This paper was presented at the 2020 Munich Security Conference at a side event hosted by the Armed Conflict Location & Event Data Project (ACLED) in partnership with the Geneva Centre for Security Policy (GCSP). It is an updated and expanded version of Global Conflict and Disorder Patterns: 2019.
In the past 10 years, the world has witnessed a decline in global cooperation and security. This downturn is manifest through multiple internationalized wars and massive humanitarian crises, rising nationalism from global powers, transnational terror organizations using sophisticated recruitment techniques, cyber-attacks orchestrated by marginalized states, sustained levels of violence in nominally ‘post-conflict’ countries, and a drastic rise in the number of non-state violent agents. An intensification of violence and risk has accompanied these notable shifts. Drawing on the ACLED dataset of almost a million political violence and protest events across over 100 countries, we can discern four broad patterns that summarize the current conflict landscape and indicate how disorder is likely to evolve in the future:
(1) Political violence is rising and manifesting as disorder in multiple forms. It is persistent and dynamic, consistently adapting to changing political circumstances and opportunities, rather than dissipating. For these reasons, it is best to understand political violence not as a failure of states, but as a volatile and flexible feature of political systems.
(2) Political violence is rising most quickly in developed states: Russia, Mexico and Turkey are key examples of how specific forms of political violence find an outlet in relatively wealthier states. Continued conflicts in the Democratic Republic of Congo, Somalia and Afghanistan demonstrate the intractable nature of wars in states with inconsistent government control and capacity across territory. Conflict is most persistent in poorer states, but even in these environments, it is a tool of the powerful, rather than the poor and aggrieved.
(3) The fallout from many externally imposed peace-building and stabilization efforts, forced elections, and corruption is unprecedented levels of militia and gang violence. Rather than a descent into chaos, this trend is tied directly to the domestic politics of states and the economic benefits of conflict. The form and intensity of such conflict adapts to political competition within states. As a result, we should expect a continued rise in militias, gangs and violence across most states.
(4) Finally, demonstrations are increasing drastically — but most peaceful protests have no effect on political structures and elite politics. State security forces continue to intervene violently in protests, and mobs — often hired by politicians — are responsible for a significant and deadly increase of rioting in South Asia and beyond.
Disorder over wars
Violent political instability and disorder are widespread, multidimensional and adaptive. Disorder takes many forms: election violence in India; gangs that offer security and services for a price in Central America; death squads that have ties to security forces in the Philippines; community vigilantes who operate identity-based security services across the Sahel; cyber-encouraged lone-wolf attacks across the US and Europe; and political assassinations and violence targeting political party supporters and journalists in South America. It can result in conflicts like those in Africa’s Great Lakes region; divided states in Yemen and Somalia; organized criminal violence in Central America; and politicized mob violence in India.
The lesson from these examples is that political violence and disorder do not arise from breakdowns of political control, and they are not examples of ‘exceptionalism’. As demonstrated by the earlier map, these are common, persistent, widespread forms of conflict. They arise and are shaped by local political and economic competition. In turn, understanding shifting power dimensions and elite networks can provide clear insights into how different forms and intensities of violence are incentivized by various political systems.
There is some form of violent, political disorder occurring in almost every state
Over the last 10 years, more than half the world’s population lived in direct contact with, or proximity to, significant political violence.1 We can no longer ask ‘which state has the riskiest drivers of violence?’ because many states are sites of significant disorder. The ‘center of gravity’ for political violence is not sub-Saharan Africa, where conflict levels have stayed relatively high and stable. Rather, the greatest rise in political violence in recent years has been in developed states including Syria, Brazil, Mexico, the Philippines, Ukraine, Libya and Egypt.
Modern political violence is no longer a scourge of low-income states. It is not solely a result of ‘state weakness’ or ‘fragility’: it can occur in states with a robust governing apparatus and economic growth. Conflicts in middle-income countries have similar aims as those in low-income countries: to secure control of the state, territory and rents in the hands of select political interests. Political violence is a tool of control used by the most powerful, a trend exacerbated by state building projects that have sought to ‘recentralize’ power after two decades of decentralizing and liberalizing efforts. Rather than working to resolve conflicts, it is often powerful elites that encourage shifts and increases in violence in order to cement their positions or to take advantage of changing political circumstances. There are clear benefits to engage in violence, and chief amongst them is that it is an effective weapon through which to garner political power. In short, politics causes political violence.
The overall effect is that economic growth and prospects are no longer direct indicators or outcomes of conflict. Middle-income states can be overtaken by violence, and low-income states with and without large-scale conflicts have failed to meet globally agreed-upon development goals, or significantly advance towards those goals in the past decade. These bleak facts suggest that it is difficult, if not impossible, to ‘develop out’ of conflict.
There is a violence market for militias and gangs
Political violence is increasingly transactional rather than motivated by group grievances. There are strong economic parallels between labor markets and the violence market, and it is equally important to focus on the buyer of violence as well as the supplier of force. The suppliers of force are predominately militias and gangs, and much of the recent rise and dissemination of violence is driven by these groups. The demand for low-level, intermittent conflict has risen, as the terms of exchange between gangs and politicians are cheap, transactional and shallow.2 The result is a hierarchy of armed groups that is reflected in price, actions and motives. The market perspective strips political violence of its grievance ‘motivation’ and instead concentrates on exchange terms, and the ways purveyors of violence can distinguish themselves within a crowded field of competitors. Violent ‘brands’ are ways that a group can do so: territorial, identity, religion, age, party, previous associations, costs, and modality of violence are each ways that groups compete for sporadic patronage.3,4 The buyer decides the cause; the seller provides the conflict. As the figure below demonstrates, across Africa, activity by militias/gangs and state forces accounts for the greatest increase in violence.
Politicians hire gangs to engage in violence around elections, to kill and threaten competitors and their supporters, seize territory, secure illicit trading routes, and increase their political leverage through force. Examples include vigilante killings of government opponents in the Philippines and the 2015 election violence in Burundi perpetrated by Imbonerakure militias recruited from ‘unemployed youth’ by the regime’s CNDD party. Communities are often forced into protection rackets where they are ‘taxed’ for ‘security’ (see recent patterns in Brazil5).
Outside of direct political engagements, violent groups may be motivated by explicit economic interests like securing control of illicit trade, resource and migration routes. Cartels are an extreme example of a parallel system with political and security agendas interwoven with economic and financial outcomes. Consider two examples: in Mexico, cartels create significant security vacuums and high levels of civilian harm (see figure below); in Mali, an extensive organized criminal network operating in 2012, coupled with a serious national security crisis and active transnational political groups, generated the current Sahel crisis. As a result, over 100 distinct militia groups are now operating across the Sahel.6 When these types of groups do collectively organize, and can fund activities through myriad illegal avenues, it leads to widespread violence and destruction.
State-building increases risks of violence against civilians
States and regimes remain uneven and biased in the security of their citizens and territory. While the international community engages in extensive state-building and stabilization efforts, the security of states has not resulted in the security of people. Attempts at recentralizing power amongst a growing circle of elites has increased the risk of violence and created new challenges for citizens. States and elite networks are not neutral parties: strong states use violence strategically to further entrench power; in weaker states, violence is perpetrated by many, equally powerful agents; and authoritarian states are populated with political elites who use violence to sustain their leverage and challenge competitors. A focus on achieving stability through the inclusion of violent elites can reduce conflict levels in the short term, but may increase the incentives for violent competition in the long term.
There is a great range in how states and citizens experience security: in places like Mexico and Burundi, active and latent groups dominate the security environment, while in Iran, Turkey and Ukraine, the level of per capita civilian killing is low, but perpetrated by the same small range of state, and state-associated groups. In countries like Ethiopia and Pakistan, the possibility of high numbers of ‘re-activated’ groups mean that civilians are at a heightened risk of mass violence, should the political environment change suddenly.
Demonstrations seem unable to effect structural political change
Peaceful protests and riots increased significantly in 2019 compared to 2018, across all regions. African rates of peaceful protests and riots almost doubled, while a third more protest activity occurred in South Asia. In recent years, the focus on citizen engagement, civil society and ‘popular uprisings’ suggests that this modality of political engagement can turn around increasingly authoritarian politics. Indeed, the successes in removing leaders in Sudan and Algeria in 2019 contributed to this global sentiment. But there are also reasons to be skeptical and reticent about the impact of popular movements: the ability of protesters to form a coherent political agenda is limited by internal fractures and state attempts at repression; in over 12% of all protests, demonstrators are attacked by police; nearly 7% with excessive violence by state security forces. Further, most protests movements have occurred in authoritarian states with significant levels of intra-regime divisive politics, and elites are prepared to use movements as leverage to oust leaders. Leadership changes in such governments often occur without significant protest movements — such as in Zimbabwe — and when they co-occur, elites can ‘adopt’ the public mandate to push for personally beneficial changes, as the recent government changes in Sudan suggests.
Finally, politicians have readily used the same tactics against civil society through their hiring of mobs to drown out or disassemble alternative voices and peaceful movements. Regimes are co-opting the same tools as civil society, but with far higher rates of violence. This is evident in the increased rates of mob violence where vigilantes, and those linked or hired temporarily by political parties, target civilians. Fatalities due to these riots rose by 22% in 2019, and in extreme cases, like India, they are responsible for nearly 2,000 civilian targeting events. Regimes have also responded by using technologies to track and target dissidents, break up movements, black out the internet during violent crackdowns, etc. Modern protest movements have not resolved these issues, despite technology advancements and global goodwill. They remain the most damning factors influencing social movements’ success.
Prediction, measurement and containment
A pressing question is whether we are caught up in short term, limited scrambles for power amongst many local groups, or whether this is a prelude to longer-term competition across small and larger powers, proxy wars and extensive instability. This brief serves as a diagnosis of the changes currently underway. It suggests that ignoring the new realities of conflict, or relying on previous policy prescriptions, will fail to contain and abate this disorder.
In the case of short-term scrambles, successful political settlements may decrease violence, but at the cost of reinforcing the ‘benefits’ of conflict for the powerful. This can lead to cycles of violence and limited political development. These short-term scrambles are often calculated and strategically aligned attempts to claim positions, voters, rents and territories. A significant problem will emerge if regimes, elites and militias see violence as the most successful way to compete. In longer term competitions, the protracted use of violence possible if those in power do not prioritize domestic stability or civilian safety. There are few incentives for peace in these environments without international support and cooperation aimed at changing the violent calculus of power.
Because peace does not necessarily follow conflict, and there exist so many varieties of, markets for, and benefits to political violence, a reasoned prediction for the future is difficult.7 More active groups make violence rates and geography less predictable, in part because most active groups have vastly different goals and behavior, organizational potential, financial incentives and prospects. Yet, the benefits of violence as a political and economic tool are clearly demonstrated in disorder. While political motives are linked to group formation, economic incentives allow for continuation and adaptation. Expanding our interpretation of how funding and financial opportunities heighten violence, and assessing how security environments have evolved, is a first step. A second step will be to grapple with how the fallout from two decades of intervention, stabilization, liberalization and power decentralization has created the opportunities for these forms of violence to proliferate. A central theme to violent disorder is that it adapts to the local political environment: these environments must incentivize lower violence through reliable alternatives for employment and competition, and they must impose sanctions on those who use and buy violence by denying them access to office, rents, territory and impunity.
1 This is assessed by the active presence of more than one armed, organized non-state, politically motivated violent group.
2 Kars de Bruijne (2019) “The making of a market: politicising gangs in Sierra Leone.” http://matsutas.com/big-men/the-making-of-a-market-politicizing-gangs-in-sierra-leone-by-kars-de-bruijne/ . See also recent work by De Waal (2016).
3 Dowd, Caitriona (2016) The emergence of violent Islamist groups: branding, scale and the conflict marketplace in sub-Saharan Africa. https://reliefweb.int/sites/reliefweb.int/files/resources/ACLED-Working-Paper-on-Violent-Islamist-Groups_August-2012.pdf ; https://rusi.org/publication/rusi-newsbrief/how-east-africa%E2%80%99s-terrorists-build-their-brand-strength
4 A market system co-exists with a more formal ‘patrimonial’ system of politicians with associated militias; and there is evidence of both state and non-state organizations hiring and disseminating this ‘hired labor’.
5 The Wilson Center (2019) “Brazil’s Violence Map Shows Alarming Trends.” https://www.wilsoncenter.org/blog-post/brazils-violence-map-shows-alarming-trends
6 ACLED (2019) “Press Release: Political Violence Skyrockets in the Sahel.” https://www.acleddata.com/2019/03/28/press-release-political-violence-skyrockets-in-the-sahel-according-to-latest-acled-data/; ACLED (2019) Mid-Year Update: Ten Conflicts to Worry about in 2019 https://www.acleddata.com/2019/08/07/mid-year-update-ten-conflicts-to-worry-about-in-2019/
7 Prediction efforts often use patterns from the past 50 years to suggest the general future direction. We suggest that post WWII and post-Cold War patterns present something of an aberration, and more recent patterns (post 2010) present more reliable ‘prior’ violence patterns. Prediction also strongly depends on data choice, and if data does not pick up smaller group actions, or local patterns, it is insufficient to explain a drastic rise in these very outcomes.