Most contemporary lethal violence does not occur in conflict zones, the majority of states most affected by lethal violence are not at war, and the levels of lethal violence in many nonconflict settings are higher than in war zones. Much of this nonwar violence is organized, not random, and political in nature. A narrow focus on wars and formal armed conflicts thus obscures the high levels of everyday violence and insecurity around the world. This essay makes the case that adopting a broad understanding of political violence – including violence committed by the state and its agents, and nonphysical violence as the violation of basic rights – is essential to gain insight into the causes and consequences of, and to frame appropriate responses to, war and violence in the twenty-first century.

On December 17, 2010, Mohamed Bouazizi set himself alight in a small city south of Tunis as a violent and ultimately suicidal protest against the repeated humiliation and harassment he suffered from local officials. Street demonstrations broke out the next day in Sidi Bouzid and spread to Tunis ten days later, and on January 14, 2011, Tunisian President Ben Ali resigned. Demonstrations spread across the Middle East, from Libya to Yemen. In Egypt, President Hosni Mubarak resigned on February 11, and after a brief democratic experiment, the military seized power and drastically curtailed vocal opposition. The mid-February protests in Libya spiraled into civil war, international intervention, and insecurity and state collapse. Syrian protests between March and July 2011 also spiraled downward into civil war, which has since mutated into a regional conflict involving Iraq, Syria, the Islamic State, and various proxies, third parties, and Western volunteers and recruits. More than 140,000 people – and possibly up to 400,000 – have been killed to date.1

This synopsis illustrates the challenge in our quest to explain the causes and consequences of armed conflict, or to prevent and resolve conflicts and mitigate their effects. International relations as a discipline focuses on the “war” side of this situation – in Syria, Libya, and Northern Iraq – and starts its analysis when large-scale violence has already occurred. This, coupled with a weak understanding of how war is related to the broader backdrop of political violence (and violence in general), obscures the mechanisms and processes through which everyday “dynamics of contention” can underlie and lead to large-scale outbreaks of violence.2 An exclusive focus on war means we know little about how we get from such things as state repression or group violence to civil war – from Sidi Bouzid to Syria – and what the consequences might be for international and regional order.

There are four good reasons for moving, empirically and conceptually, “beyond war” to study political violence in general terms and from a holistic perspective. A narrow empirical focus on war obscures the scope and scale of intentional harm associated with “nonwar” forms of violence. It understates the human costs and consequences of war-related violence. It limits the scope of debate on moral and legal responsibility to forms of violence covered by just war principles and international humanitarian law, while obscuring the morally equivalent responsibility that governments should face for other forms of violence and harm committed in their name. Finally, it hinders understanding of the way different forms of violence may be linked through processes that escalate and exacerbate conflicts, and that may have broader impacts on state formation, state disintegration, and regional order.

In this essay, I will unpack these claims, and make the case for adopting a three-dimensional understanding of political violence – defined as violence used for explicitly stated political ends, or that undermines and challenges the state's legal monopoly over the legitimate use of force, or that implicates the state and its repressive apparatus – as essential for gaining insight into the causes and consequences of, and framing appropriate responses to, war and political violence in the twenty-first century.

Four facts about contemporary violence make good starting points to broaden our perspective: 1) most lethal violence does not occur in conflict zones; 2) the majority of states most affected by lethal violence are not at war; 3) the levels of lethal violence in some nonconflict settings are higher than in war zones; and 4) much of this violence – but we do not know how much – is organized, nonrandom, and in some sense political.

The first three claims can be substantiated by approaching violence from a sociological, criminological, or public health perspective. An average of 508,000 people died violently around the world each year between 2007 and 2012; only about seventy thousand – or 15 percent of them – died in wars or formal armed conflicts.3 The remainder – more than four-hundred thousand – died in nonwar contexts, and a significant proportion of these (around 5 percent) died at the hands of the state or its agents.4 Even if (as I argue below) this picture of the number of war-related violent deaths is misleadingly low, it shows that war is only one piece of a much larger puzzle of lethal violence.

Figure 1 standardizes violent deaths in the thirty most violent states. While there may be uncertainty around the specific rankings, these numbers are conservative, and are based on aggregating conflict and nonconflict violent death data.

Figure 1

Violent Deaths per 100,000 of Population, Annual Average, 2007–2012

Figure 1

Violent Deaths per 100,000 of Population, Annual Average, 2007–2012

Close modal

Some of the most violent countries in the world, such as Syria, Afghanistan, South Sudan, Pakistan, and Yemen are in war zones. But some Latin American and Caribbean countries that are not at war, including Honduras, Venezuela, and El Salvador, are more dangerous places to live than Afghanistan. Other countries with high levels of lethal violence, including Brazil, Colombia, Mexico, South Africa, and the Democratic Republic of the Congo, are also not formally at war.

For the fourth claim, there are few systematic overviews of the scope of political violence in nonwar settings, and we lack common definitions of what constitutes communal violence, terrorist attacks, politically motivated violence, organized criminal and gang violence, riots, and so on.5 Cross-national comparisons of the scale and distribution of violence within states are uncommon, and most country-level and microlevel work is disconnected analytically from a larger picture.6 Adopting a narrowly criminological or legal perspective and labeling all nonconflict deaths as “homicides” is also misleading, since “homicide” conjures up a form of interpersonal violence that is individual, unorganized, relatively random, not linked to broader dynamics, and essentially apolitical (and very rare in advanced industrialized states). This is an inadequate way to think about the more than fifty thousand violent deaths in cartel-related warfare in Mexico, or land-rights disputes in Yemen that claim several thousand lives a year (and which have now escalated into full-scale war).7 Violence in many global hot spots can and does have large-scale sociopolitical consequences, and is not marginal or caused by deviant individual behavior but rather part of a dynamic of political contention that erupts into violence. A focus only on lethal violence in war thus gives us a highly misleading picture of the current global scale and intensity of violence in general, and of politically salient violence in particular. It also narrows our normative gaze in ways that diminish responsibility and accountability.

Zooming in on war-related violence also highlights limitations in the way in which the human and social costs of war are presented. Most trend analysis of wars is based on a threshold of either one thousand or at least twenty-five battle-related deaths per year.8 Based on this, we obtain the picture of recent trends shown in Figure 2.

Figure 2

Armed Conflict by Region, 1946–2014

Figure 2

Armed Conflict by Region, 1946–2014

Close modal

In this account, interstate war is all but obsolete. There were no interstate wars in 2013, and only seven ongoing wars with more than one thousand battle deaths. Lowering the threshold to include conflicts with twenty-five or more battle deaths, there were thirty-three ongoing armed conflicts in 2013, of which twenty-four were internal conflicts and nine were “internationalized internal conflicts” in which external parties were actively engaged.9 The overall number of wars has also declined to around thirty per year. And the human costs of war have also allegedly declined, with the annual total of battle deaths in these data sets hovering around sixty thousand per year, although this figure has risen since 2013 due to intense fighting in Syria and Iraq.

There are roughly four hundred thousand malaria deaths and 1.24 million road traffic deaths worldwide per year, and from this viewpoint, war is a relatively minor and declining cause of human suffering, a form of deviant behavior less relevant today than at any time in human history.10 I will return to some criticisms of this declinist argument below, but for now, note that harm intentionally committed against fellow humans has significantly different consequences, practically and ethically, from the accidental or natural disasters that befall us. Large-scale violence and insecurity destroys the fabric of communities, erodes social and human capital, negatively impacts economic development, and undermines political legitimacy and stability in a qualitatively different fashion from disasters or accidents. Violence is a psychologically intense human interaction and scars victims, witnesses, and perpetrators in ways that are difficult to capture and quantify. Political institutions are responsible for protecting people against intentional harm, whether committed by state agents or other individuals, a responsibility that goes to the heart of their raison d'être as a means to escape the state of nature.

A focus on the numbers can highlight two important ways in which widely cited figures systematically underestimate the impact of armed conflict. The first is the undercounting of the direct victims of lethal violence within armed conflicts, which arises from methodological choices and data limitations of conflict data projects that rely on media reports or official figures. The second, much larger shortcoming is the lack of attention to the large-scale indirect consequences, some of which are lethal, of conflict and violence. Controversies over war deaths in Iraq illustrate well the problem of systematic undercounting. For the period 2003 to 2015, Iraq Body Count (ibc) documents between 144,384 and 166,085 civilian deaths from conflict-related violence, and a total of 220,000 deaths, including combatants.11 The Uppsala Conflict Data Program (ucdp) data, however, record only 53,361 deaths over this period – around one-third of the ibc's total.12

The most violent year in that conflict (2006) provides an even more divergent picture. ucdp data estimate battle-related deaths in Iraq at 3,931, with a higher total of 5,840–7,028 deaths, including “the fighting between the government and unclear perpetrators, such as … the victims of roadside bombs where no group claimed responsibility.”13 The United Nations, on the other hand, using information from morgues, hospitals, and municipal authorities, concluded that there were more than 34,000 violent deaths in Iraq in 2006, a figure that includes all types of violence, such as conflict deaths, attacks on civilians, homicides, criminality, and domestic violence.14 We thus have numbers that range from three thousand to thirty thousand – an entire order of magnitude – depending on what you are counting.15 Since much of the violence afflicting Iraq was undoubtedly linked to the ongoing conflict, it is difficult to grasp the impact of violence and insecurity on Iraqi politics and society based on the lower and misleading figures.16

The large-scale indirect and lethal impact of conflict on populations is discussed in Paul Wise's contribution to the companion to this issue of Dædalus (forthcoming winter 2017).17 What is important to note here is that the indirect impact of conflict does not rest on a narrow concept of agency: identifiable perpetrators, acts causing immediate harm (use of force, threats, and displays of force), and individual victims. Instead, it includes diffuse actions by corporate or collective agents, such as the state or armed militias, which result in large-scale population displacement and/or loss of access to basic needs, such as food, shelter, or basic health services and treatment. This results in increased mortality, especially among the young, old, and otherwise vulnerable, and can be measured by the increase in mortality over what would be expected if the conflict and violence had not erupted (usually based on a prewar estimate). In several recent conflicts, including in South Sudan, Darfur, Burundi, Iraq, and Liberia, indirect deaths have represented up to ten times the number of direct victims of violence.18 In the Democratic Republic of the Congo, perhaps the most dramatic recent case, a reported 3.3 million people died in the violent conflict between 1998 and 2002.19 While that figure may be too high, only a relatively small percentage (perhaps no more than 10 percent) of the victims were killed violently; the remainder died of easily preventable causes triggered by forced displacement, loss of access to life necessities, and the lack of basic care. These deaths are still narrow “physical harm” with lethal consequences, and are only indirect in their agency.

Numbers tell only one part of the story, whether we are concerned with trend analysis, causes and consequences, or legal and ethical responsibilities. But although scholars have moved away from a focus on large-scale organized violence to analyze such things as terrorism, nonstate armed groups, the microdynamics of civil war, and subnational and transnational violence, the political dimension of violence itself remains underconceptualized. The question “what makes violence political?” has no simple and unambiguous meaning and is wrapped up with questions of legitimacy and morality. Most scholarship assumes that political violence can be identified and categorized by focusing on the degree and scale of organization of the violent actors, the meaning and motivation or purpose of the acts, or the nature of the act itself. None of these criteria by themselves are sufficient, however, without clarifying what we mean by “violence” and “political.”

The dominant focus has been on the physical nature of the act of violence causing death or injury, as the data sources used for conflict analysis demonstrate. This minimalist conception of violence has some important (and unhelpful) consequences for how we can think about the relationships between different forms of violence.20 Criminologists and sociologists go beyond lethal violence, but still concentrate on criminal behavior, gang violence, riots, demonstrations, and other acts that cause physical harm and destruction. All of these approaches share a somatic understanding of violence revolving around “the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community, which either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment, or deprivation.”21

Attempts to move beyond a somatic understanding of violence have included psychological violence, sexual and gender-based violence (in which the sex of the victim is intrinsic to the choice of acts, some of which do not involve physical harm), violence by deprivation, neglect or omission, and such things as systemic, structural, or symbolic violence. These broadenings are, however, almost never linked to the somatic or physical harm conception.22 In almost all cases, albeit with some important exceptions, the study of violence has been inextricably linked to the illegitimate use of physical force to cause harm, with the issue of legitimacy shaping the choice of words: illegitimate acts are “violent”; legitimate acts are merely “use of force.”

It is probably not desirable to take on board all of these different potential meanings of violence. But if we wish to understand the links between different kinds of violence (war and nonwar), processes of escalation, and social and political consequences, at least two kinds of neglected violence should be brought into the picture. The first is violence that has been “made legal”; the most widespread forms of legal violence being the use of force by authorized agents of the state – police, gendarmes, paramilitaries, and others – especially when this goes beyond what would be considered the legitimate use of force by such agents. For example, the “police in Nigeria commit extrajudicial killings, torture, rape, and extortion with relative impunity … routinely carry out summary executions of persons accused or suspected of crime; rely on torture as a principal means of investigation; commit rape of both sexes … and engage in extortion at nearly every opportunity.”23 Hundreds of Nigerians are killed each year by the police, and in Jamaica, the Dominican Republic, Rio de Janeiro, Sao Paolo, and Nairobi, police and other extrajudicial executions are commonplace, representing between 13 and 43 percent of all violent deaths.24 Deaths in violent political unrest, such as in Egypt during the 2011 uprisings, also fall in this category, since they are seldom prosecuted.25 Of course, the bloody twentieth century was rife with what Zbigniew Brzezinski called “politically motivated carnage,” which, according to his estimates, took upward of eighty million lives outside of actual combat.26 Linked to the excessive but legal forms of violence is what philosopher Vittorio Bufacchi has termed “violence as violation”: forms of violence that implicate the state, but which do not involve brute physical force to wreak harm. This form of violence includes the entire apparatus of repression and the range of “personal integrity rights … concerned with individual survival and security, such as freedom from torture, ‘disappearance,’ imprisonment, extrajudicial execution, and mass killing.”27

Large-scale “legal violence” has two important consequences. First, when perpetrated in nonconflict contexts, it undermines respect for state security institutions, creating a vacuum in which other violent actors can operate with relative impunity and even some rough legitimacy.28 Second, it is often wrapped up in pre-conflict dynamics, as the weakening legitimacy or efficiency of state institutions facilitates the resort to violence by diverse actors to resolve conflicts or express discontent and opposition. The resort to violence in response to “legal” forms of violence can, under certain conditions – as the Tunisian case illustrates – have powerful social and political effects.

A broad conception of violence – including its political dimension – is thus critical to understand how different forms of violence may be linked to war and armed conflict. Yet most authors do not explicitly define political violence (or war), and simply work with categories such as communal conflict, ethnic conflict, civil war, or interstate war in an additive approach. Some have been opportunistic in their case studies and research.29 Others have defined political violence as being “explicitly for a designated and reordering purpose: to overthrow a tyrannical regime, to redefine and realize justice and equity, to achieve independence or territorial autonomy, to impose one's religious or doctrinal beliefs,” focusing on violent nonstate actors contesting the legitimacy of the existing order.30 Such approaches are either too unfocused or too narrow, and Christian Davenport has it right when he points out that “researchers have paid far more attention to the evils done against governments (and citizens) by dissidents, rebels, and terrorists than to the evils done by presidents, the police, military, secret service, national guard, and death squads against those within their territorial jurisdiction.”31 Political violence, including war and armed conflict, should be defined as including violence used for explicitly political ends, or that directly undermines and challenges the state's legal monopoly over the legitimate use of force, or that implicates the state and its repressive apparatus. These various forms of political violence have often profound consequences on the legitimacy and functioning of the state, and engage the responsibility of state and nonstate armed actors.

Extending how we think about political violence has implications for just war theory, and for the ethics of violence. The legal case around challenges to and expansions of just war theory – to cover such things as nonstate actors, undeclared wars and asymmetric warfare – is cogently argued by Allen Weiner and Seth Lazar in the Winter 2017 issue of Dædalus.32 As they note, just war doctrine developed to deal with state parties in a formally declared war, and war and armed conflict are defined as occurring “whenever there is … protracted armed violence between governmental authorities and organized armed groups or between such groups within a state” that crosses a minimum threshold of intensity and whose participants possess “a certain command structure and have the capacity to sustain military operations.”33 This mirrors the more restrictive definitions to determine what counts as a war, and is silent on other forms of violence, including the indirect victims of armed conflict. It also often legitimizes the use of lethal force by states by putting “a conceptual and moral gulf between the resort to such force … for political purposes by state agencies and its political employment by nonstate actors.”34

We lack, however, clear and integrated concepts to help us understand “just and unjust political violence.”35 Such concepts would have to deal in a consistent and robust fashion with normative and legal issues along five dimensions in addition to the just war doctrine: state violence against citizens; the state's “responsibility to protect” them; the international community's responsibility, if any; the responsibility of nonstate armed actors before national and international criminal law; and the responsibility of state and nonstate actors for the indirect (but still attributable) consequences of violent or repressive acts.

Such a reflection is beyond the scope of this essay. But the international legal toolkit is not empty, and piecemeal normative reflections abound. There may be, for example, a trend toward taking the long-term consequences of war into account when assessing proportionality. The extensive literature on the “responsibility to protect” (explored by Jennifer Welsh in this issue) is rooted in just war reflections. There are also generally agreed upon principles, such as the “Basic Principles on the Use of Force and Firearms by Law Enforcement Officials” to govern legal interventions.36 The Genocide Convention, which includes “deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part” (article 2), can be seen as implying that indirect forms of violence short of mass murder should be part of deliberations about responsibility and accountability. The UN Convention against Transnational Organized Crime (2003) and its protocols partly address common legal standards, and a patchwork of national criminal laws deal with participation in, or membership of, organized violent groups (gangs, organized criminal groups) even when these are not engaged in large-scale violence.

But in general, the focus of the ethical debate remains on war and not on other kinds of violence, and we have few reasons to think the patchwork of legal doctrine and ethical reflection adds up to a robust and consistent normative framework. Sadly, the framework for holding state agents responsible and accountable to citizens, to provide security and protection, and to refrain from unjustly harming them is weak – as recent cases of excessive use of force by police officers in the United States illustrate. Broadening our perspective to examine the many forms of violence beyond war opens a window on this lacuna.

Linking war and other forms of political violence, broadening our understanding of political violence, and adopting a holistic approach to measuring and monitoring draws our attention to phenomena that are traditionally ignored in war and conflict studies.37 But beyond this, it has to add value to our understanding of the roots, dynamics, and consequences of contemporary political violence, especially since war is not distinct from other violence dynamics within states and societies, and all forms of violence have generative effects that “constitute, uphold and organize existing social relations.”38 Traditionally, however, scholars have focused on the consequences of large-scale interstate or internal conflicts between formally organized actors, and considered lower-level violent interactions to be separate from, or the product of, these macrolevel processes.

But the dynamics of violence could work both ways: microlevel violence can feed upward into large-scale conflict: from Sidi Bouzid to the Islamic State. We know that the metanarrative of violent conflict (often political or ideological) can encompass a host of microlevel and localized disputes that are not directly connected with the metanarrative, and Yale professor Stathis Kalyvas paints a compelling picture of how the macro/political and micro/private forms of violence are intertwined.39 There is thus good reason to conclude that large-scale violence can escalate upward from a host of deeply entrenched and enduring microlevel violent exchanges or struggles for power.40 We also know that genocide and state violence usually unfold as part of other violent interactions, including interstate and civil war. Understanding how this can be so has important implications for conflict resolution and post-conflict peacebuilding, or for strategies of intervention.

In Pakistan, for example, while violence is concentrated in the Northwest Frontier Province and Karachi, there are at least twenty-two separate “dyads” or “triads” of conflict, most involving the state, Lashkar-e-Islam, and/or different Taliban factions. This intense violence has recently resulted in two to three thousand violent deaths annually, and is accompanied by high levels of urban political violence in major cities. Total deaths from political violence in Pakistan run around five thousand per year, including “casualties in terrorist attacks, operations by the security forces and their clashes with militants, ethnopolitical violence, drone attacks, inter-tribal and intermilitants clashes, sectarian clashes, religious/communal violence, cross-border attacks and clashes, criminal gangs' clashes with one another and with the security forces.”41 These are not disconnected forms of violence, nor is the struggle with radical Islam necessarily the most important. The different forms of violence are linked in complex ways, and add up to a worrying picture in which the security forces of the state are deeply implicated. Pakistan may be extreme, but it is not unique.

A second important generative consequence of violence is temporal, and can be seen in the relationship between conflict and post-conflict violence. The idea that different forms of violence may be linked over time is not new: observers of the American Civil War and post – World War I Europe postulated that rising criminal violence in America and Britain was the product of the social and cultural dislocative effects of, respectively, the Civil War and World War I.42 A similar story can be told about high levels of post-conflict violence in parts of Central America, such as El Salvador and Guatemala, now linked to gang-related violence, vigilantism, and state violence.43 Again, these are extreme, but not unique, cases.

A third form of mutual reinforcement can be seen where violent actors pursuing political and private (criminal) goals interact and support each other, making a separation between the study of political and criminal violence difficult. The two are linked in places such as Afghanistan and Colombia, where drug trafficking and control of smuggling routes are objects of contestation between armed groups, who in some cases forge links with criminal groups and cartels.44 In West Africa, major extractive and predatory enterprises have fed the emergence of warlords in Liberia, Sierra Leone, and the Eastern Democratic Republic of the Congo, who do not necessarily pursue ideological or state-building aims, but seek to capture the state to pursue predatory or neopatrimonial ends.45 In places like Iraq, such “dual-purpose violence” has been characteristic of both the politically motivated violence of insurgents and organized criminality since 2003. Individuals, often with a criminal background, financed future insurgent activities by participating in the looting shortly after the fall of the Ba'ath regime, and impoverished looters targeted the political elite “in acts of political revenge but also to satisfy long accumulated material needs.”46 One cannot understand the dynamics of politically motivated armed conflicts without seeing how they are tied up with nonpolitical violence and criminal activities.

Finally, seemingly disconnected forms of violence can be linked in complex ways. Sexual violence in and after conflicts, especially in parts of West Africa, is connected to broader conflict and violent dynamics, and “the specific, often exclusive, focus on sexual violence … hampers our understanding of the relationship between sexual violence and other (supposedly) ‘ungendered’ violence…. These forms of violence are … manifestations of the same systemic failures and mechanisms.”47 This is not a unidirectional chain where conflict violence causes higher levels of sexual and gender-based violence. There are deeper mechanisms and processes at work, as states with lower levels of gender equality and higher levels of violence are more likely to initiate the use of force in interstate disputes, to be involved in interstate conflicts, to be less peaceful internationally, to be less compliant with international norms, and to be less likely to have good relations with neighboring states.48

The nature and impact of contemporary political violence cannot be measured solely by such things as changes in levels of lethality or trend analysis of the number of ongoing conflicts. While extensive debates around the “end of war” or the “end of violence” may shed some light on the place of war and lethal violence in social life, they tell us little about the order-creating or order-destroying effects of war and violence. They may also be highly misleading, as numerous critics have pointed out, both in their presentation of data and their interpretation of causes. Battle deaths (and homicides, too) may have declined precipitously, at least in the developed world, due to better medical care and interventions.49 Figures on the lethality of war now focus on direct conflict deaths, excluding the indirect deaths (starvation, disease) that accounted for huge numbers of victims in previous periods. And the age distribution of populations (increases in life expectancies), coupled with the fact that most violence is committed by men between the ages of fourteen and twenty-nine, means that lethal violence levels will inevitably decline as populations age and life expectancy increases.

If all the trends are positive, why are persistent expressions of unease and insecurity at the highest levels so widespread?50 The source of unease stems not from the numbers, but from the changing nature and unclear consequences of contemporary violence. Even low-level but sustained violence can have long-term systemic and structural consequences that affect processes of state-building and decay, create cross-border sources of regional instability, and distribute power away from state institutions. The erosion of the state's practical monopoly over the use of violence, the steady proliferation of more powerful and sophisticated weapons to nonstate armed actors, the relative ease with which “violence entrepreneurs” can operate in many parts of the world, and the weak and fragile nature of many state institutions intended to provide safety and security are all worrying trends.

In order to think clearly about the impact of war, armed conflict, and political violence in the twenty-first century, we thus have to go beyond war, terrorism, and civil conflict to look at all sources, causes, and consequences of lethal violence. Many of these causes, such as governance failure or state collapse, or the process of state and society-building, are intensely political, have national and international implications, and are interlinked. To make sense of trends in contemporary war and armed conflict and to understand the sustainability or generalizability of these trends, we also need a serious analysis of the scope conditions that governed the emergence of this relatively peaceful state (and states).

Author's Note: Thanks to Scott Sagan, Seth Lazar, Barry Posen, and participants at the American Academy of Arts and Sciences' workshops on “New Dilemmas in Ethics, Technology and War” for useful comments on an earlier draft.


The lower figure is the total of “martyrs” and “regime fatalities” from the Violations Documentation Center in Syria; the higher figure is from Staffan de Mistura, UN special envoy for Syria. See the Violation Documentation Center in Syria, “Statistics for the Number of Killed,”; and Staffan de Mistura quoted in “Syria Death Toll: UN Envoy Estimates 400,000 Killed,” Al Jazeera, April 23, 2016,


See Doug McAdam, Sidney Tarrow, and Charles Tilly, Dynamics of Contention (Cambridge: Cambridge University Press, 2001); and Lorenzo Bosi, Charles Demetriou, and Stefan Malthaner, eds., Dynamics of Political Violence (Farnham, United Kingdom: Ashgate, 2014).


These figures do not cover the most intense period of fighting of the Syrian conflict, and conflict death totals are much higher since 2012.


See Geneva Declaration Secretariat, Global Burden of Armed Violence 2015: Every Body Counts (Cambridge: Cambridge University Press, 2015).


One exception is Monty G. Marshall, Third World War (Oxford: Rowman and Littlefield, 1999). See also Monty G. Marshall, “Major Episodes of Political Violence, 1946–2015,” Center for Systemic Peace, May 25, 2016,


Laia Balcells and Patricia Justino, “Bridging Micro and Macro Approaches to Civil Wars and Political Violence: Issues, Challenges and the Way Forward,” Journal of Conflict Resolution 58 (8) (2014): 1343–1359; and Patricia Justino, Tilman Brück, and Philip Verwimp, eds., A Micro-Level Perspective on the Dynamics of Conflict, Violence, and Development (Oxford: Oxford University Press, 2013).


See United Nations Office of Drugs and Crime, Global Study on Homicide 2013 (Vienna: United Nations, 2014). On Yemen, see Small Arms Survey, Under Pressure: Social Violence over Land and Water in Yemen (Geneva: Small Arms Survey, October 2010). On Mexico, see Paul Kenny and Mónica Serrano, eds., Mexico's Security Failure: Collapse into Criminal Violence (London: Routledge, 2012). See also International Institute for Strategic Studies, Armed Conflict Survey 2015 (London: International Institute for Strategic Studies, 2015), which identifies forty-two active conflicts and 180,000 fatalities in 2014, and acknowledges this by including the deaths in El Salvador, Guatemala, Honduras, and Mexico as conflict-related deaths.


See Uppsala Conflict Data Program, Department of Peace and Conflict Research, “Definition of Armed Conflict,” Uppsala Universitet,


Uppsala Conflict Data Program, Department of Peace and Conflict Research, “Uppsala Conflict Data Program,” Uppsala Universitet,


For World Health Organization estimates, see World Health Organization, “Global Health Observatory (gho) Data: Road Traffic Deaths,”; and World Health Organization, “Media Centre: Malaria,” For the declinist thesis, see Steven Pinker, The Better Angels of our Nature: The Decline of Violence in History and its Causes (London: Penguin, 2011); and Joshua Goldstein, Winning the War on War: The Decline of Armed Conflict Worldwide (London: Penguin, 2011).


Iraq Body Count, “Documented Civilian Deaths from Violence,” (accessed October 15, 2015).


The result of the sum of “war and minor conflict,” “nonstate conflict,” and “one-sided violence” categories. See Uppsala Conflict Data Program, Department of Peace and Conflict Research, “Number of Conflicts: 1975–2015,” Uppsala Universitet,


The Uppsala Conflict Data Program also pointed out that “battle-related deaths were just a small part of the violence in Iraq during the year, as much violence was directed against civilians.” See ibid.


United Nations Assistance Mission for Iraq, Human Rights Report 1 November–31 December 2006, United Nations Iraq, Iraq Body Count figures for 2006 total 29,451 civilian deaths.


Iraq is not a unique case. In the Bosnian War (1992–1995), Uppsala Conflict Data Program data tabulate 29,113 deaths (all forms of conflict), while the authoritative post-conflict studies estimate 104,732 deaths. See Ewa Tabeau and Jan Zwierzchowski, “The 1992–95 War in Bosnia and Herzegovina: Census-Based Multiple System Estimation of Casualties Undercount,” paper for the International Research Workshop on “The Global Costs of Conflict,” February 1–2, 2010,


Penny Green and Tony Ward, “The Transformation of Violence in Iraq,” British Journal of Criminology 49 (2009): 609–627.


Paul H. Wise, “The Epidemiologic Challenge to the Conduct of Just War: Confronting Indirect Civilian Casualties of War,” Dædalus 146 (1) (forthcoming winter 2017).


For a range of estimates for several recent conflicts, see Geneva Declaration Secretariat, Global Burden of Armed Violence (Geneva: Geneva Declaration, 2008), 40, 46–47.


There is controversy over these figures, with estimates for the later (less violent) period of 2001–2007 ranging from 2.8 million excess deaths to “only” 860,000. See Human Security Research Group, Human Security Report 2009/2010: The Causes of Peace and the Shrinking Costs of War (Oxford: Oxford University Press, 2011), 123–131.


Vittorio Bufacchi, “Two Concepts of Violence,” Political Studies Review 3 (2005): 193–204. See also Mary Jackman, “Violence in Social Life,” Annual Review of Sociology 28 (2002): 387–415.


Etienne Krug et al., World Report on Violence and Health (Geneva: World Health Organization, 2002), 5.


Tim Jacoby, Understanding Conflict and Violence (London: Routledge, 2008), chap. 3.


Open Society Institute, Criminal Force: Torture, Abuse, and Extrajudicial Killings by the Nigeria Police Force (New York: Open Society Institute, 2010), 53. See also Amnesty International, Killing at Will: Executions and Other Unlawful Killings by the Police in Nigeria (London: Amnesty International, 2009).


See Todd Foglesong and Christopher Stone, “Measuring the Contribution of Criminal Justice Systems to the Control of Crime and Violence: Lessons from Jamaica and the Dominican Republic,” Faculty Research Working Papers series (Cambridge, Mass.: John F. Kennedy School of Government, 2009), 16–17.


Amnesty International, Egypt Rises: Killings, Detentions and Torture in the “25 January Revolution” (London: Amnesty International, 2011).


His estimate of the twentieth-century total for war deaths (civilians and combatants), indirect conflict-related deaths, and “political motivated carnage” is between 167–175 million. See Zbigniew Brzezinski, Out of Control: Global Turmoil on the Eve of the 21st Century (New York: Touchstone, 1993), 15–17.


Christian Davenport, “State Repression and Political Order,” Annual Review of Political Science 10 (2007): 2; and Bufacchi, “Two Concepts of Violence.”


Jovana Carapic, Order and Authority within and Beyond the State (PhD diss., Graduate Institute of International and Development Studies, Geneva, 2013).


The cases chosen by Tilly et al. include the French Revolution, the American civil rights movement, Italian post-1968 unrest, Mau Mau violence in Kenya in the 1950s, the uprising in the Philippines in the 1980s, Hindu-Muslim conflict in India, the anti-apartheid movement in South Africa, and Soviet disintegration. See also the cases in Bosi et al., Dynamics of Political Violence.


David Apter, ed., The Legitimization of Violence (New York: New York University Press, 1997), 5.


Davenport, “State Repression and Political Order,” 1.


Allen S. Weiner, “Just War Theory & the Conduct of Asymmetric Warfare,” Dædalus 146 (1) (forthcoming winter 2017); and Seth Lazar, “Evaluating the Revisionist Critique of Just War Theory,” Dædalus 146 (1) (forthcoming winter 2017).


International Committee of the Red Cross, “How is the Term ‘Armed Conflict’ Defined in International Humanitarian Law?” icrc Opinion Paper, March 2008, 3–4; and International Criminal Tribunal for the Former Yugoslavia, The Prosecutor v. Dusko Tadic, Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction, October 2, 1995, para. 70.


C. A. J. Coady, Morality and Political Violence (Cambridge: Cambridge University Press, 2006), 3.


I am indebted to Scott Sagan for the term and the insights in this paragraph.


Adopted by the Eighth United Nations Congress on the Prevention of Crime and the Treatment of Offenders, Havana, Cuba, 1990,


There are of course major exceptions. See, for example, Benjamin Valentino, Final Solutions: Mass Killing and Genocide in the 20th Century (Ithaca, N.Y.: Cornell University Press, 2004); and Martin Shaw, War and Genocide (Cambridge: Polity, 2003).


Claudio Colaguori, “Symbolic Violence and the Violation of Human Rights: Continuing the Sociological Critique of Domination,” International Journal of Criminology and Sociological Theory 3 (2) (June 2010): 392.


Stathis Kalyvas, “The Ontology of ‘Political Violence’: Action and Identity in Civil Wars,” Perspectives on Politics 1 (3) (2003): 475–494.


Charles Brockett, Political Movements and Violence in Central America (Cambridge: Cambridge University Press, 2005); and Andreas Wimmer, “War,” Annual Review of Sociology 40 (2014): 173–197.


Pak Institute for Peace Studies, Pakistan Security Report 2012 (Islamabad: Pak Institute for Peace Studies, 2013), 7.


Edith Abbott, “The Civil War and the Crime Wave of 1865–70,” Social Service Review 1 (2) (1927): 212–234; and Dane Archer and Rosemary Gartner, “Violent Acts and Violent Times: A Comparative Approach to Postwar Homicide Rates,” American Sociological Review 41 (6) (December 1976): 937–963.


Ellen Jane Sharp, Vigilante: Violence and Security in Postwar Guatemala (PhD diss., University of California, Los Angeles, 2014).


Jonathan Goodhand, “Corrupting or Consolidating the Peace? The Drugs Economy and Post-Conflict Peacebuilding in Afghanistan,” International Peacekeeping 15 (3) (2008): 405–423.


Séverine Autesserre, The Trouble with the Congo: Local Violence and the Failure of International Peacebuilding (Cambridge: Cambridge University Press, 2010).


Green and Ward, “The Transformation of Violence in Iraq,” 618–619. Notorious organized crime figures also reportedly helped insurgent cells fund their activities through kidnappings, bribery, and highway robberies.


Maria Eriksson Baaz and Maria Stern, The Complexity of Violence: A Critical Analysis of Sexual Violence in the Democratic Republic of Congo (DRC) (Uppsala: The Nordic Africa Institute, 2010), 13. See also Dara Kay Cohen, “Explaining Rape during Civil War: Cross-National Evidence (1980–2009),” American Political Science Review 107 (3) (2013): 461–477.


Mary Caprioli, “Primed for Violence: The Role of Gender Inequality in Predicting Internal Conflict,” International Studies Quarterly 49 (2) (2005): 161–178; and Valerie M. Hudson, Mary Caprioli, Bonnie Ballif-Spanvill et al., “The Heart of the Matter: The Security of Women and the Security of States,” International Security 33 (3) (2009): 7–45.


I have not space to present the critique by anthropologists, sociologists, and historians of Pinker's claims. See Tanisha Fazal, “Dead Wrong? Battle Deaths, Military Medicine, and Exaggerated Reports of War's Demise,” International Security 39 (1) (2014): 95–125; Anthony Harris, Stephen H. Thomas, Gene A. Fisher et al., “Murder and Medicine: The Lethality of Criminal Assault, 1960–1999,” Homicide Studies 6 (2) (2002): 128–166; and Benjamin Zeimann, “Histories of Violence,” Reviews in History, April 12, 2012, For a supportive review, see Human Security Research Group, Human Security Report 2013 (Vancouver: Human Security Press, 2013).


Including, for example, Michael Ignatieff, “The New World Disorder,” New York Review of Books, September 25, 2014; or statements by various high-ranking American security and intelligence officials.