1. A Decline in Political Violence?
On 20 May 2011, the Norwegian daily Aftenposten (2011) observed that ‘in the last six months, supporters of democracy in Egypt and Tunisia have achieved exactly what Al-Qaida could not – without resorting to terrorism or extensive acts of violence.’ Is this part of a global trend towards non-violence? If so, can we identify policies and institutions that can encourage this development?
Recent studies argue that many forms of political violence indeed have declined globally (Goldstein, 2011; Pinker, 2011). The frequency of armed conflicts has dropped by 40% since the end of the Cold War (Themnér & Wallensteen, 2011), and the long-term trend in coups d'état is also downwards (Powell & Thyne, 2011). Research conducted at the Centre for the Study of Civil War, PRIO, shows that the number of people killed in such conflicts declined markedly after World War II (Lacina & Gleditsch, 2005). This reduction is even more remarkable considering that the global population has nearly tripled over the period.
However, the tendency to equate political violence with major interstate and civil war obscures the diversity of the phenomenon. Recent years have seen bloody terrorist attacks such as those in London and Mumbai; fierce sectarian fighting has upset Somalia; popular uprisings have been met with indiscriminate repression in e.g., Libya and Syria; sexual and gender-based violence continue to mar societies in the aftermath of war; and organized crime in Latin America today may be as lethal as the civil wars that raged the continent during the 1980s. Do these types of political violence follow the same declining pattern?Or are we witnessing a transformation of violence, where ‘new threats [..] have supplemented continued preoccupations with conventional war between and within countries’ (World Bank, 2011)?
This proposal outlines a Centre of Excellence on Transforming Political Violence (TPV) that will document, evaluate, and explain the trends in political violence. TPV will apply a broad definition of political violence, i.e. violence that is carried out by or on behalf of a social group or a political organization with the explicit aim to affect relations between groups or between groups and a government. Criminal violence will be relevant to TPV only to the extent it is clearly linked to political conflicts. Hosted at PRIO, the Centre will address the following overarching questions:
· Is there a consistent decline in political violence?
· What explains variations in the decline across types and regions?
· What are the main challenges to non-violent politics and institutional stability?
· How can institutions be designed to promote non-violent politics?
Institutions are key to all four research questions. Institutions are central in many explanations of the decline in political violence, and the increase in conflict-managing institutions is an outcome of this trend. Linking institutions to the transformation of political violence represents a significant break from much contemporary research, which relates violence to structural conditions such as ethnic polarization, inaccessible geography, or widespread poverty or inequality. These conditions are important, but change only slowly. Institutions are established as means to political ends and may be modified for similar reasons. Malleable and potentially important peace-inducing institutions include representation of ethnic minorities, institutions for transitional justice, redistribution and land reform, bodies overseeing security force behavior, or UN peacekeeping engagements. In the long run, institutions may also change structural conditions such as poverty and inequality.
The research carried out in Transforming Political Violence will not only increase our theoretical understanding of political transformation, but is also inherently policy relevant. TPV will thus interact closely with policy communities and explicitly seek to inform applied politics. The seven designated Working Groups will address complementary dimensions of the violence-institution nexus that will guide the development of targeted policy prescriptions on ‘what works.’
2. Beyond the Research Frontier
This section briefly surveys the relevant literature on the four broad research questions and outlines TPV’s theoretical points of departure.
2.1 Is there a consistent decline in political violence?
The picture of a gradually more peaceful world is increasingly accepted with the publication of influential reports such as the Human Security Report (HSRP, 2005) and World Bank (2011). The evidence is clear for conventional wars, and several studies indicate similar trends for other forms of political violence such as domestic and transnational terrorism (Enders, Sandler & Gaibulloev, 2011; HSRP, 2008). The continued wave of democratization suggests that in many parts of the world, violent conflict has been replaced by non-violent politics in a broad sense of the term.
To conclude that this necessarily implies a decline in all types of political violence is premature, however. TPV will explore three central caveats. First, armed conflicts exhibit significant geographic variation (Buhaug & Gleditsch, 2008). Indeed, whereas successful conflict resolution has dramatically reduced the number of armed conflicts in Latin America and Southeast Asia since the Cold War era, Africa, the Middle East, and South Asia show little improvement. Likewise, Islamist political violence appears on the rise (HSRP, 2010). TPV will study whether regional variation also applies to other types of violence, and why.
Second, for most forms of violence, firm conclusions cannot be drawn without better data and more research. Some types of political violence lack systematic global data. Single-case reports aside, we know very little about female participation in violent movements and general patterns in sexual abuse and gender-based violence. For one-sided violence, possibly the most lethal form of political violence in the 20th century, reliable and updated data span only the two most recent decades (Eck & Hultman 2007). There is also a scarcity of research at the opposite end of the spectrum: the emergence and success of non-violent political opposition movements (Cunningham, 2011; Stephan & Chenoweth, 2008). Although less than one-fourth of the world now remains autocratic (Marshall & Cole, 2009), nominal democracy may be undermined by widespread corruption and electoral fraud. Modern technology may aid in anti-governmental mobilization, but may also facilitate effective governmental monitoring and coercion. TPV will collect new and better data on political institutions and other aspects of governance.
Third, the decline in some forms of overt political violence may be accompanied by an increase in more subtle forms of violent coercion and dissent – in other words, we might be seeing a substitution effect. Many governments rely on violence to retain power, and human rights violations may be increasing in some parts of the world (Gibney, Cornett & Wood, 2010). Urban demonstrations and riots, exemplified by the 2011 ‘Arab spring,’ appear to be on the rise in Asia and Africa (Buhaug & Urdal, 2011). Likewise, new data from the Uppsala Conflict Data Program (UCDP) show an increase in inter-communal conflicts since 1989 (Eck, Kreutz & Sundberg, 2010), and a study of political protests and civil unrest suggests a similar trend (Nardulli, Leetaru & Hayes, 2011). These trends are consistent with potential substitution effects and may substantiate the claim in World Bank (2011) that ‘21st century violence doesn’t fit the 20th-century mold’ or accounts that highlight the emergence of ‘new wars’ (Kaldor, 1999).
Four dominant sets of explanations provide insights into the variation in certain forms of political violence across regions and, consequently, also some of the trends we observe.
Social and economic development: Today, political violence is least prevalent in societies characterized by high income, widespread education, diversified economies, and heavy reliance on human and financial capital. Four related hypotheses may explain this regularity. First, extensive specialization renders individuals and states dependent on one another. As damage inflicted on one actor ripples through the web of interdependent economies, conflict becomes comparably costly (Rosecrance, 1986). Second, development increases the importance of assets difficult to appropriate by means of force, making wars of conquest less profitable and democratization less threatening to elites (Acemoglu & Robinson, 2006; Boix, 2003). Third, development helps governments to enforce rules that preserve a non-violent, productive environment (Fearon & Laitin, 2003). Fourth, in industrial societies with a flexible labor market and a well-functioning welfare system, the income forgone by joining a rebellion is likely to outweigh material benefits (Collier & Hoeffler, 2004).
Democratization: The democratic peace literature (Doyle, 1986; Gleditsch & Hegre, 1997) shows that democratic states rarely if ever fight wars against each other. Democracies, moreover, are less repressive (Carey, 2009), use less violence against civilians (Eck & Hultman, 2007), and suffer fewer battle-related deaths during civil wars (Gleditsch, Hegre & Strand, 2009). When citizens have control over their governments, they restrain governments’ propensity to use violence. Accountability restrains governments from engaging in conflict-escalating bluff games (Fearon, 1994). Countries with better female political representation see less political violence (Caprioli, 2000; Melander, 2005a; 2005b).
Less is understood, however, about the relationship between democratic institutions and other forms of violence. The combination of restraints on government repression and opportunities for organized opposition facilitates clandestine movements, and democratization in the absence of economic development may in fact increase the risk of many forms of political violence (Collier & Rohner, 2008; Hegre et al., 2001). There is no ‘democratic civil peace’ when it comes to terrorism. However, whereas most suicide attacks appear motivated by Western (especially US) interference in the Muslim world, these events rarely strike consolidated democracies (Pape & Feldman, 2010).
Norms: Prevailing norms regarding the use of violence for political goals have clearly changed. The international norm against states’ territorial expansion had its roots in the Westphalian peace, and very few states have attempted to violate this norm after World War II. Indiscriminate killing of civilians now justifies ‘humanitarian interventions’ by dominant powers, as most recently seen in the UN resolution to protect civilians in Libya. Likewise, gender equality is slowly becoming a universally accepted principle. Just as dueling and slavery have become obsolete (Mueller, 1989), norms are increasingly turning against political violence. Human cultures may have the capacity to retain, learn, and spread ideas that are beneficial – changing norms may change societies. However, our theoretical understanding of what drives changes in norms and how norms gain acceptance remains incomplete.
Increased international activism: Both HSRP (2005) and Goldstein (2011) see UN peacekeeping operations as the prime source of the decline in internal armed conflicts after the Cold War. In parallel, we have seen an increase in international mediation efforts (Beardsley et al., 2006); NATO engaging in out-of area operations; involvement of regional organizations such as the African Union; a dramatic rise in international election monitoring (Kelley, 2008); the establishment of the International Criminal Court in 2002; and more targeted initiatives, such as the UN-adopted Kimberley Process Certification Scheme from 2003 to put an end to so-called ‘blood diamonds’ that fuelled atrocities and wars in West Africa during the first post-Cold War decade.
These four theoretical approaches offer complementary explanations for the decline in state-based armed conflict, but have not been applied equally systematically to other forms of violence. Many forms of political violence, however, have been declining in parallel. TPV will examine the conditions under which these explanations apply to forms of violence other than armed conflict.
2.3 What are the main challenges to non-violent politics and institutional stability?
TPV will address three major global phenomena that may provide challenges to securing societal stability and security:
Vulnerability of the global economy: The recent fuel, food, and financial crises have revealed the vulnerability of many societies to falling stock markets and fluctuations in international commodity prices, and could fundamentally alter the nature of the globalized economy and hinder continued growth (Rodrik, 2010). As indicated above, a reversal of past growth trends is likely to also threaten the decline in political violence.
Demographic trends: The world population, presently at 6.9 billion people, is projected to reach 9.3 billion within 2050 (UN, 2011). A key concern in this regard is food security, as the combined effect of absolute population growth and shifting consumption habits may put non-sustainable pressure on subsistence resources (Goldstone, 2010). Moreover, within the next 40 years at least 3 billion will be absorbed by cities, straining societies’ ability to provide public services like adequate housing, electricity, water supply, health care, education, and jobs. Transnational migration may amplify existing pressures on services and resources, instigate host-newcomer tensions, and lead to a ‘brain drain’ from the South. Changing age compositions posit another demographic challenge. Youth bulges are robustly linked to various forms of political violence (Urdal, 2006). In other regions, an increased number of elderly people will put stress on public services (Goldstone, 2010).
Climate change: Global warming is widely regarded as the ‘defining challenge of our time’(Ki-moon, 2008). A number of physical changes associated with global warming may be harmful to socio-economic growth and political stability, such as sea-level rise, more extreme weather-related events, melting of glaciers, drying of rivers and lakes, and increased soil evaporation. There is little evidence to date that armed conflict is a likely consequence of climate variability (Buhaug, 2010), but little is known about long-term implications of environmental change or the influence of resource scarcity on other forms of political violence. The cost of adapting to these risks – as well as mitigating climate change – constitutes a massive challenge (Stern, 2006).
TPV will study how these challenges are likely to affect trends in violence by relating observed patterns of violence in the past with economic recession, rapid shifts in commodity prices, population growth and mobility, and environmental change. TPV will use this information in combination with forecasting techniques (Hegre et al., 2011) to evaluate likely implications of these challenges.
2.4 How can institutions be designed to promote non-violent politics?
The final research question has a practical ambition: To formulate knowledge-based policy recommendations. Suggested interventions must be causal in the sense of involving factors that can be manipulated exogenously with the desired effect. Structural explanations do not lend themselves to direct policy intervention; the focus here will be on factors within the control of the policymaker.
Effectively designed institutions can potentially affect aspects of all four explanations indirectly: Institutions may stimulate economic growth, ensure representation, improve human rights, and facilitate third-party intervention. Current knowledge is insufficient for effective policy design. For example, democratic institutions appear to have important direct conflict-reducing effects and may also be instrumental in ensuring growth-conducive policies (Doucouliagos & Ulubasoglu, 2008). However, citizens’ struggle to gain political representation in government is often a cause of political violence in itself, as recently seen in Egypt and Libya. Moreover, democracy is essentially an institutionalized agreement between social groups not to use violence or coercion to arrive at political decisions. As such, this pact is what we need to explain. The impact of democratization cannot be seen in isolation from the other explanations reviewed, such as economic development and cultural-normative changes. Still, it is possible that the partial institutionalization of such agreements between groups may spill over into other relations with other groups in the political system, or pave the way for more comprehensive agreements. Our aim is to be able to identify, for example, exactly when and how various institutions of democracy should be promoted.
Central to this effort will be to move away from dichotomous concepts and a simplified, uni-dimensional democracy–autocracy continuum. Various institutional configurations vary considerably in terms of their effectiveness (Fjelde, 2010; Gates et al., 2006; Hegre et al., 2001). In order to understand this variation, the Centre will disaggregate the institutions that together define ‘democracy’ and ‘autocracy’, by distinguishing between institutions of selection, constraints, elections, decentralization, electoral fraud, etc., and look into related institutions such as rules related to gender equality and minority representation, judiciaries and federal arrangements. Collecting such data will allow TPV to advance research on how regime openness has developed over time and how this process is associated with political violence.
In addition to focusing on democratic institutions, TPV will address other institutional ‘prescriptions’ that are related to democracy but less formal in nature. These will form the basis for the thematic Working Groups of TPV and are detailed in section 4.1.
3. Our approach
A set of important common guiding assumptions and principles that will guide the research at TPV are summarized below. Some theoretical implications of these assumptions are detailed in section 4.1, and methodological implications in section 4.2.
Actors and motivation: Much conflict research focuses on structural opportunities. However, violence cannot simply be understood as an outcome that afflicts particular countries but must be understood as purposive strategic choices in conflict between specific actors, for example central governments, specific rebel organizations or organizations that purport to represent communities such as ethnic groups (Cederman & Gleditsch, 2009). If conflict is costly, there should be incentives for avoiding violence; accounting for the resort to violence must thus outline the mechanisms that prevent non-violent settlements between actors. TPV will conduct detailed case studies to achieve this, but also develop quantitative designs along the lines of Buhaug et al. (2008) and Raleigh & Hegre (2009) to be able to model the choices of actors statistically.
Substitutes vs. complements: Alternative forms of violence are possible substitutes and complement each other. The debate on the transformation of violence and purported explanations raises questions about how incentives shape the choice of specific actions and how different outcomes may be related to one another. For example, greater political openness may allow non-violent action to substitute for rebellion, while repression can encourage rebel groups to resort to terrorism to complement conventional attacks. Frequently, objections to the alleged decline in violence surmise that a decline in conventional war is accompanied by an increase in other forms of violence, such as terrorism or one-sided violence. Hence, alternative actions and outcomes must be studied jointly.
Formal and informal institutions and incentives for violence: Political institutions include the broad range of rules and conventions that structure interactions and influence political outcomes. Particularly important are the formal, governmental institutions that regulate recruitment to political office and participation in this selection process, and ensure that actors abide by the rules and refrain from changing them in unconstitutional ways. Examining how political contests within the institutions engender or prevent violence, and how institutions develop and are sustained, will be central to all research at TPV. Less formal institutions, however, can also affect the incentives of the actors and the functioning of formal institutions. For example, perceived legitimacy, individual trust, and social norms play important roles in the decision to resort to violence. Dark features such as patronage or discriminatory institutions can motivate violence – but also coopt actors and pre-empt violence (e.g., Fjelde, 2010).
Endogeneity and institutions: To understand the potential relationship of institutions to peace, and to be able to give advice to policy makers, we need to consider the origins and possible endogenous nature of institutions (Przeworski, 2004). Many specific power-sharing institutions have emerged as a compromise to conflict between actors, and may as such be endogenous to the settlement of the conflict. If so, efforts to transfer such institutions to other conflicts do not necessarily decrease the incentives to use violence (e.g., Snyder, 2000). Understanding the prospects for conflict resolution mechanisms through institutions and the limits of institutional engineering requires us to consider endogeneity. We will handle this through research on how institutions are formed and on their effects, linking these two processes through a set of research strategies: discussing them in conjunction with thick descriptions of particular cases of institutionalization, seeking ‘natural experiments’ where institutions arise for reasons that are clearly unrelated to particular instances of political conflict, and employing statistical methods that are designed to handle endogeneity.
Integration of the micro and macro-level: While many studies adopt a ‘closed polity’ model to individual conflicts, the strategies of actors and incentives for violence in individual conflicts are very much shaped by links to the outside world. Gleditsch (2007) and others have pointed to the importance of transnational linkages such as transnational actors, external support, and outside pressure on the actors. International institutions such as peacekeeping operations, the International Criminal Court, or human rights monitoring agencies are designed with the explicit intention of changing the incentives of actors and decrease the attractiveness of violence. More generally, actors adopt strategies based on learning from the experiences from other conflicts, and their anticipated reactions from international actors. Understanding violent conflict dynamics and how possible international responses can shape the incentives of the actors thus requires us to link the micro level, specific actors and conflict dynamics with the macro level or the transnational context.
The research in the first five years of TPV will be structured around seven themes representing conditions or quasi-institutional prescriptions that facilitate the transformation from violent to non-violent politics. Many of these relate to each other, and all relate to the four research questions as well as to the study of formal political institutions. In addition to these seven thematic Working Groups, TPV will have a set of cross-cutting groups that ensure communication between the thematic groups concerning issues that are of interest to all research at TPV.
4.1 Thematic Working Groups
All researchers at TPV will be member of at least one Working Group (WG). WG leaders are responsible for outlining the research agenda, recruiting TPV core staff, recruiting research associates and international experts to the WG, and arranging annual workshops. The WG leaders will also advise the TPV directors on important matters and assist in hiring procedures.
WG 1. Non-violent strategies (Kristian Skrede Gleditsch)
WG 1 will study trends in the prevalence of non-violent protests. Nonviolent activities and protest are integral to well-functioning democratic political systems, but are also common in non-democratic societies (Sharp, 1973). There is an increasing global salience of nonviolent protest and renewed interest in its effectiveness (e.g., Stephan & Chenoweth, 2008). Studies of conflict and protest have been limited by an excessive focus on violence, and the observed decline in political violence may be due in part to substitution and the emergence of alternative means and more effective non-violent movements. There has been little research on how incentives for violence and nonviolence are shaped by the interaction between governments and non-state actors, as well as other social characteristics such as ethnic relations and elite factionalization (Svensson & Lindgren, 2011). WG 1 will explore explanations for this shift by studying the incentives for opposition groups to employ political violence compared to non-violent actions. Core research questions include explaining the decisions of actors to use nonviolent versus violent strategies, and how social characteristics and political institutions influence the relative effectiveness of violent and nonviolent strategies.
WG 2. Human rights (Sabine Carey)
WG 2 will explore another possible transformation of violence, namely from overt group conflict to latent violence in the form of government repression. In political systems that keep large segments of the population outside the formal political channels, leaders typically rely on repression and human rights violations as a strategy to retain power. The most indiscriminate repression (such as in Kampuchea under Pol Pot) seems to be in decline, but more limited repression may be on the rise. How has the nature and scope of repression changed, and why? Authoritarian regimes vary in the brutality of their repressive behavior, and repression occurs also in semi-democratic political systems. This WG will study how institutions of accountability can restrain the behavior of political elites. Accountability can be provided by domestic institutions such as strong parliaments and independent judiciaries, but also by international organizations like the African Union or the UN, as well as by domestic NGOs and media. Since repression affects the behavior of the opposition, WG 2 will work closely with WG 1 to study the escalatory repression-dissent dynamics of violence. This WG will look at when governments seek to avoid responsibility by resorting to pro-government militias, and how globalization, migration, and international focus on human-rights issues may change governments’ incentives for using repression.
WG 3. Governance (Scott Gates)
WG 3 will study the importance of changes in governance – the organizational and behavioral characteristics of how decisions made in the formal institutions are implemented – for the decline in political violence. Variation in governance may be more important than variation in the decision-making institutions themselves, to the extent that ‘the quality of governance peace’ may be more important than the ‘democratic peace’ (Råby & Teorell, 2010). Every political system faces issues of oversight and control of the bureaucracy. Despite significant efforts to constrain them, bureaucrats possess significant degrees of discretion. Bureaucrats make political decisions in the implementation of policy, affected by their own policy preferences. Poor governance in this sense may undermine the formal institutions regulating representation. Of particular interest is the security sector. Excessive use of force, rape, false arrest, illegal search, and the misuse of authority constitute important issues of accountability of the police. A decision to employ a ‘hearts and minds’ strategy as opposed to indiscriminate use of force may be seriously undercut if soldiers rape and kill. This WG will explore which institutional designs maximize decision-makers’ control and oversight of the security apparatus, how institutions holding key decision-makers accountable may be complemented by institutions that prevent agency loss in the executive branch, and look at how international trade and aid conditionality affects the governance-violence relationship.
WG 4 will study the interaction of distributive issues, political violence, and transitions to robust peace. Can the transformation of political violence be attributed to changes in distributional patterns? Conflict over the distribution of economic and political power – between ethnic groups, classes, men and women, clans, parties, and other social groups – contributes to political violence, and redistribution is often an explicit demand. Moreover, sustained political violence frequently transforms the political economy. How political violence ends further shapes prospects for political violence, on the one hand, and robust peace, on the other. Redistributive institutions consequently have a potential to shift politics into non-violence at various critical junctures, particularly when non-violent demands for redistribution shift toward violent ones, as the terms of conflict resolution are set, and in the immediate aftermath of conflict termination. Such institutions are often distinct from the formal democratic institutions – they can be one-off land reforms as in Korea and Taiwan, or the development of welfare policies targeting those making redistributive claims. WG 4 will address questions such as, what role do formal distributive institutions play compared to informal institutions in reducing political violence? Which redistributive institutions and processes are more likely to lead to and sustain robust peace? And how do redistributive institutions (particularly land reform institutions) affect conflicts over land use, tenure, and distribution during and after conflict?
WG 5. Trust and tolerance (Ola Listhaug)
A possible decline in political violence must be seen in conjunction with the emergence of ‘civic orientations’ – a cultural transformation characterized by support for democracy, political and social trust, acceptance of minority groups, and support for human rights (Welzel & Inglehart, 2005). To what extent is this observation part of a larger trend in attitudinal changes underlying the decline in political violence? Few studies of political violence take popular attitudes sufficiently into account. Is a violent attempt at conflict resolution a viable strategy for an actor if large or important segments of the population find violence morally unacceptable? This question is equally relevant for governments as for opposition groups, be they political parties, ethnic minorities, or would-be military coup makers. Attitudes and norms are often assumed to be essential for the sustainability of institutions (Almond & Verba, 1963; Sullivan & Transue, 1999; Welzel & Inglehart, 2005). How do individuals perceive the institutional mechanisms in place to handle intergroup conflict (e.g. power-sharing or consociational arrangements)? How do post-conflict institutions affect attitudes toward and trust between groups that have been in conflict as well as within such groups? This WG will study how peaceful norms toward conflict resolution emerge, to what extent attitudes toward violence are functions of formal institutions or previous violence, and how citizens manage the transition to a more peaceful society.
Much political violence falls along lines defined by easily identifiable group characteristics. Such identities may be determined by differing creeds, classes, castes or ethnic affiliations, and many conflicts are linked to the exclusion from political power of such groups. Ethnic groups often have histories of antagonisms or define particular forms of political exclusion and are clearly relevant to a study of the decline in political violence (Gurr, 1993; 2000). WG 6 will study how variation in political violence is related to group-based exclusion from power. Gurr (2000) points to four main factors: the strengthening of group rights, democratization, the decline of uncompromising and discriminatory policies, and the growing importance of international institutions and interventions. A central topic will be to study how political institutions should be designed to avoid ethnic exclusion through power-sharing or power-dividing institutions, and investigate the problems associated with such arrangements (Rotchild & Roeder, 2005). This WG will study the effectiveness of these institutions as well as the conditions under which they are sustainable. A sudden increase of political participation can lead to great disruptions within a political system (Cederman, Hug & Krebs, 2010; Mansfield & Snyder, 2005) and newly elected elites might wish to reduce inclusiveness to maintain their hold on power. An important point of departure are new data on ethnic exclusion from power (Cederman, Wimmer & Min, 2010). Key areas of inquiry will be the role of ethnic or religious voting, in elections as well as within parliaments, for the regulation of political conflict, and how ethnicity interacts with political institutions such as decentralization, political rights, and rules regulating political competition.
WG 7. Conflict resolution (Erik Melander)
WG 7 will explore the extent to which the transformation of political violence is due to the emergence of new conflict resolution mechanisms. Violent inter-group conflicts are typically resolved by an agreement to refrain from using violence: explicit (e.g., in the form of a clearly spelt out agreement) or implicit (e.g., when violence has already declined and parties assure each other of their preference for peaceful relations), formal or informal. Conflict resolution is typically carried out within an institutional framework such as ‘truth and reconciliation commissions’, mediators appointed by the UN, a country’s formal institutions, or within traditional local institutions such as councils of village elders. The output of the resolution may have the form of a peace agreement. WG7 will study the conditions for and dynamics of such conflict resolution processes, in what form and under what conditions peace agreements are sustainable, and investigate how they interfere with other institutional arrangements in a political system. Particular attention will be given to local, traditional conflict resolution mechanisms that have been given scant attention in the political violence literature, and to international mediation institutions such as those mandated by the UN, by major power actors, and regional institutions. The WG will look at general mechanisms whereby formal and informal conflict resolution alter the incentives of actors to use or return to violence, and how the UN and other international institutions may contribute to limiting violence by means of peacekeeping operations, mediation, and other measures.
4.2 Cross-cutting groups
TPV will have seven cross-cutting groups (CCG) as discussion fora for a set of thematic and methodological concerns important to all thematic WGs. They will seek to expose the WGs to methodological innovations as well as conduct research to advance their topics. The CCGs will recruit researchers from all WGs and thereby strengthen integration within TPV. The CCGs will hold regular workshops, seminars, short courses, as well as separate subsections on the TPV webpage.
CCG 1. Gender equality and political violence (Ragnhild Nordås)
The decline in political violence has been accompanied historically with the emergence of gender equality. However, it is unclear which way the causal arrow goes, and also which exact mechanisms are at play. To improve our understanding of the role of gender equality, this CCG will 1) conceptualize gender equality more broadly than in earlier research, including various levels of analysis and institutional levels; 2) explore how (attitudes towards) gender equality and violence relate in order to inform the more general explanations of political violence; and 3) collect systematic data for empirical investigation of the gender-political violence nexus.
CCG 2. Qualitative research techniques (Kristine Höglund)
This CCG will stimulate debate and development of qualitative research methods in use at TPV. Qualitative methods – broadly defined to include interviews, participant observation, ethnographic mapping, recording of oral histories, as well as aspects of surveys (particularly less structured protocols) and experiments (debriefing after experiments often takes the form of qualitative interviews) – are frequently essential to substantiating causal inference by testing whether the claimed causal process is indeed exerting significant causal leverage. To address the rapid advances in qualitative research, the group will explore a broad range of qualitative techniques which can be used to generate valid causal inference, tools for assessing the veracity of information gathered via qualitative research methods, and opportunities and challenges for multi-method research.
Across all areas of investigation, TPV will combine methods as needed for making valid causal claims, including an emphasis on the evaluation of alternative explanations, the careful designation of the case and the universe of which it is an instance, the specification of observable implications of both the theory advocated and its rivals. CCG 2 will collaborate with CCG 3 to allow for communication between qualitative/historiographical and quantitative research, and to provide a forum where the use of particular methods can be challenged by researchers with diverse backgrounds and expertise.
This CCG will take up challenges related to the statistical methods employed at TPV, focusing in particular on the relationship between theoretical models and the statistical methods employed to evaluate the predictions from the models (e.g., Bas, Signorino & Walker, 2008; Granato, Lo & Wong, 2010), as well as the ability of statistical methods to measure dynamic latent variables like the state of the political system and level of development (Høyland, Moene & Willumsen, 2011). The group will advise members of TPV on statistical issues and introduce new methods and developments in political methodology.
Particularly important are methods for causal inference, including matching methods, instrumental variable estimation and difference-in-difference strategies (see Angrist, Imbens & Rubin, 1996; Ho et al., 2007). These inferential strategies help researchers carefully measure the potential impact of policy changes on violent behavior without making strong assumptions about exogeneity or functional form. Further, developing and fostering expertise in this growing area of methodology can provide leverage across the thematic Working Groups as factors that influence institutional development in democracies become key matching variables in analyses that attempt to measure the impact of those institutions on violent and non-violent outcomes.
Proper integration of theoretical and statistical models will be facilitated by promoting Bayesian methods. The group will also stimulate systematic use of out-of-sample prediction for model evaluation. This is particularly important for models purporting to underpin policy recommendations (Ward, Greenhill & Bakke, 2010). TPV will also expand current forecasting approaches (Hegre et al., 2011) to a wider range of violence. Other approaches that will be encouraged are non-parametric models and quantitative text analysis as well as maintaining a good understanding of more standard approaches.
Experimental methods, with their commitment to randomization of a given ‘treatment’ across individuals or groups, enable scholars to overcome thorny issues of causal inference that typically arise when working with observational data. Scholars have, for example, used behavioral games to assess the effects of co-ethnicity on collective goods provision in the slum city of Kampala, Uganda (Habyarimana et al., 2009), the effects of media on intergroup prejudice and norms in post-genocide Rwanda (Paluck, 2009), and the impact of development aid on social cohesion after the civil war in Liberia (Fearon, Humphreys & Weinstein, 2009). Scholars have also combined the experimental tradition with survey work to generate new insights about dynamics of conflict. Blattman (2009), for example, draws on surveys to assess the post-conflict willingness of abducted child soldiers to involve themselves in collective goods behavior, including voting. Particularly promising is the new trend toward using survey-based experiments to measure sensitive attitudes in conflict settings. CCG 4 will help to introduce these methods into new fields by organizing workshops that expose TPV researchers to these methods and by facilitating co-authorships.
A primary focus of this CCG is on agent-based modeling (ABM), a simulation technique that represents social actors as individual computational entities (the ‘agents’). These entities then interact to generate emergent patterns at the societal level. ABM has successfully been applied to various questions related to TPV, such as the size of wars (Cederman, 2003), ethnic mobilization (Bhavnani, 2006), and collective mobilization (Siegel, 2009). A cutting-edge development in the use of ABM is the incorporation of spatial data, such that the model runs on a real-world geographic space (Weidmann & Salehyan, 2010). The group will explore agent-based and other types of simulations both as theoretical devices and forecasting tools.
This CCG will coordinate data collection projects and provide all TPV researchers with standardized routines for transforming and combining different data sources. The group will also critically examine the quality of data collected and used at TPV, including possible bias arising from the increasing use of events datasets and surveys.
CCG 6 will benefit from and contribute to PRIO’s RCN-supported INFRASTRUKTUR project, the Advanced Conflict Data Catalogue (ACDC), which provides external third-party data collection efforts with a set of resources to increase the interoperability of these datasets and assist such projects in focusing their limited resources on gathering information by providing a high-quality framework for data collection. The first product from this project, the PRIO-GRID (Tollefsen, Strand & Buhaug, 2012) will be released in 2012. The coordination of data collection efforts through the ACDC infrastructure ensures that researchers within and outside TPV have access to large amounts of compatible, high-quality data.
TPV will collect data primarily on political violence and political institutions. The close link to the Department of Peace and Conflict Research, Uppsala University will be extremely useful for this task given their unusually solid experience with coding of political violence data. For data on political institutions, TPV will continue to update the MIRPS dataset introduced in Gates et al. (2006), collaborate with Patrick Regan to extend the time series of the Institutions and Elections Project (Regan, Frank & Clark, 2009) to cover the 1960–2010 period, and with Michael Coppedge and associates at the University of Notre Dame to code detailed characteristics of all government institutions. TPV will also expand data on repression, building on the data used in (Carey, Gibney & Poe, 2010).
TPV will compile and publish an index of prevalence of political violence – the Globa Political Violence Index – building on the various data collection efforts, and reflecting the conviction that different forms of political violence have related explanations and similar dynamic patterns. Existing indices such as the World Peace Index and the Global Peace Index simply throw together a lot of indicators with little if any consideration of theoretical relevance and relative importance. TPV research will permit a state-of-the-art science-based index of use to policy makers and scholars alike.
CCG 7 Research dissemination (Gudrun Østby)
This CCG will work to stimulate dissemination at TPV. The primary outlet for the research will be international peer-reviewed journals, top international book publishers, and participation in academic conferences as those arranged by the ISA or the ECPR. CCG 7 will also oversee the dissemination of research results by means of a website, with separate subsections for each WG and CCG.
In line with TPV’s deliberate focus on the policy relevance of the research, it will actively reach out to a broader audience and policy-makers through the following strategy:
· Networking: Many TPV members are engaged in consulting with policy-makers through research done for the UN, the World Bank, etc., including considerable contributions to the World Development Report 2011. TPV members also have close contact with the Human Security Report Project at Simon Fraser University in Vancouver which specializes in presenting research to international policy-makers. The strategy will deliberately maintain and extend this network of contacts and seek new commissions for policy reports.
· Writing for policy-makers: In addition to requiring scholars to write short ‘policy briefs’ targeted at policy makers, TPV will commission free-lance journalists to summarize research done at TPV.
· Media strategy: TPV will actively liaise with journalists to make them aware of research at TPV, and to make researchers better aware of the topics that interest journalists.
· Targeted presentations: TPV will offer targeted briefings of central new research results to key international and Norwegian decision-makers, partly in the form of policy conferences.
In addition, TPV scholars will publish in Norwegian targeted at a general, non-specialist audience.
TPV will be an administrative unit of PRIO, in accordance with the guidelines set by the Research Council of Norway, and located at PRIO’s premises. As stipulated in the RCN framework agreement with host institutions, the TPV Director will have a high degree of autonomy in scientific matters and recruiting. The TPV Director reports to the PRIO board, which also serves as board of TPV.
PRIO is among the world’s leading research institutes in the area of peace and conflict. PRIO researchers are widely cited and PRIO is home to two leading IR journals, the Journal of Peace Research and Security Dialogue. PRIO research is dependent on short- and medium-term funding from multiple sources, and has insufficient resources to initiate large, concentrated research efforts except through the RCN’s CoE framework.
In combination with the pre-established reputation of PRIO research, the CoE funds will allow us to bring together a set of excellent scholars that currently are located at various institutions throughout Scandinavia, Europe, and the US. The various strands in the research agenda reviewed above are currently studied in partial isolation from each other, and the formation of a sizable CoE to bring these strands together will release a large potential for cross-fertilization and innovative research. The location of such a centre in Oslo will also stimulate Norwegian research on political violence, strengthen education in conflict research and political science, and provide important input to current debates of central interest to Norwegian and international policy makers.
The research proposed here expands on the work at the Centre for the Study of Civil War (CSCW) which has had CoE funding since 2002, but its research agenda departs significantly from CSCW. While CSCW focused on a wide set of causes of civil war, TPV will look at how a wide spectrum of political violence is avoided, with a particular focus on the causal effect of institutions. Moreover, TPV will have much greater focus on policy relevance. In terms of personnel, TPV will benefit greatly from the expertise gained from running the CSCW by retaining key administrative staff, but the scholarly personnel will shift to reflect the difference in focus.
Several PRIO projects will be incorporated into TPV, some of them emerging out of the CSCW after the termination of CoE funding in 2012: TPV Director Håvard Hegre will start up a 5-year project on Political Institutions and Armed Conflict. Former PRIO Director Stein Tønnesson will work on a project on the East Asian Peace funded by the Swedish Riksbankens Jubileumsfond. This project will be based partly at Uppsala University, partly at PRIO, with considerable matching funding from TPV. Ragnhild Nordås will continue to work on a project on Sexual Violence in Armed Conflict; and Håvard Strand’s ACDC project will continue through 2013. The editor (Henrik Urdal) and several editorial committee members of the Journal of Peace Research will participate in TPV.
TPV will be directed by Håvard Hegre (currently Professor of Political Science at the University of Oslo). Hegre is Associate Editor of Journal of Peace Research and was editor of Norsk Statsvitenskapelig tidsskrift (Norwegian Journal of Political Science) up to 2010. Hegre has directed several major research projects. In 2001–03, he coordinated the World Bank Project on ‘The Economics of Civil War, Crime, and Violence’ under the direction of Paul Collier and is a co-author of Collier et al. (2003). He has since directed a RCN ‘Outstanding Young Investigators’ [YFF] project as well as several other research projects, and currently leads one of the CSCW working groups. Hegre has published close to 30 peer-reviewed articles in international journals, including most of the top-ranked journals in political science and international relations. He has contributed to several policy reports and publications targeted at policy-makers, among others the African Development Report 2008, the 2009 edition of Global Crises, Global Solutions, World Development Report 2011, and a UN ESCWA report on governance and conflict relapse (forthcoming). Hegre was ranked among the 10 most cited scholars world-wide in research on armed conflict by ISI Web of Knowledge in 2006. As of 1 June 2011, the ISI Web of Knowledge has recorded 828 citations to Hegre’s journal articles. Ten articles are cited 30 times or more.
Halvard Buhaug (CSCW/PRIO) will be Deputy Director and coordinator of CCG activities. He is leader of one of the CSCW working groups, and has coordinated CSCW cross-cutting activities. He currently directs the RCN project on ‘Security Implications of Climate Change.’ Past project leader experience includes a World Bank-commissioned project on climate change and conflict and the PRIO node for the ERC project ‘Disaggregating Civil Wars.’ Buhaug has published 15 articles in international peer-reviews journals. As of 1 June 2011, his articles are cited 215 times. He is frequently consulted by media as an expert on security aspects of climate change, as seen in the wide media coverage of his 2010 PNAS article.
The WG leaders are all leading scholars in their fields, recognized for their innovativeness, and have extensive publication records. They teach at leading universities in Norway, Europe, and the US, and are in the position to recruit excellent young researchers to the center as well as pull in other leading scholars in the field. Together, they bring together expertise on various forms of political violence as well as the development of institutions. While the WG leaders have been chosen for their proven expertise, the CCG leaders are all young, promising researchers specializing in the topics of their groups. CCG leadership will provide these researchers with leadership experience and significantly add to their professional development. TPV researchers will be recruited in part from PRIO and CSCW.
The WG structure incorporates international cooperation with several leading environments for research on political violence: The conflict research groups at the Universities of Essex, Geneva, Mannheim, and Uppsala; ETH Zürich, and NTNU are among the most prominent of their kind in Europe, as is the program on ‘Order, Conflict, Violence’ at Yale University. International cooperation will take the form of working group meetings at TPV, scholar exchange, joint participation in international conferences, and extensive co-authoring. TPV will also work closely with the Fulbright Foundation to facilitate exchange with US institutions.
TPV will involve scholars from the global South and establish cooperation with among others Instituto de Estudios Políticos y Relaciones Internacionales at the National University of Colombia, the University of Cape Town, and the Bangladesh Institute of Peace and Security Studies.
CCG 1 explicitly applies a gender perspective to the academic content across the WGs. In terms of research staff, women remain a minority among top international scholars in the field. This is reflected in the composition of the senior scholars at TPV. The gender distribution among younger scholars is much more even. In recent years, about three quarters of the PhD students affiliated with PRIO have been women, several of whom will be recruited to TPV. Our target is that half of the junior scholars at TPV be women, and our gender strategy will focus on actively encouraging and creating opportunities for female junior scholars to develop their research careers and achieve professor competence. Based on the high level of productivity so far among the female scholars listed in this application, we expect at least two of them to reach this goal within 2017, which means that the proportion of women in leading positions after the first five years of the TPV should increase significantly.
8. Ethical considerations
The intention of our research is that it be valuable to citizens of violent countries, although parts of it may be controversial to their authorities. Our approach here is to adhere to the strictest scientific standards in terms of quality, transparency, and replicability of data collection as well as clarity, stringency, and objectivity in the analysis of data and in presentation of the results. Much information collected by TPV regarding sensitive issues such as electoral fraud will be taken from official sources and incident reports from elections observers. We will not engage in any activities that exacerbate the vulnerability of the participants, and interviews and experiments will be based on informed consent and results will be anonymized.
TPV will have a relatively small number of researchers in permanent positions. We will seek to gradually develop and extend this group by hiring a few scholars with excellent academic and funding potential, but not beyond what is sustainable after the end of RCN CoE funding. Other scholars will be hired on a temporary basis. All external working group leaders will have temporary part-time contracts. As indicated in the budget, TPV has a deliberate strategy to raise funds from other sources (the RCN, EU, the MFA, etc) and expect between a third and a half of the budget to be non-CoE funds. See note to the funding plan for further details. In combination with a flexible staff structure, we expect the phasing-out to be smooth.
10. Research training and recruitment
Training of junior staff and students is a central component in TPV’s portfolio of activities and an important instrument to ensure recruitment of the best candidates to senior researcher positions at PRIO. To this end, TPV will collaborate closely with, and contribute actively to, the PRIO-based Research School in Peace and Conflict. The strong international network of TPV will strengthen the research school’s leading international place in training of scholars in the field of peace and conflict research.
Over the 10 years, the TPV budget will fund 30 person-years of PhD and post-doc stipends. The research school will benefit from the close contact with the host institutions of the working group leaders (e.g. Yale, ETH, Essex, and Uppsala University) by facilitating student exchanges and joint seminars and workshops. TPV will contribute to teaching at the three Scandinavian universities associated with the Centre.