The limits of peace in Latin America

ABSTRACT Latin American governments frequently emphasise the democratic and peaceful order in the region. These claims are based on two developments: First, except for Cuba, the region has experienced processes of democratisation since the early 1980s. Second, since the 1990s, a series of long-lasting civil wars have ended with negotiated settlements and without a relapse into war. Based on such a superficial analysis, Latin America can be perceived as a successful example of the liberal peacebuilding model. However, although Latin America has democratised and ended wars, it is still the most violent region in the world. This article argues that democratisation and peacebuilding focussed rather on formal changes than on dealing with the structural problems reproducing different manifestations of violence. A focus on the interaction between both processes provides evidence for the possibilities as well as the limitations of change.

However, the empirical evidence on Latin America's peacefulness is mixed at best. The region shows low levels of interstate war, but militarised interstate disputes are common. 2 Examples abound: In 1978, the military regimes in Argentina and Chile were on the edge of war over the control of the Beagle channel; in the 1980s, Central American internal conflicts came close to escalating into a regional war; and currently, the border between Venezuela and Colombia is the site of a significant amount of sabrerattling.
A focus on intrastate or civil war provides an even darker picture as the twentieth century was characterised by a large number of very violent conflicts in most Latin American countries. 3 The Mexican Revolution (1910)(1911)(1912)(1913)(1914)(1915)(1916)(1917)(1918)(1919)(1920) and the Colombian 'Violencia' (1948)(1949)(1950)(1951)(1952)(1953)(1954)(1955)(1956)(1957) are just the most deadly examples. 4 While most civil wars and armed conflicts have ended, 26 of the 54 countries classified as 'dangerous places' by SIPRI (Stockholm International Peace Research Institute) are in Latin America and the Caribbean due to the high number of violent deaths there (including homicides and battle-related deaths). 5 In the twenty-first century, three forms of violence have so far been dominant in Latin America: First, there has been an upsurge of violence mostly classified as 'criminal'. 6 Second, repressive state policies and the militarisation of public security have emerged in countries with and without a prior history of civil war. 7 Last but not least, Latin America has stood out as a region with high levels of selective political violence against human rights defenders, representatives of social movements and independent journalists, among others. 8 The murders of city councillor, Marielle Franco, in Rio de Janeiro (2018) and human rights and environmental activist, Berta Cáceres, in Honduras (2016) are just the most prominent cases. While 'criminal' violence dominates the public discourse, repression by the state and selective political violence are mostly neglected or justified. 9 A similar contradiction between official claims and empirical reality can be observed regarding democracy. Latin American experiences shaped the debate on the third wave of democratisation. 10 Starting with Ecuador in 1978, Latin American countries (except Cuba) have introduced competitive elections, increasing the possibilities of political participation for formally excluded groups from the left. In 2001, the OAS adopted the American Democracy Charter, the 28 articles of which oblige the member states to respect democracy and human rights and strengthen existing OAS instruments for actively defending democracy. 11 While a series of democracy's deficits such as weak institutions 12 and high levels of corruption 13 have been highlighted, the relationship between democracy and peace has not been debated much. 14 This article argues that the minimalist concepts of peace and democracy in Latin America are caused by and contribute to the neglect of the structural drivers of violence and authoritarianism. Framing the present as peaceful and democratic allows states to criminalise actors advocating fundamental change. However, neither the termination nor the absence of organised political violence in the form of war leads to violence reduction but to shifting patterns of violence which ultimately reproduce the status quo.
All social orders need to develop formal and informal institutions to control and limit violence. 15 The possibilities for the prevention and containment of violence vary according to the context and may include the rule of law, control or repression. At least theoretically, the process of democratisation should mean that the rule of law and its application to all actors become more important. But the related reforms, such as the creation of an independent judiciary or a democratically accountable police force, do not happen from scratch; they interact with existing structures and behaviours shaped by history, culture and the experience of war and widespread violence. 16 A focus on the link between the processes of democratisation and peacebuilding in Latin America shows that reforms of the institutions crucial for democratic violence control have been limited and mostly path dependent. As a consequence, security policies have been repressive, undermining civil and political rights and thus democratisation. The article proceeds as follows: The next section identifies the relevant factors linking democratisation and peacebuilding. The subsequent section assesses Latin American experiences, as well as how they have varied. Across the region, we can observe a mix of liberal, hybrid and authoritarian policies of violence control. The final section inserts Latin American experiences into the more general debate on peacebuilding, thereby contributing to a better understanding of the problems and challenges international peacebuilding strategies face.

The links between peacebuilding and democratisation
The debates on democratisation and peacebuilding have obvious overlaps. 17 Theoretically, democracy can be seen as a method of civil and constructive conflict transformation. 18 The use of violence should be limited to specific exceptions, such as the cases of self-defence or where democratic institutions are allowed to exercise it. 19 The democratic peace debate echoes these claims, arguing that consolidated democracies do not go to war with each other and are able to prevent and control violence inside their borders by democratic means. 20 The idea of a liberal peace draws on the experiences in Western Europe and the establishment of a central state with a legitimate monopoly on violence. 21 The state is responsible for controlling violence via institutions such as the military (external security) and the police (public security), and for the sanctioning of the illegal use of violence through the judicial system. In these countries, this process of state formation historically decreased war and armed conflict 22 as well as other forms of violence such as homicide. 23 Since the United Nations adopted the Agenda for Peace 24 in 1992, the main focus of the international peacebuilding debate has been the so-called liberal paradigm. 25 The main elements of liberal peace are negotiated forms of war termination, 26 postwar democratisation 27 and state-building. 28 Pointing to a lack of empirical evidence, the critique of the liberal peacebuilding strategy has focused on the (im)possibility of reproducing this model in non-Western historical and cultural contexts and under the current globalised conditions. 29 The debate on the lack of 'local ownership' has provided interesting insights into the problems of external intervention, such as a lack of recognition of local orders, of the role of local authorities and traditions and of their relevance for peacebuilding. 30 Due to limited empirical evidence of a successful shortcut to the Western liberal peace in current post-war societies, many authors highlight the hybrid character of post-war orders 31 or the need for 'adaptive' or 'pragmatic' approaches. 32 Others emphasise the contradictions and challenges of democratisation in the aftermath of war. 33 With a focus on 'illiberal' and 'authoritarian' post-war contexts, some authors claim that liberal peacebuilding came to an end in the twenty-first century. 34 In these cases, Sri Lanka and Central Asia among others, wars were ended via military victory and war recurrence has been prevented by authoritarian politics of control. Here the main mechanisms of violence control are related to discursive practices, practices of spatial control and a political economy of corruption and clientelism. In the short term, this approach might prevent war recurrence but it is rather unstable in the mid-to-long term as it produces new conflicts. Despite these variations, international actors such as the United Nations continue to promote inclusive, just and peaceful societies as one of the 17 Sustainable Development Goals (no. 16). 35 These debates have introduced new perspectives on peacebuilding dynamics. First, they highlight that peace means many different things to different people. 36 Second, peacebuilding is a complex process and external interventions might be part of the problem rather than part of the solution. Last not least, there is no 'one-size-fits-all' strategy and the outcome of peacebuilding is shaped by the interaction between local conditions and global contexts as well as the agency of a variety of actors. While this is important, the focus on peacebuilding classified by adjectives (liberal, authoritarian, hybrid and local) runs the risk of creating new silos and complicates comparisons across these contexts. Processes of peacebuilding and democratisation are historically openthat is, they are neither linear nor is their outcome necessarily the same as that in Western Europe. Based on this assumption, the analysis of peacebuilding in specific contexts needs to take into account the whole range of possibilities from liberal to authoritarian and hybrid strategies. Analysing peacebuilding and its outcomes along a spectrum has several advantages: First, this approach is not normative and does not predict a specific outcome. This is important as we can observe a mix of various elements in specific contexts. At the same time, the spectrum is a useful tool for comparative assessment and the explanation of variations in outcomes.
Second, the idea of a spectrum also broadens our concept of peace beyond the absence of war, putting the reduction of different forms of direct physical violence (e.g. state repression and homicide) at the core of peacebuilding. While this adheres to Galtung's 'negative peace' concept, 37 it makes the comparison between post-war societies and other societies with different levels of violence possible. 38 We can thus include a peace perspective on transformation processes beyond transformations out of war.
Third, the idea of a spectrum acknowledges the fact that a clear-cut distinction between 'liberal' and 'illiberal' or 'war' and 'peace' is often not possible. 39 Without assuming a temporal sequence, it is quite obvious that the factors important for war onset or termination can differ from those important for non-war-recurrence and 'quality peace'. 40 Fourth, as an analytical tool, the spectrum enables us to assess the dynamics and the direction of peacebuilding processes over time and beyond transitions from war and armed conflict.
Starting from these assumptions, Latin American countries are very interesting cases for studying the interaction of democratisation and peacebuilding. With the exception of Cuba, Latin American political systems are (or were until recently) rated as democratic. They hold regular free and fair elections to select government; provide a minimum of cheques and balances between the executive, the legislative and the judiciary and guaranteeat least on paperfundamental civil and political rights. 41 Although there is growing debate on possible reversals regarding democracy, 42 compared to other regions of the Global South, Latin America is still perceived as rather successful. With the end of the region's longest internal war in Colombia in 2016, 43 the cycle of internal wars might have ended. Latin America shows not only high levels of variation but also some commonalities between those countries that have experienced war and those that have not: on the surface, Latin American experiences seem to mirror a liberal peace, as most countries have remained inside the electoral democracy frame and wars have ended. At the same time, there is evidence across the region of a mix of liberal, hybrid and illiberal policies to control violence. The following section analyses this variation.

Varieties of peacebuilding in Latin America
Peacebuilding and democratisation are complex and non-linear processes. The debates on liberal, hybrid and illiberal peace outlined above identify changes in the political system, the economy and societal relations that are relevant for the prevention and control of violence. What are the core elements discussed in these approaches?
The liberal peacebuilding approaches emphasise the importance of competitive elections 44 as a major element to increase political participation, while executive restraints 45 are important between elections. The rule of law and accountability for all actors, including elites, 46 should also contribute to liberal peace. In the economy, at least some intention to work towards the reduction of existing inequalities must exist as democratic regimes are based on the promise of equality. 47 Authoritarian conflict management 48 is not just based on the absence of these 'liberal' conditions or the dominance of 'illiberal' elements but also relies on a set of specific control strategies including state propaganda and the control of information, military and civilian modes of controlling space, and the hierarchical distribution of economic resources for specific constituencies.
The following Table 1 illustrates the variations in some of these conditions for peacebuilding across Latin America with a snapshot for the year 2018 based on comparative quantitative data. 49 A more in-depth study would need to look at time series and changes in time, but this goes beyond the scope of this article. As I do not discuss causality but want to illustrate the variation, the following Table 1 includes data on elections, the rule of law, welfare regimes, corruption and inequality, as well as on the perception of security and the levels of lethal violence.  49 These data come from mainstream liberal organizations and institutions as we lack alternative data at the regional level. The analysis of authoritarian mechanisms is usually based on case studies with qualitative data.
objective levels but illustrate the comparison of higher or lower scores within this sample. An interesting picture emerges illustrating the variations of peace and democracy: • The most peaceful and democratic countries, Chile, Uruguay and Costa Rica, seemed to be able to change their development path despite dictatorship (Chile 1973-1990and Uruguay 1973-1985 and civil war (Costa Rica 1948). Today, they not only conduct free and fair elections but also receive high ratings for the rule of law and have the most progressive welfare regimes. Levels of corruption and violence are low. Uruguay, moreover, is a country with low inequality in the Latin American context. These three countries seem to resemble patterns highlighted currently in the liberal peacebuilding approach. What is astonishing is that despite comparatively low levels of violence in all three countries, perceptions of danger regarding personal safety are highly prevalent. • Nicaragua exhibits some patterns of successful authoritarian conflict management as there are high levels of corruption and the democratic governance scores are low, but until 2017, violence has not been a major issue. In 2018, however, at least 300 persons were killed and 2000 injured when government forces repressed antigovernment demonstrations. 50 • Another more or less consistent hybrid pattern is related to the most violent societies, Venezuela, Colombia, Brazil, Mexico, Guatemala, Honduras and El Salvador. Here, we can find elements of both the liberal as well as the authoritarian approaches. Elections are more or less fair and free, but they coexist alongside high levels of impunity and clientele politics, corruption and discourses of fear. An interesting outlier is the comparatively low level of inequality in El Salvador showing that there is no direct causal relation between inequality and violence. • The other Latin American countries carry out relatively fair and free elections but don't have equivalent levels of rule of law and welfare regimes. They also have high scores for corruption, inequality, insecurity and violence. • In relation to the historical experience of authoritarianism versus civil war or armed conflict, no clear pattern exists.
How can we explain this variation? At the crossroads between peacebuilding and democratisation, there are two commonalities across the region: negotiated transitions and the neoliberal development model. Theoretically, negotiated transitions out of war and authoritarianism can be seen as a step towards the development of a consensus on the future. The debate on the influence of the mode of transition on peace and democracy is inconclusive. 51 A recent DfID (Department for International Development) study 52 on elite bargains emphasises the importance of the underlying power relations for different outcomes. Studies based on historical institutionalism show that reforms do not emerge from scratch but are shaped by prior experiences and societal contexts. 53 Latin America's transitions from authoritarian to democratic systems, as well as its out-of-war transitions (except for Peru's), were based on elite pacts. While the ambitions outlined in the peace agreements in El Salvador and Guatemala were high, the reforms were layered and gradual rather than profound. Impunity is a good example of the effect of elite pacts on the rule of law in both transitions. Many Latin American societies made huge efforts to document the gross human rights violations that took place during the wars or the authoritarian regimes, through the establishment of truth commissions. But holding the perpetrators accountable was a difficult process as many countries passed amnesties. 54 As a consequence, high levels of personnel continuity persist within the state's security institutions (military, police and judiciary). Even where peace agreements included provisions on the vetting of military officers, as in Guatemala or El Salvador, change was slow and came only after significant external pressure. 55 Across the region, judicial institutions are weak and subject to political influence and corruption; problems of access to justice persist. 56 A second important condition across the region was the dominance of the neoliberal development model during the transition to democracy. Across the region, this reduced the already meagre capacities of Latin American states for social redistribution and the reduction of inequality, although specific results in relation to welfare systems vary according to historic trajectories. In 1994, the United Nations' mediator in El Salvador, Alvaro de Soto, and economist, Graciana del Castillo, warned against the negative consequences of neoliberal policies for peace. 57 Despite advances in poverty reduction and the resource boom of this century's first decade, Latin America still is the most unequal region of the Global South. 58 Seen from the perspective of both developments, path dependency prevailed over the impetus for change and traditional elites were able to maintain or modernise their power. 59 Both developments directly affected peacebuilding: Inequality is a major driver of criminal and social violence 60 ; elite pacts supported impunity and contributed to the militarisation of public security across the region. 61 Sending the military to the streets not only to combat crime but also to crush social protest violates fundamental political and civil rights. In mid-2018, this was even true for countries that seemed to have reformed their security institutions successfullyfor example, Nicaragua, where the police squashed student protests, or Argentina, where the government planned to send the military to the street against G20 protesters. 62 Violence on the part of state security institutions is highly consequential for the democratisation process as it produces cycles of violence and repression that undermine democracy. 63 Selective political violence targeting representatives of social movements, human rights defenders, independent judges, attorneys and journalists is a consequence. Figure 1 illustrates the increase in murdered human rights activists documented in the annual reports of Amnesty International and Human Rights Watch between 2007 and 2017.
Although there are significant differences in these developments across the region, they are mostly in degree rather than substance. Some problems are more profound in countries with a history of civil war and armed conflict, where the promises of war termination rarely translated into a peace beyond the absence of war. The peace agreements in El Salvador and Guatemala, for example, did help to open up participation in the political system. In El Salvador, candidates from the former guerrilla FMLN (Frente Martí para la Liberación Nacional) were even elected to the office of president in 2009 and 2014. However, the peace processes and democratisation were unable to change the patterns of social marginalisation, especially regarding poor people, youth and indigenous groups. 65 As part of Latin America's minimalist understanding of peace, governments highlight the peacefulness and democratic character of politics in the region. CELAC, the OAS and Uruguay's ambassador to the United Nations are just a few examples. Seen from a government perspective, hijacking the discourse of peace and democracy serves a double purpose: First, it helps criminalise social and political opponents (and their actions) who are not limited to formal institutions such as elections or who contest the social status quo. At the same time, political elites tend to downplay the political influence of criminal organisations despite calling for '"tough on crime" or "law and order" policies'. 66 Latin America provides extensive evidence of the related processes of escalation of repression and violence by state and non-state actors.
Second, the discourses are an instrument to keep external intervention regarding democracy and peacebuilding to a minimum. Current examples abound: Venezuela and its allies have rejected a debate within the OAS on the increasing authoritarianism in the country. On 31 August 2018, Nicaragua withdrew its invitation to the Office of the United Nations High Representative for Human Rights to investigate the violence between the police and protesters. And last not least, Guatemala's president has tried to get rid of the United Nations mission to combat impunity and corruption (Comisión Internacional  64 Contra la Impunidad en Guatemala, CICIG). In all three cases, the presidents and their governments are themselves the subject of investigations related to the violation of national as well as international democratic norms through violence and corruption.
In a nutshell, Latin American experiences appear to provide evidence of the limits of gradual change and the persistence or reproduction of historically engrained authoritarian conflict management practices. Using the spectrum of peacebuilding strategies as an analytical tool helps to identify the underlying contradictions and dynamics. Elites exploit international discourses on democracy and peace to further their economic, political and social interest. They criminalise the political opposition against the status quo as well as marginalised groups such as youths or indigenous communities. The final section summarises what we can learn from Latin American experiences for global approaches to peacebuilding.

Latin American contributions to the global debate
Latin American experiences show how specific peacebuilding approaches are not only shaped by local contexts but also by the patterns of macro-regional development and path dependence regarding institutional change. Democratisation as progress towards conflict-processing institutions is limited to contexts or windows of opportunity where local actors support or promote an agenda of change. In Guatemala, for example, advances in the rule of law were possible as long as external support for such measures had allies inside local institutions (the Attorney Generals Claudia Paz y Paz and Thelma Aldana). When this changed in 2018, the President moved to dismantle the CICIG and has not really encountered international resistance. Most countries formally comply with minimum international standards (elections) but sideline or undermine those institutions or processes that could change the existing power relations. The corruption charges against the Guatemalan president and the withdrawal of the CICIG head's visa are an interesting illustration of this. Path dependence prevails most of all in relation to social and political conflicts framed as 'criminal'.
Accordingly, two interrelated arguments can be made: First, neither democratisation nor the termination of the civil wars has changed the deeply ingrained social inequality in many parts of the region. 67 Participation in violence was a means of political change during the twentieth century but petty crime or drug trafficking is today a means of survival or even upward social mobility. Despite many differences between these periodssuch as the lack of ideology and differences in the global environmentyoung, marginalised youth have provided the bulk of the rank and file in organised violence. Second, state repression, selective political violence and corruption have allowed traditional elites to maintain and reproduce their unequal economic and social status quo despite democratisation. The violence against human rights defenders and social activists (see Figure 1) is just the tip of the iceberg. The result of these developments has been framed as 'perverse state formation' 68 ; as 'criminal governance', 69 from a more anthropological and local perspective, or as 'violent democracies'. 70 In the context of minimalist concepts of peace, formally democratic regimes and high levels of violence not only coexist but are also intertwined shaping the limits of peace in Latin America.
The comparative analysis of Latin American experiences shows that a broader conceptualisation of peace as violence reduction is fruitful as it allows for the identification of continuity and change. A crucial element of enhancing the scope of peace in Latin America (and probably elsewhere) is the recognition of full citizenship. 71 The current patterns of selective political violence in Latin America are obviously linked to the process of change. The problem with liberal peace is not its emphasis on universal human rights but the lack of protection for those who claim and advocate for these rights. The gap between the formal guarantees of civil and political rights and their acknowledgement and protection must be closed. This would provide a basis for inclusive policies able to reduce the high levels of political and social inequality. The current backlash against formerly marginalised groups (women, youth, indigenous and LGTBI) in Brazil and elsewhere shows how contentious these changes are. Independent judicial institutions are key to subordinating elites from all political backgrounds to the rule of law. Given the legacies of war and violence, even selective political violence endangers non-violent actions for change.
Politically this means that despite their limited influence, international actors need to insist on the implementation of peace accords and the guaranteeing of civil and political rights to prevent these being circumvented or undermined. While this challenges the sovereignty and the state-centric approach, non-enforcement may lead to renewed escalation of violence along political lines. The high levels of selective political violence in Central America and Colombia (see Figure 1) show how dangerous this might be.
Across academic and policy debates, Latin American experiences connect with those from other world regions and the recent innovations in the peacebuilding debate. 72 But Latin American experiences highlight that internal and external actors need a broader perspective of peace beyond the absence of war as a compass for the future. Otherwise, local power relations may instead support a relapse into the authoritarian and violent practices of the past or the countries may remain stuck in transition. Peacebuilding choices are not necessarily dichotomous and limited to liberal or illiberal choices. Latin American experiences show that there is an important grey zone and high levels of variation. But the end of war or of an authoritarian regime is not necessarily a tipping point towards change. Theory and policy approaches need to systematically take into account the fact that neither peacebuilding nor democratisation proceed as linear processes but are rather highly contested. Outcomes are shaped by the context-specific interaction between global norms, local contexts and the diverging goals and interests of a multitude of actors.
Peace and democracy need broad coalitions based on the acknowledgement of human rights for all. The right to physical integrity stands at the core and policies of comprehensive violence reduction are a necessary basis for inclusive, just and peaceful societies as envisioned in Sustainable Development Goal no. 16 73 and the United Nations agenda of sustaining peace. 74 These documents have developed from and transcended the liberal tradition and as such are an important reference to facilitate change towards a peace beyond the absence of war. While change in this direction is a long-term endeavour, violence is a powerful instrument for reproducing the status quo.