The Army Field Manual 3-24, Counterinsurgency, provided a much-needed course change for American forces in Iraq and Afghanistan by focusing the attention of commanders on factors that are not traditionally the concern of the American military. While many commanders had already recognized that conventional tactics were ill-matched to dealing with insurgencies and had adapted accordingly, others were still fighting the insurgents on an ad hoc and counterproductive manner in 2006. The “Neo-Classical” framework that underpins the FM 3-24, however, is based on political science about the revolutionary insurgencies of the Cold War. This “classical” school of Cold War–era counterinsurgency focused on defeating communist and anti-colonial insurgencies by strengthening weak governments that are seen by a critical mass of people in the host nation as illegitimate.

In line with Maj. Jon Bate’s exhortation to get the military and social scientists back together, it is useful to look at the political theory that underlies the FM 3-24. Why are people fighting in the first place? What’s the deeper problem that we, the counterinsurgents, have to solve? Is the problem really the same as it was during the Cold War? I argue that, in many situations, the COIN framework might not be sufficiently complete or appropriate to the ethnically based intrastate conflicts that have been prevalent since the end of the Cold War, in which case a different approach is needed.

FM 3-24: Revolutionary Warfare

FM 3-24 frames insurgency as a contest between insurgents and governments over an undecided population, a contest whose outcome is principally determined by the relative capability of each side to govern people. The undecided civilian population, the manual’s theory suggests, will support the side that they think can best provide services. Initially, the government’s poor track record at providing basic and essential services—public safety, infrastructure, even trash collection—fosters either active or passive support of the insurgents.

While the manual does state that people might have grievances other than lack of state capacity (see section 3-73, for example), the “Logical Lines of Operation” (LLOs) are overwhelmingly targeted at state capacity deficits. The only LLO that gets its own chapter is developing host nation forces (chapter 6), underlining the importance of state capability and belying the claim that information operations is the most important LLO. Still, the LLOs are in line with the overall theory that lack of state capacity leads to state illegitimacy, which in turn leads to insurgency.

Ethnic Combat

Is COIN just as applicable in ethnic conflicts as in non-ethnic ones? This is an important question given that conflicts based on ethnic identity have become more common since the end of the Cold War; in 1953 only 15 percent of conflicts were classified as “ethnic” in nature, compared to 60 percent in 2005. Most scholars take a broad view of ethnicity as being based on certain ascriptive characteristics like language, race, or religion: significantly for any state-building enterprise, it means that people have an identity other than the state to which they feel loyalty. While early quantitative analyses of civil wars found predictors of civil war outbreak that were of dubious value to policymakers (such as that civil wars are more likely to occur in poorer countries), newer attempts to study the predictors of civil war have found that higher “horizontal inequalities” between groups is associated with increased likelihood of civil war outbreak. Most military participants in the US intervention in Iraq would find it hard to describe the course of events without reference to Sunnis, Shiites, and Kurds. Compared to this, except for some references to the Hmong, a scholar of Vietnam might entirely omit any discussion of ethnicity, indicating a fundamental difference between those conflicts.

In the case of an ethnically divided polity, a government might be seen as illegitimate not because it is weak (as COIN theory would predict) but because it is dominated by one ethnic group. Stephen Biddle, for example, made this point explicitly in 2006: “But if the debate in Washington is Vietnam redux, the war in Iraq is not. The current struggle is not a Maoist ‘people’s war’ of national liberation; it is a communal civil war with very different dynamics.” As Donald Horowitz argued, strengthening an ethnically captured state is cause for great alarm to members of the ethnic minority, who then have reason to support the insurgency all the more vigorously. Some scholars, like Stathis Kalyvas, would argue that there is no fundamental difference between the civil wars of the Cold War and those afterwards. And certainly, if there is no fundamental difference between ethnic and non-ethnic wars, then the prescriptions we make in both cases should be the same. If, however, we believe that ethnic conflicts are different than non-ethnic ones, then we risk being not only unproductive, but counter-productive.

One criticism of focusing on ethnicity might be that deal-making between ethnic groups needs to be accomplished at a higher level than the military is responsible for. If we tried to foster a grand bargain between the Sunnis and Shiites, for example, we would have usurped a function that is appropriate to the State Department. It’s therefore important to show that viewing conflicts through the lens of ethnicity is just as important for small-unit leaders trying to implement counterinsurgency tactics on the ground as it is for political decision-makers in the air-conditioned conference rooms of the host nation capital.

Looking at the only LLO to get its own chapter in the FM 3-24, Working with Host Nation forces, we can see one area where our recommendations might be very different in an ethnic war than in a revolutionary struggle. If the central security forces are dominated by one ethnic group, such as happened to Iraq’s Interior Ministry, strengthening their capabilities will strengthen the grievances that motivate insurgents in the first place. If insurgents are Sunni Arabs who are afraid that a Shiite-dominated government will persecute them, strengthening that government will not solve the problem; in fact, it will exacerbate the problem because Sunni Arabs will have a valid reason to fear and resist the central government that we are building. This sounds intuitive, but the fact remains that is an outcome not adequately accounted for in FM 3-24’s approach to counterinsurgency.

The focus on ethnicity defies simple additions and deletions to a checklist, however: sensitivity to ethnic concerns must pervade all aspects of COIN decision-making. Say we’ve identified a demand for more schools through our METT-TC analysis: what neighborhood are we going to build it in? If it is a neighborhood that is dominated by the government-aligned ethnic group, this could be a propaganda coup for the insurgents. “See how the Americans only help the Shiites! We have to push them out and reclaim our country.” The same questions apply to decisions about what contractor builds the school, who teaches there, etc. The most pertinent question in a revolutionary war would simply be how to provide the maximum benefit at the lowest cost, but this is obviously wholly inadequate when individual loyalties are influenced more by group identity than by beliefs about government effectiveness.

Consider also information operations, the LLO that ties all other LLOs together; FM 3-24 rightly emphasizes the importance of messaging in countering any form of insurgency. But who delivers the message is just as important in ethnic contexts as what the message is. Having radio broadcasters or news anchors who are uniformly of one ethnic group would send a clear message about whom the Americans are aligned with. Just as local Sunnis who joined the police in Anbar could easily distinguish a local Iraqi from, say, a Tunisian (who was likely to be affiliated with AQI) while many Americans struggled to do so, we might employ a messaging team that sounds normal to us, unaware of the shibboleths that loudly proclaim ethnic affiliation to locals.

Why This Matters

Overall, FM 3-24 was important in encouraging commanders to think about the population and the impact of their units’ operations on civilians in the Iraq and Afghanistan conflicts. But if certain contexts require different solutions, we require a different field manual for different conflicts, or even different areas in the same conflict. If we’re in an ethnically homogenous area of southern Iraq and people are angry because they don’t have electricity and the security services can’t keep them safe, then implementing FM 3-24 and strengthening the host nation’s capacity makes sense. If we’re in Sunni Arab–dominated Anbar province and people are afraid that the Shiite-dominated central government will oppress them, then bringing in well-trained Shiite and Kurdish soldiers won’t help the situation; in fact, it will inflame tensions. Al Qaeda in Iraq was defeated not when the central government was strengthened, but when the sheikhs were co-opted and their young men integrated into local police units during the Awakening; this deal-making with mid-level elites and devolution of power is not envisaged by the current COIN doctrine. The twenty-first century’s battlespace is one full of nuance: we need our doctrine to be just as nuanced in order to be victorious there.


Matthew Cancian is a PhD student in Security Studies and International Relations at MIT. He formerly served as an artillery officer in the Marine Corps, deploying to Sangin, Afghanistan as a forward observer.

The views expressed in this article are those of the author and do not reflect the official policy or position of the Department of the Army, the Department of Defense, or the US government.


Image credit: Staff Sgt. Samuel Bendet, US Air Force