow Declaration?s call for a provisional government aside from the existing KPG raised his ire (Ibid). Kim?s attempt to rectify the situation was laughably unsuccessful and met with no efforts on behalf of the AMG to punish him for his activity (Cumings 1981). Assassination was a favored tactic of right-wing extremist and ultranationalist Kim Ku. It began with Hyon-Hu?s December 1945 assassination Song Chin-u for his advocating of trusteeship (McCune 1947b, Cumings 1981, Choi 2002) and continued with the murder of KCP leader Yo Un-hyong in July 1947 (after a failed attempt two months earlier), also by associates of Kim Ku (Cumings 1981, Oh 2002a, Oh 2002b, Matray 1985, Sook 2002). Kim Ku?s followers claimed another victim in December; right-wing politician Chang Tok-su fell after coming out in favor of foreign supervision of Korean elections (Matray 1985). Although not on a scale as large as that of Weimar Germany, paramilitary forces were powerful and consequential elements in the South Korean political scene in the years before the war. By mid-1946, there were two competing right-wing youth groups competing for recruits: the National Youth Movement, under the leadership of Syngman 261 Rhee, and the National Youth Corps, under the tutelage of Yi P?m-s?k (Cumings 1981, Han 1974). Four months after their formation in June, the Korean Youth Corps had 20,000 recruits (Han 1974). By January 1947, the National Youth Movement?which Syngman Rhee had just placed under the control of a Chinese Nationalist with ties to Chiang Kai-shek?boasted 30,000 recruits and the support of wealthy landlords and businessmen (Matray 1985). Following elections and Rhee?s ascent to power, Rhee sought to merge the competing youth groups and place them under his control. Along these lines, Rhee founded the Korean Youth Corps in December 1948, and merged the KYC with the National Youth Corps in January 1949 (Han 1974). The KYC would serve as Rhee?s personal police force (Ibid). Civil unrest was common during the occupation period. As noted above, the interregnum between the end of the Japanese occupation and the beginning of the American occupation saw the rise of independent political factions. Naturally, these factions posed some problems for the outgoing Japanese authorities and incoming American military authorities (who later declared these elements illegitimate). Immediately following Japan?s announcement of surrender, the historical record indicates hundreds of attacks on Japanese and Koreans who collaborated with the occupation (K.- K. Kim 1995). In October, there was a small clash between elements of the Nanw?n People?s Committee (affiliated with the CPKI and later the KPR) and AMG troops (Cumings 1981). These incidents were but a harbinger of the challenges the occupation authorities would face from both left and right. For all the right-wing extra-parliamentary activity in occupation-era South Korea, the most impressive examples were leftist: the Autumn Harvest Uprising of 1946 and the 262 Jeju Rebellion of 1948-9. Tensions between right-wing factions that controlled the national police and left-wing factions that opposed the AMG and KPG had simmered in the early years of the occupation. The period saw an increase in right-wing youth groups. One of the most prominent was the National Youth Corps, which anti-Japanese resistance fighter Yi P?m-s?k established in June (Han 1974). His group competed for recruits with Syngman Rhee?s Korean Youth Corps, which retaliated against left-wing youths protesting in front of the Soviet Consulate a month earlier (an attack that led to renunciation by the AMG and some short jail terms for the assailants; Cumings 1981). Simultaneously (between April and August), the Korean National Police carried out a campaign against leftist political leaders. Most prominent leftists ended up in jail or went into hiding by September (Cumings 1981). One of the raids included the headquarters of the National Council of Korean Labor Unions in August (Ibid). In response to the raids, 8,000 rail workers walked off the job in protest on September 23. Within a few days the strikes spread to printers, electricians, and student workers. Students refused to attend classes, and newspapers supported the strikes (Ibid). The unrest that began in Pusan would soon spread throughout the south, expanding into the Autumn Harvest Uprising of 1946. The initial strike became a general protest by all those who claimed grievances under the occupation government; the list of demands the strikers sent to General Hodge included increases in rice rations, higher wages, housing and rice for jobless workers and returnees, betting working conditions, freedom to organize, a labor law along the lines of that in the north, the release of political prisoners and an end to reactionary terror, and the transfer of power back to the people's committees (Cumings 1981). On September 29, the protest turned violent: 3,000 263 strike breakers joined with 3,000 police and youth group members and attacked 3,000 strikers in the Seoul rail yards (Ibid).80 The violence escalated on October 9 when a bloody battle between rioters favoring the strikers and police?supported by AMG soldiers?left twenty-four dead (Ibid, Hart-Landsberg 1998). It was this event that triggered the revolt that spread throughout the south. Most of these attacks were retributive against local police and administration (Ibid, Park 2002), but some of the attacks were directed at the AMG for having coordinated the strikebreaking activity (Hart-Landsberg 1998). Hodge, in support of the provisional government, sent in tanks as reinforcement and declared martial law, but the riots raged well into November (Ibid, Cumings 1981, Park 2002). Ten of thousands of South Koreans participated in the riots, resulting in attacks on thousands of police officers, hundreds of installations and residences of the police and local administrators. While Hodge blamed the attacks on the Korean Communist Party, its leader?Yo Un-hyong? denied responsibility; the gradual and diffuse nature of the riots indicates a spontaneous contagion-like movement rather than a coordinated attack (Hart-Landsberg 1998). The rebellion on the island of Jeju in 1948 began as a direct protest against the elections and grew into an outright guerilla war against the South Korean police and military (Hart-Landsberg 1998). The violence took 40,000 lives, and left only 170 of the 400 villages on the island inhabitable (Ibid). Anti-government activity began in March 1948, peaked in the autumn of that year and continued well into the first months of 1949. Between March and June 1948, terrorists operating under the banner of Communism torched and destroyed nearly 350 government buildings, killed 147 rightist candidates, 80 The mishmash of opponents to the strike continued to patrol industrial areas, wielding blunt weapons, and attacked publications that supported the strike (Ibid). 264 campaigners, officials and their families, and left 600 wounded (Han 1974). A guerilla army of four thousand established control over the central villages on the island by the summer (Hart-Landsberg 1998). South Korean forces launched their counterattack in May, with mixed results. Police forces managed to arrest 3,000, and attempted to control the insurgency by forcing civilians into guarded towns and defoliating wooded areas in which the rebels might hide (Hart-Landsberg 1998). The situation took a turn for the worse in Jeolla, when communists who had infiltrated the National Police Force successfully mutinied and drew the support of 3,000 villagers in nearby Yosu (Ibid, Matray 1985, Han 1974). This new wave of the rebellion established kangaroo courts and summarily tried and executed policemen and military officers. Sympathizers in Sunch?on raised the flag of the Democratic People?s Republic of Korea and pledged their allegiance to the North (Matray 1985). South Korean police and military forces, now officially an organ of the Republic of Korea, responded by executing suspected sympathizers without trial under the protection of a recently-passed National Security Act. The National Security Act declared martial law over the island of Jeju and permitted the government to dismiss educators and journalists suspected of disloyalty, engage in prior censorship of publications, and imprison editors (Ibid). Low-level fighting continued after this crescendo, and slowly petered out by May 1949 (Hart-Landsberg 1998) Rhee?s eventual dominance in South Korean politics is largely tied to the right wing?s monopolization of extra-parliamentary power, specifically through the police force and paramilitary elements. The AMG put the National Police Force (NPF), a remnant from the colonial era, under the control of anti-communist Cho Py?ng-ok (Han 265 1974, Cumings 1981). Working in conjunction with Rhee and Kim Ku, the NPF forced the KCP into hiding (Cumings 1981). From the very start, the NPF helped secure votes for Rhee and his allies through violence against potential competitors (Ibid). In the lead- up to the first elections in May 1948, hundreds of electoral officers and candidates died at the hands of police (Matray 1985, Gordenker 1958). The employment of police and paramilitary forces to influence electoral outcomes is the ultimate example of Rhee?s disdain for democratic politics and the antidemocratic nature of the first Republic of Korea. Summation The AMG?s attempts to establish a democratic regime failed. While the threat environment was more than favorable, the American Military Government in Korea passed up its opportunity to produce marginal occupation value. While they did help arm the provisional government, the threat environment demanded a higher commitment than the United States was willing to supply. Moreover, by not sufficiently distancing themselves from the former Japanese colonial government, and perhaps by strengthening their ties with Korea?s most recent and hated enemy, the US increased the security deficit it needed to overcome. By not producing marginal security, the US lacked a public good that it could have distributed in order to build support for the occupying authorities and for the regime they fostered. This hamstrung any efforts to narrow the spectrum of acceptable political contestation in such a way that would have empowered moderate elements. Moreover, the AMG?s misguided support for coalitions that were outside the optimal winning coalition, driven largely by miscalculation of these domestic coalitions? 266 aims and methods, empowered a constituency that was uninterested in building a democratic regime. The results reflect the expectations of the working hypotheses. A brief history of uprisings, and the response by the South Korean government, clearly indicates that the new regimes were unable to consolidate their authority without undermining democratic institutions. The leader of that government was a man who openly shared his disdain for democratic procedures. The democratic elements that did vie for power ran up against a system that was rigged against them. Executive and legislative recruitment were anything but open, as reflected in South Korea?s Polity IV score during that period. The AMG and the government they imposed failed to narrow the range of acceptable political contestation in a way that was favorable to democratic consolidation. Reconciliation of Competing Hypotheses Hypothesis 1?that there will be fewer protests and less political violence when the occupier is offering improvements in living conditions and establishing itself as an ally?is difficult to test in this case because it is difficult to find a period that reflects these conditions. The period most characteristic of this condition would have been the spring and summer of 1947. This was the period during which Acheson and the State Department were calling for increased aid to Korea and when relations between the US and USSR started to break down, but before the US embarked upon the reverse course in Japan and declared that it would not offer any substantial security commitment to South Korea. So was the American occupation zone more stable during this period? It does 267 seem that this was a period of relative calm. The Taegu Uprising and the general strikes that accompanied it faded by spring 1947; the Jeolla Rebellion would not occur until the following year. This period was not perfectly stable?it saw two assassination attempts on Yo Un-hyong, one of them successful. It was, however, relatively calm across the span of the case. If there is a problem with this finding, it is whether the greatest perceived external threat was the Soviet Union or Japan. If it is the latter, then this period does not fit the conditions set by the hypothesis. If it is the former, then this claim makes sense. At worst, the findings fail to reject the null hypothesis. Furthermore, aside from their attempts to build a provisional government, the AMG took a generally laissez faire position on domestic Korean politics, and so it is difficult to say which US policies were detrimental, beneficial or neutral. There were food riots and peasant rebellions in response to skyrocketing food prices, largely due to reforms that the AMG implemented and then quickly abandoned. However, food shortages that result in food riots are not a strong confirmation of the Marginal Value Model. The best example is the outbreak of anti-trusteeship protests, assassinations and Kim Ku?s attempted coup against the AMG that followed the Moscow Communiqu?. The overwhelming preference of Koreans in the south was for immediate independence and unification, and the announcement that the USSR and US would stick to their wartime plans for an interim of tutelage provoked understandable rage. Beyond these examples, there?s no significant correlation between US policy in its occupation zone and the worst periods of unrest, during the Taegu and Jeolla revolts (perhaps because there is not a significant amount of variance in AMG domestic policy in this case). 268 However, the history indicates an increase in domestic instability as the occupier makes friends with the occupied state?s enemies. While the Soviet Union steadily became a greater threat as the occupation progressed, Koreans continued to harbor fear and loathing towards Japan throughout the occupation period (and long after). The last period of relative calm during the Korean occupation?the interlude of spring and summer 1947?ended just as the US began to openly support the remilitarization of Japan. It was also at this time that it became clear the US would not risk its own resources to defend Korea from the Soviet Union or the puppet regime in the north. The period between 1945 and mid-1947 was not perfectly calm by any means, but it was measurably more stable than the period between late 1947 and mid-1949. The data here support the working hypothesis. Liem (1949) and Cumings (1981) supply some insightful theories as to why the first Republic of Korea was undemocratic. It is more than plausible that backing ultranationalists, empowering landlords who had reason to fear redistributive policies, and a lack of planning, understanding and respect for the actual situation ?on the ground? had an adverse effect on political freedom. These hypotheses are perfectly compatible with the Marginal Value Model, particularly in terms of managing the occupation relationship. Moreover, while the Liem piece is remarkable for its prescience and Cumings? history is one of the most comprehensive about the US occupation of Korea, neither provides a testable, social-scientific foundation for their conclusions. In sum, neither of these hypotheses are substantial rivals to the working hypotheses. If anything, they reinforce the Marginal Value Model. 269 Economic Drivers As in the other cases, we should not ignore economic variables. As is the case with Korean electoral politics, there is a dearth of economic data during the period of the case. What we do know about the occupation-era economics of South Korea is that the situation was bleak. Inflation grew out of hand. Agricultural prices skyrocketed. Industrial production was lacking, although it?s important to note that the southern half of the Korean peninsula was not an industrialized region to begin with (and wouldn?t seriously begin to industrialize until the 1960s). It is also important to note the connection between the economic conditions of the period and the policies that foreign powers? both Japan and the United States?imposed on the region. Although Hodge?s government in Korea took a different tack than MacArthur?s in Japan by ending agricultural price controls early on, occupation agricultural policy in South Korea was initially just as unsuccessful (Cumings 1981). Aware of the possibility of runaway inflation, skyrocketing rents and hoarding, occupation officials issued Ordinance #9 on October 5, limiting rents to one-third of the harvest. The popular measure went unenforced while landlords frequently disregarded the decree (Ibid). Massive inflation in food prices ensued, and the AMG felt it necessary to rescind Ordinance 9 and re-impose the colonial- era system of forced collection and redistribution (Cumings 1981). The situation was dire enough for Secretary of State Acheson to attack the ROK leadership in April 1950, chastising their failure to rein in inflation and reindustrialize (Farley 1950). Inflation was certainly a problem, but one could hardly place all the blame on Syngman Rhee & Co. In the first year of the occupation, rice prices exploded by a factor of nearly 300. They tripled again in the following year (Cumings 1981). The 270 general cost of living surged by a factor of 140 in that first year, while overall food prices increased hundredfold (McCune 1947b). Inflation was the result of a massive increase in the supply of the Korean yen (which was pegged to the Japanese yen and controlled by the colonial authorities) during WWII. The Korean yen supply increased from K?300 million in mid-1939 through K?8 billion by the time of Japan?s surrender in order to subsidize the wartime economy (Ibid). As brutal as the colonization of Korea may have been, most of the economic growth of the past half-century was the result of Japanese investments in industry and agriculture. Korean industrial production increased threefold between 1929 and 1938, but said production was concentrated in the northern reaches of the peninsula (L and G 1946). Between 1932 and 1943, the population of Korean industrial workers tripled (Cumings 1981). Meanwhile, over 90% of large factories were the property of Japanese owners (K.-K. Kim 1995). The industrialization of Korea was largely for the benefit of Japan, which purchased Korean goods and consumed Korea?s agricultural produce. During this general period, 70-80% of Korean exports went to Japan, although Japan?s dependence on Korean rice declined after the Japanese invaded Indochina (N. G. Kim 1997, Cumings 1981). Moreover, millions of Koreans worked (often involuntarily) in Japan in support of the war effort. In the final year of the war, 32% of the Japanese workforce was in fact Korean, while a majority of the mineworkers were Korean born (Cumings 1981). When the Japanese occupation ended, demand for Korean labor plummeted. Korean workers returning from Japan represented an additional burden when they returned home. At the same time, hundreds of thousands of Japanese repatriated to the main islands after WWII, draining the territory of the industrialists and managers who 271 had been responsible for Korea?s rapid industrialization (McCune 1947b). Finally, due to the northerly distribution of light and heavy industry on the peninsula, much of the growth potential and power generation in Korea belonged to the North and the Soviets, leaving the South to deal with subpar infrastructure and shortages in power, industrial goods and?crucial to an agrarian economy?fertilizer (Matray 1985). One should not ignore Korea?s economic situation, which likely contributed to the domestic unrest. Likewise, one should also not ignore the lack of interest in the occupying government to rebuild (or build, as it may have been) the economy in the region. We cannot assume that deprivation in this case was endogenous to the South Korean economy. In stark contrast to US actions in West Germany and Japan, the US made no substantial efforts to construct an industrial base in the occupation zone, to modernize their currency or to manage inflation. Such efforts could have produced goods that endeared the population to the occupying authorities and cleared the path towards a consolidated democracy. Summation of Findings and Conclusion The intra-case study of the US occupation of Korea and the attempted establishment of a democracy in South Korea conforms to the expectations of the working hypotheses. The period of greatest domestic unrest occurred during the period when the AMG grew more comfortable with Korea?s historic enemy and failed to balance against the new one. Events indicating instability were less common when the US looked to be on the verge of enhancing its commitment to Korea and before the 272 improvement in military ties between the US and Japan. The Marginal Value Model sufficiently encompasses the expectations of rival hypothesis. Despite an external threat level critical enough to produce marginal security, Washington failed to do so when it refused to signal its intentions to align with Seoul against Moscow, Beijing and Pyongyang. Even if the AMG had achieved these goals, it could easily have erased them all by promoting domestic actors outside of the optimal winning coalition whose preferences conflicted with those of the US. Finally, this case study contradicts Edelstein?s interpretation of the US occupation of Korea. While he also comes to the conclusion that the occupation failed, he claims this is due to an absence of a foreign threat. An examination of the objective threat environment during the occupation indicates that he is?to quote Don Herbert??right for the wrong reasons.? It is not the absence of an external threat that undermined the occupation. Indeed, the external threat level was objectively critical. Rather, it was the failure of the United States to demonstrate a credible commitment to defend Korea against that threat that doomed the endeavor. So long as the United States failed to persuade domestic elements of the added value of the occupation, Korea viewed the occupiers as the primary threat to their sovereignty and territorial integrity. This is understandable considering Korea?s experience with colonial subjugation, but it is nevertheless inaccurate in light of how vulnerable South Korea became once American soldiers left. Yes, Edelstein is right that Koreans did not perceive the threat north of the 38th. What Edelstein fails to do is explain why. This dissertation fills that gap. 273 Chapter 7: Review of Findings and Conclusion Review of Hypotheses and Findings Through four case studies the working intra-case hypotheses have held up well. H1: Events indicating regime instability will be less common among cases in which the occupiers increase marginal occupation value through the provision of security and other goods than they will be among cases in which occupiers reduce the value of the occupation. There were fewer events indicating regime instability in West Germany when occupation policy abandoned reparations and factory dismantlement and when the Western Allies brought West Germany into their security perimeter. In West Germany, what little political instability occurred seem to do so in the earliest years of the occupation, before the Allies ended their policies of demontage and fully renounced the desire to pastoralize Germany. Afterwards, the western occupation zones and the Federal Republic were generally calm. Electoral support for the optimal winning coalition remained generally strong throughout. There are no situations in which the Western Allies occupy with any obvious enemies of the Federal German Republic after WWII, or in which the US allies with enemies of Japan. In Japan, civil unrest is more common prior to the Korean War and in the earliest days of the occupation when the nation was on the brink of famine. However, electoral indicators do not conform as well as the extra-parliamentary indicators. For instance, the Communist Party?s success in the 1949 election directly contradicts H2, as the US seemed to be providing domestic wellbeing as well as aligning itself with Japan at that 274 point. The late success of the Japanese Communist Party in 1949 also seems to contradict the hypothesis?the optimal winning coalition should have been strong at this moment. One might want to attribute these anomalies to lag. In the Weimar Republic, domestic unrest appears to have dissipated during periods matching the description above?specifically after the Locarno agreement. The optimal winning coalition was more successful in capturing the Reichstag during these periods, as well. In Korea, this hypothesis is difficult to test due to a lack of variation on the independent variables. However, political unrest seems to have died down during the fleeting moments when it appeared the US might be willing to commit to Korea in the long run, and before the US set its sights on remilitarizing Japan. In the occupation-era Weimar Republic, the optimal winning coalition suffered its worst losses in response to the punitive nature of the Versailles treaty and its harsh enforcement. This era also saw a significant surge in paramilitary recruitment and political violence. There is a clear-cut case in Weimar Germany of the occupation forces pursuing security cooperation with a current or recent enemy of the occupied state. Through 1923, the AAF (specifically the French and Belgian authorities) sponsored Rhenish separatist movements and a Polish insurgency. During this period, we see an increase in paramilitary recruitment and massive losses for the optimal winning coalition. There is no situation in which the US strived for an improvement of living standards in the Korean case. There is, on the other hand, a clear cut situation in which the US sided with one of Korea?s enemies. Beginning in 1948, the US began to signal to the USSR and anyone else who might be listening?including those in the southern occupation zone in Korea?that they would rearm Japan and bring them into the US 275 security perimeter. This period correlates with one of the most extreme periods of political violence. In addition to measuring how well the model holds up within each case, we can apply the Marginal Value Model to comparative case histories. These hypotheses hold up against the inter-case comparison. H2: Events indicating regime instability with be less common within cases when the occupiers increase marginal occupation value than they will be when the occupiers reduce the value of the occupation. Without question, the two calmest cases in terms of domestic unrest are Japan and post-WWII Germany, while the two deadliest cases are the Weimar Republic and South Korea. If one were to rank the cases in terms of domestic unrest, we would see the following ordering: 1) South Korea; 2) Weimar Germany; 3) Japan; 4) West Germany. While both South Korea and Weimar Germany experienced periods of tumult and upheaval, only South Korea experienced two legitimate uprisings during the occupation. Japan was only slightly more unruly than West Germany, largely due to the volume of (typically nonviolent) mass protests prior to 1950. However, if were to rank the cases by paramilitary organization, we would flip the Weimar and Korea cases. If we were to focus on success of pro-democracy parties, we would see a reverse ordering: Japan and post-WWII Germany would be at the forefront, Korea and Weimar Germany would bring up the rear. Regardless of how you measure regime instability, there is a clear dichotomy: West Germany and Japan were exceptionally stable; Korea and Weimar Germany were exceptionally unstable. This mostly correlates with the H2 in the direction that the Marginal Value Model expects. The US failed to directly promote the welfare of Koreans 276 during its occupation, while Washington and made no effort to establish a Korean- American alliance until after the occupation concluded and the Korean War began. In Germany, the US, UK and France pursued a policy of reconstruction and paired reconstruction with a major defense commitment. The same applies to the American occupation of Japan. The case that stands out is the Weimar Republic, where the AAF eventually did act to promote domestic welfare and acted in support of the occupied state, but did so after pursuing the exact opposite policy for five years. One can consider this a case in which the Weimar Republic does not fit the specified population or one in which it fails to support the hypothesis. Either way, the findings strongly support the hypothesis in three cases and may support all for. The cases in which the allies pursued the most constructive occupation policies were Japan and Germany, with the occupiers outlaying substantial resources to rebuild political and economic institutions. The most deleterious policies occurred in the Weimar Republic with reparations and the occupation of the Ruhr. The US neither pursued a constructive nor destructive occupation policy in Korea from any objective standpoint. However, the US did align itself with Japan, Korea?s arch-enemy at that point in history. Just the same, the AAF aligned itself with the Polish insurgents and Rhenish separatists in the Weimar. In contrast, the occupiers allied themselves with Japan and post-WWII Germany against the Soviet Union. The occupiers established a mutual security pact with the occupied territory in both German cases and the Japan case. In the Weimar case, however, the occupied territory concluded a secret pact with Russia before the more public agreement at Locarno. This is also the case with the higher level of regime 277 instability. The case history supports the hypothesis. The US made no attempt to establish a security pact with Korea before the conclusion of the occupation. H3: Democratic regimes will survive only in cases where the occupiers manage to provide economic aid and commit materially to defending the new regime from a foreign threat before the occupiers aid an enemy of the occupied state and/or before the occupied state forms an alliance with another, non-occupying state. In all cases, the occupiers offered some sort of security pact to the occupied state at some point during the observation period. However, only in two cases did the occupiers offer security assistance before assisting the occupied state?s enemies or before the occupied state found another security partner. In the Weimar Germany case, the occupiers secured a mutual security pact with the German government, but only after the occupiers aided Rhenish separatists and Polish insurgents, and only after these threats and the disarmament mandates drove Germany into Soviet Russia?s arms. In the case of South Korea, the US offered security assistance only after the end of the occupation in response to the Korean War (and after the US had begun to recruit Japan as the first line of defense in the Cold War in Asia). Only in the Federal Republic and Japan did the occupiers offer a security guarantee early enough to mitigate the threat posed by the occupation itself. Not coincidentally, these were the successful cases. 278 Reviewing Rival Hypotheses Demilitarization The four cases indicate that, contrary to the demilitarization thesis proposed by Linz (1990) and Stepan (1991), remilitarization is far more important. The two successful cases?Japan and Germany?were cases in which the military was initially disbanded but then rebuilt in such a way that it bolstered institutions that favored the optimal winning coalition. In the failed cases, either demilitarization failed or was not necessary in the first place. Disarmament was a key component of the Weimar occupation and it failed in a spectacular fashion. This highlights the fact that a weakened state in a threatening anarchic system always has some demand for security. If a state and its constituencies prefer survival, they will seek external security from one source or another. In the case of Weimar Germany, they sought security from the remnants of Kaiser Wilhelm?s II regime and from Russia. They sought security from these sources because the occupation relieved the Weimar state of its own capacity to produce its own security goods without providing an alternative source. Demilitarizing the Weimar Republic endangered the regime rather securing it, and it was the remilitarization of the FRG and Japan that secured the new regime and empowered the optimal winning coalitions. Finally, in the case of South Korea, demilitarization was irrelevant to the consolidation of the regime? the Japanese had largely completed that task long before the Americans arrived. The ?debellicization? thesis does a poor job of explaining success and failure in these cases. Political Re-education 279 Political re-education through educational policy and institutional learning may play a significant role in consolidating democracies in the long run, but it is not a process that is applicable to all cases. It is conceivable that the banishment of anti-democratic actors from the domestic scene and reconfiguration of educational policies to engender democratic values did increase pro-democracy affect in future generations. This theory is problematic considering that weak democracies often fail to survive a full generation. This was the case for Weimar Germany and South Korea. Political re-education might strengthen democracies, but only democracies that are on the path to consolidation. Therefore, re-education processes shed no light on the weakest cases and no light on half the cases in this dissertation. Institutional Design The character of the institutions that develop under occupation may in fact have a major impact on the survival or collapse of a democratic regime, but the literature on institutional design and its correlation (or lack thereof) with success and failure in these cases sheds little light. It is difficult to separate the effect of the institutions from the effect of a polarized electorate. However, we have established that electoral constituencies in South Korea and Weimar Germany were more radical than those in Japan and the Federal Republic. We have also established that radical groups in South Korea and the Weimar Republic directed their energies towards extra-parliamentary politics. In other words, institutions themselves do not marginalize radical elements, although they can deny them access to the formal reins of power. It is important not to empower radical groups, but it is first important to marginalize them before they can 280 seriously threaten the new regime. This is the joint responsibility of the occupation forces and the new regime. In order for electoral and representative institutions to be successful in a young regime imposed from above, powerful actors must marginalize extra- parliamentary threats to those institutions. This is what the occupation forces and their domestic partners accomplished in the Federal Republic and Japan, and it is what they failed to accomplish in Weimar Germany and South Korea. Delegitimation via Defeat In the three cases in which the occupiers ousted a pre-existing regime of domestic origin (the two German cases and Japan), delegitimation of the previous regime does not seem to play a significant role in democratic consolidation (or its absence). The defeat of the Third Reich plausibly delegitimized the NSDAP, however this does not explain the marginalization of non-Nazi radical parties on either the right or left after 1945. In Japan, defeat did not lead to the immediate delegitimation of the old regime?administrators and legislators from the old regime remained out of politics only so long as MacArthur?s occupation forbade them from participating. In Weimar Germany, defeat delegitimized the coalition that supported Kaiser Wilhelm II and the Great War. However, these parties remained illegitimate only until the imposition of the Treaty of Versailles largely delegitimized its signatories?the optimal winning coalition that brought democracy to Germany. This illustrates the primary deficiency of the delegitimation thesis: it only applies until the behavior of the occupation forces begins to have an effect on domestic constituencies. Domestic constituencies reject the ancien r?gime because it became a threat to their continued existence. If the new regime also becomes a threat due to its 281 association with destructive occupation policies, then that regime will lose legitimacy as well. This is what happened in the Weimar case, and it is similar to the failure of the American military government?s attempts to establish a legitimate democratic government in South Korea. There seems to be no correlation between the type of delegitimation process described by Sa?Adah (2006) and the consolidation of democratic regimes in these cases. Geographic Division The argument that an occupation can excise antidemocratic elements by dividing the territory may be a sound one but it is difficult to test in this dissertation. There are only two cases in which the occupiers divided the territory: West Germany and South Korea. One case was a spectacular success, while one was a terrible failure. However, in the division of the Korean Peninsula, it was the industrialized north that fell to the Soviets and the agrarian south that fell to the Americans. Perhaps, in an alternate universe, where South Korea was initially more industrialized and filled with capitalists and proletarians, the results may have been different. However, we cannot test this case nor can we test the possible effects of dividing Weimar Germany or postwar Japan. What is evident, however, is that attempts to divide a country against its will tend to turn its people against the occupiers. Koreans and Germans both opposed unification; Germans East and West clamored for unification until it finally occurred at the end of the Cold War, while Koreans still clamor for unification today. Syngman Rhee openly disobeyed General Hodge and American policymakers stateside in his attempts to reunify the peninsula. Likewise, French attempts to separate the Ruhr and Saar from Germany drew protests in 282 both postwar cases. Division would likely do more harm than good if it signals that the occupiers are enemies rather than allies. Nevertheless, these cases are not an appropriate sample to test this hypothesis. Retention of Institutions Coyne?s (2008) hypothesis about how the retention of pre-existing institutions smoothed the transition from autocracy to democracy is problematic for several reasons. In the Japanese case, it assumes that the Japanese people would be unaware that the Japanese government was serving the interests of its American occupiers. This is an absurd assumption that underestimates the understanding of domestic constituencies. Moreover, it ignores the fact that coopting Japanese institutions dispossessed the previous elite class, meaning that the previous elite is no more or less likely to oppose the occupation. In the case of the Federal Republic, despite Coyne?s claims to the contrary, this is simply not what took place. The occupation forces did not leave significant democratic institutions intact in Germany following WWII?the Allies erased preexisting military institutions and rebuilt them from scratch; replaced institutions that had been ruthlessly centralized with a federal system; discarded the legislative body that Hitler had rendered irrelevant and replaced it with a new system with electoral and representative rules that differed significantly from the Weimar-era Reichstag. The Allies neither intended to preserve, nor did they preserve, the character of either the Nazi or Weimar institutional arrangements. In South Korea, the United States largely did retain Japanese 283 institutions from the colonial era?including property laws, restrictions on expression, administrators and police forces?with less than optimal results. If there is one case in which Coyne?s thesis may hold up, it may in fact be Weimar Germany. The limited involvement of the Allied and Associated Forces in destroying the old regime and constructing the new meant that the revolutionaries worked with the institutions that the Kaiser left them. The primary contributions of the Weimar Constitution were the enumeration of new rights, granting the Reichstag new powers and the formal elimination of the aristocracy. That said, the socialists and democrats who took power in the early years of the Weimar Republic largely ruled through pre-existing institutions just as the Americans ruled through Japanese ones. Of course, this contributed to the ruin of the regime when Adolf Hitler successfully manipulated the powerful executive (itself a relic of the previous regime). Moreover, the retention of institutions did nothing to stem the rising tide of radical nationalist and communist forces that saw no legitimacy in the Weimar Constitution. In none of these cases does Coyne?s hypothesis seem to adequately explain consolidation or failure, and in one key case it seems to explicitly contradict the historical record. Prewar Modernization That Japan modernized before WWII?in other words, that it transitioned towards a market economy with industrial modes of production and a political system capable of distributing public goods?may certainly have played a roll in the successful democratization of that country. Pempel (1992) emphasized the role of the Meiji Restoration and the industrialization of the Tokugawa Era, building on work by Lipset 284 (1959) and Moore (1966) indicating the importance of modern values and social structures to the evolution and consolidation of democracy. However, this hypothesis fails to explain why Japan evolved more like the Federal Republic and less like the Weimar Republic, both of which evolved after Germany became a centralized state with broad redistributive capabilities and an industrialized economy. It also fails to explain why Weimar and Federal Germany evolved so differently. It also fails to explain what to expect in new democracies with both an autocratic and democratic history (such as the Federal Republic). Prewar modernization does not provide a satisfactory explanation for understanding the diversity in processes and outcomes in the four cases here. Economic Growth I have stressed several times in this dissertation that one should not discount the role of economic growth in democratic consolidation, in regimes imposed or otherwise. It is most important to remember that growth is dependent on good policy, and rarely are indicators of growth truly independent variables. It is tempting to refer to the Japanese and German postwar growth stories as ?miracles,? but this ignores the intentional role the occupations assumed in rebuilding the economies of these states, recapitalizing their markets, and plugging them into a global trading regime that created an outlet for their goods and access to cheap materials. One of the primary reasons the occupation forces were able to facilitate consolidation and marginalize antidemocratic constituencies in the Federal Republic and Japan was due to the massive amounts of economic aid and the construction of open economies in these cases. It is possible that the failure to privilege democratic constituencies in Weimar Germany was due to the fact that the Allied and 285 Associated Forces did the opposite: they isolated Germany with customs barriers and sucked capital out of the market through reparations. In South Korea, the antidemocratic coalition that eventually took power was deeply aligned with a landlord class that opposed market reforms and had no special interest in opening or industrializing their economy, while American intervention into markets primarily took the form of abortive liberalization attempts without any plans for global integration or recapitalization. Economic growth is important in consolidating imposed regimes, but growth is a function of occupation policy: a prerequisite, just like producing marginal value through providing goods of security and public welfare. In other words, economic growth is not a rival hypothesis: it is a companion hypothesis that the Marginal Value Model fully incorporates. Summation This dissertation considers several alternative hypotheses that attempt to explain the consolidation and collapse in each case. Some of these alternative approaches apply universally to all cases of democratic consolidation, while some specifically address the case at hand and make no claims of generalizability. In the cases where the authors claim to explain consolidation in general, their hypotheses either fail to explain these cases in a satisfactory manner or do so in such a way that complements, rather than invalidates, the Marginal Value Model. In the cases where the authors claim to explain a specific case, the models prove their limited utility by demonstrating their lack of generalizability and their inability to explain what the Marginal Value Model does not. All in all, the Marginal Value Model both explains democratic consolidation and collapse in the cases 286 at hand and does so in a generalizable manner while explaining as much or more than its rivals. Brief Application of the Marginal Value Model to Ongoing Cases The Marginal Value Model yields predictions about the prospects of three salient ongoing democratization projects: Iraq following Operation Iraqi Freedom, Afghanistan following Operation Enduring Freedom and the Palestinian Authority under Israeli occupation. There is not space to deal with these in the detail applied to four cases in this dissertation. Nonetheless, a cursory glance can tell us what the Marginal Value Model expects about the future of the regimes in these three cases. As established constantly throughout this thesis, there are three requirements for a successful consolidation of a democratic regime under occupation: the presence of an external threat apart from the occupier, the provision of security goods by an occupier to defend against said threat, and additional goods of a sort that make the occupied state better off than it would be otherwise. Let us examine whether these three conditions exist in each case. Palestine Few occupations in history have drawn closer scrutiny and greater opprobrium than Israel?s occupation of the West Bank and the Gaza beginning in 1967 (Kassis 2001). The occupation has been characterized by sporadic violence, including two broad uprisings from 1987-93 and 2000-05 (the First and Second Intifadas; Hilal 2003). It was 287 not until twenty-seven years after the occupation began that it became an exercise in democratization. The Palestinian Authority was established in 1994 as a result of the Oslo Accords (formally known as the on Declaration of Principles on Interim Self- Government Arrangements), which provided a roadmap for the Israeli-Palestinian peace process, reconstruction of the West Bank and a transition to a democratic Palestinian government (Shikaki 1996; Hilal 2003). The United States injected itself into the process of democratization following the Declaration of Principles, and then once again as part of the George W. Bush Administration?s broader Middle East policy beginning in 2002 (Veliotes 2002). Is there an external threat greater than the state of Israel? There does not appear to be. The Palestinian people find themselves in their present situation primarily due to the existence of Israel. Palestine was denied statehood as a result of the victory of Israeli forces in 1948, and predominantly Palestinian areas came under Israeli occupation following the pre-emptive 1967 war against Egypt, Jordan and Iraq (McWilliams and Piotrowski 2009). Palestine and allied forces failed to significantly dislodge Israel following the 1973 Yom Kippur War (Ibid). While the relationship between Israel and various Palestinian opposition groups has varied over the years, the current state of affairs primarily stems from the results of these two wars. It would be hard to imagine any situation in which a foreign threat could pose a greater threat to Palestine than Israel itself. Without a strong external threat, it is doubtful that Israel could produce security goods that would compensate for the threat of the occupation itself. Highlighting this fact is the comparatively large amount of aid that arrives from non-Israeli sources, such as the 288 European Union (Stetter 2003). The recent economic deprivation following the Second Intifada and Israeli blockade has had no success in narrowing the range of acceptable political contestation around pro-democracy constituencies; instead they have eradicated an already weak Palestinian middle class and driven support from the moderate Fatah faction to the radical Hamas (McWilliams and Piotrowski 2009). In the absence of a strong external threat apart from Israel, no capacity for Israel to provide security goods, and occupation policies that have done more harm than good to the most important Palestinian constituencies, the Marginal Value Model would predict failure of consolidation for any democratic regime under the Palestinian Authority?or any authority in the creation of which Israel plays a major role. Afghanistan The US-led operation in Afghanistan, officially Operation Enduring Freedom, commenced on October 7, 2001, and evolved into a multinational peacekeeping and state-building operation led by NATO?s multinational International Security Assistance Force (ISAF; Goodson 2005). It is difficult to find a foreign threat in the Afghan case unless one stretches the definitions of ?foreign? and ?threat.? Technically, the Taliban grew out of the Pakistani Mujahedeen, enjoyed (enjoys?) Pakistani sponsorship, and overthrew the post-Soviet Islamic State of Pakistan (replacing it with the Islamic Emirate of Afghanistan in 1996; Ibid, Price 2012). After an initial defeat at the hands of US and NATO forces, the Taliban resurged from 2004 onward, resulting in the majority of Afghan civilian deaths and threatening to disrupt the 2009 elections (Worden 2010). The problems with classifying the Taliban as a foreign threat are A) it emerged from a faction 289 that helped expel the Soviet Union following the 1980s occupation, and B) it is composed primarily of Pashtun tribesmen who are native to Afghanistan as well as Pakistan and move relatively freely across the mountainous border region (Ibid). However, for the purposes of this exercise, let us consider the Taliban a pressing foreign threat to Afghanistan. How well have the United States and ISAF managed the Taliban threat since their intervention? At first, very well; the coalition mustered its forces quickly and drove the Taliban out of Kabul and Kandahar. The intervention came with billions of dollars in aid and a program of militarization intent on building a centralized, 70,000-strong Afghan National Army (Goodson 2003, Goodson 2005, Rubin 2006). Moreover, coalition forces have made a determined effort to modernize and democratize Afghanistan. The provisional government, led by Hamid Karzai and installed by the coalition, transitioned to a constitutional, presidential republic in 2004 (Rubin 2004, Worden 2010). Through 2010, the US and ISAF participant states spent over $4 billion on infrastructure, electoral mobilization and arming local warlords to fight the Taliban and allied militias (Ibid; Rubin 2006; Nixon and Ponzio 2007). Unfortunately, commitments of force and economy have not produced the desired results. The first elections in 2004 went smoothly but produced victories for local warlords and war criminals whose only virtue is that they opposed the Taliban (Hill 2010; Worden 2010). Even these efforts have been largely unsuccessful. While the dollar commitment was large, the US and ISAF never deployed more than 20,000 troops (Goodson 2005)?not much for a country the size of Alaska and a population as large as California?s, but as much as the United States was willing to commit while administering 290 simultaneous occupations. Moreover, US and ISAF missions have often been at cross purposes, one (the US) funding anti-Taliban militias that are often as deadly as their enemy, and the other (the ISAF) attempting to keep the peace by disarming warlords (Rubin 2006). The result is an unstable country wherein large swaths of territory are still under the control of private militias (Ibid). The most recent elections were far less promising: registration and vote counting were rife with fraud, Karzai?s most serious challenger withdrew at the last moment, and only a last minute deal prevented Taliban- affiliated terrorists from turning the poll into a bloodbath (Worden 2010). While it is clear there has been a major effort to state-building and democratization in Afghanistan, and the US and ISAF have signaled their commitment through significant aid disbursements, it does not appear that the coalition has produced marginal occupation value, nor have they successfully narrowed the spectrum of political contestation to include democrats and exclude antidemocratic factions. The meager force commitment was enough to bed down the Taliban early in the occupation, but Islamist radicals have rebounded since, in many cases joined by equally violent factions that once enjoyed US or NATO support. Moreover, while electoral institutions are in place, these institutions are often nothing but a conduit for local warlords to exercise their paramilitary influence through more formal channels and protect their interests. Rather than exclude antidemocrats, the introduction of elections and a representative parliament (the Loya Jirga) have served only to institutionalize the power structure that emerged in 2002. While more hopeful than the Palestinian case, the Marginal Value Model does not predict democratic consolidation in Afghanistan. 291 Iraq The most recent confrontation between Iraq and a US-led coalition began on May 20, 2003, ostensibly in response to Iraq?s alleged noncompliance with UN Security Council Resolution 1441 (banning the production of weapons of mass destruction and requiring the admission of weapons inspectors; McWilliams and Piotrowski 2009). Coalition forces defeated Saddam Hussein?s regime quickly, declaring victory on April 9, 2003 and establishing the Coalition Provisional Government (Diamond 2005). The US appointed Paul Bremer to lead the Coalition Provisional Government until the coalition officially returned sovereignty to Iraq in June 2004 (Dodge 2010). The civilian occupation of Iraq ended with the departure of US forces on December 15, 2011 (Dodge 2012). The Marginal Value Model does not predict success in this case. The greatest potential foreign threat to Iraq at this time would be Iran, with which they fought a bloody war during the 1980s (McWilliams and Piotrowski 2009). However, there is no evidence that Iran poses an existential threat to the Iraqi state at this time. The most prominent threats to personal security in Iraq following the 2003 invasion were internal, not external. Among the more prominent was the uprising led by Muqtada al- Sadr, which drew support from the disenfranchised Shia population in Baghdad and Basra (although the uprising drew Shia insurgents from Iran as well as other neighboring states; Dodge 2010). Also contributing to violence in occupation-era Iraq was the dissolution of the Revolutionary Guard and Bremer?s de-Ba?athification policy, aggrieving the predominantly Sunni population which enjoyed a privileged position in Hussein?s government (Diamond 2005; Moon 2009). These uprisings, both Sunni and Shia, produced violence that significantly hampered reconstruction efforts and, through 292 2009, made Iraq a far more dangerous place than it was before the invasion (Dawisha 2005). In other words, the greatest threat to Iraqi security post-2003 was internal violence that emerged as a result of?not in spite of?intervention by the United States, United Kingdom and the rest of the coalition. The intervention is not without its silver linings. A surge in the US troop presence, reconstruction of the Iraqi armed forces, and a ceasefire between Sunni and Shia insurgents ended the civil war by 2009 (Dawisha 2010; Dodge 2010). While voting in the first elections in 2005 was divided along ethnic lines, the March 2010 poll produced support for secular and nationalist factions (Dawisha 2010). Civil society in the form of private, active, local organizations seems to be rebounding from the oppression of Hussein?s regime (Dawisha 2005). At the moment, the regime is nominally democratic with generally fair elections and a representative legislature (Ibid). However, the potentially dictatorial ambitions of Prime Minister al-Maliki are cause for concern. Maliki has retained control of the premiership since 2006, during which he has sponsored the proliferation of private ?death squads? and secured control over a counter-terrorism force that is accountable only to him (Dodge 2010). The inability of either the coalition or the new regime in Baghdad to limit the reach of violent extra-parliamentary politics in Iraq for several years is troubling, an indicator of the incapacity of the occupation force to successfully narrow the spectrum of political consolidation. The rise of a leader who is reluctant to submit his capacity for violence to rule of law is indicative of the same. The eventual end to sectarian violence is a point in the coalition?s favor, but the fact that the civil war began as a result of the intervention and the decapitation of the Ba?athist regime indicates that the occupation did 293 not yield marginal value. The absence of a clear foreign threat, the introduction of a domestic threat by the occupiers run counter to what the Marginal Value Model indicates is necessary for the successful consolidation of an imposed democratic regime. Moreover, any US-led coalition must also account for their responsibility for the economic deprivation resulting from the imposition of sanctions following 1991?s Operation Desert Storm. In order to demonstrate that the US could be a credible ally to Iraq, it must produce a strong enough signal to overcome the damage the United States has inflicted since they forced Iraq?s withdrawal from Kuwait. There is no evidence in this case that this has occurred, and so the Marginal Value Model does not expect that this regime will consolidate successfully. The dictatorial tendencies of the Maliki regime point in this direction as well. Conclusion While the model does not fit the observed outcomes perfectly, the patterns are reasonably clear. Within cases and across cases, regime instability is more common when the occupying states engage in behaviors that indicate the production of marginal security. When the occupying power is protecting the occupied state against a foreign threat, the electoral and parliamentary fortunes of the optimal winning coalitions improve, paramilitary expansion and recruitment trends downward, and domestic insecurity dissipates. We see the same when the occupying power supplements these behaviors with occupation policies that improve the welfare of the electorate. Moreover, we witness increased regime instability when the occupier sides with the enemies of the 294 occupied regime. When occupiers produce more security than they consume, they improve the fortunes of pro-democracy constituencies by eliminating grievances and demonstrating the value of democratic institutions and parties. This narrows the spectrum of acceptable political contestation, bolstering moderates and marginalizing radicals. The result is a society wherein groups that rely on democratic institutions thrive while groups that oppose democratic institutions are absorbed by pro-democracy coalitions or fade into irrelevancy. By ensuring that the democratic regime has significantly more allies than enemies, this process safeguards democratic institutions for the long haul. Contributions to the Literature I see this thesis as a first attempt to contribute to multiple literatures in international relations. First and foremost, this study contributes to our understanding of foreign imposed regime change, specifically those of the democratic type. To understand whether or not it is feasible to impose democracy in foreign powers, one must have some understanding of how nascent democratic regimes fare under the constraints of foreign occupation. Moreover, the first step toward assessing the practical value of such a policy begins with understanding what types of occupation policy are more or less conducive to the long-term success of the new democratic regime. This remains a small, methodologically homogenous literature, and it is here where I believe this work can have its biggest impact. Second, this work addresses the equally small literature examining the success of foreign occupations. Albeit, this thesis addresses a particular goal?democratization? whereas states consider multiple explicit and implicit goals when occupying foreign 295 states, but it picks up where Edelstein left off by narrowing the focus, defining more clearly the a priori conditions of success and failure, and relying on objective rather than subjective variables in both the independent and dependent variables. In looking more closely at occupation outcomes in future scholarship, we should consider the role of external and internal threats, as Edelstein argues. But we also have to consider the capacity of the occupiers to respond to that threat in a way that is favorable to the occupied state. This is the mechanism by which threat dynamics should influence success and failure. Third, I see this work as the latest contribution to the broader catalog of work examining democratic consolidation. While scholars have expended a great deal of energy attempting to explain why some democratic regimes consolidate and others do not, a relatively minuscule segment of the literature is dedicated to understanding the role of external security dynamics. This is critical in the case of imposed regimes, or simply democratic regimes that arise under foreign occupation. In such situations, the regime is constrained both from above and below. Since the institutional setting for political competition is significantly different when states exercise their powers in foreign states, it is unreasonable to simply assume that consolidation will fail or succeed in these cases as well as it does in others. In other words, there is a gap in the consolidation literature, a gap I believe this project begins to fill. Finally, I believe this approach to the issue of imposed democratization be an improvement on the existing approaches spearheaded by Coyne, Enterline and Greig. As noted in the first chapter. The statistical approaches by Enterline and Greig warp the concepts of imposition and consolidation so grotesquely that their findings cease to be 296 relevant to real-world processes. On the other hand, Coyne?s comparative historical approach only scratches the surface of the cases he studies, missing important processes and assuming the presence of others for which there is no evidence. This thesis attempts to forge a middle path, although one that skews closer to Coyne than Enterline and Greig. The Marginal Value Model clearly defines what is meant by democratization, consolidation and occupation, and then seeks to investigate these phenomena through historical process tracing on multiple levels of analysis. In doing so, we unearth a deeper understanding of the process of democratic consolidation in pre-sovereign states that retains the potential for generalization once applied to other cases. Going Forward There is still much work to be done along the lines of this dissertation. First, there is room to improve the depth of each case study. By drilling deeper into lower levels of analysis one might uncover evidence of the actual ideological and psychological processes that are taking place in the minds of constituents elites. Second, I need to include more cases. There are at least nine other cases of democratic regimes arising under military occupations. If the Marginal Value Model is accurate, then we should be able to uncover the same relationships in Cuba and the Philippines as in Korea, in Austria and Italy as in Germany. Finally, there are ongoing cases that this model may be able to shed some light upon. Most obvious are the cases of Iraq and Afghanistan, but also applicable is the case of the occupation of the Palestinian territories by Israel. Will democratic institutions in these regimes resemble the success of West Germany and Japan, will they flounder as they did in the Weimar Republic and South Korea? The 297 answer comes down to the prevailing threat environment, measures to provide security, and the general tenor of relations between the occupiers and the occupied. This would be the next step for future investigation into the matter of democratization in occupied territories. 298 Bibliography Abe, Hitoshi, Muneyuki Shind?, and Sadafumi Kawato. The Government and Politics of Japan. Translated by James W White. Tokyo: University of Tokyo, 1994. Adenauer, Konrad. "Konrad Adenauer in the Bundestag on the European Coal and Steel Community, 12 July 1952." In Politics and Government in the Federal Republic of Germany: Basic Documents, by Carl-Christoph Schweitzer, Detlev Karsten, Robert Spencer, R. Taylor Cole, Donald Kommers and Anthony Nicholls. New York: Berg, 1984b. ?. "Memorandum of Chancellor Konrad Adenauer to the Allied High Commission on the Security of the Federal Republic, 29 August 1950." In Politics and Government in the Federal Republic of Germany: Basic Documents, by Carl- Christoph Schweitzer, Detlev Karsten, Robert Spencer, R. Taylor Cole, Donald Kommers and Anthony Nicholls. New York: Berg, 1984c. ?. "Speech by Konrad Adenauer, Chancellor of the Federal Republic, at the Petersberg Ceremonies, 21 September 1949." In Politics and Government in the Federal Republic of Germany: Basic Documents, by Carl-Christoph Schweitzer, Detlev Karsten, Robert Spencer, R. Taylor Cole, Donald Kommers and Anthony Nicholls. New York: Berg, 1984a. Alexander, Gerard. "Institutionalized Uncertainty, The Rule of Law, and the Sources of Democratic Stability." Comparative Political Studies 35, no. 10 (December 2002): 1145-1170. ?. "Institutions, Path Dependence, and Democratic Consolidation." Journal of Theoretical Politics 13, no. 3 (July 2001): 249-270. Alvarez-Rivera, Manuel. Election Resources on the Internet: Elections to the German Bundestag - Results Lookup. January 13, 2012. http://electionresources.org/de/ (accessed January 22, 2012). Armstrong, Charles K. "The Cultural Cold War in Korea, 1945-1950." The Journal of Asian Studies 62, no. 1 (February 2003): 71-99. Aruga, Tadashi. "Japan and the United States: A Half-Century of Partnership." In Japan and the United States: Fifty Years of Partnership, by Chihiro Hosoya. Tokyo: Japan Times, 2001. Auer, James E. "Article Nine of Japan's Constitution: From Renunciation of Armed Force 'Forever' to the Third Largest Defense Budget in the World." Law and Contemporary Problems 53, no. 2 (Spring 1990): 171-187. 299 Banchoff, Thomas. "The Enduring Transformation of Postwar German Foreign Policy." In The Postwar Transformation of Germany: Democracy, Prosperity, and Nationhood, by John S. Brady, Beverly Crawford and Sarah Elise Wiliarty. Ann Arbor, MI, USA: University of Michigan, 2002. Bandelj, Nina, and Bogdan Radu. "Consolidation of Democracy in Postcommunist Europe ." Center for the Study of Democracy 6, no. 4 (2006): 1-23. Barak, Oren. "Dilemmas of Security in Iraq." Security Dialogue 38, no. 4 (2007): 455- 475. Barnet, Richard J. The Alliance: America, Europe, Japan, Makers of the Postwar World. New York: Simon and Schuster, 1983. Barnhart, Michael A. "A Secondary Affair: United States Economic Foreign Policy and Japan." In Democracy in Occupied Japan: The U.S. Occupation and Japanese Politics and Society, by Mark E Caprio and Yoneyuki Sugita. New York: Routledge, 2007. Bendix, Rheinhard. "Social Stratification and Political Power." In Class Status and Power, by Rheinhard Bendix and Seymour Martin Lipset. Glencoe, IL: The Free Press, 1953. Bensahel, Nora. "Mission Not Accomplished: What Went Wrong with Iraqi Reconstruction." Journal of Strategic Studies 29, no. 3 (August 2006): 453-473. Berman, Sheri. "Civil Society and the Collapse of the Weimar Republic." World Politics 49, no. 3 (April 1997): 401-429. Bernhard, Michael. "Democracy in Germany: A Reappraisal." Comparative Politics 33, no. 4 (July 2001): 379-400. Bernhard, Michael, Christopher Reenock, and Timothy Nordstrom. "Economic Performance And Survival In New Democracies: Is There a Honeymoon Effect?" Comparative Political Studies 36, no. 4 (May 2003): 404-431. Boix, Carles. Democracy and Redistribution. Cambridge, UK: Cambridge University, 2003. Borton, Hugh. "American Occupation Policies in Japan." Proceedings of the Academy of Political Science 22, no. 4 (January 1948a): 37-45. ?. "Occupation Policies in Japan and Korea." Annals of the American Academy of Political and Social Science 255 (January 1948b): 146-155. ?. "Preparation for the Occupation of Japan." The Journal of Asian Studies 25, no. 2 (February 1966): 203-212. 300 Braibanti, Ralph JD. "Occupation Controls in Japan." Far Eastern Survey 17, no. 18 (September 1948): 215-219. Bresser Pereira, Luiz Carlos, Adam Przeworski, and Jos? Mar?a Maravall. Economic Reforms in New Democracies: A Social Democratic Approach. Cambridge, UK: Cambridge University, 1993. Bueno de Mesquita, Bruce, Alastair Smith, Randolph M. Siverson, and James D. Morrow. The Logic of Political Survival. Cambridge, MA, USA: Massachusetts Institute of Technology, 2005. Buhite, Russell D. "'Major Interests': American Policy toward China, Taiwan, and Korea, 1945-1950." Pacific Historical Review 47, no. 3 (August 1978): 425-451. Cagan, Phillip. "The Monetary Dynamics of Hyperinflation." In Studies in the Quantity Theory of Money, by Milton Friedman. Chicago: University of Chicago, 1956. Caprio, Mark E. "Resident Aliens: Forging the Political Status of Koreans in Occupied Japan." In Democracy in Occupied Japan: The U.S. Occupation and Japanese Politics and Society, by Mark E Caprio and Yoneyuki Sugita. New York: Routledge, 2007. Caprio, Mark E, and Yoneyuki Sugita. "Introduction: The U.S. Occupation of Japan? Innovation, Continuity, and Compromise." In Democracy in Occupied Japan: The U.S. Occupation and Japanese Politics and Society, by Mark E Caprio and Yoneyuki Sugita. New York: Routledge, 2007. Carey, John M. "Parchment, Equilibria, and Institutions." Comparative Political Studies 33, no. 6/7 (August/September 2000): 735-761. Catudal, Honor? Marc. Kennedy and the Berlin Wall Crisis : A Case Study in U.S. Decision Making. Berlin: Berlin-Verlag, 1980. Cha, Victor D. Alignment Despite Antagonism: The United States-Korea-Japan Security Triangle. Stanford, CA: Stanford University, 1999. Childers, Thomas. "The Social Bases of Nation of the National Socialist Vote." Journal of Contemporary History 11(4) Special Issue: Theories of Fascism (October 1976): 17-42. Choi, Sang-Yong. "Trusteeship Debate and the Korean Cold War." In Korea Under the American Military Government, 1945-1948, by Bonnie BC Oh. Westport, CT: Praeger, 2002. Cohen, Frank S. "Proportional versus Majoritarian Ethnic Conflict Management in Democracies." Comparative Political Studies 30, no. 5 (October 1997): 607-630. 301 Cohen, Jerome B. "Japan: Reform vs. Recovery." Far Eastern Survey 17, no. 122 (June 1948): 137-142. Cohen, Theodore. Remaking Japan: The American Occupation as New Deal. New York: The Free Press, 1987. Colaresi, Michael, and William R Thompson. "The Economic Development- Democratization Relationship: Does the Outside World Matter?" Comparative Political Studies 36, no. 4 (May 2003): 381-403. Collier, Paul. Wars, Guns and Votes: Democracy in Dangerous Places. New York: HarperCollins, 2009. Collier, Paul, and Anke Hoeffler. "Aid, Policy and Growth in Post-Conflict Societies." European Economic Review 48, no. 5 (October 2004): 1125-1145. Conradt, David P. "Changing German Political Culture." In The Civic Culture Revisited, by Gabriel Almond and Sidney Verba. Boston: Little, Brown, 1980. Cornebise, Alfred E. Causality, Chaos Theory, and the End of the Weimar Republic: A Commentary on Henry Turner's Hitler's Thirty Days to Power. Vol. 38. Carbondale, IL, USA: Southern Illinois University Press, 1983. Coyne, Christopher. After War: The Political Economy of Exporting Democracy. Stanford, CA, USA: Stanford University, 2008. ?. "Reconstructing Weak and Failed States: Foreign Intervention and the Nirvana Fallacy." Foreign Policy Analysis 2, no. 4 (October 2006): 343-360. Crespo, Jos? Antonio. "The Liberal Democratic Party in Japan: Conservative Domination." International Political Science Review / Revue internationale de science politique 16, no. 2 (April 1995): 199-209. Croissant, Aurel. Korea (Republic of Korea/ South Korea). Vols. II: South East Asia, East Asia and the South Pacific, in Elections in Asia and the Pacific: A Data Handbook, by Dieter Nohlen, Florian Grotz and Christof Hartmann. Oxford: Oxford University, 2001. Cromwell, Richard S. ""Rightist Extremism in Postwar West Germany"." The Western Political Quarterly 17, no. 2 (June 1964): 284-293. Cumings, Bruce. The Origins of the Korean War: Liberation and the Emergence of Separate Regimes, 1945-1947. Princeton, NJ: Princeton University, 1981. Dawisha, Adeed. "The Prospects for Democracy in Iraq: Challenges and Opportunities." Third World Quarterly 26, no. 4/5 (2005): 723-737. ?. "Iraq: A vote against sectarianism." Journal of Democracy 21, no. 3 (2010): 26-40. 302 Dawson, William Harbutt. Germany Under the Treaty. London: George Alllen & Unwin, 1933. Destler, I M, Hideo Sato, Priscilla Clapp, and Haruhiro Fukui. Managing an Alliance: The Politics of U.S.-Japanese Relations. Washington: Brookings, 1976. Diamond, Larry Jay. "Three Paradoxes of Democracy." Journal of Democracy 1, no. 3 (Summer 1990): 48-60. ?. "Promoting Democracy." Foreign Policy 87 (Summer 1992): 25-46. ?. "Lessons from Iraq." Journal of Democracy 16, no. 1 (2005): 9-23. Diehl, James A. Paramilitary Politics in Weimar Germany. Bloomington, IN: Indiana University Press, 1977. Dionisopoulos, P Allan. "Revisionist Tendencies in Post-Occupation Japan." The Western Political Quarterly 10, no. 4 (December 1957): 793-801. Diskin, Abraham, Hanna Diskin, and Reuven Y. Hazan. "Why Democracies Collapse: The Reasons for Democratic Failure and Success." International Political Science Review 26, no. 3 (July 2005): 291?309. Dobbs, Charles M. "Korea and Berlin: A Hypothesis." The International History Review 7, no. 3 (August 1985): 415-21. Dodge, Toby. "The Ideological Roots of Failure: the Application of Kinetic Neo liberalism to Iraq." International Affairs 86, no. 6 (2010): 1269-1286. ?. "Iraq's Road Back to Dictatorship." Survival 54, no. 3 (2012): 147-168. Dower, John W. Embracing Defeat: Japan in the Wake of World War II. New York: WW Norton, 1999. Dulles, Allen W. "Alternatives for Germany." Foreign Affairs 25, no. 3 (June 1947): 421- 432. Dunbabin, JPD. The Cold War: The Great Powers and Their Allies. Edinburgh Gate, UK: Pearson Education Limited, 2008. Edelstein, David. Occupational Hazards: Success and Failure in Military Occupation. Ithaca, NY, USA: Cornell University, 2008. ?. "Occupational Hazards: Why Military Occupations Succeed or Fail." International Security 29, no. 1 (Summer 2004): 49-91. 303 Eichengreen, Barry. "Mainsprings of Economic Recovery in Post-war Europe." In Europe's Post-war Recovery, by Barry Eichengreen. New York: Cambridge University, 1995. Eiji, Takemae. Inside GHQ: The Allied Occupation of Japan and its Legacy. New York: Continuum, 2002. Elgie, Robert. "The Perils of Semi-Presidentialism. Are They Exaggerated?" Democratization 15, no. 1 (February 2008): 49?66. Emanuel, Kerry A. Divine Wind: The History and Science of Hurricanes. Oxford, UK: Oxford University, 2005. Enterline, Andrew J, and J Michael Greig. "Against All Odds? The History of Imposed Democracy and the Future of Iraq and Afghanistan." Foreign Policy Analysis 4 (2008b): 331-347. ?. "Perfect Storms?: Political Instability in Imposed Polities and the Futures of Iraq and Afghanistan." Journal of Conflict Resolution 52, no. 6 (September 2008a): 880- 915. Esposito, Chiarella. "Marshall Plan Lessons for Contemporary Aid Donors." In Europe's Post-war Recovery, by Barry Eichengreen. New York: Cambridge, 1995. Eyck, Erich. A History of the Weimar Republic. Translated by Harlan P Hanson and Robert GL White. New York: Atheneum, 1970. Eyre, John D. "Post-Occupation Conditions in Rural Japan." Annals of the American Academy of Political and Social Science 308 (November 1956): 113-120. Fails, Matthew D., and Heather Nicole Pierce. "Changing Mass Attitudes and Democratic Deepening." Political Research Quarterly 63, no. 1 (March 2010): 174-187. Farley, Miriam S. "Crisis in Korea." Far Eastern Survey 19, no. 14 (August 1950): 149- 156. Fearon, James D. "Signaling Foreign Policy Interests: Tying Hands versus Sinking Costs." Journal of Conflict Resolution 41, no. 1 (February 1997): 68-90. Fisher, Charles A. "The Role of Korea in the Far East." The Geographical Journa 120, no. 3 (September 1954): 282-298. Freedom House. "Freedom in the World | Freedom House." Freedom House. 2011. http://www.freedomhouse.org/sites/default/files/inline_images/FIWAllScoresCou ntries1973-2011.xls (accessed February 29, 2012). Frieden, Jeffrey. Global Capitalism: Its Fall and Rise in the Twentieth Century. New York: Norton, 2007. 304 Fuchs, Steven J. "Food Policy, Land Reform and Japan's Economic Recovery." In Democracy in Occupied Japan: The U.S. Occupation and Japanese Politics and Society, by Mark E Caprio and Yoneyuki Sugita. New York: Routledge, 2007. Gaddis, John Lewis. The United States and the origins of the Cold War, 1941-1947. New York: Columbia University, 2000. Gallicchio, Marc. "The Kuriles Controversy: U.S. Diplomacy in the Soviet-Japan Border Dispute, 1941-1956." Pacific Historical Review 60, no. 1 (February 1991): 69- 101. Gasiorowski, Mark J, and Timothy J Power. "The Structural Determinants of Democratic Consolidation: Evidence from the Third World." Comparative Political Studies 31, no. 6 (December 1998): 740-771. Gatzke, Hans W. "Russo-German Military Collaboration During the Weimar Republic." American Historical Review 63, no. 3 (April 1958): 565-597. Geddes, Barbara. "What Do We Know About Democratization after Twenty Years?" Annual Review of Political Science 2 (June 1999): 115?44. German Democratic Republic. "Decree of the Minister of the Interior of the German Democratic Republic, East Berlin, 12 August 1961." In Politics and Government in the Federal Republic of Germany: Basic Documents, by Carl-Christoph Schweitzer, Detlev Karsten, Robert Spencer, R. Taylor Cole, Donald Kommers and Anthony Nicholls. New York: Berg, 1984. German National Assembly. "Weimar constitution." Wikisource. August 8, 1919. http://en.wikisource.org/wiki/Weimar_constitution (accessed April 4, 2009). Gibney, Alex. "Six Days to Reinvent Japan." The Wilson Quarterly 20, no. 4 (Autumn 1996): 72-80. Gillingham, John. "The European Coal and Steel Community." In Europe's Post-war Recovery, by Barry Eichengreen. New York: Cambridge University, 1995. Gleditsch, Kristian. All International Politics Is Local: The Diffusion of Conflict, Integration, and Democratization. Ann Arbor, MI, USA: University of Michigan Press, 2002. Glees, Anthony. Reinventing Germany: German Political Development Since 1945. Oxford: Berg, 2003. Goertz, Gary. "Choosing Cases for Case Studies: a Qualitative Logic." Qualitative and Multi-Method Research Newsletter, 2008: 11?14. Goertz, Gary, and Harvey Starr (eds.). Necessary Conditions: Theory, Methodology, and Applications. Lanham: Rowman and Littlefield, 2003. 305 Goodson, Larry P. "Afghanistan's long Road to Reconstruction." Journal of Democracy 14, no. 1 (2003): 82-99. ?. "Bullets, ballots, and poppies in Afghanistan." Journal of Democracy 16, no. 1 (2005): 24-38. Gordenker, Leon. "The United Nations, the United States Occupation and the 1948 Election in Korea." Political Science Quarterly 73, no. 3 (September 1958): 426- 450. Government of the USSR. "Note from the Government of the USSR to the Government of the United States, 27 November 1958." In Politics and Government in the Federal Republic of Germany: Basic Documents, by Carl-Christoph Schweitzer, Detlev Karsten, Robert Spencer, R. Taylor Cole, Donald Kommers and Anthony Nicholls. New York: Berg, 1984. Government Printing Bureau. The Constitution of Japan | Birth of the Constitution of Japan. November 3, 1946. http://www.ndl.go.jp/constitution/e/etc/c01.html (accessed April 14, 2012). Grimm, Sonja. "External Democratization after War: Success and Failure." Democratization 15, no. 3 (June 2008): 525?549. Grumm, John G. "Theories of Electoral Systems." Midwest Journal of Political Science 2, no. 4 (November 1958): 357-376. Guthrie-Shimizu, Sayuri. "Occupation Policy and Postwar Sino-Japanese Relations." In Democracy in Occupied Japan: The U.S. Occupation and Japanese Politics and Society, by Mark E Caprio and Yoneyuki Sugita. New York: Routledge, 2007. H, S. "Allied Policy in Japan." The World Today 5, no. 5 (May 1949): 190-201. Haftendorn, Helga. Coming of Age: German Foreign Policy Since 1945. Lanham, MD, USA: Roman & Littlefield, 2006. Haggard, Stephan, and Robert R Kaufman. The Political Economy of Democratic Transitions. Princeton, NJ, USA: Princeton University, 1995. Halbrook, Stephen P. "Nazi Firearms Law and the Disarming of German Jews." Arizona Journal of International and Comparative Law 5 (2000): 483-535. Han, Sungjoo. The Failure of Democracy in South Korea. Berkeley, CA: University of California, 1974. Hanrieder, Wolfram F. "West German Foreign Policy, 1949-1979: Necessities and Choices." In West German Foreign Policy, 1949-1979, by Wolfram F. Hanrieder. Boulder, CO, USA: Westview, 1980. 306 Hart-Landsberg, Martin. Korea: Division, Reunification, and U.S. Foreign Policy. New York: Monthly Review, 1998. Heberle, Rudolph. "The Political Movements Among the Rural People in Schleswig- Holstein, 1918 to 1932." Journal of Politics 5, no. 1 (February 1943): 3-26. Heidenheimer, Arnold J. "Federalism and the Party System: The Case of West Germany." The American Political Science Review 52, no. 3 (September 1958): 809-828. Hilal, Jamil. ?Problematizing Democracy in Palestine.? Comparative Studies of South Asia, Africa and the Middle East 23, nos. 1 & 2 (2003): 163-172. Hill, Lewis E, Charles E Butler, and Stephen A Lorenzen. "Inflation and the Destruction of Democracy: The Case of the Weimar Republic." Journal of Economic Issues 11, no. 2 (June 1977): 299-313. Hill, Matthew Alan. "International actors and Democratisation: Can USAID Deliver a Democratic Culture to Afghanistan?." International Relations 24, no. 2 (2010): 155-174. Hillenbrand, Martin J. "The United States and Germany." In West German Foreign Policy, 1949-1979, by Wolfram F. Hanrieder. Boulder, CO, USA: Westview, 1980. Hiroi, Taeko, and Sawa Omori. "Perils of Parliamentarism? Political Systems and the Stability of Democracy Revisited." Democratization 16, no. 3 (June 2009): 485- 507. Hitchcock, William I. France Restored: Cold War Diplomacy and the Quest for Leadership in Europe, 1944-1954. Chapel Hill, NC, USA: University of North Carolina, 1998. Holt, Charles A. and Susan K Laury. ?Risk Aversion and Incentive Effects.? The American Economic Review 92, no. 5 (December 2002): 1644-1655. Hunsberger, Warren S. "Japan's Economy: From Resource Constraints to External Surpluses." In Japan's Quest: The Search for International Role, Recognition, and Respect, by Warren S Hunsberger. Armonk, NY: M E Sharp, 1997. Hunt, Richard N. German Social Democracy, 1918-1933. New Haven, CT: Yale University Press, 1964. Huntington, Samuel P. The Third Wave: Democratization in the Late Twentieth Century. Norman, OK, USA: University of Oklahoma, 1991. Hyland, William G. "The Soviet Union and Germany." In West German Foreign Policy, 1949-1979, by Wolfram F. Hanrieder. Boulder, CO, USA: Westview, 1980. 307 Irwin, Douglas A. "The GATT's Contribution to Economic Recovery." In Europe's Post- war Recovery, by Barry Eichengreen. New York: Cambridge University, 1995. Kaes, Anton, Martin Jay, and Edwarg Dimendberg. The Weimar Republic Sourcebook. Berkeley, CA: University of California Press, 1994. Kang, Seonjou, and James Meernik. "Civil War Destruction and the Prospects for Economic Growth." Journal of Politics 67, no. 1 (February 2006): 88?109. Kassis, Mudar. ?Civil Society Organizations and Transition to Democracy in Palestine.? Voluntas: International Journal of Voluntary and Nonprofit Organizations 12, no. 1 (2001): 35-48. Kelleher, Catherine McArdle. "Germany and NATO: The Enduring Bargain." In West German Foreign Policy, 1949-1979, by Wolfram F. Hanrieder. Boulder, CO, USA: Westview, 1980. Kennan, George F. Memoirs: 1925-1950. New York: Little, Brown, 1967. Killick, John. The United States and European Reconstruction. Edinburgh: Keele University, 1997. Kim, Kyung-Kook. The Division of Korea and the Alliance Making Process: Internationalization of Internal Conflict and Internalization of International Struggle, 1945-1948. Lanham, MD: University Press of America, 1995. Kim, Nam G. From Enemies to Allies: The Impact of the Korean War on U.S.-Japan Relations. Bethesda, MD: International Scholars, 1997. King, Gary, Gary Keohane, and Sidney Verba. Designing Social Inquiry: Scientific Inference in Qualitative Research. Princeton: Princeton University Press, 1994. Kolb, Eberhard. The Weimar republic. Translated by P.S. Falla. London: Unwin Hyman, 1988. Kommers, Donald P. "The Basic Law of the Federal Republic of Germany: An Assessment after Forty Years." In The Federal Republic of Germany at Forty, by Peter H. Merkel, 133-159. New York: New York University, 1989. Krieger, Wolfgang. "Was General Clay a Revisionist? Strategic Aspects of the United States Occupation of Germany." Journal of Contemporary History 18, no. 2 (April 1983): 165-184. Krishnan, R R. "Early History of U.S. Imperialism in Korea." Social Scientist 12, no. 11 (November 1984): 3-18. 308 Kruger, Peter. "The Federal Republic as a Nation-State." In The Postwar Transformation of Germany: Democracy, Prosperity, and Nationhood, by John S. Brady, Beverly Crawford and Sarah Elise Wiliarty. Ann Arbor: University of Michigan, 2002. Kuechler, Manfred. "The NSDAP Vote in the Weimar Republic: An Assessment of the State-of-the-Art in View of Modern Electoral Research." Historical Social Research 17, no. 1 (1992): 22-52. Kvistad, Gregg O. "Buiding Democracy and Changing Institutions: The Professional Civil Service and Political Parties in the Federal Republic of Germany." In The Postwar Transformation of Germany: Democracy, Prosperity, and Nationhood, by John S. Brady, Beverly Crawford and Sarah Elise Wiliarty. Ann Arbor, MI, USA: University of Michigan, 2002. L, H, and M K G. "Korea Past and Present." The World Today 2, no. 4 (April 1946): 175- 192. Lai, Brian, and Ruth Melkonian-Hoover. "Democratic Progress and Regress: The Effect of Parties on the Transitions of States to and away from Democracy." Political Research Quarterly 58, no. 4 (December 2005): 551-564. Lauterbach, Alfred T. "Economic Demobilization in a Conquered Country: Germany 1919-1923." The Journal of Politics 6, no. 1 (February 1944): 28-56. Lee, Jongsoo. The Partition of Korea after World War II: A Global History. New York: Palgrave MacMillan, 2006. Levi, Werner. "International Control of Japan." Far Eastern Survey 15, no. 9 (September 1946): 299-302. Lichbach, Mark I. The Rebel?s Dilemma. Ann Arbor, MI USA: University of Michigan, 1998. Liem, Channing. "United States Rule in Korea." Far Eastern Survey 18, no. 7 (4 1949): 77-80. Lindenfeld, David F. "Causality, Chaos Theory, and the End of the Weimar Republic: A Commentary on Henry Turner's Hitler's Thirty Days to Power." History and Theory 38, no. 3 (October 1999): 281-299. Lindlar, Ludger, and Carl-Ludwig Holtfrerich. "Germany's Export Boom at Fifty--An Enduring Story?" In The Postwar Transformation of Germany: Democracy, Prosperity, and Nationhood, by John S. Brady, Beverly Crawford and Sarah Elise Wiliarty. Ann Arbor, MI, USA: University of Michigan, 2002. Lindseth, Peter L. "The Paradox of Parliamentary Supremacy: Delegation, Democracy, and Dictatorship in Germany and France, 1920s-1950s." Yale Law Journal 113, no. 7 (May 2004): 1341-1415. 309 Linz, Juan J. "Democracy Today: An Agenda for Students of Democracy." Scandinavian Political Studies 20, no. 2 (June 1997): 115-134. ?. "The Perils of Presidentialism." Journal of Democracy 1, no. 1 (Winter 1990): 51-69. ?. "The Transition to Democracy." Washington Quarterly 13, no. 3 (1990): 148. Linz, Juan J, and Alfred Stepan. Problems of Democratic Transition and Consolidation. Baltimore, MD, USA: Johns Hopkins University, 1996. ?. The Breakdown of Democratic Regimes. Baltimore, MD, USA: Johns Hopkins University, 1978. Lipset, Seymour Martin. Political Man: The Social Basis of Politics. Garden City, NY: Anchor Books, 1963. ?. "Some Social Requisites of Democracy: Economic Development and Political Legitimacy." The American Political Science Review 53, no. 1 (March 1959): 69- 105. Lo, Nigel, Barry Hashimoto, and Dan Reiter. "Ensuring Peace: Foreign-Imposed Regime Change and Postwar Peace Duration, 1914?2001." International Organization 62, no. 4 (October 2008): 717?736. Loewenberg, Gerhard. Parliament in the German Political System. Ithaca, NY, USA: Cornell University, 1967. MacDonald, Giles. After the Reich: The Brutal History of the Allied Occupation. New York: Basic Books, 2007. Mahoney, James, and Dietrich Rueschemeyer. Comparative Historical Analysis in the Social Sciences. New York: Cambridge University Press, 2003. Mahoney, James, and Gary Goertz. "A Tale of Two Cultures: Contrasting Quantitative and Qualitative Research." Political Analysis 14, no. 3 (2006): 227?249. Maki, John M. "Japan: Political Reconstruction." Far Eastern Survey 16, no. 7 (April 1947): 73-77. Mark, Eduard. Revolution by Degrees: Stalin's National Front Strategy for Europe, 1941- 1947. Washington: Cold War International History Project, Woodrow Wilson International Center of Scholars, 2001. Markovits, Andrei S., and Simon Reich. "The Contemporary Power of Memory: The Dilemmas for German Foreign Policy." In The Postwar Transformation of Germany: Democracy, Prosperity, and Nationhood, by John S. Brady, Beverly Crawford and Sarah Elise Wiliarty. Ann Arbor, MI, USA: University of Michigan, 2002. 310 Marshall, Monty, and Keith Jaggers. "Polity IV Project: Political Regime Characteristics and Transitions, 1800-2007." 2009. http://www.systemicpeace.org/polity/polity4.htm. Matray, James I. The Reluctant Crusade: American Foreign Policy in Korea, 1941-1950. Honolulu, HI: University of Hawaii, 1985. McAdam, Doug, Charles Tilly, and Sidney Tarrow. "Methods for Measuring Mechanisms of Contention." Qualitative Sociology 31, no. 4 (2008): 307-331. McCune, George M. "Post-War Government and Politics of Korea." The Journal of Politics 9, no. 4 (November 1947a): 605-623. ?. "The First Year of Liberation." Pacific Affairs 20, no. 1 (March 1947b): 3-17. McGlothlen, Ronald. "Acheson, Economics, and the American Commitment in Korea, 1947-1950." Pacific Historical Review 58, no. 1 (February 1989): 23-54. McNelly, Theodore. "Japan?Potential for Peace." World Affairs 128, no. 1 (April, May, June 1965): 28-35. McWilliams, Wayne C, and Harry Piotrowski. The World since 1945: A History of International Relations. Denver: Lynne Rienner, 2009. Mee, Jr., Charles L. The Marshall Plan: The Launching of the Pax Americana. New York: Simon and Schuster, 1984. Meernik, James. "United States Military Intervention and the Promotion of Democracy." Journal of Peace Research 33, no. 4 (November 1996): 391-402. Mendel, Jr., Douglas H. "Japan Reviews Her American Alliance." The Public Opinion Quarterly 30, no. 1 (Spring 1966): 1-18. Merkl, Peter H. "The German Response to the Challenge of Extremist Parties, 1949- 1994." In The Postwar Transformation of Germany: Democracy, Prosperity, and Nationhood, by John S Brady, Beverly Crawford and Sarah Elise Wiliarty. Ann Arbor, MI, USA: University of Michigan, 2002. ?. The Origin of the West German Republic. New York: Oxford University, 1963. ?. "The Role of Public Opinion in West German Foreign Policy." In West German Foreign Policy, 1949-1979, by Wolfram F. Hanrieder. Boulder, CO, USA: Westview, 1980. Merritt, Richard L. "American Influences in the Occupation of Germany." Annals of the American Academy of Political and Social Science 428 (November 1976): 91- 103. 311 Mitchell, Otis C. The Cold War in Germany: Overview, Origins and Intelligence Wars. Lanham, MD, USA: University Press of America, 2005. M?ller, J?rgen, and Svend-Erik Skaaning. "Stateness First?" Democratization 18, no. 1 (February 2011): 1-24. Mommsen, Hans. The Rise and Fall of Weimar Democracy. Translated by Elborg Foster and Larry Eugene Jones. Chapel Hill, NC, USA: University of North Carolina, 1996. Moon, Bruce E. "Long time coming: Prospects for democracy in Iraq." International Security 33, no. 4 (2009): 115-148. Moore, Jr., Barrington. Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. Boston: Beacon, 1966. Moore, Ray A, and Donald L Robinson. Partners for Democracy: Crafting the New Japanese State under MacArthur. New York: Oxford University, 2002. Morley, James W. Japan and Korea: America's Allies in the Pacific. New York: Walker and Company, 1967. Mousseau, Michael. "Market Prosperity, Democratic Consolidation, and Democratic Peace." Journal of Conflict Resolution 44, no. 4 (August 2000): 472-507. Murdoch, James, and Todd Sandler. "Civil Wars and Economic Growth: A Regional Comparison." Defence and Peace Economics 13, no. 6 (September 2002): 451- 464. Nelson, Keith L. Victors Divided: America and the Allies in Germany, 1918-1923. Berkeley, CA: University of California Press, 1975. Nicholls, AJ. Weimar and the Rise of Hitler. London: Macmillan Press, 1991. Niclauss, Karlheinz. "Political Reconstruction at Bonn." In From Dictatorship to Democracy: Coping with the Legacies of Authoritarianism and Totalitarianism, by John H. Herz, 29-56. Westport, CT, USA: Greenwood, 1982. Nixon, Hamish, and Richard Ponzio. "Building democracy in Afghanistan: The Statebuilding Agenda and International Engagement." International Peacekeeping 14, no. 1 (2007): 26-40. North Atlantic Council. "Communique Issued Following Ministerial Meeting of the North Atlantic Council, Paris, With Appended Declaration on Berlin, 16-18 December 1958." In Politics and Government in the Federal Republic of Germany: Basic Documents, by Carl-Christoph Schweitzer, Detlev Karsten, Robert Spencer, R. Taylor Cole, Donald Kommers and Anthony Nicholls. New York: Berg, 1984. 312 Nozaki, Yoshiko. "Educational Reform and History Textbooks in Occupied Japan." In Democracy in Occupied Japan: The U.S. Occupation and Japanese Politics and Society, by Mark E Caprio and Yoneyuki Sugita. New York: Routledge, 2007. O?Donnell, Guillermo, and Philippe Schmitter. "Tentative Conclusions about Uncertain Democracies." In Transition from Authoritarian Rule: Prospects for Democracy, by Guillermo O?Donnell, Philippe C. Schmitter and Laurence Whitehead. Baltimore, MD, USA: Johns Hopkins, 1986. Oh, Bonnie BC. "Introduction: The Setting." In Korea Under the American Military Government, 1945-1948, by Bonnie BC Oh. Westport, CT: Praeger, 2002a. ?. "Kim Kyu-sik and the Coalition Effort." In Korea Under the American Military Government, 1945-1948, by Bonnie BC Oh. Westport, CT: Praeger, 2002b. Okudaira, Yasuhiro. "Forty Years of the Constitution and Its Various Influences: Japanese, American, and European." Law and Contemporary Problems 53, no. 1 (Winter 1990): 17-49. Oliver, Robert T. "Transition and Continuity in Korean-American Relations in the Postwar Period." In Korean-American Relations: 1866-1997, by Yur-Bok Lee and Wayne Patterson. Albany, NY: State University of New York, 1999. O'Riordan, Elspeth. "The British Zone of Occupation in the Rhineland." Diplomacy and Statecraft 16 (2005): 439-454. Park, Chan-Pyo. "The American Military Government and the Framework for Democracy in South Korea." In Korea Under the American Military Government, 1945-1948, by Bonnie BC Oh. Westport, CT: Praeger, 2002. Park, Richard L. "Transition in Japan." Far Eastern Survey 18, no. 19 (September 1949): 225-226. Parliamentary Council. "Articles of the Basic Law." In Politics and Government in the Federal Republic of Germany: Basic Documents, by Carl-Christoph Schweitzer, Detlev Karsten, Robert Spencer, Donald Kommers, Anthony Nicholls and R. Taylor Cole. New York: Berg, 1984. Pawley, Margaret. The Watch on the Rhine: The Military Occupation of the Rhineland, 1918-1930. New York: IB Taurus, 2007. Peceny, Mark. "Forcing Them to Be Free." Political Research Quarterly 52, no. 3 (1999): 549-582. Pei, Minxin. "Lessons of the Past." Foreign Policy 137 (2003): 52-55. Pempel, T J. "Japanese Democracy and Political Culture: A Comparative Perspective." PS: Political Science and Politics 25, no. 1 (March 1992): 5-12. 313 Pevehouse, Jon C. "Democracy from the Outside-In? International Organizations and Democratization." International Organization 56, no. 3 (Summer 2002a): 515? 549. ?. "With a Little Help from My Friends? Regional Organizations and the Consolidation of Democracy." American Journal of Political Science 46, no. 3 (July 2002b): 611-626. Plischke, Elmer. The Allied High Commission for Germany. Washington: Historical Division, Office of the Executive Secretary, Office of the U.S. High Commissioner for Germany, 1953. Plumper, Thomas, and Christian Martin. "Democracy, Government Spending, and Economic Growth: A Political-Economic Explanation of the Barro-Effect." Public Choice 117, no. 1-2 (October 2003): 27-50. Pollock, James K. "A Territorial Pattern for the Military Occupation of Germany." The American Political Science Review 38, no. 5 (October 1944): 970-975. Porch, Douglas. "Occupational Hazards." National Interest 72 (2003): 35-47. Post, Jr., Gaines. "Reflections on the German Question." In International Politics and German History: The Past Informs the Present, by David Wetzel and Theodore S. Hamerow. Westport, CT, USA: Praeger, 1997. Przeworski, Adam. "Some Problems in the Study of the Transition to Democracy." In Transition from Authoritarian Rule: Prospects for Democracy, by Guillermo O?Donnell, Philippe C. Schmitter and Laurence Whitehead. Baltimore, MD, USA: Johns Hopkins, 1986. Przeworski, Adam, and Fernando Limongi. "Modernization: Theories and Facts." World Politics 49, no. 1 (January 1997): 155?183. Przeworski, Adam, Michael Alvarez, Jos? Antonio Cheibub, and Fernando Limongi. "What Makes Democracy Endure." Journal of Democracy 7, no. 1 (1996): 39-55. Price, Colin. "Pakistan: A Plethora of Problems." Global Security Studies (2012): 53-62. Quigley, Harold S. "Evaluating the Japanese Occupation." Far Eastern Survey 20, no. 17 (October 1951): 176-178. Ray, Donald P. "Our Progress in Japan." World Affairs 110, no. 2 (Summer 1947): 109- 113. Reday, Joseph Z. "Reparations from Japan." Far Eastern Survey 18, no. 13 (June 1949): 145-151. 314 Reichlin, Lucrezia. "The Marshall Plan Reconsidered." In Europe's Post-war Recovery, by Barry Eichengreen. New York: Cambridge University, 1995. Reifenberg, Jan. "Germany and the Atlantic Community." In West German Foreign Policy, 1949-1979, by Wolfram F. Hanrieder. Boulder, CO, USA: Westview, 1980. Richardson, Bradley. Japanese Democracy: Power, Coordination, and Performance. New Haven, CT: Yale University, 1997. Richter, Michaela. "From State Culture to Citizen Culture: Political Parties and the Postwar Transformation of Political Culture in Germany in the Federal Republic of Germany." In The Postwar Transformation of Germany: Democracy, Prosperity, and Nationhood, by John S. Brady, Beverly Crawford and Sarah Elise Wiliarty. Ann Arbor, MI, USA: University of Michigan, 2002. Rohrschneider, Robert. Learning Democracy: Democratic and Economic Values in Unified Germany. Oxford: Oxford University, 1999. Rubin, Barnett R. "The Political Economy of War and Peace in Afghanistan." World Development 28, no. 10 (February 2000): 1789-1803. ?. "Crafting a Constitution for Afghanistan." Journal of Democracy 15, no. 3 (2004): 5- 19. ?. "Peace Building and State-Building in Afghanistan: Constructing Sovereignty for Whose Security?." Third World Quarterly 27, no. 1 (2006): 175-185. Rupieper, Hermann-Josef. "'Der Bund fuer Burgerrechte': Transnational Relations and the Problem of Democratization in West Germany." In International Politics and German History: The Past Informs the Present, by David Wetzel and Theodore S. Hamerow. Westport, CT, USA: Praeger, 1997. Russett, Bruce, and John Oneal. Triangulating Peace: Democracy, Interdependence, and International Organizations. New York: Norton, 2001. Sa'Adah, Anne. "Regime Change: Lessons from Germany on Justice, Institution Building, and Democracy." The Journal of Conflict Resolution 50, no. 3 (June 2006): 303-323. Schedler, Andreas. "Measuring Democratic Consolidation." Studies in Comparative International Development 36, no. 1 (Spring 2001b): 66?92. ?. "Taking Uncertainty Seriously: The Blurred Boundaries of Democratic Transition and Consolidation." Democratization 8, no. 4 (Winter 2001a): 1-22. ?. "What is Democratic Consolidation?" Journal of Democracy 9, no. 2 (April 1998): 91-107. 315 Schiff, Victor. The Germans at Versailles, 1919. London: Williams and Northgate, 1930. Schmidt, Gustav. "Divided Europe-Divided Germany (1950-63)." Contemporary European History 3, no. 2 (July 1994): 155-192. Schmitt, Carl. Crisis of Parliamentary Democracy. Translated by Ellen Kennedy. Boston: Massachusetts Institute of Technology, 1988. Schwartz, Thomas A. "The 'Skeleton Key': American Foreign Policy, European Unity, and German Rearmament, 1949-54." Central European History 19, no. 4 (December 1986): 369-385. Schweitzer, Carl-Christoph, Detlev Karsten, Robert Spencer, R. Taylor Cole, Donald Kommers, and Anthony Nicholls. Politics and Government in the Federal Republic of Germany: Basic Documents. New York: Berg, 1984. Seawright, Jason, and John Gerring. "Case Selection Techniques in Case Study Research: A Menu of Qualitative and Quantitative Options." Political Research Quarterly 64, no. 2 (2008): 294-308. Shepard, Walter James. "The New Government in Germany." American Political Science Review 13, no. 3 (August 1919): 361-378. Shikaki, Khalil. ?The Peace Process, National Reconstruction, and the Transition to Democracy in Palestine.? Journal of Palestine Studies 25, no. 2 (Winter 1996): 5- 20. Shively, W Phillips. "Party Identification, Party Choice, and Voting Stability: The Weimar Case." American Political Science Review 66, no. 4 (December 1972): 1203-1225. Snyder, Louis L. Encyclopedia of the Third Reich. Chatham, Kent, UK: McGraw-Hill, 1998. Sook, Jeon Sang. "U.S. Korean Policy and the Moderates during the U.S. Military Government Era." In Korea Under the American Military Government, 1945- 1948, by Bonnie BC Oh. Westport, CT: Praeger, 2002. Stachura, Peter D. "National Socialism and the German Proletariat, 1925-1935: Old Myths and New Perspectives." Historical Journal 36, no. 3 (September 1993): 701-718. Starr, Harvey, and Christina Lindborg. "Democratic Dominoes Revisited: The Hazards of Governmental Transitions, 1974-1996." Journal of Conflict Resolution 47, no. 4 (August 2003): 490-519. Stepan, Alfred. "Paths toward Redemocratization: Theoretical and Comparative Considerations." In Transition from Authoritarian Rule: Prospects for 316 Democracy, by Guillermo O?Donnell, Philippe C. Schmitter and Laurence Whitehead. Baltimore, MD, USA: Johns Hopkins University, 1986. Stepan, Alfred, and Cindy Skach. "Constitutional Frameworks and Democratic Consolidation: Parliamentarianism versus Presidentialism." World Politics 46, no. 1 (October 1993): 1-22. Stetter, Stephan. "Democratization without Democracy? The Assistance of the European Union for Democratization Processes in Palestine." Mediterranean Politics 8, no. 2-3 (2003): 153-173. Stiglitz, Joseph E. "Interest Rates, Risk, and Imperfect Markets: Puzzles and Policies." Oxford Review of Economic Policy 15, no. 2 (1999): 59-76. Stueck, William. "The Coming of the Cold War to Korea." In Korea Under the American Military Government, 1945-1948, by Bonnie BC Oh. Westport, CT: Praeger, 2002. Svolik, Milan. "Authoritarian Reversals and Democratic Consolidation." American Political Science Review 102, no. 2 (May 2008): 153-168. Tang, Min. "Examining the Lagged Effect of Economic Development on Political Democracy: A Panel-VAR Model." Democratization 15, no. 1 (February 2008): 106-122. Third Army. American Military Government of Occupied Germany, 1918-1920: Report of the Officer in Charge of Civil Affairs. Department of War, Third Army and American Forces in Germany, Washington: US Government Printing Office, 1943. Thompson, Edmund. "National Competitive Advantage and the Roles of Economic and Political Freedom: Evidence from Hong Kong." Public Choice 120, no. 3/4 (September 2004): 401-437. Truman, Harry S. "Truman Doctrine: President Harry S. Truman's Address Before a Joint Session of Congress, March 12, 1947." The Avalon Project: Documents in Law, History and Diplomacy. March 12, 1947. http://avalon.law.yale.edu/20th_century/trudoc.asp (accessed January 30, 2012). Tsuzuki, Chushichi. The Pursuit of Power in Modern Japan: 1825-1995. Oxford: Oxford University, 2000. Ulfelder, Jay. "International Integration and Democratization: An Event History Analysis." Democratization 15, no. 2 (April 2008): 272?296. Ulfelder, Jay, and Michael Lustik. "Modeling Transitions to and from Democracy." Democratization 14, no. 3 (June 2007): 351?387. 317 van Ree, Erik. Socialism in One Zone: Stalin's Policy in Korea, 1945-1947. Oxford: Berg, 1989. van Riel, Arthur, and Arthur Schram. "Weimar Economic Decline, Nazi Economic Recovery, and the Stabilization of Political Dictatorship." The Journal of Economic History 53, no. 1 (March 1993): 71-105. Veliotes, Nicholas A. ?The Bush "Vision" for Palestine: Realistic or Apocalyptic?? Mediterranean Quarterly 13, no. 4 (Fall 2002): 11-20. von Hindenburg, Paul. "The Stab in the Back. Testimony delivered in Berlin on 11/18/1919." In The Weimar Republic Sourcebook, by Anton Kaes, Martin Jay and Edwarg Dimendberg. Berkeley, CA: University of California, 1994. von Hippel, Karin. Democracy by Force: US Military Intervention in the Post-Cold War World. Cambridge, UK: Cambridge University, 2000. Voth, Hans-Joachim. "Did High Wages or High Interest Rates Bring Down the Weimar Republic? A Cointegration Model of Investment in Germany, 1925-1930." Journal of Economic History 55, no. 4 (December 1995): 801-821. ?. "With a Bang, not a Whimper: Pricking Germany?s ?Stock Market Bubble? in 1927 and the Slide into Depression." Journal of Economic History 63, no. 1 (March 2003). Walt, Stephen M. Revolution and War. Ithaca, NY, USA: Cornell University, 1996. ?. The Origin of Alliances. Ithaca, NY, USA: Cornell University, 1987. Washburn, John N. "Russia Looks at Northern Korea." Pacific Affairs 20, no. 2 (June 1947): 152-160. Webb, Steven B. "Fiscal News and Inflationary Expectations in Germany After World War I." Journal of Economic History 46, no. 3 (September 1986): 769-794. Webster, Andrew. "From Versailles to Geneva: The Many Forms of Interwar Disarmament." The Journal of Strategic Studies 29, no. 2 (April 2006): 225-246. Wells, Roger H. "Interim Governments and Occupation Regimes." Annals of the American Academy of Political and Social Science 257 (May 1948): 57-71. West Berlin Commandants. "Statement issued by West Berlin Commandants, 29 August 1950." In Politics and Government in the Federal Republic of Germany: Basic Documents, by Carl-Christoph Schweitzer, Detlev Karsten, Robert Spencer, R. Taylor Cole, Donald Kommers and Anthony Nicholls. New York: Berg, 1984. Whitehead, Laurence. "International Aspects of Democratization." In Transition from Authoritarian Rule: Prospects for Democracy, by Guillermo O?Donnell, Philippe 318 C. Schmitter and Laurence Whitehead. Baltimore, MD, USA: Johns Hopkins University, 1986. Wilkins, Mira. "American-Japanese Direct Foreign Investment Relationships, 1930- 1952." The Business History Review 56, no. 4 (Winter 1982): 497-518. Willamson, David G. The British in Germany, 1918-1930: The Reluctant Occupiers. New York: Berg, 1991. Williams, Justin. "Completing Japan's Political Reorientation, 1947-1952: Crucial Phase of the Allied Occupation." The American Historical Review 73, no. 5 (June 1968): 1454-1469. Willis, F. Roy. The French in Germany, 1945-1949. Stanford, CA, USA: Stanford University, 1962. Wilson, Woodrow. Fourteen Points. Washington, DC, January 8, 1918. Windsor, Phillip. German Reunification. London: Elek Books Limited, 1969. Wolf, Holger C. "Post-war Germany in the European Context." In Europe's Post-war Recovery, by Barry Eichengreen. New York: Cambridge University, 1995. Worden, Scott. "Afghanistan: An election gone awry." Journal of Democracy 21, no. 3 (2010): 11-25. Wright, Quincy. "The Armistices." American Political Science Review 13, no. 1 (February 1919): 128-132. 319