Democracy Without Parties? Political Parties and Regime Change in Fujimori's Peru Steven Levitsky Maxwell A. Cameron ABSTRACT Political parties are critical to Latin American democracy. This was demonstrated in Peru, where an atomized, candidate-centered party system developed after Alberto Fujimori's 1992 presidential self-coup. Party system decomposition weakened the democratic opposition against an increasingly authoritarian regime. Since the regime collapsed in 2000, prospects for party rebuilding have been mixed. Structural changes, such as the growth of the informal sector and the spread of mass media technologies, have weakened politicians' incentive to build parties. Although these changes did not cause the collapse of the party system, they may inhibit its reconstruction. n much of Latin America, democracy faces a critical problem: one of its central pillars, the political party, is increasingly viewed with dis- satisfaction by citizens and, in some cases, by politicians. Yet most stu- dents of Latin American politics continue to share E. E. Schattschneider's view (1942, 1) that democracy is "unthinkable" without parties. Indeed, recent scholarship suggests that parties remain critical to the achieve- ment (Corrales 2001), performance (Mainwaring and Scully 1995; Main- waring 1999), and stability (Mainwaring and Scully 1995; Gibson 1996; McGuire 1997) of democracy in Latin America. Few countries highlight the difficulties of achieving and sustaining democracy in the absence of parties more clearly than contemporary Peru. In the 1990s, the Peruvian party system decomposed to a degree that surpassed even the most notoriously fragmented systems in Latin America (Tanaka 1998, 1999; Lynch 1999; Conaghan 2000; Planas 2000). Throughout the decade, electoral politics was dominated by political "independents" and candidate-centered parties, many of which did not survive beyond a single electoral cycle. At the same time, Peru's demo- cratic regime succumbed to a 1992 autogolpe, or self-coup, led by Pres- ident Alberto Fujimori. Over the course of the decade, a weak and frag- mented opposition failed to prevent Fujimori from dismantling institutional checks on his power; ultimately, internal tensions caused by a corruption scandal brought about Fujimori's fall.
LATIN AMERICAN POLITICS AND SOCIETY This article analyzes the relationship between party collapse and political regimes in Peru. It examines three central questions. First, it seeks to explain the decomposition of the Peruvian party system. It argues that although the deep structural crisis of the 1980s weakened established parties and created an opening for antipolitical establish- ment outsiders, the full-scale decomposition of the party system and its replacement with an atomized, candidate-centered system were ulti- mately the product of Fujimori's political success in the wake of the 1992 autogolpe. Peruvian politicians drew two lessons from this success: that public opinion would not reward the defense of formal democratic institutions; and that parties were not necessary for (and might impede) career advancement. In light of these lessons, scores of politicians abandoned both established parties and the democratic opposition in favor of an "independent" strategy centered on individualized and short-term elec- toral goals. The second question is the regime implications of party system col- lapse. This article argues that although the established parties had clearly lost democratic support in Peru (in the sense that they were mas- sively rejected by voters), their demise had far-reaching implications for Peruvian democracy. Party collapse and the proliferation of political "independents" weakened the democratic opposition in two ways. First, by focusing on short-term electoral gain rather than challenging the increasingly authoritarian regime, key political figures ceased to serve as instigators of what O'Donnell (1994) calls horizontal accountability. Second, the proliferation of candidate-centered movements, further- more, eroded the opposition's capacity to act collectively or mobilize against the regime. The third question in this study concerns the prospects for rebuild- ing parties in post-Fujimori Peru. In contrast to scholarly approaches that locate the roots of party failure in either government repression or institutional design-both of which offer potential remedies in the post- Fujimori period-we present a more cautious, historical-structural view. Strong parties are products of particular historical, sociological, and technological conditions that are absent or only weakly present in the contemporary period. Because of long-term structural changes, such as the growth of the urban informal sector and the increased influence of mass media technologies, contemporary politicians may lack both the incentive and the capacity to build new party organizations. Although these structural changes did not cause the collapse of Peru's party system, they may inhibit its reconstruction. 2 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES WHY PARTIES MATTER IN A DEMOCRACY Political parties, to paraphrase John Aldrich, make democracy "work- able."' For voters, parties make democracy workable by providing crit- ical information about what candidates stand for and how they can be expected to govern (Downs 1957). Voters use party labels and platforms as cues or shortcuts, and parties may be evaluated based on their past performance, either in government or in opposition. Where parties are weak or noninstitutionalized, voters must confront a dizzying array of (often short-lived) electoral options, which limits their capacity to eval- uate candidates retrospectively, associate them with known labels or ideologies, or even differentiate among them (Mainwaring and Scully 1995, 25; Mainwaring 1999, 324-27). Parties also make democracy workable for politicians (Aldrich 1995). Politicians are self-centered and shortsighted animals who, when left to their own devices, have little incentive to think beyond the next election or their own electoral district. Consequently, they confront a variety of coordination problems, both in their pursuit of public office and in government (Aldrich 1995). Parties are critical to solving these problems.2 Because they exist beyond a single election and must com- pete on a national scale, parties develop longer-term priorities and broader goals than individual politicians do. To the extent that parties discipline politicians, then, they can reshape politicians' incentives in ways that induce them to act in a more farsighted and collective manner. Yet parties do more than simply make democracy "workable" for voters and politicians. They also help make democracy viable for soci- ety as a whole. In Latin America, parties have contributed to democratic stability in a variety of ways. They help to protect the interests of socioe- conomic elites who have the capacity to "kick over the chess board" (Boron 1992, 76). When powerful socioeconomic actors cannot protect their interests in the electoral arena, they are more likely to support non- democratic alternatives or engage in praetorian tactics that put demo- cratic institutions at risk. Parties serve as an important means of pro- tecting powerful interests in the electoral arena. As Edward Gibson (1996) has argued, when strong conservative parties have protected elite economic interests in Latin America, democratic regimes have tended to be stable. Similarly, when organized working classes are strong, links to labor-based parties may be critical to regime stability (Collier and Col- lier 1991; McGuire 1997). Strong parties are also essential to democratic governability. By serving as a bridge between executives and legislatures, parties provide a critical mechanism for overcoming gridlock. Without the disciplining function of parties, legislatures may degenerate into chaos or, worse, a marketplace for influence peddling. In Latin America, weak parties have 3
LATIN AMERICAN POLITICS AND SOCIETY been associated with legislative inefficiency, executive-legislative con- flict, policy ineffectiveness, and, not infrequently, regime crises (Main- waring and Scully 1995; Mainwaring 1999). By contrast, where parties have been strong and party systems institutionalized (Chile, Costa Rica, Uruguay), executive-legislative relations have tended to be smoother and governability crises less frequent (Mainwaring and Scully 1995). Parties help hold elected leaders accountable to democratic institu- tions. Many Latin American democracies are characterized by weak or ineffective systems of checks and balances (or "horizontal accountabil- ity"), which allow executives to govern at the margin of other demo- cratic institutions and actors (O'Donnell 1994, 1998). Caesarist behavior is strongly associated with weak parties and poorly institutionalized party systems (Mainwaring and Scully 1995, 22-23; Mainwaring 1999, 328). Parties are important instigators of horizontal accountability. Where they are weak and politicians gain power through direct, unmediated appeals, executives tend to govern in a personalistic and anti-institutional manner, often violating the "unspoken rules of the game" that underly republican institutions (Mainwaring and Scully 1995, 22). Where parties are strong, by contrast, politicians must work through them in order to obtain higher office and, when in office, must cooper- ate with them in order to remain there. The primary arena through which parties check executive power is the legislature. By resisting the tendency to fuse legislative and execu- tive power, parties provide a bulwark against the despotism of an over- weening executive. Through the right of inquiry, censure, and oversight, they use the legislature to constrain executive power and prevent its abuse. Autonomous legislatures also help to guarantee the independ- ence of the judiciary by ensuring that courts are not stacked and judges not summarily dismissed. They may even help to ensure the integrity of the electoral process by defending the independence and transparency of electoral institutions.3 Parties play an important role in recruiting and socializing democratic elites, and thus in limiting the space available to political outsiders. They provide the foundation for a democratic political class. Political classes vary considerably with respect to their openness, their coherence, and their links to society, and this variation may have important consequences for democracy. When they are oligarchic or cartellike, channels of access and the scope of competition are reduced, and the gap between elites and mass publics tends to widen. In such a context, citizens may con- clude that politicians are unrepresentative, corrupt, or unconcerned with the public interest, and this perception may erode democratic legitimacy.4 Yet when a political class is absent, politics becomes a world of amateur or "outsider" politicians, many of whom lack experience with-and in some cases, commitment to-democratic institutions. 4 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES Historically, the most effective means of maintaining a political class that steers clear of both oligarchy and amateur politics has been the political party. Strong parties broaden and diversify the elite recruitment process. In many European countries, for example, social democratic parties played a central role in recruiting members of the working class into the political system; and in the United States, urban Democratic Party machines provided channels of access to various immigrant and ethnic groups.5 Parties also socialize office seekers into democratic politics and pro- vide them with training and experience. Although party politicians are hardly exempt from irresponsible and even authoritarian behavior, on the whole, they are more likely than outsiders to have experience with (and be oriented toward) democratic practices, such as negotiation, compro- mise, and coalition building. They are also more likely to value demo- cratic institutions, or at least to have a stake in their preservation. The absence of strong parties often gives rise to outsider or "neopopulist" can- didates who are elected on the basis of direct and often antisystem appeals (Roberts 1995; Weyland 1999). Because they are amateurs at dem- ocratic politics, outsiders constitute a "shot in the dark" in terms of their capacity to govern and their commitment to democracy. Indeed, in Latin America, the election of political outsiders has frequently resulted in inef- fective, irresponsible, and in some cases undemocratic governments. Beyond democratic viability, political parties may play an important role in achieving democracy (Corrales 2001; Franklin 2001). They do so in at least two ways. They facilitate collective action; the ability of dem- ocratic oppositions to maintain a united front is often critical to their suc- cess (Corrales 2001). Fragmented opposition movements tend to be weak, to lack coherence, to have limited mobilizational capacity, and to be highly vulnerable to cooptation and the "divide and rule" strategies of autocratic incumbents (Corrales 2001, 95-96). By facilitating coordination among and, when necessary, imposing discipline on individual leaders, strong parties help to avoid such problems. Strong parties also serve as "mobilizing structures" for opposition movements (Tarrow 1994; Franklin 2001), providing them with the organizational resources necessary to sus- tain a mass-based democracy movement. When parties are weak, prodemocratic elites often lack national-level infrastructures or strong ties to society, and civic groups tend to lack the kind of horizontal ties that facilitate macrolevel collective action (Roberts 1998, 72-73). In summary, political parties are essential to achieving, maintaining, and improving the quality of democracy. Where they are weak, class actors tend to have less of a stake in electoral politics, legislatures are less able to oversee the executive, antisystem candidates are more common and more successful, and societies are less well equipped either to resist authoritarian encroachments or to remove autocratic governments. 5
LATIN AMERICAN POLITICS AND SOCIETY DEMOCRATIC BREAKDOWN AND PARTY SYSTEM COLLAPSE IN PERU During the early 1990s, Peru experienced the collapse of both its party system and its democratic regime. Each of these developments can be traced largely to the profound political and socioeconomic crisis that hit Peru in the late 1980s. Yet party weakness and the Fujimorazo also interacted in important ways. Party weakness made Fujimori's election possible, and the election of an antipolitical-establishment outsider exacerbated Peru's political-institutional crisis and increased the proba- bility of a coup. In turn, Fujimori's success in the wake of his 1992 auto- golpe accelerated the process of party system decomposition by creat- ing an incentive for politicians to abandon existing parties and pursue office as "independents." These defections decimated an already debil- itated party system. Party Crisis, Outsider Politics, and the 1992 Autogolpe Throughout most of the 1980s, Peru possessed a relatively coherent (if weakly institutionalized) four-party system, consisting of the leftist United Left (IU), the populist American Popular Revolutionary Alliance (APRA), the centrist Popular Action (AP), and the conservative Popular Christian Party (PPC). Although the strength of these parties has been the subject of debate, all possessed national structures, discernible programs or ide- ologies, and identifiable social bases.6 In the 1985 presidential election and the 1986 municipal elections, the four parties collectively accounted for more than 90 percent of the vote (Tanaka 1998, 55). Beginning in the late 1980s, however, Peru's established parties fell into crisis (Cameron 1994; Cotler 1994; Tanaka 1998; Lynch 1999; Planas 2000). The demise of the old party system is attributable to a variety of factors, including aspects of the electoral system (Schmidt 1996; Tuesta 1996, 1998) and the mistakes of the parties themselves (Lynch 1999).7 The principal roots of the crisis, however, lay in the profound struc- tural challenges the parties faced during the 1980s. The elimination of the last barriers to full suffrage, large-scale urban migration, and the expansion of the urban informal sector radically altered the electoral landscape. The growth of the informal economy, which encompassed more than 50 percent of the economically active population by 1990, weakened class-based organizations, eroded collective and partisan identities, and produced a growing pool of politically unattached voters (Cameron 1994; Lynch 1999, 160). At the same time, a deep economic crisis and the brutal insurgency of the Shining Path guerrilla movement limited the capacity of the estab- 6 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES lished parties to build enduring linkages to this emerging electorate. In the wake of the successive failures of the AP government of Fernando Belainde Terry (1980-85) and the APRA government of Alan Garcia Perez (1985-90), the political center collapsed, leaving a large sector of the elec- torate available for outsider appeals (Cameron 1994, 1997, 45-50). The first manifestations of party system crisis emerged in the 1989 municipal elections, when Ricardo Belmont, a radio personality who had formed an "independent movement" called Obras (Works), was elected mayor of Lima. A year later, the top two finishers in the first round of the 1990 presidential election, novelist Mario Vargas Llosa and the virtually unknown Alberto Fujimori, were both political amateurs who appeared from outside the established parties. Vargas Llosa, whose candidacy was backed by AP and the PPC, finished first in a fragmented field in the first round, but Fujimori overwhelmingly won the runoff. Although it was forged in the crisis of the late 1980s, however, Fuji- mori's victory was also facilitated by several contingent factors. The IU, which might otherwise have capitalized on the crisis of the centrist par- ties, suffered a debilitating schism in 1989. Fujimori received behind-the- scenes support from Alan Garcia, who sought to weaken APRA candi- date and internal rival Luis Alva Castro (Schmidt 1996, 342; Lynch 1999, 191). Fujimori also benefited from the mistakes of Vargas Llosa, whose radical neoliberal platform and alliance with traditional conservative elites helped to consolidate an "anyone but Vargas Llosa" vote.8 Finally, the election of an outsider was facilitated by Peru's majority runoff system, which allowed Fujimori to capture the presidency despite fin- ishing second in the first round (Schmidt 1996; Tanaka 1998). Fujimori's election had devastating consequences for democracy. The election of an antipolitical-establishment outsider exacerbated what had already become a serious regime crisis (Kenney 1996; McClintock 1996; Cameron 1997, 1998). Hyperinflation and mounting political vio- lence had generated broad public dissatisfaction with the political status quo. Although such conditions would have posed a severe challenge to any government, Fujimori was particularly ill equipped to respond to them in a democratic manner. A political amateur, he had no real party behind him, no program ready for implementation, and no team to staff the government. His supporters held less than a fifth of the seats in the congress, and his initial support from the left and APRA quickly evapo- rated. He was opposed, moreover, by leading sectors of the political, economic, and religious establishment.9 Lacking experience with the give and take of democratic politics, Fujimori opted for an authoritarian strategy for political survival, designed by his security adviser, Vladimiro Montesinos: the April 1992 autogolpe. Although there is no guarantee that the regime would have survived under a party-backed politician, it is reasonable to suggest that 7
LATIN AMERICAN POLITICS AND SOCIETY Fujimori, an antipolitical-establishment outsider without any socializa- tion in the political process, was particularly open to an authoritarian alternative. Authoritarian Success and the Rise of Political "Independents" Although the crisis of the 1980s badly weakened the Peruvian party system, it was Fujimori's success in the wake of the 1992 autogolpe that proved more devastating. Although widely discredited, the established parties survived the 1990 election (Tanaka 1998, 194-95; Planas 2000, 337). APRA won a surprising 25 percent of the legislative vote, and AP, the PPC, and the IU all retained an important presence in the congress. What ultimately sealed the fate of the party system was the outcome of Fujimori's high-stakes battle with the established parties. As Martin Tanaka (1998, 198-200) has argued, outsider presidents Fernando Collor de Mello of Brazil and Jorge Serrano of Guatemala lost similar battles during this period. Had Fujimori been impeached like Collor, or had he failed in his autogolpe attempt like Serrano, Peru's party system might have ended up similar to those of Brazil and Guatemala: weak and dis- credited but essentially intact.10 Fujimori, however, succeeded. His claim that the autogolpe was necessary to rid Peru of a "false democracy" (Paredes Castro et al. 1992) dominated by party cliques was, by and large, accepted by Peruvians. Public support for Fujimori jumped from 53 percent in March 1992 to 81 percent after the autogolpe (Tanaka 1998, 219), and over the next three years, the president's average approval rating was a whopping 66 per- cent (based on data from Tanaka 1998, 219). Fujimori's popularity helped to create a broad base of support for the new regime. Fuji- morista forces easily won the constituent assembly elections held in November 1992: Fujimori's New Majority/C-90 won 49 percent of the vote, compared to 9.8 percent for the second-place PPC. The following year, a new constitution was approved-albeit with greater diffi- culty-via referendum. In 1995, Fujimori was easily reelected, winning a stunning 62.4 percent of the vote and gaining an absolute majority in the new congress. The autogolpes success was based largely on its timing (Tanaka 1998, 220-21). During the months immediately preceding and following the coup, Fujimori vanquished two forces that Peruvians had come to find unbearable: hyperinflation and the cycle of protest and violence that had culminated in the terrorism of Shining Path. The autogolpe was carried out at a time when economic stabilization was taking hold and Fujimori's popularity was on the rise (Tanaka 1998, 220-21). The cap- ture of Shining Path leader Abimael Guzman several months later 8 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES helped to consolidate that support. By late 1992, a clear majority of Peruvians had decided that Fujimori represented the sort of strong leader for whom they had been yearning. This success distinguished Peru from other Latin American cases in the 1990s. Other elected outsiders in the region (Collor, Serrano) failed to resolve their countries' deep political and economic crises and were ultimately defeated, and other hyperinflationary crises in the region (Argentina, Bolivia, Brazil) were resolved by established parties. Only in Peru was an antiparty president responsible for ending the crisis (Tanaka 1998, 52-53). Fujimori's successes effectively buried the established parties. The autogolpe created a new partisan cleavage: Fujimori versus "the opposi- tion." Given the president's popularity, this cleavage worked to the par- ties' great detriment. Anti-Fujimori politicians became collectively known as la oposici6n, a derogatory moniker connoting self-serving opponents of the government. Worse, established parties found them- selves on the wrong side of public opinion when they defended dem- ocratic institutions. Although surveys found substantial opposition to many of the abuses committed by the Fujimori government (Tanaka 1999, 10-14; Carri6n 2000), it is also clear that much of the electorate viewed those abuses as the price of strong and effective leadership. This was particularly true of working-class and rural Peruvians, many of whom had experienced only limited access to justice and tended to associate the preexisting rule of law with corrupt judges and police offi- cers. In general, then, a staunch defense of the rule of law was a prior- ity only for relatively privileged sectors of society (students, profession- als, unionized workers), and mainly in Lima. The opposition's defense of democratic institutions in the immedi- ate aftermath of the coup thus failed to win mass support. When Con- gress, meeting in secret, impeached Fujimori and appointed vice presi- dent Maximo San Roman to replace him, San Roman's presidential aspirations made him a laughingstock, and legislators were derided when they appeared in public. When constituent assembly elections were held in late 1992, APRA, AP, and the IU boycotted them as illegit- imate; yet the boycott failed to prevent the constitutional process from moving forward, and the vacancy left by the established parties was quickly filled by political newcomers. Eighty-five percent of the vote went to either Fujimorista or "independent" candidates. Although the established parties participated in subsequent elec- tions, they fared poorly. In the 1993 municipal elections, APRA, AP, the PPC, and IU collectively won just 33.3 percent of the vote, and in the 1995 presidential election, the four parties accounted for just 6.3 percent of the vote. In the 1995 race, no candidate who ran under an established party label won even 5 percent of the vote. Many regime opponents ral- 9
LATIN AMERICAN POLITICS AND SOCIETY lied behind the candidacy of former U.N. Secretary-general Javier P&rez de Cuellar, who headed the Uni6n por el Peru (Union for Peru, or UPP). Although it presented itself as independent, the UPP was widely per- ceived as another incarnation of la oposici6n. Perez de Cuellar received just 21.8 percent of the vote, barely a third of Fujimori's total. Peruvian politicians drew two lessons from the initial failures of the democratic opposition, both of which had far-reaching implications for the party system and for democracy. First, they concluded that a defense of the political status quo ante was not a viable electoral strategy. Not only did public support for Fujimori make the defense of democratic institutions unprofitable, but such a strategy associated politicians with the discredited old guard elite that led the oposici6n. Confronted with this problem, ambitious non-Fujimorista politicians began to distance themselves from prodemocratic forces and to define themselves as "independents." Many of them adopted an ambiguous position in regard to the Fujimori regime, refusing to condemn the autogolpe and avoid- ing, whenever possible, direct conflict with the president. Instead of challenging the regime itself, they sought to advance their careers within it, focusing on developing personal reputations as effective administra- tors. In so doing, the new "independents" essentially foreswore their role as agents of horizontal accountability. The second lesson politicians drew was that they no longer needed parties. During the 1980s, independent, candidate-centered electoral strategies were rarely an effective alternative to parties. Politicians who defected from major parties, such as Hugo Blanco, Miguel Angel Mufarech, and Andres Townsend, generally failed in the electoral arena (Tanaka 1998, 96-97). This situation changed considerably in the 1990s. Fujimori's success suggested that established party labels and organiza- tions were no longer necessary for (and might be a hindrance to) a suc- cessful political career. Fujimori himself invested little in party organization. His original party, Cambio 90 (Change 90, or C-90), lacked a program, a national structure, and a minimal activist base. Although Fujimori might have used his popularity to transform C-90 into an organized party, he ignored C-90 and even actively impeded its consolidation (Planas 2000, 347-51). Instead, he substituted state agencies for party organization (Roberts 1995, 2002; McClintock 1999).1 With the assistance of friendly television networks, he also relied heavily on direct mass appeals (Roberts 1995; Conaghan 2000, 282). In preparation for the 1992 con- stituent assembly elections, Fujimori created a second, equally person- alistic party: Nueva Mayorfa (New Majority, NM). Neither C-90 nor NM ever developed an organizational life of its own. Three years later, Fuji- mori created a third party, Vamos Vecino (Let's Go Neighbor), to com- pete in the 1998 municipal elections. The pattern was repeated in 2000, 10 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES when the creation of Peru 2000 brought the total number of Fujimorista parties to four. Fujimori thus established a new model of electoral organization: the disposable party. Although often referred to as "independent move- ments," such parties were actually little more than electoral labels or candidate-centered vehicles. During the 1990s, this model became gen- eralized as politicians of all ideological stripes borrowed Fujimori's orga- nizational strategy. The result was a massive hemorrhaging of the estab- lished parties and the proliferation of personalistic vehicles. Many of the new independents were defectors from established parties. The most important of these was Alberto Andrade of the PPC. A successful three-term mayor of the prosperous Lima district of Miraflo- res, Andrade quickly detected the antiparty mood of the electorate and, in anticipation of the November 1995 metropolitan Lima mayoral elec- tion, declared himself an independent. "No candidate affiliated with a party had a chance of winning the mayoralty of Lima," he bluntly told PPC leader Luis Bedoya Reyes later (Diario La Reptblica 2000). Andrade created a personalistic organization with an apolitical label: Somos Lima (We are Lima). The move succeeded: Andrade defeated Fujimori ally Jaime Yoshiyama in the 1995 election and established himself as the country's leading non-Fujimorista politician. Central to Andrade's initial success was his ambiguity in relation to the regime. During his first term as mayor, he avoided taking positions on national issues, particularly those related to democracy, and instead sought to build a reputation as an effective administrator. Anticipating a 2000 presidential bid, Andrade transformed Somos Lima into Somos Peri (SP), which, despite some effort to create a national organization, remained personalistic and largely without programmatic content. Other politicians who became independents in the 1990s include former Social Security Institute director Luis Castafeda Lossio, an ex-AP member who created the National Solidarity Party (PSN) as a vehicle for a 2000 presidential bid; and former APRA senator Javier Valle Riestra, who briefly served as Fujimori's prime minister in 1998. At the local level, influential mayors, such as Alexander Kouri of Callao (PPC), Jose Murgia of La Libertad (APRA), Michel Azcueta of Villa El Salvador (IU), and Angel Bartra of Chiclayo (AP), formed independent movements. Many legislators also became free agents, switching parties at each elec- tion to ensure reelection. Examples include Henry Pease (IU to UPP to Peru Posible, PP), Beatriz Merino (FREDEMO to the Independent Mor- alizing Front [FIM] to SP), Maximo San Roman (C-90 to Obras to UPP), Anel Townsend (UPP to SP to PP), and Alberto Borea (Hayista Base Movement to PPC to UPP). At the same time, many new and aspiring politicians began to create their own parties instead of joining existing ones. The most important 11
LATIN AMERICAN POLITICS AND SOCIETY of these was Alejandro Toledo, a business school professor and politi- cal amateur who created Pais Posible in preparation for a 1995 presi- dential bid. He resurrected the organization before the 2000 elections, changing its name to Peru Posible (PP). PP had no raison d'etre other than Toledo's presidential candidacy. Another example was Federico Salas, who was elected mayor of Huancavelica as an independent and who cofounded (but later abandoned) Peru Ahora (Peru Now) in antic- ipation of a presidential bid. By the end of the decade, the "independent movement" had emerged as the dominant mode of electoral organization in Peru. As Catherine Conaghan observes, presidential hopefuls "do not see their relationship to parties as especially integral to their own ambitions for power. What they do regard as crucial is their relationship to the media establishment and the cultivation of a positive public image, preferably as a political inde- pendent" (2000, 182). Indeed, all the country's successful parties in the late 1990s-including both progovernment parties and opposition parties, such as Somos Pera and Peru Posible-were personalistic, candidate-cen- tered vehicles that lacked national structures or even minimal links to civil society. Without programs or ideologies to identify themselves, many of these parties simply adopted the name of the territory they sought to rep- resent. At the national level, Somos Peru, Perut Posible, Peru Ahora, UPP, and Pern 2000 emerged between 1995 and 1999. At the local level, par- ties such as We Are Huancayo, Forward Chiclayo, Ayacucho 95, Eternal Cuzco, and Let's Save Huaraz proliferated. Thus, in the 1990s, electoral politics reached a degree of fluidity and atomization that surpassed any other country in Latin America, includ- ing notoriously fragmented party systems such as those of Brazil and Ecuador. Peru's party system was created anew at each election: in 1990, the leading parties were C-90, FREDEMO, and APRA; in 1995, they were New Majority and UPP; in 1998, Vamos Vecino and SP; in 2000, Peru 2000 and PP. In 2000, the parties that had dominated electoral politics in the 1980s together received less than 2 percent of the vote; and each of the top five presidential candidates-Fujimori, Toledo, Andrade, Salas, and Castaneda Lossio-presided over a candidate-centered party. This pattern was reproduced at the local level. Each municipality devel- oped its own party system. The number of independent mayors increased from 2 in 1986 to 79 in 1993 (Planas 2000: 268); and in 1998, independent lists proliferated to such a degree that the overall number of parties soared into the hundreds. A few attempts to build national parties were made in the 1990s. In 1998, Andrade built alliances with mayors and regional leaders in an effort to expand his movement nationwide (Planas 2000, 289-93), but this party-building process never really took hold. After the 1998 munic- ipal election, SP quickly fell victim to a government campaign to coopt 12 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES pro-Andrade mayors. In August and September 1999, 23 mayors aban- doned him (Planas 2000, 390). Many of these mayors were swayed by bribes or promises of future positions in the government; others despaired of governing without the support of the national executive. Although some scholars have cited these developments as evidence that government repression made party building impossible in the 1990s (Planas 2000, 394), the failure of the SP party-building project is probably best understood as a product of the incentives and constraints facing Peruvian politicians. Because Andrade could use the media instead of party organization to launch a national candidacy, his organizational efforts never even remotely approached those that went into building par- ties such as APRA or AP. Local politicians, moreover, had few incentives to join or remain with SP. Like Andrade, many of them were independ- ents who had won (or could win) office without national party affiliation. Lacking a well-oiled party machine or a coherent identity, SP could not establish enduring linkages to these local leaders. Without either the resources or the disciplinary mechanisms to ensure that SP members remained in the fold, Andrade's coalition was inherently unstable. PARTY WEAKNESS, CAESARISM, AND THE FAILURE OF THE DEMOCRATIC OPPOSTION The consolidation of an atomized system of candidate-centered politics seriously inhibited efforts to restore democracy in Peru.12 Between 1992 and 2000, opposition parties repeatedly failed to check the Fujimori government's autocratic behavior. They played, moreover, only a mar- ginal role in the regime's eventual collapse. Scholars differ considerably in their explanations of these failures. Some analysts argue that the opposition parties failed to seize opportunities to promote a democratic transition (Tanaka 1998, 1999). Others point to the repressive tactics of the Fujimori regime (Lynch 2000). This study offers a somewhat different assessment. Like Lynch, it views the opposition's failure to check Fujimori's autocratic behavior as rooted more in its political weakness than in its strategic mistakes; yet in this view, the opposition's weakness was a product not of govern- ment repression but of politicians' adaptation to an environment in which both the defense of democracy and party building were widely perceived as unprofitable. The "independent" politicians who abandoned anti-authoritarianism in favor of individualized efforts to win elections ceased to serve as instigators of horizontal accountability. Without vigorous opposition to the abuse of power, the network of public institutions responsible for upholding the rule of law became increasingly degraded and ineffec- tual. This inaction ultimately eroded vertical accountability as well. To 13
LATIN AMERICAN POLITICS AND SOCIETY ensure that elections are fair, electoral institutions must be transparent, and Congress and the judiciary must be able to check executive abuses. The unwillingness or inability to defend these institutions ultimately left opposition politicians defenseless as the conditions for free and fair elections eroded. The collapse of parties also eroded the opposition's capacity for col- lective action and mobilization. Key decisions, such as whether or not to participate in a questionable election, how to select a single opposi- tion candidate, and whether or not to negotiate with the regime (and on what terms), require organizations that can speak for (and discipline) large numbers of politicians across the national territory. The "inde- pendent movements" that emerged in Peru in the 1990s lacked such organizations. With no raison d'etre other than the election of their founder, they had short time horizons and lacked an encompassing interest in the preservation of democratic institutions. Lacking national infrastructures, strong roots in society, or linkages to important civic and social organizations, parties were reduced to narrow circles of elites, which left them, and the democracy movement they spearheaded, with- out much mobilizational muscle. Destruction of Mechanisms of Horizontal Accountability Between 1995 and 2000, the Fujimori government grew increasingly authoritarian. It circumvented, abused, or dismantled many of the mech- anisms of horizontal accountability that the regime itself had established in the 1993 Constitution, while expanding the powers of the National Intelligence Service (Servicio de Inteligencia Nacional, SIN) (Mauceri 2000; Rospigliosi 2000; Conaghan 2001). The process of regime harden- ing began with an issue that had been at the center of the autogolpe decision: presidential reelection. In August 1996, the congress passed the Law of Authentic Interpretation of the Constitution, which stated that Article 112 of the 1993 Constitution could not be applied retroactively and therefore that Fujimori had only been elected once under the new constitution. This law established the legal foundations for his 2000 reelection bid. Despite the clear unconstitutionality of Fujimori's effort to legalize his reelection bid, the opposition was unable successfully to challenge it through institutional channels. Fujimori's legislative majority limited the utility of the congress as a vehicle for checking executive power. Other mechanisms of horizontal accountability, such as the courts and the electoral authorities, were corrupted by the vast network of espi- onage, bribery, and extortion operated by Vladimiro Montesinos out of the SIN offices. The judicial branch was brought to heel via the system- 14 45: 3
LEVITSKY AND CAMERON: PERU'S PARTIES atic removal of independent judges and their replacement by Mon- tesinos agents and "provisional" judges whose tenure hinged on gov- ernment discretion (and who therefore tended to rule in the govern- ment's favor). The government also effectively subordinated Peru's highest electoral authority, the National Election Board (JNE). In Decem- ber 1997, the congress passed legislation (popularly known as the Fraud Law) that allowed recently appointed "provisional" judges to elect mem- bers of the JNE. This rule change allowed the government to stack the JNE with Fujimori allies, and in 1998, elections for five members of the JNE produced a solid progovernment majority. The institution that perhaps posed the greatest threat to Fujimori's reelection bid was the Constitutional Tribunal (TC), Peru's highest authority on constitutional matters. In January 1997, the TC ruled, in a 3-2 vote, that the Law of Authentic Interpretation did not apply to Fuji- mori. In response, the government effectively dismembered the TC. On May 28, 1997, the Fujimorista congress voted to impeach the three TC members who had voted against the government in January. The jus- tices were never replaced, leaving the country without its highest con- stitutional authority for more than three years. Without recourse to the legislature, the courts, or the electoral authorities, the opposition attempted to call a referendum on the reelec- tion issue.13 This effort was undermined by another act of Congress. In April 1996, the congress had approved the so-called Siura Law III (named after a lawmaker close to Montesinos), which required 48 votes in Congress before a referendum could be held. This law, which was initially adopted to stop an opposition-led referendum on privatization, clearly violated the spirit of the referendum provision in the constitu- tion. It effectively meant that because the opposition lacked 48 votes in the legislature, only the government could use referenda. Nevertheless, opposition forces, under the umbrella of a civic association called the Democratic Forum, launched a campaign to collect the 1.2 million sig- natures needed to call a referendum on the reelection. It gathered more than 1.4 million signatures, but in August 1998, Congress voted 67-45 to block the referendum. In December 1999, the JNE, now dominated by Fujimoristas, officially approved Fujimori's candidacy. A large majority of Peruvians opposed the government's abuses, and the blocking of the referendum provoked nationwide protests.14 This opposition, however, was not channeled into a sustained prodemocracy movement. The protests soon fizzled, and the TC and ref- erendum issues quickly faded from the national agenda. These opposi- tion failures were rooted largely in party weakness. The demise of organized parties left the democratic opposition without effective mobi- lizing structures or strong roots in society. Umbrella groups, such as the Democratic Forum, were supported by parties from across the political 15
LATIN AMERICAN POLITICS AND SOCIETY spectrum, but because the parties lacked infrastructures and activist bases, these organizations never became more than heterogeneous assortments of individual political elites. They provided forums for indi- vidual politicians but not channels for public participation or vehicles for large-scale mobilization. Party collapse also eroded the opposition's capacity to act collec- tively. The UPP, which had brought together a large portion of the anti- Fujimori opposition in 1995, was decimated by defections after the elec- tion. The UPP congressional bloc dwindled from 17 to just 7 members, and the party quickly lost credibility as a serious opposition vehicle (Conaghan 2000, 279). The plethora of loosely organized and personal- istic groups that emerged in its place seriously undermined opposition cohesion. Efforts to create a single opposition bloc in the legislature, such as the Parliamentary Democratic Opposition Bloc and the National Coordination Front, failed repeatedly. In the former case, the bloc dis- banded soon after its formation because none of the leaders could find time to meet (Latin American Weekly Report 1997). The opposition was further weakened because the leading non- Fujimorista politicians, such as Andrade, Castafeda Lossio, Toledo, and Salas, shied away from confrontation with the government and kept their distance from the democracy movement as they prepared their presidential bids. For example, although Andrade clearly aspired to the presidency, before 1998 he generally limited his public statements to municipal issues, such as traffic congestion and the remodeling of Lima's historic center.15 The other three stayed so far from the democ- racy movement that it was often difficult to tell if they represented the government or the opposition. Consequently, even as support for the government fell significantly between 1996 and 1998, no credible alter- native to Fujimori emerged. The 2000 Transition: Opposition Weakness and Collapse from Within Having failed in the effort to block Fujimori's unconstitutional bid for a third term in 2000, opposition parties were not the protagonists in the collapse of the Fujimori regime. Only the implosion of the regime itself brought about its demise. After nearly four years of preparation, Fujimori's December 1999 decision to register his name as a candidate in the 2000 election was anticlimactic. Fujimori knew his presence was needed to ensure that the political system he had built did not collapse in disarray, leaving key members to defend themselves against accusations of corruption and human rights abuses. Opinion polls, moreover, suggested that his chances of winning were good. The government hastily formed a new 16 45: 3
LEVITSKY AND CAMERON: PERU'S PARTIES "electoral alliance" called the Frente Independiente Peru 2000, which nominally included (but in reality circumvented) C-90, Nueva Mayoria, and Vamos Vecino. Despite sharp internal differences, Fujimorista lead- ers agreed on the need to achieve reelection. There was no question about where ultimate power lay: Montesinos was the undisputed mas- termind behind the reelection campaign.16 Opposition candidates faced formidable challenges in their efforts to defeat Fujimori. Beyond the president's continued popularity (which hovered near 40 percent), they confronted an electoral playing field mined with dirty tricks and unfair incumbent advantages. Fujimori enjoyed a virtual monopoly over network television, an array of tabloid newspapers (many of which were later revealed to be subsidized and directed by SIN) that systematically assaulted opposition candidates, and the vast resources of the state. For example, SIN spied on opposition candidates and hired thugs to organize counterdemonstrations at oppo- sition rallies; and state employees, including the armed forces, cam- paigned for progovernment tickets. Although opposition leaders com- plained bitterly about these conditions, the government's control over the judicial branch and electoral authorities, such as the JNE and the National Office of Electoral Processes (ONPE), left no means of redress. The absence of institutional checks on government electoral abuses was made particularly manifest by the 1999-2000 signature scandal involving Fujimori's newest electoral vehicle, Peru 2000. Because Peru 2000 lacked even a minimal activist base, it was unable to collect the nearly five hundred thousand signatures needed to register a new party. An investigation by the newspaper El Comercio (2000) showed that as many as one million signatures had been forged by hundreds of people working in "signature factories" in late 1999, apparently under the supervision of Fujimori ally Absal6n Vasquez. The scandal should have led to the disqualification of Peru 2000, but the JNE and ONPE stonewalled and downplayed the issue. The government promised to investigate itself, but a congressional investigation by Fujimorista Edith Mellado and a judicial investigation by the notoriously biased judge Mirtha Trabucco found nothing pointing to higher-ups. Under these conditions, the opposition's best strategy might have been either to withdraw from the race or to form a broad coalition behind a single candidate (Bazo and Cameron 2000), as anti-authoritarian forces did in Chile in 1989 and Nicaragua in 1990. Yet the major parties never held serious negotiations toward either of these goals. Indeed, the lead- ing candidates never once met in person to discuss the matter. Andrade, who led the field in many opinion polls in 1997 and 1998, was the least willing to join a common anti-Fujimori front. Thus, in April 1999, when eight opposition parties signed an accord to oppose Fujimori's reelection and work toward a single opposition candidacy, Andrade demurred. 17
LATIN AMERICAN POLITICS AND SOCIETY There are several reasons for the opposition's failure to unite behind a common strategy. Each of the major opposition candidates was an "independent" who, having abandoned the struggle to defend democratic institutions, simply sought to replace Fujimori in the new political system. The candidates also faced a collective action prob- lem: although the best options may have been for all candidates to boycott the election or to run united, each individual candidate pre- ferred the option to run while all the others stood down. This prob- lem was exacerbated by the extreme volatility of the electorate, which, by making it difficult to calculate candidates' electoral prospects, convinced all candidates that they had a chance of win- ning. It was further exacerbated by Peru's majority runoff electoral system, which allowed for the possibility that a marginal candidate could finish second in the first round and then win in the second round (as Fujimori did in 1990). Most fundamentally, however, the candidates' collective action problem was rooted in party weakness. Had the major candidates belonged to institutionalized parties, longer time horizons might have encouraged the adoption of a coordinated strategy, and mechanisms would have existed to impose a single candidacy. Yet SP, PSN, and PP were little more than candidate-centered electoral vehicles. Lacking col- lective identities or goals beyond the personal ambitions of individual leaders, their time horizons did not extend much beyond the 2000 elec- tions. They therefore had virtually no incentive to give up the chance to participate in those elections. Divided, the opposition stood little chance of defeating Fujimori in 2000. Andrade, who had long been the most popular opposition candi- date, suffered repeated and intense attacks by the progovernment media and saw his support decline precipitously in 1999.17 After a similar wave of government attacks weakened Castaneda Lossio, Toledo emerged from the pack as the most viable opposition candidate; and as soon as he did, support for Andrade and Castaieda Lossio evaporated-a clear sign of the fragility of their parties and electoral bases. The official results of the first round gave Fujimori 49.9 percent of the vote-tantalizingly close to the 50 percent needed to avoid a second round-and Toledo 40.2 percent. Fujimori's vote total was almost cer- tainly inflated (Interview, Senior Electoral Officer 2002), and many observers suspect that the ONPE would have awarded Fujimori a first- round victory had it not been for intense scrutiny by national and inter- national observers and massive election-night demonstrations in support of Toledo. The outcome, however, was not inconsistent with what polls had predicted.18 The government's refusal to establish a level playing field for the second round led Toledo, now free of the collective action problems that had hindered opposition candidates in the first round, to 18 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES withdraw from the race. Running unopposed, Fujimori won the second round with 51 percent of all ballots cast (including blank and null votes) and 74 percent of the valid votes.19 Party weakness also contributed to the opposition's failure to pre- vent Fujimori's reconsolidation of power after the contested election. Following the reelection debacle, the Fujimori government faced intense protest both in Peru and in the international arena. On the domestic front, Toledo led a large-scale protest, called the March of the Four Suyos, to coincide with Fujimori's July 28, 2000, swearing-in ceremony. On the international front, both the U.S. government and the Organiza- tion of American States observer mission declared the election unfair. Although the OAS General Assembly voted not to invoke Resolution 1080 and expel Peru on the grounds that its unfair election constituted an interruption of the democratic institutional order, it sent a high-level mission to Peru to recommend ways of improving the democratic process (Cooper and Legler 2001). None of these sources of pressure was sufficient to dislodge Fuji- mori from power. Although the March of the Four Suyos constituted the largest opposition protest of the Fujimori period, the government ably used the violence (later shown to have been the work of SIN agents) to discredit the opposition and discourage further protest. By August 2000, OAS-sponsored negotiations had stalled and the government was show- ing little inclination to carry out the democratizing reforms the opposi- tion demanded. Fujimori not only survived the immediate postelectoral protests but also managed to reconstruct a new governing majority. Pro-Fujimori forces won only 52 of 120 seats in the congress in 2000. Without a leg- islative majority, the government risked losing control over key com- mittees that could begin to investigate abuses committed by the execu- tive branch. To avoid this, Montesinos employed the same strategy he had used to weaken Somos Peru in 1998: he used a combination of bribery and blackmail to coopt individual opposition members. In the weeks following the 2000 election, as many as 18 legislators changed their partisan affiliation. Toledo's PP was the hardest hit by defections, losing nearly a third of its 29-member caucus. Popularly known as trdnsfugas, or turncoats, these defectors provided Fujimori with an ample legislative majority.20 The trdnsfugas justified their opportunism in a variety of ways. Many said they wanted to distance themselves from the March of the Four Suyos and to avoid extreme positions. They stressed their desire to work "for the country" rather than for a party. Most insisted that they were not joining the government but sitting as "independents" (although this, of course, freed them to vote routinely with the gov- ernment). In reality, most of the trdnsfugas were making deals with 19
LATIN AMERICAN POLITICS AND SOCIETY Montesinos, and many signed letters of adherence to the government side in the SIN headquarters. Typically, trdnsfugas were paid hand- somely for their defection, usually substantial monthly sums in cash. Others were given important legislative committee assignments, posi- tions that brought power, prestige, and additional perquisites. In other cases, Montesinos used his influence in the judiciary to call off investi- gations of members of Congress in exchange for their abandoning their party. Max Weber once noted that in the absence of parties, parliaments would become "a mere marketplace for compromises between purely economic interests, without any political orientation to overall inter- ests." In such a context, he argued, "any public control over the admin- istration would be vitiated" (1978, 1397). This characterization aptly fits Peru's 2000-2001 Congress. The "purchase" of turncoats thus effectively substituted for partisan coalition building in a context of extreme party weakness. Perti Posible spokesperson Luis Solari attributed the wave of defections to a lack of "responsibility" among individual legislators, declaring, "this is not a crisis of a party, but of persons" (El Comercio 2000). It was a crisis of party nonetheless. The defections reflected the tenuous nature of the linkages between "independent movements" and individual politicians. Because the "independent movements" were based almost entirely on the distribution of short-term selective incentives, a substantial number of elected legislators lacked even minimal ideological or affective ties to their parties (Grompone 2000, 145-46). They were essentially free agents; their party loyalties were virtually nil. Many legislators viewed elections as a commercial venture, involving an up-front investment to get onto a winning list. Once in power, they would recover their invest- ment through the perquisites of office and the opportunities for illicit enrichment it would present.21 The parties themselves, moreover, lacked mechanisms to induce or compel their representatives to remain in the fold. Indeed, given that unsuccessful candidate-centered parties, such as Castaneda Lossio's PSN, were unlikely to survive until the next election, their representa- tives had no incentive at all to remain with them. As of August 2000, then, Fujimori appeared to have survived the reelection debacle. Antiregime protests had largely subsided, opposi- tion forces had been discredited and demoralized, and Fujimori had constructed a new legislative majority that would enable him to govern securely for another five years. Both the United States and the OAS, moreover, had accepted the third term as a fait accompli. Domestic and international pressure was decisive only after cracks appeared in the regime. The fall of the regime primarily stemmed from internal divisions; it was, in the words of former military president Francisco Morales 20 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES Bermudez, a "monster destroyed from within" (Mariella Balbi 2000). Fujimori's reliance on Montesinos both to control the armed forces and to manage his reelection campaign blurred the boundaries between intelligence activities and partisan politics. The use of the security appa- ratus rather than an organized party reinforced the president's penchant for illegal and covert campaign activities (such as espionage, bribery, and blackmail, as well as forging signatures for Peru 2000). Yet these tactics eroded the regime's legitimacy and hurt the electoral chances of those running for Congress on Fujimori's ticket. The failure to win a majority in Congress induced the government to engage in further ille- gality: the SIN's purchase of members of Congress, the exposure of which destroyed the regime. The September 2001 release of a video showing Montesinos paying off a trdnsfuga (McClintock 2001; Cameron 2002) exposed the dark side of the regime by providing irrefutable evidence that Montecinos was running a secret "government within the government"; it effectively destroyed the congress as a functioning institution and damaged the president's already questionable electoral legitimacy; and, above all, it undermined the relationship between Fujimori and Montesinos. Mon- tesinos had accumulated a video archive documenting his control over much of the political establishment-members of Congress, senior judges, election officials, journalists, and media magnates. The purpose was to ensure his own indispensability, so that Fujimori could not remove Montesinos without bringing down the entire political system they had constructed. This meant that even after the video's release-which convinced Montesinos that he was "destroyed" (Inter- view, Former Intelligence Officer 2002)-he retained the ability to blackmail Fujimori. Seeing no other way out, Fujimori resigned, and within weeks the entire regime collapsed. As opposition leaders them- selves acknowledged, if the "Vladivideo" had not been released, the Fujimori government might well have survived.22 Change and Continuity in the 2001 Elections Under the leadership of Interim President Valentin Paniagua, Peru underwent a successful democratic transition in 2000-2001. The coun- try's democratic institutions were thoroughly reformed. Functionaries of integrity replaced Montesinos agents in the ONPE and the JNE. A judi- cial renovation eliminated many members of the "Montesinos Mafia," and hundreds of military officers were purged or were obliged to offer mea culpas for their relationship with Montesinos. In addition, media independence increased enormously. The transition culminated in new presidential elections in 2001. Toledo, who had been the frontrunner throughout the campaign, fin- 21
LATIN AMERICAN POLITICS AND SOCIETY ished first in the first round. Surprisingly, ex-president Alan Garcia fin- ished second. In the second round, Toledo defeated Garcia by 53 per- cent to 47 percent. Meanwhile, the vestigial Fujimorista forces were vir- tually wiped out, gaining just 4 of 120 seats in the congress. Although the 2001 election marked a fundamental break with the authoritarianism of the Fujimori regime, it also manifested important continuities. Garcia's improbable comeback (and to a lesser extent, that of APRA) and the relatively strong performance of PPC leader Lourdes Flores notwithstanding, electoral politics remained strikingly candidate- centered (for a different view, see Kenney 2001).23 All the presidential candidates except Garcia ran on tickets that did not exist before 1990, and most of these were little more than candidate-centered vehicles.24 Of the 11 parties that won seats in Congress, 8 could readily be classi- fied as personalist. Party switching was again rampant. Within the dom- inant electoral "movements," candidate selection was done in a person- alistic and often less-than-transparent manner. Peru Posible was rocked by allegations of nepotism and backroom dealing with powerful eco- nomic interests; 1,800 individuals came forward seeking places on the PP list. Places were purchased by wealthy people or were given to can- didates with no previous ties to PP. The 2001 election was also marked by the low quality of the can- didates. Although Toledo was more committed to democratic institu- tions than Fujimori had been, in other ways his rise was remarkably similar to that of his predecessor. Like Fujimori, Toledo had little expe- rience in politics and had never held elected office. He was a marginal candidate whose second-place finish in 2000 was virtually acciden- tal-a product of a wave of strategic voting by anti-Fujimoristas des- perately seeking a viable candidate. Like C-90, Toledo's PP was, at least initially, little more than a vehicle for his presidential bid. Although he was less autocratic than Fujimori, Toledo's political style was highly personalistic. His inexperience caused repeated mistakes during the campaign. Indeed, a major reason for his 2001 victory was simply the weakness of the rest of the candidate field. Finally, like Fujimori in 1990, Toledo lacked a legislative majority, as PP gained only 45 of the 120 seats in the congress. The return to democracy in 2001 therefore was not accompanied by a renaissance of political parties. Mechanisms of representation remained weak, and personalism, electoral volatility, and party switch- ing remained high. For voters, the 2001 electoral results were not very encouraging; the choice in the second round was between a traditional politician from an established party whose track record was nothing less than disastrous, and a new "independent" politician who had opposed Fujimori but whose electoral strategy faithfully reproduced the Fuji- morista style. 22 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES THE PROSPECTS FOR PARTY (RE)BUILDING IN PERU What are the prospects for rebuilding political parties in post-Fujimori Peru? Answers to this question vary, often hinging on scholars' expla- nations of the party system's collapse. For some analysts, the failure of party-building efforts in the 1990s was largely a product of the repres- sive tactics of the Fujimori regime. As the government's assault on Somos Peru in 1998 and 1999 seemed to suggest, party building was exceedingly difficult while Fujimori was in power (Planas 2000, 389-94; also Lynch 2000). Moreover, given that democratic institutions, such as the legislature, had been rendered ineffective under Fujimori, normal party-building incentives may have been weakened. Such an approach yields a relatively optimistic prognosis for party development in the post-Fujimori era. If the primary problem in the 1990s was the Fujimori regime or the absence of democracy, then the 2000 transition should clear the way for new party-building efforts. Institutionalist approaches locate the roots of party weakness and party system fragmentation in electoral rules-such as open-list propor- tional representation (PR), the majority runoff system, and the single national district-that weaken incentives for politicians to build strong organizations (Schmidt 1996; Tuesta 1996, 1998). If these rules persist in the post-Fujimori period, institutionalists argue, there is little reason to expect the emergence of strong parties. Electoral rules can be changed, however, and politicians' incentives could potentially be reshaped in ways that would encourage party building. The 2000 democratic transition created an opportunity for such institutional engineering. The transition triggered discussion among both scholars and politicians of a wide range of constitutional and electoral reforms. One of them, the replacement of the single electoral district with a multiple-district system, was implemented before the 2001 elec- tions. Although the majority runoff system and open-list PR may prove more resistant to change, an institutionalist perspective holds out at least some hope that parties can be rebuilt in the future. A third approach, which might be characterized as historical-struc- tural, yields a more pessimistic forecast. According to this approach, strong parties are products not of electoral engineering but of particular historical, sociological, and technological conditions, many of which are only weakly present in the contemporary period. Stable party systems are often rooted in deep societal cleavages or intense political struggles (Lipset and Rokkan 1967; Collier and Collier 1991). Indeed, many of Latin America's strongest and most enduring parties emerged from civil wars (the traditionally dominant parties in Colombia, Costa Rica, and Uruguay), social revolutions (Bolivia's MNR, Nicaragua's FSLN, Mexico's 23
LATIN AMERICAN POLITICS AND SOCIETY PRI), or periods of sudden or dramatic expansion of the electorate (Argentine Peronism and Democratic Action in Venezuela). As Michael Coppedge (1998) has argued, these party-building episodes were in many respects historically bound. Most stable party systems either took shape before the advent of mass suffrage (Chile, Colombia, Uruguay) or emerged from the process of mass enfranchisement (Argentina, Mexico, Venezuela). After this period, the crystalization of party loyalties and the absence of large groups of new voters limited the electoral space avail- able to new parties (Coppedge 1998, 175). Thus, with the exception of cases in which civil wars or revolutions created new mass identities (for example, El Salvador), few stable party systems have emerged in Latin America after the establishment of full suffrage. Moreover, long-term changes in class structures and technology have reduced politicians' need for party organizations and have increased the cost of building and sustaining them. Many of Latin Amer- ica's largest and most socially rooted parties built their electoral bases on large peasantries or emerging industrial working classes. Because of their geographic concentration and lower levels of education and social mobility, these social classes were both relatively easy to organize and relatively stable in terms of their political loyalties. With the decline of peasantries and industrial working classes and the expansion of the informal sector, the social bases of party politics have grown increas- ingly unstable. Geographically fragmented and extremely heteroge- neous in terms of their work, interests, and identities, informal sector workers are difficult to organize and encapsulate (Cameron 1994; Roberts 1998). Indeed, few new parties in Latin America have built stable electoral bases on these sectors. Technological change also militates against contemporary party- building efforts. Most large-scale party organizations were established before the spread of television. In the absence of mass media and com- munications technologies, politicians were often compelled to build and maintain extensive territorial organizations in order to reach voters. Although politicians' ability to substitute state resources or corporatist structures for party organization inhibited party development in some countries (Mainwaring 1999), party organization was nonetheless critical to electoral success in most countries. In the contemporary period, the increased influence of mass media technologies has weakened politicians' incentive to invest in party organization (Katz 1990; Perelli et al. 1995). As the success of media-based candidates, such as Fernando Collor de Mello and Vicente Fox, suggests, contemporary politicians may reach millions of voters through television and may do so more quickly and at lower cost (in terms of human and organizational resources) than through party organizations. Local party structures therefore have become increasingly "vestigial" (Katz 1990; for a critique of this view, see Scarrow 1996). 24 45: 3
LEVITSKY AND CAMERON: PERU'S PARTIES These structural changes do not spell the imminent demise of party organizations. In many countries, the persistence of strong par- tisan identities and the significant human, organizational, and patron- age resource advantages enjoyed by existing parties continue to serve as important barriers to entry for political outsiders. As long as politi- cians believe that the benefits of remaining in an existing party out- weigh the benefits of defecting and competing as an outsider, estab- lished parties may endure. Once established parties fail, however, the incentives for politicians change dramatically. Mass media and the volatility of contemporary electorates may remove individual politi- cians' incentive to build new party organizations from scratch. Conse- quently, contemporary party systems may prove to be somewhat like Humpty Dumpty: in the absence of crisis, they may persist, but if they happen to collapse (for any number of historically contingent rea- sons), all the institutional engineering in the world may be insufficient to put them back together again. The challenges of party building are exceptionally great in the Andes, where exclusion from citizenship rights is an enduring legacy of colonialism and where local, indigenous cultures coexist uneasily with the individualistic notions of citizenship that underpin representative democratic institutions. There, where the gap between the pais oficial (formal political institutions) and the pais real (established habits and customs) is greatest, the obstacles to consolidating mass partisan link- ages are particularly imposing (Van Cott 2000). A historical-structural analysis thus suggests that the prospects for rebuilding parties in Peru may be rather bleak. According to this approach, the 1990s may constitute a critical juncture for the Peruvian party system. Although the collapse of the old party system was in many ways a historically contingent outcome, the opportunities and con- straints posed by the new social, structural, and technological context are such that politicians lack both an incentive and the capacity to build new party organizations (see also Conaghan 2000, 280-82). In other words, although long-term structural changes did not cause the collapse of Peru's party system, they may inhibit its reconstruction. Two additional factors may help to reinforce or "lock in" the post- 1992 party system configuration. First, the contemporary environment "selects for" candidates (such as Toledo) who can succeed at media- based, candidate-centered politics, while the kinds of politicians that are critical to building and sustaining parties-good party bureaucrats or machine politicians-find little demand for their services. To the extent that this is the case, aspiring politicians will invest in the skills necessary to win as "independents." Second, electoral rules may become an endogenous variable in this context. Most elected offices are currently held by nonparty politicians who know how to win by electoral rules 25
LATIN AMERICAN POLITICS AND SOCIETY that favor weak parties. Such politicians are less likely than party politi- cians to support reforms aimed at strengthening parties. This is not to argue that no party rebuilding should be expected in post-Fujimori Peru. Just as relatively stable parties were built from scratch in many Central European countries in the 1990s, stronger par- ties may well re-emerge in contemporary Peru. Indeed, the resurgence of APRA in 2001 and the rise of party-oriented politicians such as Lour- des Flores (PPC) and Valentin Paniagua (AP) in the wake of Fujimori's departure created an important opportunity for party building. There is some evidence that these leaders have begun to invest seriously in party organization efforts. In the aftermath of the 2001 election, APRA launched a drive to recruit new grassroots leaders, using the Casa del Pueblo, the historic party locale in Lima, to train young provincial lead- ers from all over the country.25 Similarly, both the PPC and AP have launched drives to expand their sagging memberships, especially out- side Lima.26 Given the alternative electoral strategies available to local and national politicians, however, contemporary party-building efforts are likely to encounter greater obstacles than they did in the past. To the extent that new parties emerge, these parties are likely to be loosely structured organizations with fluid electoral bases and tenuous linkages to society. Like many other new parties in the region (such as the Argen- tine Front for a Country in Solidarity and the Chilean Party for Democ- racy), they would rely more on media-based and candidate-centered appeals than on organization and activists. Although such parties would be far better for democracy than the independent movements that dom- inate contemporary Peruvian politics, their capacity effectively to chan- nel societal interests, recruit and socialize elites, and facilitate collective action and social mobilization remains open to question. Therefore, although it would be premature to predict the demise of political par- ties in Peru, party system decomposition may prove substantially more enduring than the regime that provoked it. DEMOCRACY WITHOUT PARTIES? Peruvian democracy rebounded in 2001, but thus far, political parties have not. Even if the recent rise of party politicians such as Garcia, Flores, and Paniagua breathes life into the old parties, there is reason to think that the era of well-organized and socially rooted parties is over. If this is the case, then what are the prospects for post-Fujimori democ- racy? Here the lessons of the 1990s are sobering. The challenge of making democracy work in a context of fluid and fragmented electoral politics will be difficult. To govern effectively, politicians will be compelled to innovate, particularly in the areas of 26 45: 3
LEVITSKY AND CAMERON: PERU'S PARTIES coalition building and legislative organization. Yet even in the most opti- mistic of scenarios, the likelihood of executive-legislative conflict, exec- utive abuse of power, corruption, personalism, and successful outsider and even antisystem candidacies will remain high. Peru may not be alone in this challenge: the Venezuelan party system also decomposed in the 1990s, and long-established party systems in Argentina and Colombia showed signs of severe erosion in 2001 and 2002.27 These developments highlight a paradoxical aspect of the relation- ship between parties and democracy: although parties are essential to the effective functioning of modern democracies, they are not typically created for that purpose. Parties are created by politicians in an effort to resolve coordination problems and further their own careers (Aldrich 1995). To an extent, then, parties' various contributions to democracy are felicitous byproducts of organizations that are created for other pur- poses. As long as politicians believe that they can advance their careers through parties, then the "democratic goods" provided by parties will be provided. But if politicians (or voters) decide they are better off without parties, then those democratic goods may be underprovided. In such cases, it unclear what kinds of institutions or organizations would pro- vide those goods. This returns us to the conundrum mentioned at the beginning of this article: parties are among the least credible democratic institutions in Latin America today, yet democracy without them is nearly incon- ceivable. The Peruvian experience offers stark evidence of the indis- pensability of parties as mechanisms of representation. Whether recog- nition of this evidence stimulates a renewed interest in party building in the region, however, remains to be seen. NOTES An earlier version of this article was presented at the 23d Congress of the Latin American Studies Association, Washington, DC, September 6-8, 2001. We appreciate comments from Manuel Alcantara, Fabiola Bazo, Charles Kenney, Scott Mainwaring, Cynthia McClintock, David Scott Palmer, Gregory Schmidt, and three anonymous referees. Max Cameron is grateful to the Social Sciences and Humanities Research Council of Canada for a research grant that enabled him to conduct work in Peru in December 2001 and January 2002. Research assistance was provided by Annabella Espana, Mikala Grante, and Lissette Torres. The authors alone are responsible for the content of this article. 1. Modifying Schattschneider's oft-cited claim that "democracy is unthink- able save in terms of parties" (1942, 1), Aldrich writes that "democracy is unworkable save in terms of parties" (1995, 3). 2. According to Morris Fiorina, "the only way collective responsibility has ever existed, and can exist, given our institutions, is through the agency of the political party" (1980, 26). 27
LATIN AMERICAN POLITICS AND SOCIETY 3. In this sense, horizontal accountability is critical to what O'Donnell (1994) calls "vertical accountability," or free and fair elections. 4. Such a phenomenon occurred in Venezuela in the 1980s and 1990s (Coppedge 2003). 5. Although Latin American parties have functioned less well as channels for political recruitment, parties such as Venezuela's Democratic Action, the Chilean Socialist and Communist Parties, and the Brazilian Workers' Party have served as important channels of access to the political arena for working classes. 6. For example, Planas (2000, 399) and Levitt (2000) argue that Peru's tra- ditional parties have always been weak and personalistic, while Tanaka (1998) contends that the parties were relatively strong in the 1980s. 7. Relevant features of Peru's electoral system include a proportional rep- resentation system with no minimum threshold for legislative representation; a majority runoff system for presidential elections; a single national district for leg- islative elections, which eliminated the need for national party structures; and the double preferential vote system (in which voters cast preferential votes for two candidates on party lists), which encourages candidate-centered campaign- ing (Schmidt 1996; Tuesta 1996, 1998). 8. Indeed, Fujimori's stunning rise in the polls in the days before the elec- tion is best explained by strategic shifts in voting intentions by Peruvians des- perate to avoid a runoff between Vargas Llosa and APRA (Schmidt 1996, 344-46). 9. Fujimori also faced the real possibility of a military coup. A military cabal began preparing a coup in 1989 and had planned to prevent Fujimori from taking office (Rospigliosi 2000, 74-82). 10. Even though the opposition parties were widely discredited during the 1990-92 period, Fujimori's success was far from assured. Established parties won battles with presidents in Brazil, Guatemala, and Ecuador during the 1990s, even though they lacked broad public support. Fujimori, moreover, was not in a particularly strong position during 1990 and 1991. For much of 1991, Fujimori's public approval rating was below 40 percent (Tanaka 1998, 219). 11. These included the ministry of the presidency, the armed forces (which, in lieu of party activists, were used in electoral campaigns to paint pro-Fujimori electoral graffiti), the tax collection agency, the National Intelligence Service, municipal governments, and social welfare agencies, such as the National Food Assistance Program and the National Compensation and Development Fund (FONCODES). 12. According to David Scott Palmer (2000),. in regions such as Ayacucho, emerging civil society organizations effectively substituted for parties in chan- neling citizen participation and meeting local needs. 13. The right of citizens to initiate a referendum was enshrined in the 1993 Constitution. 14. For example, a survey conducted by the Instituto de Desarrollo e Inves- tigaci6n de Ciencias Econ6micas (IDICE) found that 77.6 percent of Peruvians disapproved of the dismissal of the three TC justices (Flores 1997, 2). 15. Andrade became more active in opposing the regime beginning in 1998. 16. To make this clear, congressional candidates were required to sign let- ters of adherence to the movement in the SIN headquarters. 28 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES 17. Although this decline was largely a product of government-sponsored attacks, it may also be attributed to Andrade's gradual abandonment of a non- confrontational "independent" strategy and his increasing association with the democratic opposition. 18. The "fraudulent" character of the election lay less in the vote-counting process than in the unfair and unfree conditions leading up to the vote, which fell well short of internationally accepted norms. 19. Toledo instructed his supporters to write "no to fraud" on their ballots. 20. The exact number of turncoats is not clear because the defectors were often ambiguous about their intentions, and some defected and then repented (the so-called trdnsfugas arepentidas). A new vocabulary emerged to describe the alchemy though which the government converted members of the opposi- tion. In addition to turncoats, there were reverse turncoats (those who aban- doned one side and then another), repentant turncoats (arepentidos, those who abandoned their party and then thought better of it), neoturncoats (those who abandoned the government party after it began to lose its grip on power) and moles (topos, or double agents who remained in an opposition party while taking orders from the government). 21. One member of Congress committed a Freudian slip during his swear- ing-in ceremony, declaring his allegiance to "God and money" (Dios y la plata) instead of "God and homeland" (Dios y la patria). 22. Interviews were conducted with four of the most important party lead- ers in Congress: one representing the PPC, one from APRA, and two from the FIM. They were asked, "Which was more important in bringing down Fuji- mori: the opposition (parties, Congress, civil society, marches, the independ- ent media, international organizations and foreign states) or the regime's inter- nal problems (the Vladivideo and the dispute between Fujimori and Montesinos)?" All agreed that the latter was more important, saying "the video was the detonator," "were it not for the video we would still be under the Fuji- mori regime," or "the video brought Fujimori down" (interviews conducted January 8-14, 2002). This is consistent with survey results reported in McClin- tock 2001, 139. 23. APRA received less than 2 percent of the vote in the 2000 presidential election. Hence, its improved performance in 2001 clearly was almost entirely a personal vote for Garcia. 24. It is noteworthy that Flores, despite being a PPC member, chose to create a new ticket called National Unity for her presidential bid. 25. Exuding optimism about APRA's future, one leader insisted that the party is focused on "the next generation, not the next election," and that this commitment to cultivating its members ensures that they remain "Apristas until death" (Interview, APRA Congressional Leader 2002). 26. Much of the PPC's party-rebuilding efforts centered on winning back former business and middle-class supporters who had defected to Fujimori during the 1990s, although party leaders have also begun to think about strate- gies for building support among the urban and provincial poor. Interview, PPC Congressional Leader 2002. 27. Alcintara and Freidenberg (2001) offer a more optimistic perspective on Latin American party systems. 29
LATIN AMERICAN POLITICS AND SOCIETY REFERENCES Alcantara Saez, Manuel, and Flavia Freidenberg. 2001. Los partidos politicos en America Latina. Am6rica Latina Hoy 27 (April): 17-35. Aldrich, John H. 1995. Why Parties? The Origin and Transformation of Political Parties in America. Chicago: University of Chicago Press. American Popular Revolutionary Alliance (APRA) Congressional Leader. 2002. Author interview. Lima, January 9. Bazo, Fabiola, and Maxwell A. Cameron. 2000. Dilemmas of the Opposition, Peru Election 2000: A Public Education Website. Accessed June 2003. Boron, Atilio. 1992. Becoming Democrats? Some Skeptical Considerations on the Right in Latin America. In The Right and Democracy in Latin America, ed. Douglas A. Chalmers, Maria do Carmo Campello de Souza, and Atilio Bor6n. New York: Praeger. 68-95. Cameron, Maxwell A. 1994. Democracy and Authoritarianism in Peru: Political Coalitions and Social Change. New York: St. Martin's Press. -- . 1997. Political and Economic Origins of Regime Change in Peru: The Eigh- teenth Brumaire of Alberto Fujimori. In The Peruvian Labyrinth: Polity, Society, Economy, ed. Cameron and Philip Mauceri. University Park: Penn State University Press. 37-69. - . 1998. Self-Coups: Peru, Guatemala, and Russia. Journal of Democracy 9, 1: 125-39. . 2002. Endogenous Regime Breakdown: The Vladivideo and the Fall of Peru's Fujimori. Paper prepared for the conference "The Fujimori Legacy and Its Impact on Public Policy in Latin America," sponsored by the Dante B. Fascell North-South Center and the University of Delaware Department of Political Science and International Relations. Washington, DC, March 14. Carri6n, Julio. 2000. La campana electoral y la opini6n piblica en el Peri actual. Paper presented at the 22d Congress of the Latin American Studies Associ- ation, Miami, March 16-18. Collier, Ruth Berins, and David Collier. 1991. Shaping the Political Arena: Crit- icalJunctures, the Labor Movement, and Regime Dynamics in Latin Amer- ica. Princeton: Princeton University Press. El Comercio (Lima). 2000. Aseguran que no hay crisis de partido dentro de Peri Posible. August 3. Conaghan, Catherine M. 2000. The Irrelevant Right: Alberto Fujimori and the New Politics of Pragmatic Peru. In Conservative Parties, the Right, and Democracy in Latin America, ed. Kevin J. Middlebrook. Baltimore: Johns Hopkins University Press. 255-84. -. 2001. Making and Unmaking Authoritarian Peru: Re-Election, Resistance, and Regime Transition. North-South Agenda Paper No. 47. Coral Gables: North-South Center. Cooper, Andrew, and Thomas Legler. 2001. The OAS in Peru. Journal of Democ- racy 12, 4: 123-36. Coppedge, Michael. 1998. The Evolution of Latin American Party Systems. In Politics, Society, and Democracy: Latin America, ed. Scott Mainwaring and Arturo Valenzuela. Boulder: Westview Press. 171-206. 30 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES . 2003. Venezuela: Popular Sovereignty versus Liberal Democracy. In Con- structing Democratic Governance, 2d ed., ed. Jorge I. Dominguez and Michael Shifter. Baltimore: Johns Hopkins University Press. 165-192. Corrales, Javier. 2001. Strong Societes, Weak Parties: Regime Change in Cuba and Venezuela in the 1950s and Today. Latin American Politics and Soci- ety 43, 2 (Summer): 81-114. Cotler, Julio. 1994. Politica y sociedad en el Perui. Lima: Instituto de Estudios Peruanos. Diario La Repzblica (Lima). 2000. Especial de elecciones: Alberto Andrade Carmona. February 6. Accessed June 3, 2002. Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper and Row. Fiorina, Morris. 1980. The Decline in Collective Responsibility in American Pol- itics. Daedalus 109 (Summer): 24-45. Flores, Ines. 1997 Desaprobaci6n a destituci6n de magistrados sube al 77.6% S61o el 39.5% aprueba a Fujimori. Diario La Repzblica, June 1: 2. Franklin, James C. 2001. The Role of Party-led Dissent in Redemocratization. Party Politics 7, 5: 567-80. Gibson, Edward. 1996. Class and Conservative Parties: Argentina in Compara- tive Perspective. Baltimore: Johns Hopkins University Press. Grompone, Romeo. 2000. Al dia siguente: el fujimorismo como proyecto incon- cluso de transformaci6n politica y social. In Elfujimorismo: ascenso y cafda de un regimen autoritario, ed. Grompone and Julio Cotler. Lima: Instituto de Estudios Peruanos. 77-174. Katz, Richard S. 1990. Party as Linkage: A Vestigial Function? European Journal of Political Research 18: 141-61. Kenney, Charles. 1996. iPor que el autogolpe? Fujimori y el congreso, 1990-1992. In Tuesta Soldevilla, ed. 1996. 75-104. - . 2000. Presidents, Parties, and Horizontal Accountability in Latin America. Paper presented at the 22d Congress of the Latin American Studies Associ- ation, Miami, March 16-18. --. 2001. Guess Who's Coming to Dinner: the Death and Rebirth of Tradi- tional Party Politicians in Peru, 1978-2001. Paper presented at the 23d Con- gress of the Latin American Studies Association, Washington, DC, Septem- ber 6-8. Latin American Weekly Report. 1997. Fujimori Promises Big Social Projects: Pres- ident Seeks to Halt Decline in Popularity. June 24: 297. Levitt, Barry Steven. 2000. Continuity and Change in Peru's Political Parties, 1985-2000. Paper presented at the 22d Congress of the Latin American Studies Association, Miami, March 16-18. Lipset, Seymour Martin, and Stein Rokkan. 1967. Cleavage Structures, Party Sys- tems, and Voter Alignments: An Introduction. In Party Systems and Voter Alignments: Cross-National Perpsectives, ed. Lipset and Rokkan. New York: Free Press. 1-64. Lynch, Nicolas. 1999. Una tragedia sin hroes. La derrota de los partidos y el origen de los independientes. Peru 1989-1992. Lima: Fondo Editorial Uni- versidad Nacional Mayor de San Marcos. 31
LATIN AMERICAN POLITICS AND SOCIETY . 2000. Los partidos en los noventa en el Peru: Ique paso y que pasara? Paper presented at the 22d Congress of the Latin American Studies Associ- ation, Miami, March 16-18. Mauceri, Philip. 2000. Unchecked Power: The Presidency Under Fujimori and Beyond. Paper presented at the 22d Congress of the Latin American Stud- ies Association, Miami, March 16-18. Mainwaring, Scott. 1999. Rethinking Party Systems in the Third Wave Democra- tization: The Case of Brazil. Stanford: Stanford University Press. Mainwaring, Scott, and Timothy R. Scully. 1995. Introduction: Party Systems in Latin America. In Building Democratic Institutions: Party Systems in Latin America, ed. Mainwaring and Scully. Stanford: Stanford University Press. 1-34. McClintock, Cynthia. 1996. La voluntad politica presidencial y la ruptura consti- tucional de 1992 en el Perui. In Tuesta Soldevilla, ed. 1996. 53-74. - . 1999. iEs autoritario el gobierno de Fujimori? In Eljuegopolftico: Fujimori, la oposici6n y las reglas, ed. Fernando Tuesta Soldevilla. Lima: Fundaci6n Friedrich Ebert Stiftung. 65-95. - . 2001. The OAS in Peru: Room for Improvement. Journal of Democracy 12, 4: 137-40. McGuire, James. 1997. Peronism Without Per6n: Unions, Parties, and Democ- racy in Argentina. Stanford: Stanford University Press. O'Donnell, Guillermo. 1994. Delegative Democracy. Journal of Democracy 5, 1: 55-69. . 1998. Horizontal Accountability in New Democracies. In The Self-Restrain- ing State: Power and Accountability in New Democracies, ed. Andreas Schedler, Larry Diamond, and Marc Plattner. Boulder: Lynne Rienner. 29-51. Palmer, David Scott. 2000. Direct Democracy at the Grassroots in Peru: "Infor- mal Politics" in Ayacucho. Paper presented at the 22d Congress of the Latin American Studies Association, Miami, March 16-18. Paredes Castro, Juan, Hugo Guerra Arteaga, and Juan Zuniga Sanudo. 1992. Entrevista al presidente Alberto Fujimori: "ni falsa democracia, ni dictadura militar." El Comercio, May 3: 4-5. Perelli, Carina, Sonia Picado, and Daniel Zoviatto, eds. 1995. Partidos y clase politica en America Latina en los 90. San Jose: Instituto Interamericano de Derechos Humanos. Peru. Former Intelligence Officer, member of SIN. 2002. Author interview. Lima, January 9. Peru. Senior Electoral Officer with ONPE. 2002. Author interview. Lima, January 8. Planas, Pedro. 2000. La democracia volatil: movimientos, partidos, liderespoliti- cos y conductas electorales en el Peru contemporaneo. Lima: Fundaci6n Friedrich Ebert. Popular Christian Party (PPC). Congressional Leader. 2002. Author interview. Lima, January 8. Roberts, Kenneth M. 1995. Neoliberalism and the Transformation of Populism in Latin America. World Politics 48, 1: 82-116. _. 1998. Deepening Democracy? The Modern Left and Social Movements in Chile and Peru. Stanford: Stanford University Press. 32 45:3
LEVITSKY AND CAMERON: PERU'S PARTIES -. 2002. Do Parties Matter? Lessons from the Fujimori Experience. Paper pre- pared for the conference "The Fujimori Legacy and Its Impact on Public Policy in Latin America," sponsored by the Dante B. Fascell North-South Center and the University of Delaware Department of Political Science and International Relations. Washington, DC, March 14. Rospigliosi, Fernando. 2000. Montesinos y lasfuerzas armadas. Lima: Instituto de Estudios Peruanos. Scarrow, Susan. 1996. Parties and TheirMembers: Organizingfor Victory in Ger- many and Britain. New York: Oxford University Press. Schattschneider, E. E. 1942. Party Government. New York: Rinehart. Schmidt, Gregory D. 1996. Fujimori's 1990 Upset Victory in Peru: Electoral Rules, Contingencies, and Adaptive Strategies. Comparative Politics 28, 3: 321-55. Tanaka, Martin. 1998. Los espejismos de la democracia: el colapso del sistema de partidos en el Peri. Lima: Instituto de Estudios Peruanos. --. 1999. Los partidos politicos en el Perut, 1992-1999: sobrevivencia, estatal- idad y politica medidtica. Lima: Instituto de Estudios Peruanos. Tarrow, Sidney. 1994. Power in Movement: Social Movements, Collective Action, and Politics. New York: Cambridge University Press. Tuesta Soldevilla, Fernando. 1996. El impacto del sistema electoral sobre el sis- tema politico peruano. In Tuesta Soldevilla, ed. 1996. 129-68. -- . 1998. Instituciones, reformas, y representaci6n politica en el Peru. Paper presented at the 21st Congress of the Latin American Studies Association, Chicago, September 26-30. Tuesta Soldevilla, Fernando, ed. 1996. Los enigmas del poder: Fujimori 1990-1996. Lima: Fundaci6n Friedrich Ebert. Van Cott, Donna Lee. 2000. Party System Development and Indigenous Popula- tions in Latin America: The Bolivian Case. Party Politics 6, 2: 155-74. Weber, Max. 1978. Economy and Society. Vol. 2. Ed. Guenther Roth and Claus Wittich. Berkeley: University of California Press. Weyland, Kurt. 1999. Neoliberal Populism in Latin America and Eastern Europe. Comparative Politics 31, 4: 379-401. 33
