By Christian Davenport and Scott Gates
Historically, those of us who rigorously study political conflict and violence have adopted the approaches and rhetoric of, and found positions within, the International Relations field (e.g., the Conflict Processes section at APSA or the Peace Science Society). This work has been associated with “interstate” war and more recently civil war. Somewhat less prominently, research also addressed “intrastate” activity (e.g., ordinary resistance, domestic spying/covert action, strikes, protest, terrorism, human rights violation/state repression, revolution, and genocide). This work is associated more with comparative as well as, in some cases, American politics. We note the fields here because quite frequently scholars (when they are searching for something to cite or someone to hire), students (as they look for someone to read), journalists (as they look for something to orient their studies around or someone to quote), and ordinary citizens (when they want to understand what is taking place around them) adopt the language, ideas, and approaches of the interstate approach without acknowledging that it might not be appropriate or effective at addressing other forms of conflict. Indeed, quite frequently on this blog someone will mention that “international scholars blah blah blah” without noting that some interested in the topic are comparativists or Americanists and do not adopt or even accept the International Relations take on the subject of inquiry.
This leads us to ask, why is this the case? Part of the reason why most individuals draw upon interstate work is simply because it has been around longer and has been somewhat more prominent in terms of the number of leading scholars, books, articles, grants and, very importantly, its connection with US foreign policy interests. Another part of the reason why most individuals do not draw upon the intrastate community, however, is a lack of awareness — it is a vast area of research and there is simply too much to read already without venturing into something new.
To try and change this situation, we wish to identify four reasons why IR/”interstate” scholars don’t read/consider/discuss more comparative/American/intrastate work and why they are wrong. Two are discussed in Part 1 and two are discussed in Part 2. We do this in order to begin, or rather continue, the process of breaking boundaries within the conflict community, something that is finding more advocates: e.g., Armstrong, Lichbach, and I, Joseph Young, James Ron, Lemke and Cunningham, and James Fearon, but which has not been completely adopted by the majority of people interested with the topic.
The scholars mentioned above have begun to blur the lines between interstate and intrastate work in a fruitful manner. More is needed.
Misconception #1 – “Intrastate Conflict Is Just Not as Important as Interstate Conflict”
It used to be the case that interstate war was frequent as well as extremely deadly. Indeed, whole generations were defined by their participation in these conflicts. The impact of this reality followed accordingly. With the global importance of interstate war, most individuals who focused, studied, discussed, and generated policy about conflict would highlight what states did to each other in the international realm. With regard to this topic, data and detailed histories were compiled and a large number of articles and books were written. For years, this was one, if not the most, important subject in the world generally and in social science specifically.
Over the course of the last 50 years, the tides have changed. At present, war between nations is now a very rare event. For example, since the end of the Second World War, the number of ongoing interstate conflicts involving at least 25 battle casualties has ranged from zero to six. Moreover, the trend has been one of decline. In the 1990s, only one or two interstate conflicts were fought each year, with the conflict between Eritrea and Ethiopia involving the highest number of casualties. From 2003 to 2008, no interstate wars were fought. This interlude ended in 2008 with the onset of an interstate conflict between Djibouti and Eritrea (Harbom and Wallenstein 2009: 577).
In contrast, the number of civil conflicts over the same period (restricted to those cases involving the government and a non-state actor engaged in mutual violence, either in tandem [civil war, counter-terrorism, and protest policing] or individually [genocide, politicide, asymmetrical violence, and terrorism]) has far exceeded the number of interstate conflicts. Indeed, while the number of interstate wars has gradually decreased, the number of intrastate conflicts has grown steadily over the course of the Cold War, reaching a peak of 51 in 1991 (Harbom and Wallenstein 2009: 578). The number declined and subsequently rose slightly over the course of the 1990s and 2000s, but civil conflicts for the entire period of study have occurred much more often than interstate war.
Counting the number of conflicts, one could argue, does not adequately assess the significance of a phenomenon. Another way to account for conflict is to estimate the number of people killed. In this regard, the relative importance of intrastate behavior is even clearer. Civilian deaths, in particular, are especially relevant. For expositional purposes, here we will limit our discussion to fatal conflict; in fact, we limit our analysis to 25 battle deaths, which is the threshold employed by the Uppsala Armed Conflict Data Project.
Figure 1. Battle Deaths by Type, 1900—2005 (Buhaug et al. 2007).
Figure 1 shows the number of battle deaths from 1900 to 2005 using data collected by Lacina et al. (2006). As expected, the two World Wars tower over all other conflicts; these two wars account for the vast majority of battle deaths (and presumably destruction). The Korean War, the Vietnam War, and the Chinese Civil War are dwarfed in comparison. The difference in battle deaths across types of wars is not extremely clear. Today’s most severe conflicts, in Iraq and Afghanistan, while generating massive media attention have incurred relatively few battle deaths.
If we were to include incidences of genocide and large-scale state repression, which would include the Rwandan genocide, the Nazi Holocaust, the Kampuchea ‘killing fields’, and the Chinese Cultural Revolution, the numbers would clearly exceed all conflicts – including those of the World Wars. In Rwanda alone between 800,000 and 1 million died (Harff 2009). In Germany, at least 6 million perished at the hands of the state. The Khmer Rouge is purported to have killed 2 million (Harff 2009). The Cultural Revolution led to the death of an estimated 35 million. Stalin’s collectivization may have killed even more.
In terms of scope, intrastate conflict also involves more countries than interstate violence, especially now. Figure 2 shows two trends: the absolute number in interstate and civil conflict (left-hand vertical axis) and share of countries (right-hand vertical axis) involved in relevant behavior.
Three distinct peaks are evident in Figure 2. In terms of the share of UN members states in conflict, the greatest proportion occurred in the early 1950s with the Korean War, which involved 20 countries. A second peak occurred in 1991. The Gulf War, an interstate conflict involving Iraq vs. the US-led coalition of 29, contributes to about half of this total. Several post-Soviet post-Yugoslav intrastate conflicts account for most of the remainder. In 2004-06, Afghanistan involved a coalition of 39 intervening in a civil conflict; Iraq had 33 countries involved in what begin as an interstate war and evolved into an internationalized civil war; in addition, the US partnered with 17 countries in the so-called “War on Terror” against al-Qaeda.
Looking at the trends over time, in terms of absolute numbers, more countries are getting involved in armed conflict. In contrast, the proportion of countries involved in war exhibits no obvious trend, since the number of countries in the world has increased. Moreover, conflict involvement also means very different things for different countries. The US and UK have been heavily involved in both Iraq and Afghanistan, but these figures also include countries such as Tonga, which has contributed only a token number of troops to the Iraqi War.
Four countries account for the highest levels of involvement in conflict not taking place on their own territory, giving support either to governments or rebel groups. Since 1989, the UK has sent troops to five conflicts; France to five; the US to three; and Russia to two. China, it should be noted, has been involved in no conflicts (Uppsala Armed Conflict Data Program 2009).
While there is no doubt that many great minds have focused their attention on interstate conflict, some of the biggest theorists in history have been concerned with intrastate conflict. The likes of Pareto, Hobbes, Machiavelli, Durkheim, Weber, and Marx all developed theories of internal war. The attention given to this work has clearly lagged behind that devoted to interstate war; this has changed recently but there is still some way to go. Part of the solution is to acknowledge the reality of how conflict has changed and to embrace the work – all of the work – being done on the diverse forms, searching for the dynamics that animate them all.
In short, we need to acknowledge that intrastate conflict is now essentially the bigger game in town, it has been one of the most important games for some time, and it will probably continue to grow in importance over time.
Misconception #2 – “Intrastate Conflict Has No Relevance to Interstate Conflict”
One should not come away from the first point with the opinion that interstate conflict should be ignored and intrastate conflict should be exclusively highlighted. Quite the contrary, we think that attention to both is essential. In arguing this, we acknowledge that theories of interstate conflict have been exported to study intrastate conflict, especially civil conflict. As such, notions of the security dilemma (Walter 1997), the role of territory (Regan 2009), balance of power (Butler and Gates 2009), and the domestic “democratic peace” (Davenport 2007) have all drawn upon international scholarship as a touchstone for their work on domestic conflict and the insights provided by these applications have been essential for the development in these areas. This effort has been largely one-sided, however. Few theories of intrastate conflict have been exported to the field of interstate conflict. Realists are particularly prone to view states’ internal politics as inapplicable to the study of the conflict between nations. Yet the two forms of conflict are inextricably linked and our understanding should be mutually informing.
Internationalized civil conflicts are particularly interesting because they exhibit qualities of both interstate and intrastate conflict. For example, the proxy wars of the Cold War (such those fought in Nicaragua, El Salvador, Angola, Mozambique, and Afghanistan) were part of a larger global conflict fought as internationalized intrastate wars. Similarly, there is new work which highlights that the conflicts fought in Rwanda, Uganda, and the Democratic Republic of Congo were also intricately connected to one another; to understand the conflict in any one country, one had to understand what was taking place within as well as between the others.
Within existing scholarship, factors associated with the onset and prolongation of civil wars is located beyond the boundaries of the nation-state. Sources can be international, having to do either with the direct actions of other states as the intervention of states resulting in an internationalized intrastate conflict, or they may be located in more complex interactions within the system of states as has been evident in the changes from the Cold War to the post-Cold War era. Transnational factors that are related to the various forms of non-state organizations or networks that span state boundaries are also relevant. Indeed, the interplay between state and non-state factors also challenges the very definition of civil war.
International factors frame the setting of civil conflict. The nature of the international system, particularly the emergence of the US as the world’s only superpower, serves as one supranational factor. Intrastate wars are rarely contained solely within the borders of a single state. For example, the conflicts in Afghanistan or the Democratic Republic of Congo manifest themselves in a larger neighborhood of states. Political and economic changes affect entire regions. Civil war does not occur in isolation but affects entire neighborhoods (e.g., through refugees [Davenport et al. 2003]). Moreover, reliance on contraband financing and extortion requires tacit compliance by neighboring countries, or their inability to enforce borders. This compliance is available in a shrinking number of states. Comparatively little work has been done on the transnational support system of state repression and human rights violation but there is clearly something to such an analysis (e.g., see della Porta et al. 2006).
Transnational non-state actors are also relevant. ‘Refugee Warriors’ – military organizations operating in their country of origin but sustained by settlements in exile – exemplify this. The effectiveness of refugee warriors depends both on the protection of the international refugee regime, on the support of the host state, and on existing forms of organization and leadership within the exile population (Harpviken 2009). The use of consultants and mercenaries for state repression as well as the subcontracting of torture to non-state actors represents yet another area where the simultaneous consideration of interstate and intrastate scholarship would also be a lucrative area for future exploration (Rejali 2007).
Networks of violent actors constitute a different type of non-state actor. For example, terrorism, as a tactic, serves as a substitute for other forms of armed struggle in situations when groups are unable to build armed forces (Butler and Gates 2009). Nevertheless, terrorism can also thrive in the context of civil war in conjunction with guerrilla tactics. Terrorist networks, when employed transnationally, such as the al-Qaeda attack on New York and Washington, also blur the distinction between the intra and inter aspects of conflict. For us to address such factors we need to change how we think about conflict.
In the post-Cold War era, these distinctions remain especially hazy in other ways. The end of the Cold War altered the role of international organizations, particularly the political climate of the UN Security Council. This new climate has allowed an expansion in the number, scope, and mandate of peacekeeping operations by the UN and other international bodies. It has also led to changes in the nature of international justice and allowed for an expanding role of the Hague regarding conflict. Not only has the ICC become more involved internationally and especially with regard to intrastate conflict (witness the case of President Bashir in Sudan), there has also been a trend toward criminalization of war. Individuals rather than states have become personally responsible for wrongdoing in the context of conflict and violations of international law. The threat of international legal consequences may deter potential belligerents; but of course it could also encourage a hard line among those who are already engaged in armed conflict (this would account for the limited impact of international human rights treaties in reducing human rights violations [Hathaway 2002; Hafner-Burton 2005]). These initiatives presume that the perpetrators will be properly identified and prosecuted, which is not always the case. Under such legal regimes, the distinction between intrastate and interstate conflict is obscured. Hard to find individuals rather than comparatively easier to find states become the parties responsible for upholding international law. This suggests that the only way to understand what is taking place will involve the intersection of interstate as well as intrastate scholarship.
 In general, deaths which result directly from armed conflict are more reliable than estimating the indirect effects of warfare such as increased mortality from disease and famine. Determining the number of people who would have died from sickness and hunger if there had not been a conflict involves a very tricky and ultimately indeterminate assessment of a counter-factual. Of course, conflict need not involve death. In the sections below, we will explore non-fatal forms of conflict.