Civil Society, Transitions, and Post-war Reconstruction in Latin America: A Comparison of El Salvador, Guatemala, and Peru

This article discusses the role of civil society in three recent and quite different cases of regime transition-cum - peace and reconstruction: El Salvador, Guatemala, and Peru. Open or ‘low intensity’ civil war characterized these three countries civil war in the 1980s and early 1990s, accompanied by severe socioeconomic problems and a difficult regime transition agenda. This poses a particular challenge to the role of civil society. Traditionally weakly articulated in these countries and battered by authoritarianism and violence, grassroots organizations, horizontal associations, and NGOs nevertheless took on an important role in the peace process in El Salvador, and especially in Guatemala. Peru, in contrast, witnessed the erosion and eventual breakdown of civil society as a result of the civil war and the related economic, social and political crises. As it turns out, the abandonment of a military and repressive logic within the state and state-related political actors is a first condition for a successful peace process and an active role of civil society. However, if the protagonists of the conflict manage to monopolize the peace process, the role of civil society in influencing the peace process and its implementation will be severely constrained. This may in turn lead to a less-than-optimal outcome and to ongoing political confrontations about the agenda of reconstruction and democratization. Este articulo analiza el papel de la sociedad civil en tres casos recientes de transicion politica cum reconstruccion post-conflicto: El Salvador, Guatemala y Peru. Estos tres paises experimentaron durante los 1980s y los primeros anos de los 90 un conflicto civil abierto de ‘baja intensidad’. Dichos conflictos fueron acompanados de graves problemas socio-economicos y de una compleja agenda de transicion politica. Este escenario influyo profundamente el papel a jugar por la sociedad civil. Organizaciones de base, ONGs y otros entes civicos se fortalecieron visiblemente durante los anos de violencia y represion. De esta manera, lograron asumir un papel importante en los procesos de paz, especialmente en El Salvador y Guatemala. En Peru, en cambio, se produjo una erosion y luego el colapso de la sociedad civil como resultado de la guerra civil y de una crisis social, economica y politica generalizada. El articulo concluye que la superacion de la logica militarista y represiva representa una primera condicion para el logro de un proceso de paz exitoso y la activacion de la sociedad civil. Sin embargo, cuando los protagonistas del conflicto logran monopolizar al proceso, la influencia de la sociedad civil resulta seriamente restringida y lleva a resultados menos favorables y a confrontaciones permanentes respecto al contenido de la agenda de reconstruccion y de democratizacion.

freedoms, and the autonomy and pluralism of civil society organizations will come out as important preconditions for a constructive and lasting role of civil society in conflict resolution and democratic consolidation in Latin America.

II. CIVIL SOCIETY, POLITICS, AND DEMOCRATIC TRANSITIONS IN LATIN AMERICA
Historically, civil society has until recently been weak within the social and political relations in Latin America. During the first century of independent statehood and nation building, Latin American countries were ruled by a succession of caudillos and oligarchic regimes. Particularly true for the period between the 1870s and 1930, a neo-patrimonial conception of the state and politics prevailed. Economic and social elites usually controlled politics, either directly through personal ties, through oligarchic political parties, or through the military. This offered very limited scope for the development of civil society. In fact, virtually the only civil organizations that were active and heeded were the rural and commercial associations of the elites themselves. Violent confrontations was the rule rather than the exception among the non-dominant classes, such that this social problem became a 'police problem' (Kruijt & Koonings, 1999). Nevertheless, civil society organizations (CSOs) started to sprout up, especially in the urban areas where trade, infra-structural development and fledgling industrialization set the stage for the first trade unions (Koonings et al., 1995). Still, the overall position of the elites vis-à-vis popular organizations remained hostile until the 1930s. In a number of countries, notably the Central American countries (with the exception of Costa Rica), Peru, Bolivia, and Paraguay, regimes and politics retained their oligarchic and exclusionary character well into the twentieth century (Hagopian, 1996). As we will see below, this was the prime structural factor behind the eruption of civil war in the 1970s and 1980s in Central America (Flora & Torres-Rivas, 1989;Torres-Rivas, 1991).
From 1930 onwards, social and political relations started to change as a result of urbanization, industrialization and the strengthening of the so-called 'popular sectors', placing on the agenda the question of the incorporation of the 'masses' into Latin American political life. In some countries, notably Argentina, Chile and Uruguay, this incorporation started even earlier (since about 1900), on the basis of slowly expanding democracy and institutional politics. In Mexico, the revolution that started in 1910 heralded the end of classic oligarchic rule but inaugurated a long-lasting period of violence. Elsewhere, the crisis of the 1930s served to hasten political changes that were already in the making. The response of political elites to the challenge of the 'incorporation of the masses' in most cases took the form of populism and corporatism (Dix, 1985). This meant that a new balance was struck between an increasingly vocal civil society, political parties, and the state.
Most analysts agree that the net results of Latin American populism between the 1930s and the 1960s amounted to the reassertion of state control over CSOs. The basic dilemma of populism was that of striking a stable balance between popular incorporation and top-down control in order to avoid radicalization of popular mobilization. Populism, although appealing to the demands and sentiments of the masses, was basically a counter-revolutionary movement aimed at bringing in new, urban-based, modernizing and moderately reform-minded regimes. This meant that the organized popular sector came to play an important role in reformulating the social base of political power, but always under tight political control; a situation that Touraine (1989) has labelled the 'hypertrophy' of the state vis-à-vis CSOs. The state came to act as a kind of institutionalized superpatrón moderating popular political aspirations by allowing for institutional space for CSOs, a certain degree of party-political or corporatist representation, and the amplification of social benefits for the organized segments of the working classes (and in rare cases the peasantry). With respect to civil society, this meant the consolidation of the so-called 'old social movements': trade unions, peasant unions, cooperatives, middle class associations, etc., but always largely dependent on political parties or populist leaders.
After 1960, a number of populist 'experiments' with democracy ran into crisis. In Argentina, Bolivia, Brazil, Chile, and Guatemala, the controlled democracy-cum-reform strategies met with a number of problems. Although it is not the intention here to set out the contours of the 'crisis of populism', it is worth mentioning that one of the ingredients was the growing difficulty populist regimes had in controlling popular organizations. In combination with rising economic problems, this led to a growing inability of the political system to absorb social tensions and political antagonism. In countries of the Southern Cone (Brazil) and Central America, this set the stage for the rise to power of conservative counter-reformist coalitions that feared escalation and radicalization of popular discontent, seen as a threat to the prevailing order, and especially to the privileges of new and old elites alike. These 'counterrevolutionary' coalitions included the military, who in most Latin American countries had been moving away from a relatively pro-populist and pro-reform stance to adopt ideologies of state security to be protected against the internal threat of radical subversion (Perelli, 1990;Rouquié, 1989;Kruijt & Koonings, 1999). During the 1960s and 1970s, in a number of countries (Argentina, Bolivia, Brazil, Chile, El Salvador, Guatemala, Uruguay), strongly politicized military institutions intervened in politics to 'de-activate' popular threats to the prevailing order (O'Donnell, 1973) and to take the lead in long-lasting authoritarian regimes (Collier, 1979).
These regimes were not only marked by authoritarianism but also by violent repression. The 'de-activation' of political opposition and the mobilized popular sectors amounted to waging internal war by the military-led regimes against their own citizenry. In some cases this took the form of 'low intensity' (if there ever was a euphemism!) but always systematic state terrorism (e.g. in Brazil or Chile) or more massive 'dirty wars' (like in Argentina). In Central America, authoritarian repression against the background of social exclusion provoked organized armed opposition erupting into outright civil war in the 1970s and 1980s. Peru has been quite a remarkable exception: counterinsurgency was waged against rural guerrillas in the 1960s by the formally democratic and mildly reformist Belaúnde government. The subsequent military regime was itself committed to an agenda of social reform and controlled popular participation (Kruijt, 1994). As we will discuss later, the end of military rule and the return to democratic politics in 1980 inaugurated a cycle of unprecedented political violence in Peru during which (especially in the 1980s) the military conducted their version of a 'dirty war'.
These conflicts systematically led to the de-activation or even destruction of civil society as it existed under the previous democratic or populist regimes. The doctrine of national security on which repression and warfare was founded made any opposition organization suspect of harboring 'subversives'. National security regimes put into place elaborate pseudo legal arrangements to outlaw political and societal opposition and to facilitate arbitrary repression of political activists, trade unionists, peasant leaders, human rights workers, and even representatives of the progressive Catholic clergy. The 'formal' repressive actions of the military, the police and the intelligence agencies in the counterinsurgency war were complemented by the ruthless operations of 'informal' armed actors like paramilitary units and death squads. In this climate (in the 1960s and 1970s in Brazil and the Southern Cone countries, and in the 1970s and 1980s in Central America) space for CSOs was widely reduced.
Nevertheless, this cycle of state terrorism paved the way for a remarkable resurgence of civil society. Enjoying increasing support from international NGOs and bilateral donors (that often exerted considerable influence), old and new CSOs emerged to protest against the authoritarian regimes. This resurgence of civil society had a threefold medium-and longterm significance. In the first place, it was of paramount importance to denounce the repression and violence unleashed by the authoritarian regimes throughout the region. The Catholic Church, especially its progressive sectors, played an important role in vociferously denouncing human rights violations and economic and social exclusion and political authoritarianism in general. The Church took the initiative in the protest itself, protected civil rights workers and supported many opposition civil organizations (Mainwaring & Wilde, 1989). As a result, during the late 1970s and 1980s, a broad spectrum of CSOs came to the fore. They can be basically distinguished in two types: the grassroots organizations at the local level, representing slum dwellers, workers, peasants, Indians, women; and the more middle class and professional or academics-linked organizations that called for the observation of human rights and social and political reform. In practice, these two types developed multiple connections, e.g. through federations of grassroots organizations, trade union federations, and a growing multitude of NGOs that often combined advocacy work, fund raising, and research activities with coordinating of grassroots and other membership organizations.
Secondly, the rise, recovery or renewal of CSOs contributed importantly to setting the stage for liberalization and democratic transitions (O'Donnell & Schmitter, 1986). In the case of institutional transitions like in Brazil and the Southern Cone countries, CSOs were active in drumming up non-violent opposition against the military dictatorships. In addition, they served as a social base for opposition political parties, providing them with a certain amount of legitimacy (especially in the absence of valid elections) and with ideological and programmatic inputs. The revitalization of CSOs also offered an alternative for left-wing revolutionary movements or armed struggle that had become increasingly unfeasible and unpopular (Angell, 1996;Castañeda, 1994). As a result, in countries such as Argentina, Bolivia, Brazil, Chile, Guatemala and Peru, a broad consensus was built between the late 1970s and early 1990s among social and political opposition forces against military regimes that the restoration of basic civil and political rights was not only an immediate end in itself and as the backdrop for the restoration of competitive and pluralist politics, but also a precondition for the realization of long overdue reforms in the economic, social, juridical and cultural domains.
Finally, these civil organizations formed the basis of a thorough reappreciation of the importance of civil society in the challenge of consolidating democracy in post-authoritarian and post-civil war Latin America. The brutal experience of military dictatorship, repressive violence and civil war had convinced the majority of social and political agents across Latin America of the vital importance of citizenship for the successful consolidation of democratic governance, the rule of law, and greater equity and social justice (Agüero & Stark, 1998;Alcántara & Crespo, 1995;Mainwaring et al., 1992). Civil society was 'discovered' as a key concept and arena for pro-active social and institutional change. Civil society was seen as the terrain for shaping a pluralist society in which non-violence and institutionality would be the prevailing norms for settling disputes and furthering variegated social interests. Civil society was to be the domain where such interests had to be expressed; as such, CSOs served to counter the influence of traditional (particularistic) politics as well as the old vice of state hypertrophy. In addition, civil society was expected to supply checks-andbalances against the legacy of state and non-state violence and against the impact of harsh but apparently unavoidable neoliberal economic adjustments.
In those countries where the stage was set for a so-called 'triple transition', the role of civil society was of particular importance and complexity. Triple transition referred to the fact that a country was not only passing from authoritarianism to democracy and from economic protectionism and statism to liberalized markets, but also from a situation of open civil war to negotiated and official peace. These interlocking transitions offered highly complex and variegated scenarios during the 1980s and 1990s for a selected number of countries: El Salvador, Guatemala, and Nicaragua in Central America, Colombia and Peru in South America, and in a sense also Mexico, after the onset of the Chiapas rebellion in early 1994 (Arnson 1999). It is to El Salvador, Guatemala and Peru that we now turn to look more closely at the cycle of war and peace and at the role of civil society in the efforts at post-war reconstruction.

III. WAR AND PEACE IN THREE COUNTRIES: EL SALVADOR, GUATEMALA AND PERU
These three countries make for an interesting comparison not only because they share a number of commonalties, but also because they show significant differences with respect to the background, nature and timing of the conflict, the peace process, and the subsequent reconstruction agenda. In El Salvador and Guatemala, civil war broke out against the background of long-term social inequality and exclusion, limited social modernization, the entrenchment of classic oligarchic elites, and the establishment of violent military regimes in the 1960s and 1970s. Failures at social reform attempts prior to the outbreak of the war were also part of this background. In both countries, the exhaustion of the armed conflict, the decision of key actors to look for peace, and the task of consolidating a comprehensively democratic order (including not only civil and political but also social and cultural dimensions) were closely intertwined dimensions. Two basic differences between El Salvador and Guatemala can be mentioned: the military balance between the regime and the opposition on the eve of the peace process; and the role of ethnicity in shaping social cleavages, the context for the armed struggle, and demands for rebuilding a post-war society. Peru shared with Guatemala the importance of ethnicity in shaping the conflict and setting the stage for the violence (but not in framing post-war demands). But it differs from both Guatemala and El Salvador in that the civil war broke out only after the full retreat of the military regime and the absence of a negotiated peace process. The nature of the conflict, its 'resolution' and the ensuing political situation in Peru had tremendous consequences for civil society in a way quite different from what happened in the two Central American countries under consideration, as will be discussed in later sections.
The cycle of war and peace in the two Central American countries can be reviewed along parallel lines (see Calvert, 1998). First, it is important to look at the historical background to the war. Social relations marked by profound inequality and cleavages coincided with a political arena that did not allow for the expression of popular demands. Especially since the 1940s, the permanent failure of social and political reforms led to increasing polarization. At the same time, the elites and the politicized military embarked on a course of repression of popular organizations, authoritarianism, and state terror. This, in turn, resulted in the mobilization of armed opposition in the 1970s and early 1980s. The civil wars can be looked at for their principal protagonists, their strategies and support base, as well as for the military outcomes and longerterm political impact.
The peace process in both countries started with key political and military factors impelling protagonists in the conflict to accept and conduct peace negotiations. The negotiation agendas themselves were complex and organized in different stages. Of key importance, was the international involvement of neighboring countries, the UN, a group of 'friend' countries, and even the US as the main supporter of the counterinsurgency campaigns. Finally, we look at the peace agreements, the issue of implementation and compliance, and the general consequences for post-war developments.
As has already been mentioned, Peru will be a different story: the longlasting oligarchic rule with military backing until the early 1960s, together with the failure of populist reform, runs along similar lines as the Central American countries. However, the early defeat of Marxist guerrillas in the 1960s, the subsequent nationalist and reformist military regime, the gradual radicalization of the extreme left in the 1970s and the subsequent eruption of civil war under a civil and democratic regime in the 1980s are uniquely different.

El Salvador
During the 1960s, tensions arising from social inequality and a closed political arena started to come to a head (Biekart, 1999;Van der Borgh, 1999). On the one hand, there was the elite (the infamous 'fourteen families') and the dominant conservative sectors of the military; on the other hand, growing social organizations (linked to peasants, students and teacher unions, the progressive Catholic Church) and rising political opposition (especially the Christian Democratic Party PDC) tried to build up alternatives to the entrenched oligarchic domination. In this endeavor they obtained occasional support from more reform-minded younger officers within the armed forces. During the 1960s and 1970s, these alternatives failed to materialize on four occasions.
First, in 1960, young reform-minded officers staged an aborted coup. The conservative countercoup failed to enlist the support of the new PDC that instead sought electoral advances with the support of civil society. Second and third, clear PDC electoral victories were jeopardized by fraud and military interventions in 1972 and 1977. Finally, in 1979, a second coup by young reform-minded officers (against the violent regime of general Romero who had taken power after the 1977 electoral fraud) was short-lived. The military reformists initially obtained the backing of the PDC and the popular organizations in the formation of a junta, but this junta failed to control the hard-line military and the representatives of the popular organizations and most of the PDC politicians backed out.
In the mean time, the military started to set up a repressive apparatus, backed by the political and financial support of the wealthy elite and the US. During the 1960s, a network of paramilitary units and pro-military informants was set up (ORDEN) to counter the rise of mass organizations in the countryside. From the late 1960s onward, paramilitary forces and the intelligence service stepped up repression and human rights violations. In response, the popular organizations started to build up a capacity for armed resistance. At the end of the 1970s, five so-called political-military organizations were active in different parts of the country, although during the 1970s, their main emphasis was on popular mobilization and non-violent action.
The failure of the reformist junta meant a final push towards open civil war. It became clear to the political opposition and the popular organizations that the political arena offered no room for peaceful change. Manipulation and fraud barred the electoral channel, and even reform-minded military putschists were unable to put a lid on the hard-lined military. Repressive violence was stepped up (resulting among other things in the assassination of the Archbishop Cardinal Romero) and right-wing militants started to set up extremist paramilitary and death squad operations, headed by former army major Roberto D'Aubuisson and supported by wealthy elite families. In response, the five existing military-popular movements united to found the Farabundo Marti Front for National Liberation (FMLN) in October 1980. The FMLN launched an offensive in 1981that led to the consolidation of opposition control over important parts of the country such as parts of the Morazán and Chalatenango provinces. Until the mid-1980s the military dimension of the conflict between the army and the FMLN prevailed; after that, the focus shifted to a more political struggle. The FMLN made efforts to set up and control a number of civil society organizations, especially in the 'liberated territories'. These so-called popular NGOs were united in the CONCERTACIÓN and maintained close links to the FMLN ( Van der Borgh, 1999: 55-56, 78-80).
Another gradual but eventually important process of change was the introduction of civil democracy and electoral politics in the course of the 1980s, precisely at the height of the civil war (Biekart, 1999). While violence and repression continued, and an early peace attempt broke down (1984), elections brought the PDC to the presidency in 1984 and ARENA (originally founded by D'Aubuisson in 1981) in 1989. ARENA president Cristiani decided to embark upon a peace process, representing more moderate conservative sectors rather than the extreme right wing followers of D'Aubuisson and the military. Slow democratic opening together with the apparent military stalemate between the army (heavily supported by the US) and the FMLN prepared the latter for a re-entry of institutional politics towards the end of the 1980s. In 1989, preliminary contacts were made between the government and the FMLN in Mexico. After a second large scale military offensive by the FMLN in that same year, bringing them into the capital city of San Salvador, had confirmed the movement's relative military strength and the existing 'strategic balance' with the army, conditions were ripe for a serious and ultimately successful peace effort.
A number of factors pushed both the Cristiani government and the FMLN to the negotiating table (Arnson, 1999;Biekart, 1999;Cañas & Dada, 1999). The moderate right wanted to end El Salvador's economic pariah status. The government as well as the army started to accept the improbability of a military victory. The FMLN realized the same thing, despite their impressive 1989 offensive display. Political parties had come to put greater trust in the political process. The US government, due to pressure from its Congress, became increasingly unwilling to back the counterinsurgent efforts of the El Salvadorian military, particularly since the Cold War was drawing to a close and, after the brutal murder in 1989 by the army of six Jesuits, their housekeeper and her daughter at the University of Central America. The FMLN in turn moved away from the perspective of establishing a victorious revolutionary regime because the Front had not only developed a greater amount of confidence in the democratization process but also because the prospects for a left-wing government were bleaker: the erosion of Sandinista rule in Nicaragua, the weakening of the Soviet Union, the likely price in terms of US opposition, and, importantly, the lack of popular support for a continuation of the armed struggle that became manifest during the 1989 offensive.
Finally, a very important role was played by international forces (Moreno, 1994;Whitfield, 1999). During the 1980s, the Central American presidents, by then all elected civilians, decided upon a regional peace strategy, known as the Esquipulas II treaty signed in 1987. This agreement called for strengthening civil rule, a cease-fire, peace negotiations, and the end of outside support for the warring parties. In El Salvador, international pressure to negotiate for peace was also exercised by European countries (largely sympathetic to the opposition) and also the US that sided with the ARENA government. Finally the UN took on a pro-active role in stimulating and moderating the negotiation process. In the end, government (with the compliance of the military) and the FMLN discussed a peace agreement in various stages. Negotiations focussed basically upon guaranteeing civil and political freedoms, allowing for the FMLN to change into a political party, transformation of the military away from counterinsurgent state terrorism and brought under civilian control, and reforms of the police and judicial system. In broad terms, agreement was also reached on the re-integration of former combatants (on donated lands), an economic and social reconstruction forum and plan, the setting up of a Truth Commission to investigate human rights violations, and the establishment of a national implementation monitoring commission (COPAZ) in which government and FMLN held equal representation (Córdova Macías, 1996).
These terms were incorporated in the Peace Agreement signed in January 1992 in the Chapultepec palace in Mexico City. Hostilities came to an effective halt. Since then, El Salvador faced the challenge of deepening the peace by consolidating democracy and addressing the country's principal problems in the field of the economy, social welfare, and the rule of law. It would go beyond the scope of this paper to discuss this in detail. According to Cañas & Dada (1999) progress has been rather limited. These authors highlight the continuous efforts by ARENA to manipulate the political process (although FMLN made electoral headway after 1997); the little space left for addressing economic and social reforms (FMLN had agreed to leave this to the post-peace civil and electoral arenas; ARENA adhered to a neoliberal adjustment agenda), and the ongoing problems with respect to judiciary reform and the consolidation of the new National Civil Police (PNC) (see also Córdova Macías, 1996;Popkin, 2000).

Guatemala
The cycle of war and peace in Guatemala bears many resemblances to El Salvador in terms of background and overall rhythm. Guatemala was a country with a huge social and cultural divide between white and ladino (mestizo) elites and middle sectors on the one hand and an impoverished indigenous peasantry (divided over more than 20 linguistic Maya groups) on the other hand. Until the 1940s bonded labour performed by the Mayan peasantry on ladino estates was common. Political rule was exercised by a closed elite, usually through military strongmen. Only between 1944 and 1954 there was a brief interlude of civil rule, aiming to carry out economic and social reforms, ending the practice of bonded labour, making public services available to the rural masses, and allowing for the growth and representation of civil society (Biekart, 1999;Dunkerley, 1991;Kruijt, 1999). This reformoriented period under the presidencies of Arevalo and Arbenz was ended through a CIA-sponsored military intervention that inaugurated more than four decades of authoritarian rule and civil conflict.
Space for opposition politics and civil society in Guatemala was even more reduced, between 1954 and the early 1980s than had been the case in El Salvador. Disgruntled military men and other opponents of the counterrevolution of 1954 took to arms as early as the beginning of the 1960s in the eastern (mainly ladino) parts of the country (Kruijt, 1999). Towards the end of the 1960s, however, this guerrilla effort was largely defeated by a national security state built up to counter the armed threat from the opposition. Although reference is often made to Guatemala's 36 years of civil war, fighting was most intense between 1962and 1972, and again between 1978and 1984(Biekart, 1999Plant, 1999). The latter phase was particularly infamous for its genocidal nature, first under the regime of General Lucas García and brought to a bloody conclusion by his successor general Rios Montt. After the defeat of the eastern guerrillas in the early 1970s new armed opposition groups were formed in the western Maya highlands, especially the Guerrilla Army of the Poor (EGP) and the Revolutionary Organization of the Armed People (ORPA). Although linked to a new radical peasant union (CUC), the Guatemalan armed opposition saw little space for non-violent strategies against the regime and its army. During the prolonged years of struggle, the Guatemalan military came to dominate the country completely, especially in the rural areas. In response, the guerrilla organizations started to build up support among the Mayan communities of the western highlands, especially after the onset of violent counterinsurgency terror by the Lucas García regime after 1977 (Kruijt, 1999).
Paradoxically, the most brutal and genocidal phase of the civil war, conducted under the regime of the protestant fundamentalist Rios Montt who had led a coup d'etat to oust Lucas in 1982, also led to the first openings towards democracy and, later, negotiated peace. Under Rios Montt, the army embarked upon a ruthless campaign of extermination and scorched earth, destroying more than 450 villages, killing ca. 100,000 people. This brought close to 800,000 people under arms in the so-called Civil Self-defense Patrols (PACs) and sent 1.5 million Guatemalans into internal displacement or exile abroad (Kruijt, 1999: 49). Rios Montt and his successor General Mejía Victores achieved a strategic victory over the armed opposition. The latter, uniting in 1982 to form the National Guatemalan Revolutionary Unity (URNG) in the image of the FMLN in El Salvador, appeared to have been making considerable military headway just a few years earlier. By the mid-1980s, the URNG was virtually defeated (although far from dismantled). The military then set out to follow up on their new counterinsurgency strategy ('beans or bullets') by carefully engineering a controlled democratic opening. This was part of a military strategy -based on a new doctrine of 'national stability' -for a 'secure' peace and a 'controlled' democracy under military tutelage but without the violent excesses and the pariah status this had brought to the country (Schirmer, 1998, forthcoming;Rosada-Granados, 1999).
The strategic defeat of the URNG and the controlled democratic opening proved to have important consequences for the prospect for peace. Not unlike in El Salvador, fledgling democracy gradually made room for nonviolent politics and created the starting conditions for peace negotiations. After modifications to the Constitution were made by a multiparty constituent assembly in 1984, elected civilian presidents prevailed (with the abortive selfcoup of Serrano in 1993 as the only interruption). Also, after 1990, the role of the military gradually diminished, as they had to relinquish the full thrust of their project of 'co-governance'. External involvement, from the regional peace initiative of Esquipulas down to a group of 'friendly' nations (that now firmly included the US) and the active role of the UN to moderate the negotiations in various locations was similarly important (Whitfield, 1999).
It is noteworthy to mention some differences between the Guatemalan and El Salvadorian process as well. The decision to accept a peace process on the part of the Guatemalan military hardly depended on outside (that is to say US) pressure. It was rather their strong position within the political arena, the military neutralization of the URNG, and the tenets of the doctrine of national stability that contributed to their acceptance of the notion of a 'strategic peace'. The military made it explicit that a formalized peace was vital for the future of the country regardless of the strength or weakness of the URNG (Kruijt & Van Meurs, 2000). In turn, the URNG tended to follow the example of neighboring FMLN, especially since their military position did not allow them any other option. In fact, a peace process was the best shot the guerrillas had at influencing the future development of the country. But the most remarkable difference within the Guatemalan peace process was the strong and vocal role of civil society. Civil society took a pro-active stance in helping bring about the starting conditions for the peace process and in influencing the agenda (Azpuru, 1999). I return to this issue in the next section.
To conclude this overview of the Guatemalan process, it can be said that the peace agreements both took more time to trash out and appeared to be more comprehensive than in the case of El Salvador. A number of partial agreements were signed prior to the signing of the final peace accords in Guatemala City in December 1996, including the cease-fire, demilitarization of the URNG, constitutional reform, indigenous rights, the Commission for Historical Clarification, the socioeconomic and agrarian question, and police and civil-military relations reform (Azpuru, 1999). Implementation was monitored by the UN Mission for Guatemala (MINUGUA, counterpart to ONUSAL in El Salvador), as well as by an increasingly active and vocal civil society. However, the reform agenda progresses rather slowly, and was even partially halted by the outcome of a 1999 referendum. The political system remained rather volatile (Torres-Rivas, 1996), especially after the failed selfcoup of president Serrano in 1993, and more recently the awkward power sharing between president Portillo and the congressional chairman Rios Montt. Many social conflicts erupted over non-resolved issues and nonpolitical violence is endemic (Azpuru, 1999). At the same time, political pluralism has been consolidated and civil society has been greatly strengthened, especially with regard to the rights and empowerment of the Mayan masses. Peru As I pointed out above, war and peace ran along a quite different course in Peru. Although the entrenchment of oligarchic rule, backed by military until the early 1960s described the situation in El Salvador and Guatemala, subsequent social and political developments were quite unique to Peru (McClintock, 1989;Kruijt, 1994). I will look briefly at the main episodes: the moderately reform oriented civilian government of Belaúnde (1963-68), the military period , the return to democracy and the civil war  and the civilian dictatorship of Fujimori (1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000).
Until the 1960s, the Peruvian political landscape was dominated by the confrontation between the populist Popular Revolutionary Alliance of the Americas (APRA) and conservative elites and the military. After a series of moderator coups to block APRA from electoral victory, the more moderate Belaúnde government set about introducing social reforms. This was a government similar to the Christian Democratic administration in Chile and Venezuela that were inspired by Alliance for Progress approaches to use reform to pre-empt radical social revolution. During this period, three small Marxist guerrilla groups took to arms in the Peruvian highlands, but the army easily defeated them.
The rise of a reform-minded political army in the 1960s was the outstanding characteristic of Peruvian politics at that time (Kruijt, 1994). Military political doctrine started to define security in terms of national development and poverty alleviation, seen as greater threats to the nation than the weak guerrilla movements. Added to this was the perception of ineffectiveness and corruption of the Belaúnde government. Towards 1968, the military made active preparations for a coup that was to inaugurate a period of military-led social transformation. The coup of 1968 brought to power a junta led by army chief general Velasco Alvarado and a military 'revolution' that lasted until 1975 when Velasco was ousted by a more conservative countercoup. The main agenda of the Velasco regime comprised economic nationalism and import substitution (state-led) industrialization, a sweeping land reform, the extension of social policies, a system of workers' co-ownership, and the active but controlled incorporation of mass organizations under the peculiar mobilization bureaucracy SINAMOS (Kruijt, 1994;Stepan, 1978). By and large, the regime was non-violent, although political society and democratic institutions were suspended.
The rapid erosion of the 'revolution' after 1975 under general Morales Bermudez was caused by economic disarray, the de-legitimization of military rule, and an activated civil society that started to protest declining standards of living, economic austerity, and the lack of political participation. The military responded by engineering an explicit pact with their erstwhile adversaries, the APRA party, to embark upon a trajectory of liberalization and return to democracy (Cotler, 1986). This was a brief period of two years (1978)(1979)(1980) during which a new constitution was drafted and preparations were held for free presidential elections in 1980. These elections brought Belaúnde and his Popular Action party back in office.
Tragically, the consolidation of democracy in Peru coincided with the 'lost decade' of the 1980s: economic decline and mismanagement, massive migration of rural poor from the Sierra (highlands) to the coastal cities, dramatic impoverishment, widespread and informalization of the chicha (urban-informal) masses. This process served as the background to one of the most obscure and violent episodes of conflict in recent Latin American history. The first bombing attack of Shining Path (SL) took place in 1980 on election day in the Andean town of Chuchi. SL had emerged from the fragmentation of the left at the regional Andean university in Ayacucho during the 1970s. University professor of philosophy Abimael Gusmán Reynoso and his following of dissatisfied and radicalized students, a part of which come from the Andean highland communities, started to prepare for armed struggle against the 'fascist and imperialist state'. SL developed an ultra-orthodox Maoist doctrine, and started base-work in the highland communities. After the start of the violence, SL mustered initial peasant support because it succeeded in imposing harsh but effective forms of order and justice in communities and zones that felt abandoned by the state (Kruijt, 1999;Palmer, 1992).
SL developed a long-term military strategy to encircle and defeat the state in two to three generations, through 'strategic defense' of its bases, followed by a more offensive phase that had to lead to the blocking and eventual take-over of the cities and the coastal zone. Eventually the conflict lasted almost 14 years, ending in 'strategic defeat' of SL shortly after the capture of its leader, Gusmán, in September 1992. In 1982, after two years of ineffective police countermeasures against the 'bandits and terrorists', the military came in, to conduct a violent dirty war in established emergency zones that led to the militarization of large parts of the Andean highlands. The backlash of this repression paradoxically facilitated SL's expansion from Ayacucho to other highland departments and to the Upper Huallaga valley, where SL took control of the cocaine trade and used the revenues to boost its operational capacities. As a result, already during the late 1980s SL attempted to move towards 'strategic balance' and the encirclement of the cities. SL increased its violent actions in Lima itself (a.o. through 'armed strikes' and the assassination of civil and community leaders) (Basombrío, 1999;Kruijt, 1999).
The final year of Garcia's government (APRA, 1985(APRA, -1990, followed up by the first three years of the Fujimori presidency brought an effective military counter strategy: this entailed a more balanced military approach in the highlands (counterinsurgency with civic action) and most importantly the expansion of rondas campesinas (counterinsurgent peasant militias). This militia was a response and a further contributing factor to the erosion of peasant support for SL. The guerrillas increasingly used arbitrary violence against village leaders and ordinary peasants, SL started prohibiting the market-oriented livelihood of the rural communities (as part of the city blockade strategy), and SL showed an overall disdain for and lack of understanding of Andean culture (Degregori, 1999b;McClintock, 1999). The army came to be seen as the lesser evil by the peasantry and the rondas as the only option for communal survival. This was capitalized upon by the Fujimori government, who regulated and armed the militias and brought them under formal control, although the rondas (or self-defense committees, as they were officially called) maintained considerable degrees of autonomy (Fumerton, 2000).
At the same time the decapitation of SL took place through the arrest of Guzmán in Lima in September 1992 by a DINCOTE (anti-terrorist police) unit. This action was part of the ongoing and successful demolishing of SL cadres, for which testimonies of repentants and trials under military law by anonymous 'faceless' judges were common procedures (McClintock, 1999). This greatly reduced SL's operational capacity, especially after Guzmán, from prison, started to call for 'peace' in 1993 (a turnaround that would be repeated, years later, by the Kurdish leader Öcalan in Turkey).
The consequence of these developments in terms of a peace process was that no peace process ever took place (Basombrío, 1999). Occasional fighting between the military and the rondas on the one hand and isolated SL bands on the other hand continued, however, until the end of the 1990s when the state of siege was finally lifted. Basombrío (1999) points out the main reasons that explain the failure of a formal peace process. SL had no interest at all in coming to terms with the state and to become part of the institutional framework of a democratic Peru. SL was committed to ideological dogmatism and a ruthless strategy of violence and terror to achieve its long-term objectives. In the process SL completely antagonized potential allies in a potential peace process: other left-wing political parties, civil society, and public opinion. When even the rural communities turned against SL, the only perspective the guerrilla had was to draw out its insurgency as long as possible, no longer as a political strategy but as a moral (and material) way of life. As such, SL became part of a spectre of 'armed actors' that were active in certain more remote regions of the country. On the other side, the state and the military were convinced, partly because of the nature of the SL phenomenon, that a solution to the war had to be military or military-cumcivic action (Kruijt, 1999).
The drama, in the case of Peru is that, while the state apparently succeeded in bringing about precisely such a solution, the costs in terms of a stunted democracy and an aborted civil society were tremendous. As we will see later on, the outcome of 'conflict resolution' in Peru was the entrenchment, for almost a decade, of a neo-populist-turned-autocratic regime (Fujimori), backed by a beefed-up intelligence apparatus based on control, blackmail and corruption, and supported by the military that had lost their former nation-building orientation (Kruijt & Pilar Tello, forthcoming).

IV. CIVIL SOCIETY AND CONFLICT RESOLUTION: PEACE PROCESS AND RECONSTRUCTION
In theory, civil society (leaving aside, for the moment, the issue of the heterogeneity of civil society) has an important role to play in conflict resolution, especially in Latin America. As Arnson (1999: 6-7) rightly observes, the conventional literature on democratic transition and consolidation in the region tends to ignore the specificities of war situations and democratization and peace building as explicitly linked endeavors. One could argue that the role of civil society in this context is at the same time more important and more complex. 'Conflict resolution' here is taken to include the process of peace building and post-war reconstruction. Peace building not only means putting an end to armed conflict (whether or not through formal negotiations and agreements) but also to address some of the principal political, economic, social, and ethnic imbalances that led to conflict in the first place (Arnson, 1999: 1). The process of post-war reconstruction entails, following upon this definition, not only the effort of rebuilding the material and institutional framework of the war-torn society but also to achieve progress in key reforms deemed necessary to do away with these root causes of conflict and to consolidate a democratic, non-violent, lawful and more just society. So, peace-building and reconstruction and democratic consolidation presuppose each other, at least in the Latin American cases, sometimes with relative success (as in El Salvador and Guatemala), sometimes with relative failure (as in Peru).
The starting point from a civil society perspective for this comprehensive challenge of conflict resolution is that civil society may even have been more seriously damaged as a result of the war than in the case of institutional authoritarianism and peaceful democratic transitions (Kruijt & Koonings, 1999). So it will depend on specific circumstances whether or not civil society in war-torn societies can play a comparable role in denouncing the status quo (i.e. open civil war) and in helping set the stage for a peace process. In particular, progress in the field of civil and political liberalization will have to be made so that civil society can be re-established prior to the start of the peace process. In addition, ongoing violence has to be sufficiently reduced or demarcated so that it does not harm the build-up of civil society. If these conditions are fulfilled, civil society can play a trigger role in demanding an end to the violence and to urging the combatant sides into negotiating. Civil society, with political parties, can help to strengthen legitimacy and support for the peace process. Civil society can influence or inform the negotiating sides and try to set part of the agenda. Civil society can provide a forum for initial stages or certain parts of the peace negotiations.
Civil society can endorse the outcomes (the agreements) and monitor and support their implementation. A key cross-sectional contribution of civil society is to supply an environment of social and political trust, in view of the legacy of antagonism and the fragility of the political and institutional framework. The principle of 'civility' or non-violence in settling disputes that have for a long time been marked by the logic of violence can be crucial for long term changes in the domain of political institutions and political culture.
With respect to the process of post-war reconstruction, it is useful to look at three specific domains in which civil society can play a key role: (1) the monitoring of the implementation of the peace agreements and the related reform agenda (e.g. security forces, rule of law, human rights, economic and social reforms); (2) strengthening democratic governance, along lines broadly parallel to the democratization agenda in the 'peaceful transitions' (channeling citizens' demands, securing transparency and accountability, addressing the exclusion of specific categories, such as ethnic groups); (3) co-organizing and implementing (parts of the) agenda for physical, economic and institutional reconstruction. Against this backdrop, in the following subsections I will briefly look at the role of civil society in conflict resolution (i.e. peace building and post war reconstruction) in El Salvador, Guatemala and Peru. Rather than trying to present a comprehensive overview for each of these three countries, I will focus upon what may be seen as the most salient feature of each case.

El Salvador: Political Polarization and the Management of Aid
The key feature of the peace process and of post-war reconstruction in El Salvador has been the polarization in the political arena. This has been a continuation of the wartime confrontation between the state (particularly the ARENA government) and the FMLN and by the fact that the peace process was first and foremost decided upon by these two political-military protagonists on the basis of their perception of the military gridlock and outside pressure. By and large, civil society was subordinated to this process, although during the 1980s both political sides made efforts to rebuild a civil society support structure but they were to a considerable extent based upon channeling political support to either the government or the FMLN or to channeling external resources to specific reconstruction programs, such as macro-economic assistance, food aid, or the resettlement of refugees made possible after the Equipulas II agreement. CSOs were in most cases closely linked to (or dependent upon) either the state or the opposition; at the same time an increasing number of 'private public' development organizations -NGOs or quasi-NGOs sprang up. (Van der Borgh, 1999;Biekart, 1999).
Civil society was largely kept at the margin of the peace negotiations, in which the FMLN assumed the role of reform party and the UN took care of some of the agenda breakthroughs without consulting civil society either. Also in the case of the Truth Commission human rights NGOs were relatively marginal to its proceedings and subsequently eclipsed by the activities of ONUSAL (Popkins, 2000: 192-3). The feeble progress in post-war reforms and democratic consolidations may to a large extent be attributed to the ongoing confrontation between ARENA and FMLN at the political stage. ARENA strove at securing its hold on power, while pressing forward a neoliberal economic agenda, while the FMLN had to come to terms with internal splits and with regaining its political position after the adverse 1994 elections. 1 As a result, civil society in El Salvador appears fragmented and dependent upon the main political forces. A second and very important factor that determines the role of civil society in El Salvador has been the massive flow of external financial assistance to the reconstruction process after 1992 (and even before). As is analyzed in great detail by Van der Borgh (1999), these flows have led to three related tendencies: tight interlocking of many non-public organizations as 'developmental NGOs' to political parties, the state apparatus, and the international donor community (especially the US, Western European countries and the UN system); their focus on resource access and program implementation and the related 'donor dependency' of the CSOs; the fragmentation at the local level where a multitude of private development-oriented agents tend to dominate the mobilization and institutional build-up of civil society and decentralized governance.

Guatemala: Civil and Ethnic Mobilization for Political Participation and Against Social Exclusion
The situation in Guatemala has been quite different: civil society is quite active and growing, pluralist and independent, as well as innovative. CSOs has played an important role in triggering the peace process, offering platforms and being a partner in the process. Neither the military nor the ruling party nor the URNG were in a position to subordinate or control civil society. As a result, CSOs have developed into major stakeholders in the implementation of the peace agreements and the deepening of democratic governance through expanding voice and strengthening peaceful consensus (Biekart, 1999).
After the constitutional reform of 1985 and the return to civil rule, one year later, space for the independent organization of civil society increased considerably. The National Reconciliation Commission (CRN), set up as a result of the Esquipulas II accord played an important role in bringing civil society to the forefront of the peace process (Arzupu, 1999). The CRN, chaired by Archbishop Quezada, started a National Dialogue; at the same time a broad spectrum of CSOs set the first stage for negotiating with the URNG. This greatly contributed to the conviction within the latter that peace should be pursued and also prepared the grounds for direct negotiations between the government (through the Peace Commission COPAZ) and the URNG. Civil society aided in blocking Serrano's self-coup in 1993 and in installing human rights ombudsman De León Carpio as the new president so that the peace process would not be jeopardized. The role of the UN as mediator and monitoring agency has often been decisive, especially after the start of the activities of MINUGUA in 1994.
A major step has been, in the middle of the peace process, the formation of the Civil Society Assembly (ASC) in 1994 (Azpuru, 1999;Kznaric, 1999;Palencia Prado, 1996). The ASC aimed at uniting civil society organizations of all kinds to work out inputs that could be presented to the negotiations. The ASC was formed as an explicit part of the 1994 framework agreement between the government and the URNG that was to regulate the final stages of the peace process. As is pointed out by Kznaric (1999:6ff.) the ASC was responsible for helping formulate key agreements and related reform agendas on. the indigenous population, economic and land issues, the role of the military and the police, repatriation of refugees, etc. The significance of ASC proved to be more lasting since it not only furthered coordination and trust among the many CSOs that had sprung up in the country, but also supported the formation of a new center-left political party (the New Guatemala Democratic Front FDNG) during and beyond the 1995 elections. Kznaric (Ibid.) explains the success of the ASC through its ability to write and submit consensus documents on important reform issues, to engineer consensus, to relate to public opinion, and to exert an autonomous influence on the URNG. In the process, the ASC enjoyed support and protection from the UN and the donors within the group or 'friend countries'.
Perhaps the most remarkable aspect related to the role of civil society in the Guatemalan peace and democratization process has been the rise of numerous indigenous organizations. These Mayan CSOs and NGOs were of extreme importance, not only to secure the inclusion of an agreement on the rights of the indigenous peoples in the peace and reform agenda, but also and particularly to break with the century-long trauma of subordination, exploitation and exclusion of the Maya communities. The peace agreement recognized the rights of the indigenous population in Guatemala as a pluriethnic and multi-cultural nation. This mobilization started with the increasing dissatisfaction among the Maya communities and leaders with respect to the URNG and the war in general, as not being in the interest of the indigenous communities while exposing them to a violent counterinsurgency campaign (Plant, 1999). After 1985, a growing number of indigenous NGOs and CSOs was set up to address a wide variety of issues: human rights, demilitarization, the problems of war widows, refugees, land issues, language and culture, etc.
Gradually, an over-arching Maya movement appeared that obtained substance in a number of umbrella CSOs that were set up from the early 1990s onward (Adams 1994;Warren 1998). These organizations such as the Civil Movement for Mayan Unity and Consensus (IUCM), the Council of Mayan Organizations of Guatemala (COMG), and the Coordination of Organizations of the Mayan People of Guatemala (COPMAGUA) were active in drawing proposals through the ASC for the peace negotiations. The ensuing indigenous agreement covered a wide range of Mayan rights issues such as identity, nondiscrimination, cultural autonomy, local indigenous governance, land rights, ethno-development, and so on (Palenzuela, 1999;Plant, 1999). Since the signing of the peace, Mayan CSOs have continued to be active in monitoring and co-implementing reforms in these fields through the participation in joint commissions and through autonomous action in the field of language, education, identity-related research, and strengthening decentralization and indigenous local governance (Plant 1998). In recent years, donor dependency has weakened the role of some of these NGOs, but new dynamics at the local level, where non-partisan Civic Committees have been elected in many communities to facilitate community-level Mayan empowerment, have been confirming the relative vigour of Guatemalan civil society alongside the unstable political arena. Now, federations of civic committees, inspired by vocal Mayan leaders such as Rigoberto Quemé, the mayor of Quetzaltenango, are acquiring national significance. Despite the complexities of the indigenous question in Guatemala and setbacks such as the rejection by referendum in 1999 of the reform process, a major historical transformation is underway largely due to the active mobilization of civil society in the peace and democratization process.

Peru: Political Violence and the Breakdown of Civil Society
If Guatemala offers reasons for optimism, and El Salvador for mixed feelings, Peru should cause dismay as to the possibilities and prospects for civil society in conflict resolution in Latin America. I already pointed at the 'no-peace' format of conflict resolution in Peru. This in itself precludes a lot of space for civil society to play a role along similar lines as happened in the two Central American countries. Dealing with the problem posed by SL meant first and foremost choosing military options of which the state maintained strict control. But it is worse: the general social and political context both during and after the conflict posed quite adverse conditions for a significant and constructive role for CSOs in Peru. Ultimately, this would lead to the breakdown of civil society under conditions of ongoing militarization, social fragmentation, political violence, and regime perversion. I will briefly substantiate these points below (based mainly on Basombrío, 1999;Fumerton, 2000;Kruijt, 1999;Kruijt & Tello forthcoming, McClintock, 1999).
In the first place, as I already mentioned, the strategy adopted by both the state and the military and SL in the conflict precluded an active role for civil society. The state nurtured its military options, in a later stage supplemented with civic action developed in close linkage to the peasant communities and the rondas campesinas. SL, in turn, would have nothing to do with CSOs or NGOs; these were made targets of terrorist attacks and assassination since SL saw them as covers for the state and also as a sector that could not be controlled. Targeting CSOs would contribute to the erosion of the institutional order and would further enhance pre-Revolutionary conditions. As a result, not only did civil society have few reference points with respect to the protagonists of the conflict, but the possibilities for the functioning of CSOs were becoming increasingly difficult.
Secondly, the disastrous performance, in economic and social terms, during the Belaúnde and García governments between 1980 and 1990 led to the overall disillusionment with respect to politics and political actors. Civil society proved unable to counter this trend, partly because synergetic relations with the party system were jeopardized, but also because of the extremely fluid and fragmented nature of social transformations during the 1980s. These are marked by impoverishment, informalization, and a flood of cholo migrants from the highlands to the towns due to decline in livelihood opportunities and social provision and also as a result of the spreading violence in the emergency zones.
Thirdly, in the rural communities, the build-up of grass roots civil structures was severely thwarted because of the breakdown of civil authority, the vanishing of developmental activities by the state and NGOs, militarization, violence, and the rise of the rondas. In a broader sense, open and endemic violence related to guerrilla actions, drug trafficking, paramilitary and vigilante tactics, and common crime contributed to an 'uncivil society' in the urban areas as well as in the highlands and Amazon regions, in which CSOs found it increasingly difficult to operate.
In the fourth place, Peru, as somewhat an exception among the Latin American countries with large indigenous populations, did not experience a rise of social movements and civil organizations on the basis of ethnicity and indigenous ethnicity. As is pointed out by Degregori (1999a) this can be explained by the traditional disdain towards the nativos, but more importantly by the efforts of various actors (the state, parties, rural movements, guerrillas) to address the rural communities in class rather than ethnic terms. This has been reinforced by the practice of the rural population itself to develop economic and social ties with the urban sector and the market economy, up to the point of massive rural-urban migration. This phenomenon tended to nurture a cholo-mestizo identity of Peruanidad rather than ethnic autonomy within a pluri-ethnic framework. The violence and erosion of civil society during the 1980s and 1990s further hampered the establishment of an indigenous civil society.
Finally, the breakdown of civil society was dealt a coup de grace by the strategy of the Fujimori regime, especially since his successful self-coup of April 1992. After that event, the neo-populism of Fujimori based on a direct appeal to the urban and rural masses gradually gave way to an authoritarian system that was based on manipulation, intimidation and violence (Basombrío, 1999;Kruijt & Tello, forthcoming). Through his right-hand, Secret Service (SIN) Chief, Montenisos, and by forcing the military into conformity and compliance, Fujimori manipulated the political process, neutralized its opponents and exerted effective control over the media, the judiciary, and civil society. In fact, Fujimori mistrusted civil society organizations and his government made efforts to hinder, control or outlaw them (by placing SIN informants in their structures). Fujimori preferred to rely on direct plebiscite and assistant-type relation with the popular masses, e.g. through the public social and infra structural program FONCODES or through granting resources directly to the peasant communities that are trying to transform their rondas into 'development committees (Fumerton, 2000). This led to the erosion of civil society and the political disarticulation of the Peruvian populace.
As a result, not only has no formal peace been reached in Peru (and has political violence not been eradicated) but the conditions for a reform and reconstruction agenda seem quite bleak, despite the recent fall of Fujimori.

V. CONCLUSION
In this paper I tried to compare the conditions of conflict, the process of peace and reconstruction, and the role of civil society in three Latin American countries that experienced open and brutal civil war and political violence during the 1970s, 1980s and 1990s. Although general political analysis of Latin American democratic transitions point at the importance of a revitalized civil society in the breakdown of authoritarianism and the transition to democratic governance, the context of open civil war and widespread political violence creates specific problems for civil society. Nevertheless, depending on circumstances, civil society may play an important role in bringing about an end to violent conflict, through peace negotiations and settlements, and in the subsequent agenda of reform and democratic consolidation.
A comparison of El Salvador, Guatemala and Peru points to a number of general findings. The abandonment of a military and repressive logic within the state and state-related political actors is a first condition for a successful peace process and an active role of civil society. Experimenting with nonviolent procedures, civil liberties, and electoral processes facilitating orderly and legitimate regime changes can set the stage for the acceptance of peace as a best option for those involved in the conflict. Civil society will need such a context in order to develop and to maneuver in favor of a peace process. However, if the protagonists of the conflict manage to monopolize the peace process, the role of civil society will not only become subordinated in a general sense, but this is likely to limit the capacity of civil society to influence the peace process and its implementation. This may in turn lead to a less-than-optimal outcome and to ongoing political confrontations about the agenda of reconstruction and democratization. El Salvador offers examples of the latter situation, in contrast to Guatemala where despite the far more gruesome nature of the conflict and the even deeper entrenchment of social exclusion, the relative autonomy of civil society (since neither the military, nor the state, nor the guerrilla were able to subordinate or control the CSOs) allowed them to play a substantial role in the peace process, contributing to the agreements and the reform agenda, and thus occupying a lasting position with respect to the challenges of democratic consolidation.
In contrast, the Peruvian case shows that even the early return to civil democracy may not be sufficient to secure peace, even when during the preceding authoritarian period political violence was not a major problem. The spiral of violence, in concert with economic decline and social fragmentation that hit the country after 1980 left no margins for a constructive role of civil society and, in fact, contributed to its virtual collapse. Instead, the state embarked upon a military 'solution' to the conflict that may have been successful in strategic terms but at a huge political cost. Political and civil society fell prey to a regime that managed to degenerate public life in less than a decade. With the fall of Fujimori and Montesinos having become a fugitive, Peru faces the challenge of reconstruction all over again. An important part of that will be the undoing of uncivil society and the re-invention of institutional politics.