On Winning Hearts and Minds: Key Conditions for Population-Centric COIN

by Gregory D. Miller

February 8, 2016 -

On Winning Hearts and Minds: Key Conditions for Population-Centric COIN

If you know the enemy and know yourself, you need not fear the result of a hundred battles. If you know yourself but not the enemy, for every victory gained you will also suffer a defeat.

-- Sun Tzu, The Art of War

There are two broad approaches to dealing with an insurgency. One is the conventional military approach, or what Dave Kilcullen refers to as the enemy-centric approach, which tends to focus efforts on the insurgents themselves.[i] There is, of course, wide variation within this approach, including whether efforts should focus on the leadership (decapitation measures) or group members (attrition), and which tools and methods a state will use to target the insurgents (covert action, law enforcement, military raids). The other is a population-centric approach, in which efforts focus on cutting off the insurgents’ lifeblood and supply lines by either providing rewards to the population for supporting the COIN, or by imposing costs on the population for supporting the insurgents. A population-centric approach may also focus on disengagement, attempting to convince current members to give up violence and to lure them away from the group.

US efforts in Iraq, especially after the 2006 publication of the US Army Field Manual 3-24, emphasized a shift from an enemy-centric to a population-centric approach.[ii] Yet despite the amount of work on the benefits of this approach, because all insurgencies are different it is critical to ask whether such an approach is always more effective. Gian Gentile and Douglas Porch are both critical of any type of COIN, regardless of the approach.[iii] But those who advocate for COIN generally favor a population-centric.[iv]

We often hear reference to winning “hearts and minds” as the critical step for a successful COIN strategy.[v] It is logical to believe that if we can just convince the population to support COIN efforts and oppose the activities of the insurgents, then we can defeat those insurgents. If we have the support of the people: the insurgents lose a pool of potential recruits; lose sources of funding, arms, and safe havens; and will have less operational security as the people provide valuable information to the government about the insurgents.

The problem with this conventional wisdom is that it overlooks important variations within insurgent movements that can make the winning of hearts and minds difficult, if possible at all. We become so preoccupied with how to win over the population that we fail to ask when attempting to do so is even appropriate. This article discusses some of the variation that exists among insurgents to identify when the hearts and minds approach is more likely to be effective and when such an approach may backfire. Although several factors influence the outcome of any COIN approach, this article focuses on some key traits of the insurgents themselves, because as the Sun Tzu quote at the beginning of this article suggests, victor requires that you know your enemy. At least as a starting point, three factors influence the likelihood of success when using COIN: the phase of the insurgency; the goal of the insurgents; and the demographics of the insurgents (especially relative to the rest of the state’s population and the COIN forces).

Winning Hearts and Minds

According to Douglas Porch, one of the first uses of the term “hearts and minds” was by a French general in Indochina countering rebellion along the Chinese border,[vi] but more significantly used by the British in Malaya.[vii] The United States engaged in similar efforts to sway the population, first in Vietnam, and more recently in Iraq and Afghanistan. The term often takes on a somewhat cynical connotation, to refer to any attempt to sway public opinion, or in strategic military terms viewing the population as the center of gravity.[viii] It is sometimes used synonymously with COIN where winning hearts and minds simply amount to defeating an insurgency.[ix] And Porch essentially equates hearts and minds to the acquisition of tactical Intel.[x] He also highlights the often-overlooked danger of a population-centric approach, in that it causes the soft target population to suffer attacks and retribution when insurgents begin to view the people to be conspiring with the government or COIN forces.

While there was a great deal written in the last two decades about population-centric strategies, and COIN more generally, most of these works focus on the ways and means that states use to implement a strategy of winning hearts and minds – in other words, conditions that are largely under the control of the state.[xi] Herman Kahn argued in 1968 that “the two most important political factors in ‘winning hearts and minds’ are looking like a winner and providing security.”[xii] The British hearts and minds strategy in Malaya primarily involved the discriminate use of force, political reforms, and improved governance.[xiii]

The very detailed RAND study by Paul et al highlights the lessons we should learn by studying the successes and failures of numerous modern insurgencies. Of the 59 cases they examine, most involve what they term “iron fist” strategies, in which the counterinsurgents focused primarily on defeating the enemy by force. Fifteen of the cases involved mixed strategies, which experienced a much higher percentage of success. Mixed strategy cases had a 73% success rate. Compared to those coded as “iron fist”, which saw success in just 17 of the 44, or 32%, of the cases. Despite this work recognizing that a mixed strategy was generally more successful than a purely military strategy, under what conditions are those mixed strategies most likely to succeed? The RAND team correctly identifies a critical factor for success – the use of a mixed strategy. Yet, there are cases where a mixed strategy was unsuccessful (Laos, 1959-1975; Kampuchea, 1978-1992; Papua New Guinea, 1988-1998; and Tajikistan, 1992-1997), and 17 where a mixed strategy was unnecessary.

Winning hearts and minds essentially means convincing the populace that the benefits of supporting the government against the insurgents outweighs the benefits of supporting the insurgents. Handing out candy and gifts to the population may be one element of a hearts and minds approach, but an effective strategy also involves anything that alters the population’s calculations. It may provide security or protect the population so that it does not have to fear the insurgents (or fear the government for that matter). It may also include punishment and retribution against the population for supporting the insurgents, at the expense of government forces. In general, the biggest distinction is that a hearts and minds approach focuses on the population while a conventional military strategy tends to focus on the insurgents. The challenge is that under some circumstances, such a focus on the population may be less effective than a strategy that emphasizes those who are engaged in the fighting. Part of the purpose of this article is to identify some of those circumstances, so that in the future we do not emphasize the wrong type of fight.

There is an assumption that winning hearts and minds is a necessary condition for success in an insurgency. If true, then that means that insurgencies are fundamentally different from conventional wars, in which hearts and minds are rarely the focus. That argument, taken to the next level, suggests that there are different types of insurgencies, and that a strategy of winning hearts and minds may be more effective in those insurgencies that are more about political victory, and less effective in those insurgencies that are more about winning the military contest.

This article focuses on those factors that are largely beyond the control of the government involved in the conflict, it better frames the strategic environment under which the insurgents and counterinsurgents both operate. Because these are largely uncontrollable conditions for the state, it is necessary to identify when these conditions exists, altering the likelihood of success for a COIN approach of winning hearts and minds. When my findings suggest hearts and minds will be less successful, that does not mean COIN will be ineffective, just that it will be more successful if it focuses on the insurgents rather than the people. When hearts and minds is likely to be more successful, the specific tactics required to achieve the greatest success still must be identified as we continue to ask these questions and seek greater insight into the phenomenon of insurgency.

What Is An Insurgency?

The U.S. Government defines insurgency as “the organized use of subversion and violence to seize, nullify, or challenge political control of a region.”[xiv] The problem with this definition is it ignores the possibility that an insurgency can also be used to protect the status quo against those advocating for change.[xv] It does allow for different categories of insurgencies, such as guerrilla warfare and terrorism, but fails to define either of those terms or explain their relationship. Previous government definitions, such as that used by the Central Intelligence Agency and RAND, see insurgency as a “protracted political-military activity directed toward completely or partially controlling the resources of a country through the use of irregular military forces and illegal political organizations.”[xvi] Again, the assertion that political organizations must be illegal for it to be an insurgency is too narrow of a perspective. The more important point is that the political organizations – whether legal or illegal – engage in illegal, violent activities.

Bard O’Neill defines it as “a struggle between a nonruling group and the ruling authorities in which the nonruling group consciously uses political resources…and violence to destroy, reformulate, or sustain the basis of one or more aspects of politics.”[xvii] Like the US government definition, this tends to assume the insurgency is always against the existing government. It also tends to ignore the possible role of third party intervention, on behalf of either side.

This article defines an insurgency as a sustained campaign of extralegal violence, intended to cause or prevent government change. Within insurgencies, groups can fall along either side of a spectrum, depending on the nature of their attacks. Guerrilla groups are those that focus their attacks on combatants – usually the armed forces of the state or of a third party involved in the country (such as an occupying power). In contrast, terrorist groups are those that focus their attacks on non-combatants. This should be a fairly clear distinction, but is really a continuum because groups that primarily targets combatants – and thus, by my definition, would be a guerrilla group – may also occasionally attack civilians. Likewise, a terrorist group that emphasizes attacks on civilians may occasionally attack combatants.[xviii] This article focuses on the more broadly defined insurgents, but one of the first determinants of whether a population-centric approach will be effective or not, is whether the insurgents focus their attacks on combatants or non-combatants.

Key Factors

Although there are many factors not addressed here that contribute to the usefulness of a hearts and minds strategy, those identified below emphasize conditions largely beyond the control of the government and its forces. These include the phases of the insurgency, the insurgents’ goals, and demographics. I discuss each of these in some detail, providing examples of both successful and failed attempts at population-centric approaches given the different conditions.

The Phases of Insurgency

Mao Tse-Tung, leader of the Chinese Communist movement, wrote about the stages of protracted war. During Phase One, when the movement is relatively weak, the group primarily organizes itself in isolated terrain and uses terrorism, both to educate the public and to further mobilize the population. Phase Two involves fewer attacks on the population and more direct attacks against the government and its armed forces, in what appears to be more guerrilla warfare.[xix] This occurs partly because the movement has more popular support, and partly because it gained the capabilities to target military forces rather than civilians. The final phase involves direct combat between the insurgents and the government’s forces. This is not a purely linear progression as a movement can move backwards or forwards. The point is that there is a difference between those movements that have the support of the civilians and those that do not.

A hearts and minds strategy is most effective when an insurgency is still in its earliest phase. Phase One activities, like terrorism, involve attacks against civilians, which new groups often use to frighten, mobilize, and “educate” the population. Support for the movement is low during this phase, not only because it is new, but because the group is attacking the very support base from which it is trying to recruit.

Once a movement progresses to the phase where it uses less terrorism, and focuses more of its attacks on the military forces, the population-centric approach will be much less effective. By this point, not only have the insurgents already gained some attention and support, but because they no longer attack civilians they will have more support from the population than during the previous phase. Thus, the opportunity to win hearts and minds from potential sympathizers will have already passed.

Audrey Cronin discusses the differences between the Islamic State of Iraq and al-Sham (ISIS) and al-Qa’ida to highlight why ISIS is not merely a terrorist group.[xx] The desire to take and hold territory suggests the group has moved beyond Phase One. Yet it still clearly engages in attacks against civilians with the intent of causing fear. Because the group is no longer in Phase One, opportunities for success using population-centric approaches are shrinking. This is also true because of the nature of ISIS’ control over territory. We can focus on Iraqi and foreign populations in an attempt to limit future support and recruits, but unless we can also protect the population under ISIS control, an approach that focuses on winning hearts and minds is unlikely to achieve long-term success.

The Insurgents’ Goal

Insurgencies form for a variety of reasons, with some more common than others during various historical eras. Three of the most prominent causes are independence/autonomy, revolution, and counter-revolution. In the case of an independence movement, the group seeks to split from the existing polity and establish a separate state. Examples include the Liberation Tigers of Tamil Eelam in Sri Lanka, the Kurdistan Worker’s Party in Iraq and Turkey, and ISIS in Iraq and Syria today. Revolutionary insurgents seek to bring about change to an existing state, often through some form of political, economic, and/or societal shift. Examples include the Chinese Communists under Mao and the FARC in Colombia. Finally, counter-revolutionary groups seek a return to some status quo ante. Ba’athists in Iraq provide one such example, many of whom joined al-Qa’ida in Iraq and later ISIS. Another is the Afrikaner Resistance Movement in South Africa, after the end of the apartheid government.

Independence movements can be unique because they often involves groups of different ethnic or religious identities, rather than the political differences that so often exist in revolutionary movements. So if one were to seek to win hearts and minds, who is the target population – those who already feel oppressed and better-off on their own, or those who do not want to lose a piece of their national territory? For this reason, it is important to identify the purpose of the insurgencts before adopting any type of population-centric approach. Then, if it is about independence, the demographics of the movement (as discussed in more detail below) become even more important.

Revolution and counter-revolution both depend on timing and phase (as discussed above) to determine the degree of popular support for the insurgents. Ba’athists were already the minority population before losing power, so their struggle was a combination of wanting to retain power and fear of what will happen to them once the majority gains power. In such situations, who is the target of the hearts and minds campaign? If the differences are ethnic, as opposed to political, then a hearts and minds strategy can work, but only if both sides are willing to compromise (less likely if the difference is a religious one). Even if the difference is political as discussed above, hearts and minds will only work in the early stages, when popular support for the movement is at its lowest level.

ISIS combines an independence movement with a counter-revolution. As stated already, many of the ISIS members are former Ba’athists hoping to restore the power they had prior to the US invasion and fall of Saddam Hussein. This is the counter-revolution component of ISIS. Yet ISIS also seeks to hold territory that is took from Iraq and Syria, and declared the Caliphate, suggesting a goal of independence. Because of this combination of goals, a purely population-centric approach will be difficult, and requires us to take a look at demographics.

Demographics

Having established that it is critical to focus on hearts and minds campaigns when the population is under attack, it is also critical to understand which portion of the population fears the insurgents. The most complex scenario is one where the insurgents are part of a separate ethnic or religious group that targets only those of the majority. This is complicated because of the multiple audiences involved. A simpler scenario, though still difficult, is when an ethnic group attacks both those of the opposing identity as well as those of their own group – commonly those working with the authorities or who are perceived to be disloyal to the cause. A still simpler scenario involves revolutionary or counter-revolutionary movements, in which there are no ethnic or religious differences among the population. In such scenarios, a hearts and minds strategies are the most likely to be effective.

Demographics matter not just for the at-risk population, but also for the insurgents and the COIN forces. The scenario most likely to lead to a successful population-centric strategy is when the insurgents are made up of foreign fighters and the COIN forces are entirely domestic. In such situations, the population is more likely to support the government forces against those viewed as outsiders. We saw the evolution of this in Iraq as local tribes became wary of the foreign fighters and began to provide more support to US and coalition forces, sometimes referred to as the Sunni Awakening.[xxi]

The most difficult scenario for population-centric approach is when the insurgents are made up of the local population while COIN forces are members of a foreign government. In such cases, the COIN forces will be viewed as occupiers and will rarely be able to gain the trust of the civilian population, no matter how much effort is put into winning hearts and minds. The various stages of the Iraq conflict illustrate these differences and how shifts to one variable will make hearts and minds more effective. In other words, it was not the adoption of hearts and minds that improved the situation in Iraq, but rather the increased use of Iraqi forces over US forces, and the increased recognition, by the Iraqi people, that many of the insurgents were foreign fighters. These shifting perceptions happened to coincide with the shift to a more population-centric approach.

The demographics of an insurgency can also change. Eric Jardine argues that populations tend to move away from violence,[xxii] which would make population-centric approaches less effective and also less necessary. But a key missing ingredient from his study draws on my previous discussion of whether the insurgents are guerrillas who rely on the people, or terrorist who target the people. If the people support the insurgents, they are less likely to leave the area. Or, in the case of ISIS, where the insurgents hold territory, people are being prevented from leaving.

Despite the high prevalence of foreign fighters in ISIS,[xxiii] most of the group’s leaders are Iraqi. The group’s Sunni beliefs create complications for a population-centric approach because it creates multiple audiences – Sunnis under ISIS control, Sunnis in the rest of Iraq and elsewhere, and other Iraqis (both Shia and Kurd). Which is the appropriate target of a population-centric approach, since the types of policies that will reassure one population may simultaneously weaken the support received from another? This does not suggest that a population-centric approach cannot work, just that it will be more difficult and less likely to succeed than if the demographics were different and the insurgents were more similar to the COIN forces, ethnically and/or religiously.

Conclusions and Recommendations

While we tend to view a population-centric approach to COIN as preferable and more humane to an enemy-centric one, it is important to identify those conditions under which each type of approach is more effective than the other. No two insurgencies are the same, and so no COIN approach should be uniform. This article examines certain conditions that affect the likelihood of success for a population-centric approach.

If one accepts the above points, then there are some logical conclusions regarding when a state should follow a hearts and minds approach during an insurgency. Population-centric approaches are most likely to be successful when an insurgency is still in its infancy, as civilians are targeted and looking for protection from the insurgents as well as reasons to oppose the insurgency. As an insurgency becomes more established, and attacks the population less frequently, an enemy-centric approach (or a mixed approach) will become more useful. ISIS, for example, is arguably already in Phase Two. While it still engages in terroristic behavior, especially towards those living under its control, it has the capability and will to engage in direct combat with Iraqi forces. This means it is no longer simply a terrorist organizations, and requires an approach closer to the enemy-centric side of the spectrum. This does not mean Iraqi and coalition forces should not still attempt to protect the population as well as win over the people, simply that the emphasis of the COIN should be on the enemy forces.

Government approaches must also depend on the goals of the insurgents. A population-centric approach is more suitable against revolutionary or counter-revolutionary movements, where a group attempts to alter the system or government from within, and thus would have a dramatic effect on the entire population. Groups fighting for independence have a more limited effect on the society, and thus necessitate a more enemy-centric or mixed approach. ISIS, despite its rhetoric about religion and the caliphate, is essentially an independence movement, wanting to take and hold territory from Iraq and Syria. This suggests a more enemy-centric approach is in order, which is consistent with the assessment above based on the phase of the ISIS insurgency.

But the existence of multiple ethnic or religious audiences will always complicate a population-centric approach. If the insurgents are of a different ethnicity from the population, especially if the insurgents are foreigners, then a population-centric approach can work. If the insurgents are the same ethnicity as the population, then such an approach will be more difficult, and may require a more enemy-centric approach. We not only see this with ISIS, but we saw the opposite of this in Iraq during the Anbar Awakening. Iraqis turned against groups like al-Qa’ida in Iraq (AQI, the forerunner to ISIS), because of its heavy contingent of foreign fighters. Much of that persists, but ISIS also includes a growing number of Iraqis, meaning that a population-centric approach will be less effective.

If the US decides to engage in the fight against ISIS, a population-centric approach may make us feel better and help reduce civilian casualties – though it is still unclear the extent to which that assumption holds true – but the evidence presented here suggests that it will not help put an end to the insurgency. Instead, a focus on the enemy itself, or at least a mixed approach, will be necessary to achieve the goal of defeating ISIS. The broader lesson is that COIN is like conventional war in at least one way; states need to have a better understanding of their enemy before trying to develop a strategy for defeating that enemy. Unlike conventional war, there may be insurgencies where winning over the population avoids the need to defeat the enemy on the battlefield. Yet we need more research to better understand when population-centric approaches are most likely to be effective.

The views expressed in this article are the author’s alone and do not reflect the position or policy of the National Defense University, the Department of Defense, or the U.S. Government.

End Notes

[i] Dave Kilcullen, “Two Schools of Classical Counterinsurgency,” Small Wars Journal Blog 27 January 2007, http://smallwarsjournal.com/blog/two-schools-of-classical-counterinsurgency(accessed 25 January 2016).

[ii] Headquarters Department of the Army. FM 3-24 Counterinsurgency (Washington, D.C.: United States Government, 2006).

[iii] Gian Gentile, Wrong Turn: America’s Deadly Embrace of Counterinsurgency (New Press, 2013). Douglas Porch, Counterinsurgency: Exposing the Myths of the New Way of War (Cambridge University Press, 2013).

[iv] Brian Burton and John Nagl, “Learning as we Go: The US Army Adapts to Counterinsurgency in Iraq, July 2004-December 2006,” Small Wars & Insurgencies 19:3 (2008), 303-327; Eric Jardine and Simon Palamar, “From Medua Past Kantolo: Testing the Effectiveness of Canada’s Enemy-Centric and Population-Centric Counterinsurgency Operational Strategies,” Studies in Conflict & Terrorism 36:7 (2013), 588-608.

[v] Several scholars contend that COIN is not a strategy, but rather a doctrine, operational method, or simply a collection of tactics. Gian Gentile, “A Strategy of Tactics: Population-Centric COIN and the Army,” Parameters (Autumn 2009), 5-17; M.L.R. Smith and David Martin Jones, The Political Impossibility of Modern Counterinsurgency: Strategic Problems, Puzzles and Paradoxes (New York: Columbia University Press, 2015). I agree, to the extent that the US has generally failed to connect ends, ways, and means, and has used its COIN doctrine primarily as an operational method to dealing with an enemy it generally fails to understand. I disagree that COIN can never be a strategy. If a state has specific ends, and ties its ways and means to achieve those ends, then one could have a COIN strategy, but it must also be flexible enough to deal with different types of insurgencies.

[vi] Douglas Porch, “Bugeaud, Gallieni, Lyautey: The Development of French Colonial Warfare,” p.394 in Peter Paret, ed., Makers of Modern Strategy: From Machiavelli to the Nuclear Age (Princeton, New Jersey: Princeton University Press, 1986).

[vii] Paul Dixon, “’Hearts and Minds’? British Counter-Insurgency from Malaya to Iraq,” Journal of Strategic Studies 32:3 (2009), 353-381 and The British Approach to Counterinsurgency: From Malaya and Northern Ireland to Iraq and Afghanistan (Palgrave, 2012); Riley Sunderland, Winning the Hearts and Minds of the People: Malaya, 1948-1960 (Rand, Memo RM-4174-ISA), September 1964; Richard Stubbs, “From Search and Destroy to Hearts and Minds: The Evolution of British Strategy in Malaya 1948-60,” in Daniel Marston, Counterinsurgency in Modern Warfare (Osprey, 2012).

[viii] Gentile, “A Strategy of Tactics,” 5-17.

[ix] Lawrence Cline makes this point in Lawrence Cline and Paul Shemella, eds., The Future of Counterinsurgency: Contemporary Debates in Internal Security Strategy (Santa Barbara, CA: Praeger, 2015), p.20. This is also largely Ivan Eland’s argument, claiming that counterinsurgency is generally ineffective without really making the clear distinction between winning hearts and minds and other forms of COIN. Ivan Eland, The Failure of Counterinsurgency: Why Hearts and Minds are Seldom Won (Santa Barbara, CA: Praeger, 2013).

[x] Porch, Counterinsurgency, 327.

[xi] Christopher Paul, Colin Clarke, Beth Grill, and Molly Dunigan, Paths to Victory: Lessons from Modern Insurgencies (Santa Monica, CA: RAND Corporation, 2013); Ben Connable and Martin Libicki,How Insurgencies End (Santa Monica, CA: RAND Corporation, 2010).

[xii] Herman Kahn et al., Can We Win in Vietnam? The American Dilemma (London, England: Pall Mall, 1968), cited in Dennis Duncanson, “Review: Vietnam-Inscrutable East?” International Affairs 44:4 (October 1968), 735-741.

[xiii] Eland, The Failure of Counterinsurgency, 93. Though Eland attributes these policies to failure in Malaya because the British ultimately granted the country independence, other scholars points to Malaya as an example of successful COIN, while still others highlight the fact that the British did not rely exclusively on a strategy of winning hearts and minds. John Nagl, Learning to Eat Soup with a Knife(Chicago, IL: University of Chicago Press, 2005); Karl Hack, “Everyone Lived in Fear: Malaya and the British Way of Counter-Insurgency,” Small Wars & Insurgencies 23:4-5 (2012), 671-699; Dixon, “Hearts and Minds?”.

[xiv] US Government Counterinsurgency Guide (January 2009), 6; FM 3-24.

[xv] Although this is not how insurgency is generally used, there are cases where governments employed insurgent tactics – extralegal violence – to target enemies of the state. I discuss this in more detail in the section on insurgent goals, as counter-revolution, but some examples of insurgency carried out by the state are the death squads in Argentina (like the Argentine Anticommunist Alliance), which was active during the 1970s, and Spain’s Antiterrorist Liberation Groups (GAL) during the 1980s. In other words, when a state employs secret and illegal means to protect itself, it is engaged in a type of insurgency. Some scholars refer to this as state terror, which is just another way of saying the state is involved in extralegal violence against its people, over the control of political territory. My preference is to define insurgency by the activities and goals of the group engaged in violence, rather than the actors engaged in the violence. The implication of this argument is that populations may occasionally need to engage in counterinsurgency efforts against their own state, which also could be either enemy- or population-centric.

[xvi] Central Intelligence Agency, Guide to the Analysis of Insurgency (Washington, D.C.: Central Intelligence Agency, n.d.), 2; Daniel Byman, Peter Chalk, Bruce Hoffman, William Rosenau, and David Brannan, Trends in Outside Support for Insurgent Movements (Santa Monica, CA: RAND, 2001), 4; William Rosenau, Subversion and Insurgency (Santa Monica, CA: RAND, 2007), 2fn4.

[xvii] Bard O’Neill, Insurgency and Terrorism: Inside Modern Revolutionary Warfare (Washington, D.C.: Brassey’s Inc, 1990), 13.

[xviii] Although I contend that the government can engage in insurgency, the ones that do are more likely to be on the terrorism end of this spectrum. Unless there is a full-blown army rising up against the state, it will be difficult for the state to engage in discriminate attacks against combatants. Instead, most examples of state violence tend to be against the population, and thus qualify as terrorism.

[xix] Mao Tse-Tung, On Protracted War (University Press of the Pacific, 2001). Mao’s other work, On Guerrilla Warfare, focuses on what he calls revolutionary war, which includes phases of both terrorism (attacks against non-combatants) and guerrilla warfare (attacks against combatants). Mao Tse-Tung, On Guerrilla Warfare (University of Illinois Press, 2000). Therefore, his use of revolutionary war and protracted war are both synonymous with my use of insurgency.

[xx] Audrey Cronin, “ISIS is Not a Terrorist Group: Why Counterterrorism Won’t Stop the Latest Jihadist Threat,” Foreign Affairs (March/April 2015).

[xxi] Joe Klein, “Is al-Qaeda on the Run in Iraq?” Time (23 May 2007), available athttp://content.time.com/time/magazine/article/0,9171,1625200,00.html (accessed 30 January 2016).

[xxii] Eric Jardine, “Population-Centric Counterinsurgency and the Movement of Peoples,” Small Wars & Insurgencies 23:2 (2012), 264-294.

[xxiii] In testimony before the US Congress, US Director of National Intelligence James Clapper, suggested that out of 31,000 members of ISIS, about 20,000 were estimated to be foreign fighters. Robert Windrem, “ISIS By the Numbers: Foreign Fighter Total Keeps Growing,” NBCNews.com (28 February 2015), available at http://www.nbcnews.com/storyline/isis-terror/isis-numbers-foreign-fighter-total-keeps-growing-n314731 (accessed 1 February 2016).