Debates over how
governments can defeat insurgencies ebb and ºow with international events.
Such debates tend to arise when the United States runs into problems in its
efforts to support a counterinsurgent government. Often the United States
approaches the problem of an armed, organized, internal nonstate political
challenge to a government as a zero-sum game. In this competition, the
state and nonstate actors compete for popular support and cooperation.
The U.S. prescription for success is twofold: to provide support for liberalizing,
democratizing reforms designed to reduce popular grievances and gain
popular support while weakening the insurgency, and to target insurgents
with military force without harming civilians. I identify this prescription as
the “good governance approach” for its focus on building a liberal, reformist
central government to quell insurgency. Others have referred to it as
the “population-centric approach,” the “comprehensive approach,” or the
“hearts-and-minds approach.”1
In this article, I reconsider the effectiveness of the good governance approach
in light of the extreme difªculty that liberal great powers have experienced
in applying it in Iraq, Afghanistan, and elsewhere. Since the al-Qaida
attacks of September 11, 2001, the United States has been trying to help smaller
Jacqueline L. Hazelton is Assistant Professor of Strategy and Policy at the U.S. Naval War College.
For their helpful comments, the author thanks the anonymous reviewers, Robert Art, Andrew
Birtle, Nick Carter, Joseph Collins, Conrad Crane, David Edelstein, William Fallon, Brendan
Green, Chaim Kaufmann, Rodrigo Javier Massi, Emily Meierding, J.E. Peterson, Daryl Press,
Jeremy Pressman, Aaron Rapport, Joshua Rovner, Karthika Sasikumar, Richard Sommers,
Benjamin Valentino, Isaac Wilson, Robert Wilson, and her interview subjects. She is also grateful to
University of Oxford Bodleian Library Senior Archivist Lucy McCann; Deborah Usher at the
Oman Archive, Middle East Center, St. Antony’s College, the University of Oxford; Paul Cornish
at the Imperial War Museum; the Liddell Hart Center for Military Archives at King’s College London;
Adam Allen at the Tameside Local Studies and Archives Center, Ashton under Lyne,
Lancashire, United Kingdom; the U.S. Army Military History Institute for a General and Mrs. Matthew
B. Ridgway Research Grant; and the Mellon Foundation.
1. See, for example, Bernard Finel, “A Substitute for Victory: Adopting a New Counterinsurgency
Strategy in Afghanistan,” Foreign Affairs, April 8, 2010, http://www.foreignaffairs.com/articles/
66189/bernard-ªnel/a-substitute-for-victory; and Andrew Exum, “Alternatives to PopulationCentric
Counterinsurgency,” Abu Muqawama blog, October 1, 2009, https://www.cnas.org/
publications/blog/alternatives-to-population-centric-coin-in-afghanistan. On the term “hearts
and minds,” see Paul Dixon, “‘Hearts and Minds’? British Counterinsurgency from Malaya
to Iraq,” Journal of Strategic Studies, Vol. 32, No. 3 (2009), pp. 353–381, doi:10.1080/
01402390902928172.
International Security, Vol. 42, No. 1 (Summer 2017), pp. 80–113, doi:10.1162/ISEC_a_00283
© 2017 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
The “Hearts and Minds” Fallacy
The “Hearts and
Minds” Fallacy
Jacqueline L.
Hazelton
Violence, Coercion, and Success in
Counterinsurgency Warfare
80
states defeat insurgents as part of the war on terror. The ªrst two U.S. military
deployments after hijacked planes slammed into the World Trade Center,
the Pentagon, and a Pennsylvania ªeld were to the Democratic Republic of
Georgia and the Philippines, where the governments were ªghting Muslim
insurgencies.2
The longest, most expensive efforts have been in Iraq and
Afghanistan, where the United States and its partner forces are still ªghting
multiple insurgent groups while the U.S. attempts to shape a more liberal
political order.3
I argue, in contrast to the good governance approach, that counterinsurgency
success is the result of a violent state-building process in which elites
engage in a contest for power, popular interests matter little to the outcome,
and the government beneªts from the use of force against civilians. My theory
of counterinsurgency success, which I call the “coercion theory,” differs in two
important ways from the conventional wisdom on good governance. First, it
identiªes armed and unarmed elites as the key actors in counterinsurgency,
rather than the populace, and it underscores the counterinsurgents’ need to accommodate
the few rather than provide reforms for all. Second, it identiªes
the government’s use of force against insurgents and civilians as an important
factor in counterinsurgency success, as opposed to limiting the use of force to
avoid civilian harm. My analysis ªnds that counterinsurgency success requires
neither good governance reforms that redistribute power and wealth among
all citizens nor popular support for the state. Rather, defeating an insurgency
has three requirements. The ªrst is the government’s relatively low-cost accommodation
of its elite domestic rivals—that is, political actors such as warlords,
regional or cultural leaders, and traditional rulers—to gain ªghting
power and information about the insurgency. The second requirement is the
application of brute force to control civilians and thus reduce the ºow of resources
to the insurgency. The third requirement is the application of coercive
force to break the insurgency’s will and capability to ªght on.
These ªndings show that, as the ªrst step in establishing relative political
stability, counterinsurgency has high moral and human costs.4
In addition,
The “Hearts and Minds” Fallacy 81
2. More recently, Georgia’s Pankisi Gorge has produced several Islamic State leaders. See Tamar
Svanidze, “Georgia’s Intelligence Service Arrests Pankisi Resident for Suspected Terror Sympathies,”
Georgia Today, July 30, 2016, http://georgiatoday.ge/news/4349/Georgia%E2%80%99sIntelligence-Service-Arrests-Pankisi-Resident-for-Suspected-Terror-Sympathies.
In the southern
Philippines, the Abu Sayyaf group and others remain active. See “Duterte Pleads with Philippine
Rebels to Rebuff Islamic State Advances,” Reuters, January 27, 2017, http://www.reuters.com/
article/us-philippines-security-idUSKBN15B0XO.
3. Alex Horton, “Report: Nearly $5 Trillion and Counting Spent on Iraq and Afghanistan Wars,”
Stars and Stripes, September 12, 2016, https://www.stripes.com/news/report-nearly-5-trillion-
and-counting-spent-on-iraq-and-afghanistan-wars-1.428664#.WR70o8m1uu4.
4. Herman Joseph S. Kraft deªnes political stability as constitutionalism, legitimacy, effectiveness,
these ªndings pose an important corrective to U.S. policy choices regarding intervention
to support a counterinsurgent partner. My argument suggests that
U.S. efforts to reduce violence in these internal conºicts by introducing political
reforms are unlikely to succeed, and that such efforts will continue to raise
human, moral, and ªnancial costs for the United States as well as for its partner
states.
In the next section, I lay out existing explanations for counterinsurgency
success, including the good governance approach and force-based or guerrillabased
approaches. I then identify their limitations and derive a rigorous theory
of good governance counterinsurgency success from the existing literature.5
I
next present the coercion theory of counterinsurgency success, explain my
qualitative research design, and compare the two theories’ explanatory power
by examining the well-known case of the Malayan Emergency and, more
brieºy, the cases of Dhofar, Oman, and El Salvador. In doing so, I identify overlooked
or underplayed evidence of limited or nonexistent reforms, government
violence against civilians, the government’s accommodation of rival
elites, and the lack of popular support for the government. I close with a discussion
of the policy implications of my ªndings.
What Causes Counterinsurgency Success?
Many explanations for why governments are able to retain power against an
insurgency assume that both political and military efforts are required. Those
who study the question disagree, however, over how much political change is
necessary and whether a government’s use of force against civilians hurts or
helps its efforts to defeat armed nonstate actors ªghting to change the distribution
of power or wealth within the state; or overturn the status quo; or divide
the state into two or more states. There are two opposing views in the counterinsurgency
literature on the role of the populace. One view is that the government
and the insurgency compete for the allegiance of the populace by
providing it with political, economic, and social beneªts—what I referred to
earlier as the good governance approach. The other view holds that the government
succeeds simply or primarily by using military force against the inInternational
Security 42:1 82
relative impermeability, and durability. See Kraft, “The Philippines: The Weak State and the Global
War on Terror,” Kasarinlan: Philippine Journal of Third World Studies, Vol. 18, Nos. 1–2 (2003),
pp. 133–152.
5. I examine the predictive power of theories, not strategies. A strategy is designed for a speciªc
place and time with speciªc actors, resources, and interests. A theory is an attempt to explain a
pattern of behavior. See Alexander L. George, Bridging the Gap: Theory and Practice in Foreign Policy
(Washington, D.C.: U.S. Institute of Peace Press, 2003), pp. 117–118.
surgency. In this section, I describe these two views in more detail and
assess their utility in explaining the causes of counterinsurgency success.6
the good governance approach to counterinsurgency success
The good governance approach to counterinsurgency success is based on the
belief that the government must provide political, economic, and social reforms
that meet the needs of the population; reduce the number and kinds of
grievances fueling the insurgency, in an effort to obtain information about the
insurgency from civilians; and use force against the insurgency with great
care to avoid civilian harm. I call this the good governance approach to underscore
its causal logic vis-à-vis the populace. “Good governance” typically
means economic growth, political representation, and efªcient administration.
According to the U.S. National Security Strategy of 2002, in states that practice
good governance, “governments must ªght corruption, respect basic human
rights, embrace the rule of law, invest in health care and education, follow responsible
economic policies, and enable entrepreneurship.”7
In this view, good governance is necessary to defeat insurgencies because it
is bad governance that causes them in the ªrst place. Greater representative
government and more public goods will build broad popular support for
the state, attract civilian cooperation against the insurgency, and marginalThe
“Hearts and Minds” Fallacy 83
6. Scholarly explanations for counterinsurgency success include regime type and the will to prevail.
Both are contested. On regime type, see Jason Lyall, “Do Democracies Make Inferior Counterinsurgents?
Reassessing Democracy’s Impact on War Outcomes and Duration,” International
Organization, Vol. 64, No. 1 (Winter 2010), pp. 167–192, doi:10.1017/S0020818309990208; and Anna
Getmansky, “You Can’t Win If You Don’t Fight: The Role of Regime Type in Counterinsurgency
Outbreaks and Outcomes,” Journal of Conºict Resolution, Vol. 57, No. 4 (August 2013), pp. 709–724,
doi:10.1177/0022002712449326. On the will to win, see Ivan Arreguín-Toft, “How the Weak Win
Wars: A Theory of Asymmetric Conºict,” International Security, Vol. 26, No. 1 (Summer 2001),
pp. 93–128, doi:10.1162/016228801753212868; and Andrew Mack, “Why Big Nations Lose Small
Wars,” World Politics, Vol. 27, No. 2 (January 1975), pp. 175–200, doi:10.2307/2009880. Neither
Arreguín-Toft nor Mack addresses counterinsurgency success generally. For an overview of recent
empirical investigations into counterinsurgency, see Eli Berman and Aila M. Matanock, “The
Empiricists’ Insurgency,” Annual Review of Political Science, Vol. 18 (2015), pp. 443–464, doi:10.1146/
annurev-polisci-082312-124553. This literature asks interesting questions and applies more rigor to
answering them than have many earlier studies of counterinsurgency. Yet it rarely asks what
causes success. It most often examines a tactic, such as an incentive provided by a counterinsurgent
to civilians in a small geographic area or the effects of counterinsurgents harming civilians.
There is little consideration of politics—such as possible roles of nationalism or of popular support—or
of external validity. Much of this research is driven by the wealth of data available from
occupying forces in Iraq and Afghanistan. Authors rarely recognize, however, that these cases are
outliers, because they involved the long-term presence of tens of thousands of foreign troops, raising
further questions about external validity.
7. Ved P. Nanda, “The ‘Good Governance’ Concept Revisited,” Annals of the American Academy of
Political and Social Science, January 2006, pp. 269–283, doi:10.1177/0002716205282847; and U.S. National
Security Council, National Security Strategy of the United States (Washington, D.C.: White
House, 2002), https://www.state .gov/documents/organization/63562.pdf.
ize the insurgents.8
In a list of requirements to defeat an insurgency, leading
counterinsurgents had as their ªrst item to “identify and redress the political,
economic, military, and other issues fueling the insurgency.”9
More recently,
Ian Beckett argues that counterinsurgency success requires government
reforms to address popular grievances.10
The 2006 U.S. Army/Marine Corps Counterinsurgency Field Manual exempliªes
the good governance approach, drawing on sixty years of Western practitioner
accounts of successful counterinsurgency campaigns.11
“Soldiers and
Marines are expected to be nation builders as well as warriors,” Gen. David
Petraeus and Gen. James Amos write. “They must be prepared to help reestablish
institutions and basic services. They must be able to facilitate establishing
local governance and the rule of law.”12
Gen. Peter Chiarelli explains
the logic based on his experience in Iraq: “A gun on every street corner, although
visually appealing, provides only a short-term solution and does
not equate to long-term security grounded in a democratic process.”13
Adopting reforms as a weapon against the insurgency, counterinsurgents become
responsible for the people’s well-being.14
Successful counterinsurgency
is “armed social work.”15
The good governance approach includes direct military targeting of insurgents,
but governments must prevent harm to civilians because harm will only
increase support for the insurgency.16
This emphasis on discriminate force is
more prominent in contemporary work than in the post–World War II literature,
but as David Galula writes, “every military move has to be weighed with
regard to its political effects, and vice versa.” “Only attack insurgents when
they get in the way,” advises the Counterinsurgency Field Manual. General
International Security 42:1 84
8. See, for example, Julian Paget, Counter-Insurgency Operations: Techniques of Guerrilla Warfare
(New York: Walker, 1967), p. 177; Robert Thompson, Defeating Communist Insurgency: Experiences
from Malaya and Vietnam (London: Chatto and Windus, 1966), pp. 50–55, 72; and U.S. Department of
the Army, The U.S. Army/Marine Corps Counterinsurgency Field Manual (U.S. Army Field Manual
No. 3-24/Marine Corps Warªghting Publication No. 3-33.5) (Chicago: University of Chicago Press,
2006), p. 51.
9. Stephen T. Hosmer and Sibylle O. Crane, Counterinsurgency: A Symposium, April 16–20, 1962
(Santa Monica, Calif.: RAND Corporation, 1962), p. iv.
10. Ian F.W. Beckett, Insurgency in Iraq: An Historical Perspective (Carlisle, Pa.: Strategic Studies Institute,
U.S. Army War College, 2005).
11. U.S. Department of the Army, The U.S. Army/Marine Corps Counterinsurgency Field Manual.
12. Ibid., p. xlvi
13. Peter W. Chiarelli and Patrick R. Michaelis, “Winning the Peace: The Requirement for FullSpectrum
Operations,” Military Review, July/August 2005, pp. 4–17, at p. 4.
14. U.S. Department of the Army, The U.S. Army/Marine Corps Counterinsurgency Field Manual,
p. 295.
15. Ibid., p. 299.
16. Ibid., p. 45.
Petraeus told troops in Afghanistan, “If we kill civilians or damage their property
. . . we will create more enemies than our operations eliminate.”17
Given the struggles of the United States and its partners in Iraq,
Afghanistan, and elsewhere, however, as well as counterinsurgency successes
that do not look much like results of the good governance approach (e.g., in
Chechnya, Turkey, and the Philippines), to what degree has this prescription
for counterinsurgency success actually achieved its objectives?
limits to the good governance approach
The internal logic of the good governance approach is solid, but the supporting
research suffers from two main problems. First, it lacks theoretical rigor.
Second, it is based on unexamined assumptions. The lack of rigor hinders consideration
of which types of government behavior are more likely to contribute
to success in which types of cases. Unexamined assumptions distort
analysis by drawing attention away from potentially important elements.
To date, there has been little research providing a rigorous theory of the
good governance approach, including speciªcation of the conditions under
which the approach holds. Research that supports the governance approach
generally does not delineate the domain in which best practices operate to best
effect.18
Further, researchers rarely engage in systematic comparisons of cases,
and they pay little attention to external validity—in one case, an author generalized
his ªndings on defeating insurgency based on his experience as a counterinsurgent
military ofªcer in two unusual campaigns.19
Others generalize
based on the experiences of a small number of intervening great powers—for
example, in a relatively rare type of case in which a great power faces an insurgency
demanding national liberation.20
Recent data-driven quantitative reThe
“Hearts and Minds” Fallacy 85
17. Here, “contemporary” refers to counterinsurgency work dealing with the U.S. loss in Vietnam
and with the U.S. struggles in Afghanistan and Iraq, in particular; “post–World War II” refers to
work on Western counterinsurgency responses to the anticolonial wars that followed the end of
World War II. David Galula, Counterinsurgency Warfare: Theory and Practice (1964; repr. New York:
Prager, 2006), pp. 5, 66; U.S. Department of the Army, The U.S. Army/Marine Corps Counterinsurgency
Field Manual, p. 300; David Petraeus, “COMISAF’s Counterinsurgency Guidance,” Washington
Post, August 1, 2010, www.washingtonpost.com/wp-srv/hp/ssi/wpc/afghanguidance.pdf;
and John Nagl, “Unprepared: Review—The Echo of Battle: The Army’s Way of War, by Brian Linn
McAllister,” RUSI Journal, Vol. 153, No. 2 (April 2008), pp. 82–89, doi:10.1080/03071840802103355.
18. See, for example, John T. Fishel and Max G. Manwaring, Uncomfortable Wars Revisited (Norman:
University of Oklahoma Press, 2006); U.S. Department of the Army, The U.S. Army/Marine
Corps Counterinsurgency Field Manual; Kalev I. Sepp, “Best Practices in Counterinsurgency,” Military
Review, May/June 2005, pp. 8–12; and David Kilcullen, The Accidental Guerrilla (New York:
Oxford University Press, 2009).
19. See, for example, Galula, Counterinsurgency Warfare.
20. See, for example, Thompson, Defeating Communist Insurgency; Galula, Counterinsurgency War-
search asks speciªc questions about the costs of the government’s use of force
and ways to gain popular support, but rarely asks what causes counterinsurgency
success.21
Moreover, much of the research that supports the good governance approach
makes potentially unwarranted assumptions about the campaigns
themselves. An example is the widespread assumption that counterinsurgent
governments deliver what they promise, thereby having positive effects on
the populace and negative effects on the insurgency.22
A related assumption
involves the role of reforms, including that the government has the capability
and will to institute them; that the populace desires them; and that the
majority has no political preferences and will side with the stronger side.23
Another problematic assumption is that foreign interveners have the ability to
decisively shape events.24
None of these assumptions is warranted by empirical
research.25
Exacerbating the problems discussed above is that many supporters of the
good governance approach have conªned their examinations to the secondary
literature on successful counterinsurgency campaigns. Although this literature
presents normatively appealing narratives, it downplays less palatable counterinsurgent
choices.26
Some research considers the government, the insurgency,
and the populace as unitary actors, occasionally also conºating the
counterinsurgent government and the intervening power backing it.27
Assuming
away domestic politics within the state experiencing the insurgency
confounds analysis, because insurgency is a domestic political problem.
The argument that good governance causes counterinsurgency success also
rests on the assumption that success requires popular support, and thus that
International Security 42:1 86
fare; John A. Nagl, Learning to Eat Soup with a Knife: Counterinsurgency Lessons from Malaya and Vietnam
(2002; repr. Chicago: University of Chicago Press, 2005); and Andrew F. Krepinevich Jr., The
Army and Vietnam (Baltimore, Md.: Johns Hopkins University Press, 1986).
21. See Berman and Matanock, “The Empiricists’ Insurgency.”
22. Karl Hack notes this problem in “The Malayan Emergency as Counterinsurgency Paradigm,”
Journal of Strategic Studies, Vol. 32, No. 3 (June 2009), pp. 383–414, doi:10.1080/01402390902928180.
23. See, for example, Galula, Counterinsurgency Warfare, pp. 8–9, 53. For challenges to this view, see
D. Michael Shafer, Deadly Paradigms: The Failure of U.S. Counterinsurgency Policy (Princeton, N.J.:
Princeton University Press, 1988); and Stathis N. Kalyvas, The Logic of Violence in Civil War (New
York: Cambridge University Press, 2006).
24. See, for example, Fishel and Manwaring, Uncomfortable Wars Revisited; and Krepinevich, The
Army and Vietnam.
25. Scholars noting these problems include Shafer, Deadly Paradigms; and Douglas J. Macdonald,
Adventures in Chaos: American Intervention for Reform in the Third World (Cambridge, Mass.: Harvard
University Press, 1992).
26. See, for example, Sepp, “Best Practices in Counterinsurgency”; and Christopher Paul, Colin P.
Clarke, and Beth Grill, Victory Has a Thousand Fathers: Sources of Success in Counterinsurgency (Santa
Monica, Calif.: RAND Corporation, 2010).
27. See, for example, Fishel and Manwaring, Uncomfortable Wars Revisited.
the populace is the center of gravity, “the hub of all power and movement, on
which everything depends.”28
It is not empirically obvious, however, that popular
support is necessary for success; the second Russian conºict in Chechnya
and the Turkish war on the Kurdistan Workers’ Party from 1984 to 1999 are
two campaigns suggesting that it may not be so. In each case, the government
experienced success without implementing reforms or limiting the use of
force. Similarly, insurgencies that rely on ªnancing from the sale of natural resources
such as gems or timber need little popular support to succeed.29
Distribution
of public goods may not be relevant if the fundamental issue is the
redistribution of power.30
Indeed, public goods may increase the ºow of resources
to the insurgency.31
These problems also appear in the leading alternative approach to counterinsurgency
success, which advocates targeting the insurgency directly with
military force and downplays questions about political efforts that may be
necessary for success.
force-based approaches to counterinsurgency success
There are two schools of thought in the post–World War II literature on counterinsurgency
success that emphasize the use of military force targeting insurgents.
One is the so-called guerrilla-centric or enemy-centric approach. The
other is the cost-beneªt approach. A lack of rigor and unquestioned assumptions
about the role of force and of politics in counterinsurgent success
limits the usefulness of both approaches.32
The “Hearts and Minds” Fallacy 87
28. Carl von Clausewitz, On War, Michael Howard and Peter Paret, trans. (Oxford: Oxford University
Press, 1976), p. 596.
29. Jeremy M. Weinstein, Inside Rebellion: The Politics of Insurgent Violence (New York: Cambridge
University Press, 2007).
30. See, for example, Shafer, Deadly Paradigms.
31. See, for example, Nathan Leites and Charles Wolf Jr., Rebellion and Authority: An Analytic Essay
on Insurgent Conºicts (Chicago: Markham, 1970); Charles Wolf Jr., “Insurgency and Counterinsurgency:
New Myths and Old Realities” (Santa Monica, Calif.: RAND Corporation, 1965), http://
www.rand.org/pubs/papers/P3132-1/; and Ken Dilanian, “U.S. Risks Wasting Billions More in
Afghanistan Aid, Report Says,” Los Angeles Times, June 17, 2011, http://articles.latimes.com/2011/
jun/17/world/la-fg-afghan-aid-20110617.
32. Scholars arguing that brutality is the primary means of defeating insurgents include Douglas
Porch, Counterinsurgency: Exposing the Myths of the New Way of War (New York: Cambridge University
Press, 2013); and David French, The British Way in Counterinsurgency, 1945–1967 (New York:
Oxford University Press, 2012). Alexander B. Downes argues that under certain narrow conditions,
indiscriminate violence can defeat an insurgency. See Downes, “Draining the Sea By Filling
the Graves: Investigating the Effectiveness of Indiscriminate Violence as a Counterinsurgency
Strategy,” Civil Wars, Vol. 9, No. 4 (December 2007), pp. 420–444, doi:10.1080/13698240701699631.
Andrew J. Birtle argues that the use of force has been more determinative than reforms. See
Birtle, U.S. Army Counterinsurgency and Contingency Operations Doctrine, 1942–1976 (Washington,
D.C.: U.S. Army Center of Military History, 2006). Gil Merom argues that democracies fail at counterinsurgency
because domestic politics forestall effective uses of force. See Merom, How Democ-
The guerrilla-centric or enemy-centric approach to counterinsurgency success
prescribes defeating an insurgency militarily. Cold War–era French army
ofªcer Roger Trinquier, for example, argues that force is the foundation of
counterinsurgency success.33
Historian Edward Luttwak notes the successful
application of terrible force, including collective punishment and reprisal killings,
in past successful counterinsurgencies.34
Retired U.S. Army ofªcer Ralph
Peters argues that “insurgencies overwhelmingly have been put down thoroughly
by killing insurgents.”35
In the 1960s, RAND researchers Charles Wolf and Nathan Leites developed
the cost-beneªt approach to counterinsurgency.36
In place of trying to gain the
allegiance of the populace through a time-consuming, expensive process of offering
reforms, they prescribed punishing civilians who do not cooperate and
providing beneªts to those who do. This carrot-and-stick approach should prevent
insurgents from obtaining resources such as food, shelter, and civilian recruits,
while military force should prevent insurgent violence.
Neither approach discussed above presents testable theories, identiªes the
conditions under which its prescription is likely to succeed, or provides a systematic
analysis of relevant cases. Each assumes that the government’s use of
violence will repress all of the insurgents’ political goals and that the use
of force will reduce rather than increase the level of popular protest.37
The Good Governance versus Coercive Theory of Success
In this section, I derive hypotheses on the good governance theory and the coercive
theory of counterinsurgency to assess their explanatory power.38
International Security 42:1 88
racies Lose Small Wars: State, Society, and the Failures of France in Algeria, Israel in Lebanon, and the
United States in Vietnam (New York: Cambridge University Press, 2003).
33. Roger Trinquier, Modern Warfare: A French View of Counterinsurgency (London: Pall Mall, 1964),
for example pp. 28, 48, 59, 63, 73–74, 84–86, 105.
34. Edward N. Luttwak, “Dead End: Counterinsurgency Warfare as Military Malpractice,”
Harper’s Magazine, February 2007, pp. 33–42.
35. Ralph Peters, “Progress and Peril: New Counterinsurgency Manual Cheats on the History
Exam,” Armed Forces Journal, February 2007, http://armedforcesjournal.com/progress-and-peril/;
and Ralph Peters, “Getting Counterinsurgency Right,” New York Post, December 21, 2006.
36. Wolf, “Insurgency and Counterinsurgency”; and Wolf and Leites, Rebellion and Authority.
37. Daniel Ellsberg and Richard Shultz note that this assumption may be particularly dangerous if
individuals are more likely to respond to force with force rather than with submission. See
Ellsberg, “Revolutionary Judo: Working Notes on Vietnam No. 10” (Santa Monica, Calif.: RAND
Corporation), 1970; and Shultz, “Coercive Force and Military Strategy: Deterrence Logic and the
Cost-Beneªt Theory of Counterinsurgency Warfare,” Western Political Quarterly, Vol. 32, No. 4
(December 1979), pp. 444–466, doi:10.1177/106591297903200408.
38. See Stephen Van Evera, Guide to Methods for Students of Political Science (Ithaca, N.Y.: Cornell
University Press, 1997), p. 26, on inferring a theory.
good governance theory
In the good governance theory of counterinsurgency success, the independent
variable is good governance and the dependent variable is insurgent defeat.
I deªne success according to the U.S. Government Counterinsurgency Guide as
the “marginalization of the insurgents to the point at which they are destroyed,
co-opted, or reduced to irrelevance in numbers and capability.”39
The theory includes two intervening variables: the discriminate use of
force against the insurgency and the implementation of reforms. I operationalize
the military variable as the systematic avoidance of harm to civilian lives
and property, even at military cost, as the Geneva Conventions require.40
I
operationalize the political variable as systematic reforms beneªting the entire
population, including elites. Systematic reforms are costly to elites because
they take wealth and power from those who rule and distribute them to those
who do not. The cause of success is the government’s gain in popular support
and the resulting ºow of information on the insurgency, and the insurgency’s
corresponding loss of support and information, as well as its military defeat.
A series of predictions ºow from the good governance theory. First, if the
targeting of insurgents is highly discriminate, sparing civilian damage even at
military cost, popular support for the government should rise. Second, if the
government introduces reforms beneªting all, its popular support should increase,
and the insurgency should be signiªcantly weakened.
If, however, the insurgency is substantially weakened or defeated through
the use of force before improvements in governance are made, or before there
is a shift in public support to the state, or if there is no shift in popular support
to the state, or if no or only highly limited reforms are implemented by the
government, then the governance theory cannot explain counterinsurgent success.
In the next section, I lay out the coercion theory of counterinsurgency
success, compare it to the governance theory, and identify scope conditions for
both theories.
coercive counterinsurgency
In the coercion theory of counterinsurgency success, the independent variable
is coercion. There are two intervening variables, one military and one politiThe
“Hearts and Minds” Fallacy 89
39. U.S. Government Interagency Counterinsurgency Initiative, U.S. Government Counterinsurgency
Guide (Washington, D.C.: U.S. Government Printing Ofªce, 2009), p. 4.
40. The Geneva Conventions require armed actors in war not to attack civilians, not to use or
threaten violence to spread terror, and to protect civilians from the effects of military operations,
including by reducing the effects of operations that could affect civilians. See Geneva Conventions
of 1949, Article 4, August 12, 1949, https://ihl-databases.icrc.org/ihl/WebART/375-590007
?OpenDocument; and Geneva Conventions of 1949, Protocol 2, June 8, 1977, https://ihl-databases.icrc.org/ihl/INTRO/475?OpenDocument
(not ratiªed by the United States).
cal. I operationalize the ªrst variable as a military campaign to destroy the insurgency’s
capability and will to continue ªghting. The campaign has two
facets, direct and indirect. The direct military effort is an attrition campaign
against the insurgency (this is also assumed to take place in the good governance
theory).41
The indirect military effort uses brute force to block the ºow
of resources from civilians to insurgents.42
I operationalize the second variable
as accommodation—that is, the use of threats and rewards to gain the cooperation
of political and military leaders outside the government in exchange for
intelligence on the insurgency and the provision of military capabilities to the
government.43
The process involves the use of coercion to break the insurgency
as a ªghting force and organization based on the use of the threat, display,
or application of force to coerce changes.44
The government gains the cooperation of elites by making accommodations.
“Accommodations” denote beneªts to some, but not all, political actors.45
The
beneªts may include direct payments; the granting of impunity for criminal
activity or violence; access to material resources such as timber or revenueproducing
checkpoints; and access to nonmaterial resources, such as the granting
of an ofªcial position (which may also prove lucrative in material terms) to
a rival elite (see table 1).
The government uses accommodations to co-opt elites because these accommodations
are far less costly to government elites than are good governance
International Security 42:1 90
41. Reducing an actor’s capability to ªght also reduces his or her will. See Robert A. Pape,
Bombing to Win: Air Power and Coercion in War (Ithaca, N.Y.: Cornell University Press, 1996).
42. On identifying a new role for a variable, see Andrew Bennett, “Process Tracing and Causal Inference,”
in Henry E. Brady and David Collier, eds., Rethinking Social Inquiry: Diverse Tools, Shared
Standards (Lanham, Md.: Rowman and Littleªeld, 2010), p. 209. Western human rights norms
should lead representatives of a great power intervener to play down the role of force in the great
power’s success. That they do not do so in contemporaneous documents indicates that force
played a useful role.
43. Charles Tilly, The Politics of Collective Violence (New York: Cambridge University Press, 2003),
p. 34; Kalyvas, The Logic of Violence in Civil War; Paul R. Brass, The Production of Hindu-Muslim Violence
in Contemporary India (Seattle: University of Washington Press, 2003); Fotini Christia, Alliance
Formation in Civil Wars (New York: Cambridge University Press, 2013); Kimberly Marten, Warlords:
Strong-arm Brokers in Weak States (Ithaca, N.Y.: Cornell University Press, 2012); Goran Peic, “Civilian
Defense Forces, State Capacity, and Government Victory in Counterinsurgency Wars,” Studies
in Conºict and Terrorism, Vol. 37, No. 2 (2014), pp. 162–184, doi:10.1080/1057610X.2014.862904;
Steffan W. Schmidt et al., Friends, Followers, and Factions: A Reader in Political Clientism (Berkeley:
University of California Press, 1977); T. David Mason, Caught in the Crossªre: Revolution, Repression,
and the Rational Peasant (Lanham, Md.: Rowman and Littleªeld, 2004); and Paul Staniland, “States,
Insurgents, and Wartime Political Orders,” Perspectives on Politics, Vol. 10, No. 2 (June 2012),
pp. 243–264, doi:10.1017/S1537592712000655.
44. Thomas C. Schelling, Arms and Inºuence (New Haven, Conn.: Yale University Press, 1966).
45. Glenn Herald Synder and Paul Diesing deªne accommodation as demands, offers, and concessions.
See Snyder and Diesing, Conºict among Nations: Bargaining, Decision Making, and System
Structure in International Crises (Princeton, N.J.: Princeton University Press, 1977), p. 195.
reforms.46
In contrast, reforms change the distribution of power and wealth
within the state. They represent major, permanent adjustments in the structures
or policies governing the distribution and exercise of power within the
state, and they affect everyone within it. Reforms include political liberalization
(i.e., creation of institutions assuring free-and-fair elections, freedom of
political expression, protection of civil and human rights, and redistribution
of property or income).47
The government needs the cooperation of rival elites because relatively few
individuals have the knowledge and skills necessary to successfully target an
insurgency militarily and politically. Civilians may have tactical information
on, for example, where they last saw insurgents or who is related to whom.
They are less likely to have politically signiªcant information on, for example,
the interests of leading insurgents and on rifts or factions within the insurgency.
Elites include defectors; sources within the insurgency; and social, intellectual,
and business leaders. They also may share nonstate military resources,
such as militias, with the government.
The coercion theory analyzes counterinsurgency as primarily a domestic political
process of violent state-building. Historically, the process has been convulsive.
Political order arises from elite efforts to reduce violent political
rivalry. Elites seek the minimum winning coalition necessary to retain power,
and they rule to protect their own interests, not those of the populace.48
The “Hearts and Minds” Fallacy 91
46. Thomas Carothers, “The ‘Sequencing’ Fallacy,” Journal of Democracy, Vol. 18, No. 1 (January
2007), pp. 12–27.
47. Roland Paris, At War’s End: Building Peace after Civil Conºict (New York: Cambridge University
Press, 2004), p. 60.
48. Charles Tilly, Coercion, Capital, and European States, 990–1992 (New York: Cambridge University
Press, 1992); Douglass C. North, John Joseph Wallis, and Barry R. Weingast, “Violence and
the Rise of Open-Access Orders,” Journal of Democracy, Vol. 20, No. 1 (January 2009), pp. 55–68,
doi:10.1353/jod.0.0060; David A. Lake, “Anarchy, Hierarchy, and the Variety of International
Relations,” International Organization, Vol. 50, No. 1 (Winter 1996), pp. 1–33, doi:10.1017/
S002081830000165X ; Bruce Bueno de Mesquita et al., “Political Institutions, Political Survival, and
Policy Success,” in Bueno de Mesquita and Hilton L. Root, eds., Governing for Prosperity (New HaTable
1. Examples of Political Change
Costly Changes Beneªting All Less Costly Changes Beneªting a Few
Government Reforms for All Government Accommodation of Selected Elites
Redistribute land to peasants Trade impunity for information
Liberalize political system Make side payments to rival elites
Protect human/civil rights Deliver basic foodstuffs/medical care on irregular
basis to some communities/individualsBuild democratic institutions
In line with the existing literature, the coercion theory recognizes that
counterinsurgency success is about political interests, as war always is.49
The
coercion theory contrasts with the good governance approach to counterinsurgency
success in three important ways, however. First, it identiªes the signiªcance
of conferring accommodations on elites rather than proposing and
implementing reforms to gain popular support. Second, it identiªes the critical
role of military control of civilians, rather than their protection from harm (see
table 2). Third, the coercion theory is explanatory, whereas the good governance
approach is prescriptive or programmatic.50
Two predictions ºow from the coercion theory.51
First, if the government accommodates
rival elites, then its intelligence and military capabilities should
increase. Second, if it uses its increased capabilities to more effectively target
insurgents and their resource ºows, including controlling the populace,
then the insurgency should be signiªcantly weakened.
The coercion theory cannot explain success if any of the following three conditions
applies: ªrst, the government does not control the populace; second,
the government implements reforms beneªting all, gaining support among
the population while reducing support for the insurgency, before using its
strengthened capabilities against civilians and the insurgency to defeat insurgent
forces; and third, the government systematically avoids harm to civilians,
International Security 42:1 92
ven, Conn.: Yale University Press, 2000), pp. 59–84; Bruce Bueno de Mesquita et al., “Political Institutions,
Policy Choice, and the Survival of Leaders,” British Journal of Political Science, Vol. 32, No. 4
(October 2002), pp. 559–590, doi:10.1017/S0007123402000236; Ronald Wintrobe, “The Tinpot and
the Totalitarian: An Economic Theory of Dictatorship,” American Political Science Review, Vol. 81,
No. 3 (September 1990), pp. 849–972, doi:10.2307/1962769; Daron Acemoglu and James A. Robinson,
Why Nations Fail: The Origins of Power, Prosperity, and Poverty (New York: Crown, 2012); Elizabeth
N. Saunders, “War and the Inner Circle: Democratic Elites and the Politics of Using Force,”
Security Studies, Vol. 24, No. 3 (October 2015), pp. 466–501, doi:10.1080/09636412.2015.1070618;
William Riker, The Theory of Political Coalitions (New Haven, Conn.: Yale University Press, 1962);
and Stathis N. Kalyvas, “The New U.S. Army/Marine Corps Counterinsurgency Field Manual as
Political Science and as Political Praxis,” Perspectives on Politics, Vol. 6, No. 2 (June 2008), pp. 347–
360, doi:10.1017/S1537592708081176.
49. Clausewitz, On War, p. 87.
50. Charles Tilly, “Mechanisms in Political Processes,” Annual Review of Political Science, Vol. 4
(2001), pp. 21–41, doi:10.1146/annurev.polisci.4.1.21.
51. Van Evera, Guide to Methods for Students of Political Science, pp. 35, 65.
Table 2. Examples of Contrasting Military Treatment of Civilians
Avoid Civilian Damage, Even at Military Cost Forcefully Control Civilians
Cancel/delay military operations Force civilians into controlled areas
Avoid use of area weapons Target communities militarily
Put military forces at risk to avoid civilian damage Limit civilian access to resources
even at military cost.52
Figure 1 presents a comparison of the predictions of the
two theories and the case outcomes. In the next section, I describe the research
design I used to gauge the relative explanatory power of these two theories.
Research Design
In this article, I inquire into the conditions under which a counterinsurgent client
government backed by a liberal great power patron succeeds in retaining
power against an insurgency.53
The universe of cases consists of internal
conºicts from 1945 to 2017 in which a liberal Western great power successfully
intervened to support a client state facing an insurgency, and in which no sigThe
“Hearts and Minds” Fallacy 93
52. Bennett, “Process Tracing and Causal Inference.” See also Nina Tannenwald, “Process Tracing
and Security Studies,” Security Studies, Vol. 24, No. 2 (April/June 2015), pp. 219–217, doi:10.1080/
09636412.2015.1036614.
53. I use the lens of patron-client relations rather than the more frequently applied principal-agent
problem for two reasons. First, the principal-agent problem makes the great power the major actor,
when in fact the counterinsurgent government is the principal, whose decisions are critical to the
campaign and who faces higher potential costs. Second, the principal-agent problem focuses on
the agent’s lack of information relative to the principal. In the relationship between a counterinsurgent
state and its great power backer, the counterinsurgent state has more information than its
backer.
Figure 1. Rival Predictions of Good Governance and Coercion Theories, Case Outcomes
Good
governance
theory
lowerlower higherhigher success
Coercion
theory
higherhigher lowerlower success
Malaya higherhigher lowerlower success
Dhofar higherhigher lowerlow
to
moderate
success
El
Salvador
lowerhigher lowerlower negotiated
success
Accommodations
benefiting
selected elites
Control
civilians
by force
Reforms
benefiting
all
Protect civilians
from harm,
even at military
cost Outcome
niªcant recurrence of violence followed within ªve years. I do not distinguish
between irregular and conventional conºict, because many insurgents and
counterinsurgents engage in both. I focus on the post–World War II period because
norms on governmental treatment of noncombatants have changed
since 1945 with the adoption of the Geneva Conventions, and because the
state’s treatment of civilians plays an important role in counterinsurgency—
either because they must be sheltered from harm, according to the good governance
theory, or because they must be tightly controlled by force, according to
the coercion theory.
The domain for this analysis consists of ªve campaigns in which the government
succeeded in preventing an insurgent takeover. Proponents of the good
governance approach claim these as exemplary for their relatively high level of
reforms and relatively low level of intentional use of force against civilians.54
These are also cases in which the liberal great power patron insisted that reforms
were necessary to defeat the insurgency and pressed its client hard to
attain them despite resistance.55
Thus, these campaigns constitute cases that
the good governance theory should be able to explain and my coercion theory
should fail to explain.
I analyze the campaigns to ask what the government did and when and
where it did it. I do not rely on what the government said it would do.56
Asking what governments actually did in their successful campaigns identiªes
choices overlooked or downplayed in most work on counterinsurgency. Public
support, meanwhile, is difªcult to assess and more so in wartime. I evaluate
popular support by examining the level of participation in government initiatives
intended to show or gain support.57
The case I discuss most fully here is the 1946–60 Malayan Emergency, in
which Britain defended its colony from the Malaya National Liberation Army
(MNLA), a group composed of communist and nationalist insurgents. I focus
International Security 42:1 94
54. On case selection according to intrinsic importance, see Van Evera, Guide to Methods for Students
of Political Science, p. 77. This criterion resembles what Aaron Rapport calls a countervailing
condition: a variable whose presence or speciªc value decreases the probability that the outcome
posited by the theory being tested will be evident. In this study, the low level of reforms and high
level of force used against civilians in Malaya make it unlikely, according to the governance theory,
that the counterinsurgency campaign would succeed—and yet it did. See Rapport, “Hard
Thinking about Hard and Easy Cases in Security Studies,” Security Studies, Vol. 24, No. 3 (2015),
pp. 431–465, doi:10.1080/09636412.2015.1070615. See also George and Bennett, Case Studies and
Theory Development in the Social Sciences, pp. 121–122.
55. See, for example, Andrew Mumford, The Counter-Insurgency Myth: The British Experience of
Irregular Warfare (London: Routledge, 2012).
56. George and Bennett, Case Studies and Theory Development in the Social Sciences, p. 86.
57. William A. Barnes, “Incomplete Democracy in Central America: Polarization and Voter Turnout
in Nicaragua and El Salvador,” Journal of Interamerican Studies and World Affairs, Vol. 40, No. 3
(Fall 1998), pp. 63–101, doi:10.1111/j.1548-2456.1998.tb00063.x.
on Malaya because the Emergency is widely believed to be an example of the
right way to do counterinsurgency.58
I also consider more brieºy the 1965–75
British-backed campaign in Dhofar, Oman, and the 1979–92 U.S.-supported
campaign in El Salvador. To do so, I examined memoirs, oral histories, and
contemporaneous documents in U.S. and British archives. I also reviewed the
secondary literature and interviewed participants.
Neither the good governance theory nor the coercion theory explains all
cases of counterinsurgency success, for two reasons. First, the scope of both
theories includes only cases in which insurgents rely primarily on the populace
for resources.59
Second, the analysis includes no cases of ethnic conºict,
which some scholars argue differ in important ways from other types of internal
conºict.60
To be sure, one might argue that both theories sacriªce ªdelity to reality for
reasons of elegance. The goal of theory building, however, is to identify causal
patterns in phenomenon across time and space, not to fully explain any single
event, act, or choice. Thus, by deªnition, theory is parsimonious. In addition,
the declining focus in political science on creating and reªning theories
increases the value of work taking on this task.61
Some critics might say that these theories are not mutually exclusive because
governments employ a mixture of strategies when ªghting an insurgency.
This criticism is unfounded, because I do not claim that governments
must choose one theory or the other. Rather, I argue that the good governance
theory errs in its identiªcation of the causes of counterinsurgency success.
The “Hearts and Minds” Fallacy 95
58. The lead author of the U.S. Army’s 2006 counterinsurgency manual, Conrad C. Crane, states
that the authors found positive examples in Malaya, El Salvador, the Philippines-Huks campaign
of 1946–55, and the 1964–2016 conºict in Colombia, along with the Civil Operations and Revolutionary
Development Support program in Vietnam. Crane, email to author, September 29, 2010.
See also Nagl, Learning to Eat Soup with a Knife; David W. Barno, “Fighting the ‘Other War’: Counterinsurgency
Strategy in Afghanistan, 2003–2005,” Military Review, September/October 2007,
pp. 32–44; Daniel Marston and Carter Malkasian, eds., Counterinsurgency in Modern Warfare (New
York: Oxford University Press, 2008); Thomas R. Mockaitis, “Resolving Insurgencies” (Carlisle,
Pa.: U.S. Army War College Strategic Studies Institute, 2011); and Anthony James Joes, Resisting
Rebellion: The History and Politics of Counterinsurgency (Lexington: University Press of Kentucky,
2004), p. 232.
59. See, for example, Weinstein, Inside Rebellion.
60. Chaim Kaufmann, “Intervention in Ethnic and Ideological Civil Wars: Why One Can Be Done
and the Other Can’t,” Security Studies, Vol. 6, No. 1 (Autumn 1996), pp. 62–103, doi:10.1080/
09636419608429300; Chaim Kaufmann, “Possible and Impossible Solutions to Ethnic Civil Wars,”
International Security, Vol. 20, No. 4 (Spring 1996), pp. 136–175, doi:10.1162/isec.20.4.136, and
Stathis N. Kalyvas, “Ethnic Defection in Civil War,” Comparative Political Studies, Vol. 41, No. 8 (August
2008), pp. 1043–1068, doi:10.1177/0010414008317949.
61. John J. Mearsheimer and Stephen M. Walt, “Leaving Theory Behind: Why Simplistic Hypothesis
Testing Is Bad for International Relations,” European Journal of International Relations, Vol. 19,
No. 3 (2013), pp. 427–457, doi:10.1177/1354066113494320; and Van Evera, Guide to Methods for Students
of Political Science, pp. 7–8, 15–16, 17–21.
I also do not consider whether a counterinsurgent tries to implement either
theory as a strategy. What matters for analyzing the relative predictive value
of the two theories is whether the process ending with counterinsurgent
success unfolds with the predicted values on the intervening variables. In
the next section, I examine how the two theories’ predictions fare in the case
of the Malayan Emergency, with occasional references to the Dhofar and
El Salvador cases.
Malaya
The British colonial administration in Malaya conducted a successful counterinsurgency
campaign from 1946 to 1960 against the MNLA, which was ªghting
for an independent communist state. The Malayan Emergency, declared in
1948, lasted longer than the insurgency, which the security forces defeated as
an organization and as a ªghting force within the ªrst few years of the campaign.
Government success was likely, given the small number of insurgents;
their lack of external support; the small, marginalized popular base they coerced
into providing support; and British advantages, including relative military
power, geography (the Malay Peninsula is relatively easy to cut off from
smuggling efforts); and British familiarity with the region and peoples dating
to the eighteenth century. At the same time, however, the government in
London and British ofªcials and military ofªcers in Malaya did not consider
success a given.
The insurgency began in 1946 with a terrorism campaign launched by
the MNLA against Malayan economic targets. At its height, the group consisted
of 5,000–10,000 lightly armed guerrillas and a political organization
based in the ethnic Chinese community.62
The insurgents drew most of their
support from tin miners and rubber tappers living in squalid camps on the
jungle’s fringes. The MNLA used extortion and coercion to dominate the isolated
camp dwellers.63
The group’s goals limited its ability to grow. Its predominantly ethnic
Chinese members sought support based not on a shared ethnicity, but on the
narrowly appealing message of communism disseminated to poor communities
and the more broadly appealing message of independence from Britain.64
International Security 42:1 96
62. Robert Taber, War of the Flea: The Classic Study of Guerrilla Warfare (Dulles, Va.: Brassey’s, 2002),
p. 139; and Robert W. Komer, “The Malayan Emergency in Retrospect: Organization of a Successful
Counterinsurgency Effort” (Santa Monica, Calif.: RAND Corporation, 1972), p. 7.
63. Komer, “The Malayan Emergency in Retrospect,” pp. 7–8.
64. Director of Operations, Review of the Emergency Situation in Malaya at the End of 1954, January
10, 1955, DEFE 11/105, National Archives, Kew, United Kingdom (TNA); and Gurney to
Residents of poor communities beyond the jungle squatters showed little interest
in politics. Independence appealed to Malaya’s elites, while communism
did not. The ethnic Chinese community, meanwhile, was fragmented by national
origin, views on the 1949 Chinese Revolution, and the kind of work in
which its members engaged.65
Malaya’s population of approximately 4.55 million
was 49 percent ethnic Malay, 38 percent ethnic Chinese, and 12 percent
ethnic Indian, with a sprinkling of other ethnicities.66
There was relatively little
anti-British feeling after the Allies’ victory at the end of World War II drove out
the Japanese occupiers.67
Ethnic Malay elites included the traditional rulers of the peninsula’s constitutive
states and English-speaking Malay administrators who feared losing
traditional Malay dominance after Malaya gained independence from
Britain.68
As the possibility of independence drew closer, ethnic Chinese and
ethnic Indian elites grew anxious about protecting their business interests
within the new state. In contrast, British authorities found that independence
mattered little to the masses. The ethnic Malay population lived mostly in rural,
self-supporting communities and was less prosperous than either the ethnic
Chinese or ethnic Indians. Ethnic Indians, like the Malays, feared that the
establishment of a liberal democracy in Malaya (the British goal) would mean
ethnic Chinese domination, because the ethnic Chinese constituted the secondlargest
ethnic population in Malaya and were its most prosperous residents.
The conventional wisdom regarding the Malayan Emergency is that after its
initial failure to defeat the insurgency through the indiscriminate use of force,
the British military turned to seeking popular support; initiated reforms, including
the granting of independence; obtained information about the insurgency;
and defeated it while avoiding intentional harm to civilians. During the
ªrst two years of the campaign, according to this view, the British relied on
massive sweeps that only alienated the populace.69
With the appointment of
Lt. Gen. Sir Harold Briggs as director of operations in 1950, the campaign
turned political. Briggs’s strategy included the creation of “New Villages,”
designed to protect civilians from insurgents; delivery of basic goods to needy
The “Hearts and Minds” Fallacy 97
Creech Jones, Secretary of State for the Colonies, Federation of Malaya Despatch No. 3 from King’s
House, Kuala Lumpur, Malaya, January 12, 1950, A. Creech Jones Ofªcial Papers IV Asia, folder 2,
box 57, MSS Brit. Emp. s. 332, Weston Library, Bodleian Library, Oxford University (Bodleian).
65. Karl Hack, Defence and Decolonisation in Southeast Asia: Britain, Malaya, and Singapore 1941–68
(Richmond, U.K.: Curzon, 2001), p. 129.
66. John Newsinger, British Counterinsurgency: From Palestine to Northern Ireland (London:
Palgrave, 2002), p. 37.
67. Julian Padget, Counter-Insurgency Operations: Techniques of Guerrilla Warfare (New York: Walker,
1976), p. 75.
68. Brownlee, Authoritarianism in an Age of Democratization, p. 57.
69. Newsinger, British Counterinsurgency, pp. 47–48.
civilians; co-optation of the insurgents’ message concerning Malayan independence;
and unity of effort among all government actors.70
Recent revisionist work underlines the forceful character of the British campaign,
however, including violence against civilians. The ªrst years of the
campaign involved the state’s use of terrorist tactics to intimidate ethnic
Chinese and others into supporting the government. From 1948 to 1949, the
government engaged in mass arrests, the destruction of property, forced population
movements, and the use of lethal force, with military assessments
ªnding these measures effective.71
A British reassessment in 1950 led to a focus
on more discriminate uses of force and an increase in population and resources
controls. In this second phase, British tactics included continued
forced resettlement and collective punishment. This time, however, the punishment
was more likely to be applied to communities actually supporting insurgents.
The ªnal phase of the campaign included the conferral of social
beneªts on some members of the populace and more targeted offensive action
against the insurgents.72
In Dhofar, British ofªcers led the ªght against the communist-nationalist insurgents
in Oman’s southernmost, poorest, and most isolated region. Analysts
often present this campaign as a model of counterinsurgency success.73
In
El Salvador, U.S. forces trained and advised Salvadoran forces and policymakers
in their battle against a nationalist-Marxist-Socialist–Social Democratic
insurgency. It, too, is often presented as a model campaign.74
El Salvador difInternational
Security 42:1 98
70. Paget, Counter-Insurgency Operations, p. 59; Komer, “The Malayan Emergency in Retrospect,”
pp. 54–56; and Newsinger, British Counterinsurgency, p. 54.
71. Huw Bennett, “‘A Very Salutary Effect’: The Counter-Terror Strategy in the Early Malayan
Emergency, June 1948 to December 1949,” Journal of Strategic Studies, Vol. 32, No. 3 (2009), pp. 415–
444,doi:10.1080/01402390902928248.
72. Karl Hack, “‘Iron Claws on Malaya’: The Historiography of the Malayan Emergency,” Journal
of Southeast Asian Studies, Vol. 30, No. 1 (March 1999), pp. 99–125, doi:10.1017/S0022463400008043;
and Bennett, “‘A Very Salutary Effect.’”
73. See, for example, Thomas R. Mockaitis, British Counterinsurgency in the Post-Imperial Era (Manchester,
U.K.: Manchester University Press, 1995); Michael Dewar, Brush Fire Wars: Minor Campaigns
of the British Army since 1945 (London: Robert Hale, 1984); John Pimlott, “The British Army:
The Dhofar Campaign, 1970–1975,” in Ian F.W. Beckett and Pimlott, eds., Armed Forces and Modern
Counterinsurgency (New York: St. Martin’s, 1985), pp. 16–45; Geraint Hughes, “A ‘Model Campaign’
Reappraised: The Counterinsurgency War in Dhofar, Oman, 1965–1975,” Journal of Strategic
Studies, Vol. 32, No. 2 (April 2009), pp. 271–305, doi:10.1080/01402390902743357; Walter C.
Ladwig III, “Supporting Allies in Counterinsurgency: Britain and the Dhofar Rebellion,” Small
Wars & Insurgencies, Vol. 19, No. 1 (March 2008), pp. 62–88, doi:10.1080/09592310801905793; Calvin
H. Allen Jr. and W. Lynn Rigsbee II, Oman under Qaboos: From Coup to Constitution, 1970–1996
(2000; repr. London: Frank Cass, 2001); and James Worrell, Statebuilding and Counterinsurgency in
Oman: Political, Military, and Diplomatic Relations at the End of Empire (London: I.B. Taurus, 2014).
74. On El Salvador as a model, see, for example, Crane, email to author; Steven Metz, “Learning
from Iraq: Counterinsurgency in American Strategy” (Carlisle, Pa.: Strategic Studies Institute, U.S.
Army War College, January 2007); Paul, Clarke, and Grill, Victory Has a Thousand Fathers; and An-
fers from the other cases discussed here in that the government remained in
control of the state despite not defeating the insurgency. The war ended with a
peace agreement.
In all three cases, as I discuss in the next section, the government survived
the insurgent challenge. In Malaya, the campaign included offensive operations
against the insurgents and systematic violence against civilians; the government
undertook no reforms; it attracted no popular support; but it did gain
elite cooperation—thus lending support to the coercion theory rather than the
good governance theory. In Dhofar, the military forcefully controlled the civilian
population, though not to the extent seen in Malaya, cutting off the ºow of
resources to insurgents. Successive sultans refused to initiate reforms, but the
military accommodated rival elites to gain information on the insurgents and
enhance its military power. In all three cases, the government failed to gain
popular support. In El Salvador, the military cleared entire rural areas without
cutting links between the insurgents and the people; the government failed
to make reforms; and it did not win over rival elites. The differences in the
El Salvador case compared with the Malaya and Dhofar cases conªrm the role
of elite cooperation with the counterinsurgency and the military’s forceful control
of civilians as identiªed in the coercion theory.
a quick military defeat of the malayan insurgency
The British use of force in Malaya, including indiscriminate aerial bombing
and the use of area weapons, as well as the forceful control of civilians and the
resources they and the insurgents needed to live, led to the 1949 defeat of
the threat posed by the MNLA.75
As of 1948, two years into the insurgency, the government’s use of violence
had forced the insurgents to break up into smaller bands. The insurgency’s
communications capabilities and ability to coordinate action were crippled; it
was short of money; and its number of attacks had declined, as had insurgent
morale.76
Captured documents reveal the consequences of the government’s
The “Hearts and Minds” Fallacy 99
gel Rabasa et al., Money in the Bank: Lessons Learned from Past Counterinsurgency (COIN) Operations
(Santa Monica, Calif.: RAND Corporation, 2007).
75. On the aim of killing bandits (insurgents) and the role of the Royal Air Force (RAF) as artillery,
see Advanced Air Headquarters, Malaya, RAF Operations in Malaya during 1950, February 9,
1951, AIR 23/8443, TNA. On the threat being eliminated by 1948, see Commissioner-General
Southeast Asia to Secretary of State for the Colonies, inward telegram, April 20, 1949, DEFE 11/32,
TNA.
76. On the poor quality of insurgent communications, see Gurney to Secretary of State for the Colonies,
inward telegram, September 18, 1948, DEFE 11/32, TNA. On the improbability of a coordinated
national insurgent offensive, see Memorandum by the United Kingdom Government,
Meeting of Prime Ministers, October 8, 1948, DEFE 11/32, TNA. On insurgent difªculties, see
Paper on the Security Situation in the Federation of Malaya, April 5, 1949, DEFE 11/32, TNA. On
actions as of March 1949, particularly its forceful resettlement of civilians into
guarded communities.77
A surrendered insurgent reported in September and
October 1949 that enthusiasm for the communists was waning and that insurgents
in his area of Selengor state had been short of ammunition for ten
months.78
MNLA leader Chin Peng dated the peak and beginning of the decline
of the insurgency to 1949–50, largely as a result of its losing contact with
civilians forced to live under military control.79
British authorities, too, recognized
the effectiveness of the military’s resettlement of civilians in reducing
the MNLA’s ability to survive.80
By 1951, the government’s use of force had shattered the insurgency as both
an organization and a ªghting force.81
With their leaders’ order in October to
withdraw to the jungle in response to the government’s tight control of the
population and resources, insurgents could no longer mass force, communicate,
or plan.82
Their ability to launch attacks had declined so dramatically that
the military changed its offensive focus from civilian resettlement and resource
controls to pursuit of the remaining rebel ªghters hiding deep in the jungle.83
International Security 42:1 100
population and food controls hurting insurgents, see Gurney to Secretary of State for the Colonies,
inward telegram, March 12, 1949, DEFE 11/32, TNA. On insurgent concern about damage done to
the organization by population controls, see Gurney to Secretary of State for the Colonies, inward
telegram, January 22, 1949, DEFE 11/32, TNA. On immediate improvement in safety in areas
where resettlement was completed, see Director of Operations to British Defence Coordinating
Committee, Appreciation of the Military and Political Situation in Malaya as of October 25, 1959,
dated November 1, 1950, FO 371/84492, TNA.
77. Gurney to Secretary of State for the Colonies, inward telegram, March 12, 1949.
78. Statement by Surrendered Insurgent S.T.S. on September 23, 1949, Statements to the Criminal
Investigation Department by Communist bandits (insurgents) during the Malaya [sic] Emergency,
B.P. Walker-Taylor Collection, one volume, MSS. Ind. Ocn. s. 257, Bodleian; and Second Statement
by Surrendered Insurgent Liew Thian Choy on September 25, 1949, Statements to the Criminal
Investigation Department by Communist bandits (insurgents) during the Malaya [sic] Emergency,
B.P. Walker-Taylor Collection, one volume, MSS. Ind. Ocn. s. 257, Bodleian.
79. Chin Peng, Dialogues with Chin Peng: New Light on the Malayan Communist Party, C.C. Chin and
Karl Hack, eds. (Singapore: Singapore University Press, 2004), p. 144.
80. On record-high insurgent casualties and resettlement damaging insurgents, see Briggs, Director
of Operations, Malaya, Progress Report on Situation in Malaya, April 26, 1951, FCO 141/15533,
TNA. On stopping the rise in insurgent incidents by 1950 and the insurgency ªghting to survive,
see Director of Operations, Progress Report on the Emergency in Malaya, October 15, 1951, DEFE
11/46, TNA. On insurgents complaining of resettlement hurting them, see Director of Operations,
Progress Report on Situation in Malaya, February 15, 1951, FCO 141/15533, TNA. On insurgent
shortages of money, food, ammunition, and medical supplies, and on the difªculty of communicating,
see Memorandum by the United Kingdom Government, Meeting of Prime Ministers, October
8, 1948.
81. High Commissioner and Director of Operations, Combined Appreciation of the Emergency
Situation, June 4, 1951, DEFE 11/45 TNA; and Director of Operations, Progress Report on the
Emergency in Malaya.
82. Chin, Dialogues with Chin Peng, pp. 144–147.
83. From General Headquarters Far East Land Forces (FARELF) to War Ofªce, Reed, Situation Report
(Sitrep), received September 20, 1954, DEFE 11/142, TNA; and Memo on Additional Parachute
Troops, July 23, 1954, AIR 23/8556, TNA.
The number of terrorist-inspired incidents fell from a monthly average of 507
in 1951 to 89 in 1954.84
In 1951, two-thirds of insurgent attacks were concentrated
in just two states, Johore and Perak.85
As one government ofªcial noted
in 1953, in some areas of Perak, insurgents “survive only in small groups of
dispirited men and women.”86
Demoralized insurgents were on the run, making
ambushes of security forces rare by 1952.87
In 1952, 7,000 man-hours of patrolling
were required to see one insurgent.88
Military analysts attributed the
dramatic fall in military contact with insurgents (from 939 contacts in 1951 to
350 in 1955) “to the elimination rates of terrorists (killed, captured, surrendered,
and died of natural causes) being higher than recruitment rates.”89
As
security forces moved deeper into the jungle, the number of insurgents killed
or captured in government ambushes rose dramatically each month, from
about 30 in May 1952 to approximately 400 in May 1953.90
The increase in the number of insurgents surrendering to the government
was an indication of their weakening will to continue ªghting. In June 1951,
the number surged about 180 percent over the three previous months, with insurgent
casualties rising 42 percent.91
In 1954, reasons given for surrender
included food shortages, pressure from the security forces, and a loss of faith
in the possibility of victory.92
There were about 2,100 insurgents in 1957, with
9,581 killed, captured, or surrendered since June 1948.93
“Every aspect of the
Emergency is dwindling,” British authorities reported in 1957. The insurgents’
“primary aim for the past four years has been to evade contact.”94
In Dhofar, after six years of barely holding on against the insurgency, the
Sultan’s Armed Forces (SAF) ramped up offensive operations in 1971. The military
moved up from the coastal area—the only part of Dhofar still held by the
sultan—into the insurgent-held mountains. The SAF took territory and cleared
The “Hearts and Minds” Fallacy 101
84. Director of Operations, Review of the Emergency Situation in Malaya at the End of 1954.
85. Director of Operations, Progress Report on the Emergency in Malaya, October 15, 1951.
86. Mentri Besar (ªrst minister), Perak state, Monthly Political Intelligence Report for Period Ending
November 20, 1953, FCO 141/7377, TNA.
87. Secretary of Chiefs of Staff Committee, Report on Visit of Dr. Cockburn’s Party to Malaya, December
10, 1952, DEFE 11/49, TNA.
88. Ibid.
89. Operational Research Section (Malaya), Technical Note No. 4/55, Some Statistics of the Emergency
in Malaya from January 1954 to June 1955, dated August 22, 1955, WO 291/1757, TNA.
90. Operational Research Section (Malaya), Memorandum No. 6/53, Statistical Survey of Activities
by Security Forces in Malaya from May 1952 to April 1953, WO 291/1731, TNA.
91. High Commissioner and Director of Operations, Combined Appreciation of the Emergency
Situation.
92. Director of Operations, Review of the Emergency Situation in Malaya at the End of 1954, dated
January 10, 1955, DEFE 11/105, TNA.
93. Lt. Gen. R.H. Bower, Review of the Emergency in Malaya from June 1948 to August 1957,
dated September 12, 1957, WO 106/5990, TNA.
94. Ibid.
the more populated Eastern Area of Dhofar, and then did the same in the
Central Area. It used indiscriminate air strikes and artillery to ªnd and attack
insurgents.95
The SAF destroyed the remnants of the Dhofar Liberation Front
in the sparsely populated Western Area, despite the insurgency’s base being
located across the border in Yemen.96
From 1979 to 1983, the Salvador government neutralized the threat of revolution,
employing a variety of terror tactics. It then used its augmented ªrepower
to break down the insurgency from a ªghting force strong enough to
threaten the government’s survival to an organization that held signiªcant
territory but was too weak to take over the state.97
The El Salvador Armed
Forces (ESAF) relied on the Batallónes de Infantería de Reacción Immediata
(Immediate Reaction Infantry Battalions, or BIRIs) to target insurgents and civilians
in the countryside.98
The BIRIs could not hold territory, but they were
able to degrade the insurgency’s military capabilities with search-and-destroy
missions over the course of the war. The Farabundo Martí National Liberation
Front (FMLN) moved from smashing conventional ESAF units in the ªeld in
1983 to seeing its main-force units shattered by U.S.-supplied airpower and
ground attacks in 1984. The FMLN returned to political organizing and economic
terrorism. Military victory was impossible as long as the state retained
U.S. backing.
forceful control of civilians
In Malaya, the British use of force targeting the insurgency, in conjunction with
authoritative control of the populace and the country’s resources, broke the insurgency’s
capacity and will to ªght. In response to insurgent efforts to coerce
material and nonmaterial support from isolated squatter settlements and other
communities, the British took steps to “protect” the population. These steps
took the form of controlling civilian behavior by force, however, rather than
International Security 42:1 102
95. John Graham, Ponder Anew: Reºections on the Twentieth Century (Staplehurst, U.K.: Spellmont,
1999), p. 347, available in the John Graham Collection, Oman Archive (OA), Middle East Center,
St. Antony’s College, Oxford; Sultan’s Armed Forces (SAF), Activity Report, January 22, 1971, FCO
8/1667, TNA; SAF, Sitrep, March 23, 1971, FCO 8/1667, TNA; SAF, Sitrep, April 4, 1971, FCO 8/
1667, TNA; SAF, Sitrep, April 13, 1971, FCO 8/1667, TNA; SAF, Sitrep, April 20, 1971, FCO 8/1667,
TNA; SAF, Sitrep, April 27, 1971, FCO 8/1667, TNA; SAF, Sitrep, May 4, 1971, FCO 8/1667, TNA;
SAF, Sitrep, May 11, 1971, FCO 8/1667, TNA; SAF, Sitrep, May 25, 1971, FCO 8/1667, TNA; John
Graham, Commander Dhofar for August–September 1972, Directive, dated August 5, 1972, folder
3, box 2, Graham Collection, OA; and Commander SAF (CSAF) Maj. Gen. Ken Perkins, Record of
Audience [with Sultan Qaboos], February 10, 1975, dated February 16, 1975, DEFE 24/1869, TNA.
96. Perkins, Record of Audience [with Sultan Qaboos].
97. William Stanley, The Protection Racket State: Elite Politics, Military Extortion, and Civil War in
El Salvador (Philadelphia: Temple University Press, 1996), p. 3.
98. Karl, “Imposing Consent?”; Stanley, The Protection Racket State, pp. 229–230; and author interview
with Max G. Manwaring, Carlisle, Pennsylvania, March 11, 2010.
protecting willing civilians from insurgent harm. The control and resettlement
campaign began in the late 1940s, but the new communities were given their
familiar name of “New Villages” only in 1950. Between 1950 and 1960, more
than 500,000 people, including about 25 percent of the ethnic Chinese population,
were forcefully resettled into heavily guarded communities, where their
behavior was closely monitored.99
“In spite of the sullenly hostile population,
we are making very good military progress by screwing down the people in
the strongest and sternest manner,” the director of operations reported
in 1956.100
Every relocation of a community relied on overwhelming force and the element
of surprise. The army descended on villages at dawn and rounded up
everyone. The villages were burned, the animals slaughtered or driven off, and
the inhabitants moved into fortiªed camps. There, they were subject to twicedaily
body searches and other strict controls.101
In other areas, civilians could
not transport food, money, or clothes; rice (a dietary staple) was rationed and
could only be sold cooked, because it spoils quickly in hot weather.102
In addition,
British authorities deported 31,249 residents between 1948 and 1955.103
Other harsh measures included detention without trial, seizure of food, and
destruction of premises used by insurgents and supporters.104
Those resettled
included ethnic Malays held in 139 communities encircled by barbed wire.105
In targeting the populace militarily, the SAF turned Dhofar’s coastal capital,
Salalah, into a prison camp surrounded by barbed wire fences, with all inhabitants
and pack animals searched to prevent the transfer of money and supplies
to insurgents.106
It tightly controlled the villages on the coastal plain.107
The military moved uncooperative civilians to Salalah, conªscating cattle
and over-chlorinating wells to make the water undrinkable.108
On the plain,
The “Hearts and Minds” Fallacy 103
99. Dixon, “‘Hearts and Minds’?”; Gurney to Secretary of State for the Colonies, April 27, 1959, FO
371/84490, TNA.
100. Hack, Defence and Decolonisation in Southeast Asia, p. 128.
101. Charles Townsend, Britain’s Civil Wars: Counterinsurgency in the Twentieth Century (London:
Faber and Faber, 1986), p. 162; and Newsinger, British Counterinsurgency, pp. 50–51.
102. Padget, Counter-Insurgency Operations, p. 60.
103. See, for example, Briggs, Progress Report on the Situation in Malaya; and Karl Hack, “Detention,
Deportation, and Resettlement: British Counterinsurgency and Malaya’s Ethnic Chinese
1948–1960,” Journal of Imperial and Commonwealth History, Vol. 43, No. 4 (2015), pp. 611–640,
doi:10.1080/03086534.2015.1083218.
104. Secretary of State for the Colonies, Memorandum, April 29, 1950, DEFE 11/35, TNA.
105. Komer, “The Malayan Emergency in Retrospect,” p. 56; and Newsinger, British Counterinsurgency,
p. 51.
106. Corran Purdon, List the Bugle: Reminiscences of an Irish Soldier (Antrim, Northern Ireland, U.K.:
Greystone, 1993), p. 269.
107. Purdon to Thwaites, September 22, 1969, folder 3, box 1, Peter Thwaites Collection, Liddell
Hart Center for Military Archives, King’s College London.
108. Ian Gordon, major in the Muscat Regiment, SAF, email to author, November 27, 2009.
the movement of foodstuffs was banned.109
Food controls were also imposed
in the mountains, home to many insurgents.110
SAF commander Brig. John
Graham ordered forces to punish Dhofaris who helped the rebels.111
The SAF
burned crops to deny resources to the insurgency.112
Depending on the
commander at the time, the SAF took greater or lesser care to avoid harm to
civilians.113
Long, fortiªed barriers in the mountains also controlled civilian
movements.Interdiction efforts culminated in completion of the mined
Hornbeam Line, stretching thirty-ªve miles from the coast across the mountains.
By the end of the war, walls and clearing operations had cut off the insurgents
in the West from 85 to 90 percent of civilians and their resources.114
In El Salvador, the government systematically targeted civilians throughout
the war because it perceived them as an existential threat. The military attacked
them using bombing and shelling; indiscriminate capture; torture;
systematic destruction of homes and crops; and forced relocation to cities and
camps, where the inhabitants were easier to control.115
It treated civilians in
insurgent-dominated areas like combatants.116
Ofªcers talked about making an
example of suspected insurgent supporters with reference to La Limpieza (the
cleansing).117
Cleansing operations spread fear,118
separating civilians from
the insurgents because residents afraid to return to their homes could not provide
support to the insurgents.119
Military sweeps depopulated entire areas. By
December 1983, an estimated 400,000 people had been internally displaced;
another 200,000 had ºed to Mexico and elsewhere in Central America (includInternational
Security 42:1 104
109. CSAF Brig. John Graham, CSAF’s Assessment of the Military Situation in Dhofar as of February
14, 1972, dated February 17, 1972, DEFE 25/293, TNA.
110. Directive for Commander Dhofar for 1972, Update, March 3, 1972, folder 3, box 2, Graham
Collection, OA.
111. Ibid.
112. David C. Arkless, The Secret War: Dhofar 1971/1972 (London: William Kimber, 1988), p. 81.
113. Ibid., p. 173; Anti-guerrilla Operations in Dhofar Training Guide, folder 2, box 2, Graham
Collection, OA; and Operational Instructions for Move to Central Area, January 1975, folder 2/3,
box 2, Edward Ashley Collection, OA.
114. Review of the Military Situation in Oman, November 26, 1973, DEFE 25/312, TNA; CSAF
Maj. Gen. Tim Creasey Talking Points, Review of the Military Situation, December 10, 1973 to January
23, 1974, dated December 19, 1973, DEFE 25/312, TNA; Col. W.J. Reed, Notes on Visit to
Oman January 20–24, 1974, dated January 30, 1974, DEFE 25/31, TNA; Ken Perkins, A Fortunate
Soldier (London: Brassey’s, 1988), p. 126; and Chiefs of Staff Committee, Defence Operational
Planning Staff, Progress of Operations in Dhofar, February 17, 1975, DEFE 24/1869, TNA.
115. Elisabeth Jean Wood, Forging Democracy from Below: Insurgent Transitions in South Africa and
El Salvador (New York: Cambridge University Press, 2000), p. 32.
116. Francisco Pedrozo, U.S. military adviser in El Salvador during the war, email to author, April
18, 2010.
117. Mark Danner, The Massacre at El Mozote (New York: Vintage, 1993), pp. 44, 52.
118. Ibid., p. 108.
119. Leigh Binford, The El Mozote Massacre: Anthropology and Human Rights (Tucson: University of
Arizona Press, 1996), pp. 151–152.
ing to refugee camps across the border in Honduras); and an estimated half a
million had entered the United States.120
In all three cases, as I discuss in the next section, government efforts to control
civilian populations did not include reforms for all. In two of the cases
(Malaya and Dhofar), efforts did include accommodations for a few.
elite cooperation, not reform
The British government had eliminated the insurgents’ “very real threat to the
security and economic recovery of Malaya” by 1949. Its success occurred before
it provided accommodations to resettled civilians. Defeat also came well
before the major political change the British government instituted in Malaya,
the granting of independence in 1957.121
The British promise and implementation
of plans for independence were not, however, intended to quell the insurgency.
British planning for an independent Malaysia began in 1942, not in
response to the Emergency.122
The British government planned to transform its
colony into a liberal democracy with a multi-ethnic military.123
Its goals included
creation of a Malaysian identity for all inhabitants.124
Ultimately, however,
the British had to put aside their plans for liberal reforms and instead
accommodate elite interests.
Britain’s ªrst effort at creating a liberal post–World War II polity in a Malaya
that was intended to graduate to self-governance was its formation of the
Malayan Union in 1946. The Union gave all residents equal rights. Ethnic elites
in the colony, however, feared that a liberal democracy based on individual
rather than ethnic rights would redistribute their power to ethnic groups other
than theirs. Malay elites, for example, feared losing status and power to the
ethnic Chinese and ethnic Indians, who made up about half the population.
Consequently, they resisted British reformist plans to grant expanded citizenship
and voting rights to non-Malays. Malaya elites also objected to further reductions
in the role of Malaya’s traditional rulers, the sultans.125
Some Malays
The “Hearts and Minds” Fallacy 105
120. Commission on the Truth for El Salvador, “From Madness to Hope: The 12-Year War in
El Salvador—Report of the Commission on the Truth for El Salvador” (Washington, D.C.: United
States Institute of Peace, March 15, 1993), pp. 26–27, https://www.usip.org/publications/1992/
07/truth-commission-el-salvador.
121. Bower, Review of the Emergency in Malaya from June 1948 to August 1957.
122. Simon C. Smith, British Relations with the Malay Rulers from Decentralization to Malayan Independence
1930–1957 (New York: Oxford University Press, 1995), p. 43.
123. Ibid., pp. 42–43; Karl Von Vorys, Democracy without Consensus: Communalism and Political Stability
in Malaysia (Princeton, N.J.: Princeton University Press, 1975), pp. 64–70; and Hack, Defence
and Decolonisation in Southeast Asia, p. 112.
124. Hack, Defence and Decolonisation in Southeast Asia, pp. 112–113.
125. GHQ Far East Land Forces, British Defence Coordination Committee Far East, to Ministry of
Defence London, for Chiefs of Staff, Top Secret Telegram Regarding Progress Report October 15,
1951 (Briggs Report), dated November 15, 1951, DEFE 11/46, TNA.
told British colonial ofªcials that “it would be preferable to continue under a
‘colonial’ regime rather than to grant the Chinese the claim for [citizen-
ship].”126
Ethnic Chinese elites, meanwhile, fearing increased Malay domination,
asked the British if they could change their minds about supporting inde-
pendence.127
Under elite pressure, Britain replaced the Union with the Federation of
Malaya in 1948. The Federation protected the special rights of the Malay sultans
(granted by the British in the colonial period), assured ethnic Malay domination
of the new state of Malaysia, and limited the electoral role of other
ethnic groups in the Federation and in Malaysia.128
Ethnic Chinese elites
were given leadership positions and the promise of future power once ethnic
Chinese residents had been given increased citizenship rights.129
In exchange,
the British received the sultans’ agreement to establish a strong, centralized
state designed to promote economic development and effective government
administration.130
In creating the Federation, Britain abandoned its reformist
plans and accepted Malay elites’ demands to maintain traditional hierarchies
and communal relationships.131
Meanwhile, British colonial authorities began systematically delivering accommodations
to selected civilian communities only in 1952, after having neutered
the insurgent threat in 1949.132
The distribution of these accommodations
was based on civilian cooperation with the authorities, such as sharing information
on the insurgency. The goods and services the British provided were
not public goods shared by all. High Commissioner Sir Henry Gurney wrote
in 1950 that the government was not implementing reforms to meet popular
needs: “We are doing little to improve the lot of [sic] mass of the population
beyond increasing taxation.”133
In the 1951–52 period, residents of the New
International Security 42:1 106
126. Ofªce of the Mentri Besar, Trengganu state, Monthly Political Intelligence Report for the Period
Ending May 20th, 1956, dated May 22, 1956, FCO 141/7379, TNA.
127. R.P. Bingham, Political Intelligence Report, Settlement of Penang, March 21 to April 20, 1956,
FCO 141/7529, TNA; and Resident Commissioner, Settlement of Penang, Political Intelligence Report,
February 21 to March 20, 1956, FCO 141/7529, TNA.
128. Von Vorys, Democracy without Concensus, pp. 77, 79–82.
129. Secretary for Chinese Affairs, Political Report for April 1956, dated May 2, 1956, FCO 141/
7378, TNA.
130. Cheah Boon Kheng, Red Star over Malaya: Resistance and Social Conºict during and after the Japanese
Occupation, 1941–1946, 2nd ed. (Singapore: Singapore University Press, 1983), p. 293.
131. Von Vorys, Democracy without Consensus, p. 73.
132. Lee Kam Hing, “A Neglected Story: Christian Missionaries, Chinese New Villagers, and
Communists in the Battle for the ‘Hearts and Minds’ in Malaya, 1948–1960,” Modern Asian Studies,
Vol. 47, No. 6 (November 2013), pp. 1977–2006, doi:10.1017/S0026749X12000741.
133. Gurney to Secretary of State for the Colonies, inward telegram, April 27, 1950, DEFE 11/35,
TNA.
Villages “enjoyed only rudimentary facilities to compensate for the trauma of
being uprooted” from their homes and forced into the camps.134
Aid such as
clothing, building materials, and cash advances was provided by charities and
the Malayan Chinese Association because of the government’s failure to do
so.135
The British Red Cross began giving New Village residents social-welfare
assistance in 1952.136
In Dhofar, the British consistently pushed the sultan to make systematic
reforms. During the war, however, none were implemented.137
The provision
of goods that did take place during the conºict, such as well-digging, occurred
primarily on the more secure coastal plain and the highland behind
the guerrilla-held mountains, so it could not have directly driven the insurgency’s
defeat.138
Sultan Qaboos refused to let Dhofaris participate in the provincial
government he had appointed.139
The British wondered in 1972 if there
was any point in continuing military aid to Oman in the absence of reforms.140
They were near despair about the lack of reforms in 1974, when the insurgency
had already lost two-thirds of its strength.141
A month after Qaboos declared
the insurgency defeated, the government showed no grasp of the importance
of meeting popular needs.142
The most important accommodation in the
conºict was Qaboos’s amnesty for insurgents, which enabled formation of
the ªrqats (militias). The ªrqats’ personal relationships, language, and particuThe
“Hearts and Minds” Fallacy 107
134. Hack, Defence and Decolonisation in Southeast Asia, p. 126.
135. R.P. Bingham, Political Intelligence Report, Settlement of Penang, November 21 to December
20, 1955, FCO 141/7529, TNA. On the Association’s efforts, see Southeast Asia Department Agrarian
Policy in Malaya, FO 371/101271, TNA; and Fight against Communist Terrorism in Malaya,
June 1951, FO 371/101271, TNA.
136. Victor Purcell, Malaya: Communist or Free? (Stanford, Calif.: Stanford University Press, 1954),
p. 81.
137. Abdel Razzak Takriti, Monsoon Revolution: Republicans, Sultans, and Empires in Oman, 1965–
1976 (Oxford: Oxford University Press, 2012), pp. 6, 194–229.
138. Tony Jeapes, SAS: Operation Oman (London, William Kimber, 1980), pp. 162–164, 209, 228; and
John Akehurst, We Won a War: The Campaign in Oman, 1965–1975 (Salisbury, U.K.: Michael Russell,
1982), pp. 78–79. On long-term development efforts, see, for example, Jeapes, interview with author,
May 15, 2009, Warminster, Wiltshire, United Kingdom; CSAF Maj. Gen. Ken Perkins, Report
to Chiefs of Staff, December 28, 1975, DEFE 11/899, TNA; Chief of Defence Staff, Notes on a Visit
to Oman, January 19 to January 22, 1976, dated January 30, 1976, DEFE 25/371, TNA; and Creasey,
Talking Points, Review of the Military Situation, December 10, 1973 to January 23, 1974.
139. Record of Meeting on Dhofar, British Residency, Bahrain, August 7, 1971, FCO 8/1667, TNA.
140. Richard Lloyd Jones, British Military Involvement in Oman, November 6, 1972, DEFE 24/
1831, TNA; and A.J. Cragg, Draft Paper for the Secretary of State, Defence Assistance to the Sultanate
of Oman, November 17, 1972, DEFE 25/294, TNA.
141. CSAF Maj. Gen. Tim Creasey, Review of the Situation, December 1973–May 1974, dated June
9, 1974, DEFE 24/57, TNA.
142. Chief of Defence Staff, Notes on a Visit to Oman, January 19 to January 22, 1976; and Chiefs of
Staff Committee, Defence Operational Planning Staff, Progress Report on Oman, draft, January 26,
1976, DEFE 25/371, TNA.
lar ªghting skills enabled the SAF to target the insurgency directly and to more
easily control and gain information from civilians.143
In El Salvador, the United States pressed for reforms it considered crucial to
defeat the insurgency.144
Government elites resisted reforms as inimical
to their interests.145
The highest-proªle reforms demanded—free-and-fair elections
and land reform—were not fully implemented, because they would have
stripped state elites of power and wealth. The so-called reforms that the government
did implement merely reinstated the limited political openness that
had existed prior to the state terrorism of the late 1970s and early 1980s. Elections
were neither free nor fair.146
They did, however, eventually help reopen
political space closed since the crackdown at the beginning of the war. The
state slowly permitted re-creation of popular and civic groups, including peasant
organizations, trade unions, women’s groups, and neighborhood associations,
though it also continued to violently target such organizations.147
In a
state where 1 percent of the population owned more than 70 percent of the
land and more than 40 percent of the rural populace comprised landless sharecroppers
or estate laborers, U.S. policymakers considered land distribution key
to defeating the FMLN.148
Under U.S. pressure, the junta announced a land reform
plan in March 1980. The plan was ambitious, but elite intransigence
severely limited its execution.149
Oligarchs remained the owners of many
International Security 42:1 108
143. Maj. Gen. Ken Perkins (CSAF 1975–1977), interview with author, May 20, 2009, Great
Bedwyn, Marlborough, Wiltshire, United Kingdom; and Jeapes, SAS, p. 78.
144. William M. LeoGrande, “A Splendid Little War: Drawing the Line in El Salvador,” International
Security, Vol. 6, No. 1 (Summer 1981), pp. 27–52, doi:10.2307/2538528.
145. T. David Mason and Dale A. Krane, “The Political Economy of Death Squads: Toward a Theory
of the Impact of State-Sanctioned Terror,” International Studies Quarterly, Vol. 33, No. 2 (June
1989), pp. 175–198, doi:10.2307/2600536; Susan Burgerman, “First Do No Harm: U.S. Foreign
Policy and Respect for Human Rights in El Salvador and Guatemala, 1980–1996,” in Debra LiangFenton,
ed., Implementing U.S. Human Rights Policy: Agendas, Policies, and Practices (Washington
D.C.: U.S. Institute of Peace Press, 2004), p. 293; Mark Peceny and William D. Stanley, “Counterinsurgency
in El Salvador,” Politics & Society, Vol. 38, No. 1 (2010), pp. 67–94, doi:10.1177/
0032329209357884; and Timothy P. Wickham-Crowley, “Understanding Failed Revolution in
El Salvador: A Comparative Analysis of Regime Types and Social Structures,” Politics & Society,
Vol. 17, No. 4 (1989), pp. 511–537, doi:10.1177/003232928901700404.
146. Karl, “Imposing Consent?”; Cynthia McClintock, Revolutionary Movements in Latin America:
El Salvador’s FMLN and Peru’s Shining Path (Washington, D.C.: U.S. Institute of Peace Press, 1998),
pp. 120–128; and Enrique Baloyra, “Elections, Civil War, and Transition in El Salvador, 1982–1994:
A Preliminary Evaluation,” in Mitchell A. Seligson and John A. Booth, eds., Elections and Democracy
in Central America, Revisited (Chapel Hill: University of North Carolina Press, 1995), p. 59.
147. Philip J. Williams and Knut Walter, Militarization and Demilitarization in El Salvador’s Transition
to Democracy (Pittsburg, Pa.: University of Pittsburg, 1997), p. 121.
148. Benjamin C. Schwarz, “American Counterinsurgency Doctrine and El Salvador: The Frustrations
of Reform and the Illusions of Nation Building” (Santa Monica, Calif.: RAND Corporation,
1991), p. 44.
149. Alberto Vargas, “El Salvador Country Brief: Property Rights and Land Markets” (Madison:
Land Tenure Center, University of Wisconsin–Madison, March 2003); Peter Shiras, “The False
large estates. A 1992 analysis found that land reform affected 20 percent of
El Salvador’s land area and 10 percent of the population, leaving 54 percent
of the workforce landless, land poor, or unemployed.150
I have not identiªed instances in which the Salvadoran government attempted
to provide accommodations to selected nonstate political or military
elites. It is possible that in a revolutionary context, such as the Salvadoran civil
war, opportunities for accommodation are unlikely.151
Not only were there no reforms in any of these cases, as predicted by the coercion
theory, but in all three, the government succeeded without gaining popular
support, as I show in the next section.
no increase in popular support
Britain never gained the broad popular support among all ethnic communities
that it believed necessary for success against the Malayan insurgency.152
Statements
ruing this fact, and the lack of civilian provision of information on the
insurgency that was supposed to follow rising popular support, appear in contemporaneous
documents throughout the Emergency.153
Elites focused on ensuring that arrangements for independence suited their
needs. The sultans lacked Britain’s concern about the communist-nationalist
The “Hearts and Minds” Fallacy 109
Promise—and Real Violence—of Land Reform in El Salvador,” in Marvin E. Gettleman et al., eds.,
El Salvador: Central America in the New Cold War (New York: Grove, 1981), pp. 163–167; Williams
and Walter, Militarization and Demilitarization in El Salvador’s Transition to Democracy, p. 106;
Schwarz, “American Counterinsurgency Doctrine and El Salvador,” pp. 47–48; and Col. Lyman C.
Duryea, U.S. Department of Defence and U.S. Army attaché in El Salvador, November 1983–
October 1985, Project 86-9 El Salvador, interviewed by Lt. Col. Emil R. Bedard, U.S. Marine Corps,
and Lt. Col. L.R. Vasquez, U.S. Army, March 4, 1986, for the AWC/MHI 1986 El Salvador Oral History
Project, U.S. Army Military History Institute, Army Heritage Education Center (AHEC),
Carlisle, Pennsylvania, pp. 66–67.
150. William C. Thiesenhusen, “Trends in Land Tenure Issues in Latin America: Experiences and
Recommendations for Development Cooperation” (Eschborn, Germany: Deutsche Gesellschaft
für Technische Zusammenarbeit, 1996), cited in Vargas, “El Salvador Country Brief” p. 10;
McClintock, Revolutionary Movements in Latin America, p. 175; Leonel Gomez, “El Salvador’s Land
Reform: A Real Promise, but a Final Failure,” in Gettleman et al., El Salvador, pp. 178–181; and
Schwarz, “American Counterinsurgency Doctrine and El Salvador,” p. 48.
151. Jeff Cole, military adviser in El Salvador during the war, email to author, April 4, 2010.
152. Note by the Secretary, Chiefs of Staff Committee, Situation in Malaya, May 1949, DEFE 11/33,
TNA; Paper on the Security Situation in the Federation of Malaya; Memorandum on Visit to Singapore
of the United States Joint Defence Survey Mission (Melby Mission): Initial, Information Discussions
with the Mission on August 8–9, 1950, DEFE 11/38, TNA; and Commissioner General,
South East Asia, Telegram to Secretary of State for the Colonies, November 24, 1950, FO 371/
84492, TNA.
153. See, for example, Director of Operations, Progress Report on the Emergency in Malaya; Secretary,
Chiefs of Staff Committee, Report on Visit of Dr. Cockburn’s Party to Malaya; Haji Mohamed
Sheriff, Mentri Besar, Kedah state, Kedah Monthly Political Intelligence Report No. 11/53, October
21 to November 20, 1953, DEFE 11/46, TNO; South East Asia Department of Colonial Ofªce,
Malaya Monthly Emergency and Political Report, September 15 to October 15, 1954, DEFE 11/142,
TNA; and FARELF to War Ofªce, Sitrep No. 338, received October 18, 1954, DEFE 11/142, TNA.
threat and remained uninterested in the Emergency.154
Ethnic Chinese residents
displayed a similar indifference to British efforts to encourage Malayan
nationalism.155
The lack of popular interest in the campaign against the MNLA
and politics more broadly frustrated the British authorities. “The public has, as
usual, done its best to disprove Aristotle’s dictum that man is by nature a political
animal,” a British ofªcial declared in 1954.156
New Village inhabitants generally did not consider themselves the fortunate
recipients of government reforms. A review of the Villages in 1954 found large
variation in levels of poverty and squalor. One was “bad in all respects. Should
be moved, preferably to Christmas Island for hydrogen bomb tests,” the reviewing
British ofªcial wrote.157
A doctor described one resettled community
in the following terms: “Four hundred beings, including children, huddled
there. . . . I shall never forget their pale and puffy faces: beriberi, or the ulcers
on their legs. Their skin was the hue of the swamp. They stank. There was no
clean water.”158
In George Town, the capital of Penang state, ethnic Chinese
considered the New Villages “small and remote concentration camps, devoid
of any attraction whatsoever.”159
In 1955, “New Villagers around Ipoh
[the state capital of Perak] still appear to be uncooperative, indifferent, or
actively hostile.”160
In 1957, British authorities found there was still considerable
discontent, rather than popular support for the government, in the
New Villages.161
Similarly, in Oman the government did not gain the popular support the
British wanted. The populace hailed Qaboos when he replaced his father,
Sa’id, but his popularity quickly faded. In June 1971, Dhofaris were disilluInternational
Security 42:1 110
154. Secretary, Chiefs of Staff Committee, Report on Visit of Dr. Cockburn’s Party to Malaya; and
High Commissioner for the Federation of Malaya to Secretary of State for the Colonies, Monthly
Intelligence Report for November 1956, FCO 141/7306, TNA.
155. Director of Operations, Progress Report on the Emergency in Malaya.
156. Bingham, Resident Commissioner, Settlement of Penang, Political Intelligence Report for the
Settlement of Penang from April 21 to May 20, 1954, dated May 31, 1954, FCO 141/7529, TNA.
157. C.H.F. Blake, Notes on New Villages in Batang Padang District [1954], Papers of C.H.F. Blake,
GB 0162 MSS. Ind. Ocn. s. 276, Bodleian.
158. Karl Hack, “Everyone Lived in Fear: Malaya and the British Way of Counterinsurgency,”
Small Wars & Insurgencies, Vol. 23, Nos. 4–5 (2012), pp. 671–699, at pp. 685–686, doi:10.1080/
09592318.2012.709764.
159. D. Gray, Ag. Resident Commissioner, Penang, Political Intelligence Report for February 21,
1955–March 20, 1955, dated March 31, 1955, FCO 141/7529, TNA.
160. Mentri Besar, Perak state, Monthly Political Intelligence Report for Period Ending March 20,
1955, FCO 141/7377, TNA.
161. Raja Haji Ahmad, Mentri Besar, Perlis state, Monthly Political Intelligence Report for the
Month Ending January 20, 1956, dated February 2, 1956, FCO 141/7339, TNA; Mentri Besar,
Selangor state, Political Intelligence Report for Selangor, November 21, 1956 to December 20, 1956,
FCO 141/7528, TNA; Mentri Besar, Negu Sembilan state, Monthly Political Intelligence Report for
the Month of March, 1956, FCO 141/7319, TNA; and Bower, Review of the Emergency in Malaya
from June 1948 to August 1957.
sioned with Qaboos and gloomy about their future.162
By November 1971,
all of Oman considered the government inefªcient and corrupt.163
At the end
of the war, Dhofari dissatisfaction was still rising over the government’s lack
of action and bad administration.164
In El Salvador, the United States considered voting a sign of popular support
for the government. Balloting fell over the course of the war, however, despite
inºated turnout totals.165
Turnout slid from 65 percent of eligible voters
in the 1984 presidential election to 48 percent in the 1985 midterms.166
Turnout
in the 1988 midterms was estimated at 48 percent and dropped to an estimated
40 percent in the 1989 presidential balloting. The turnout in the 1991
midterms was estimated at 44 percent. The decline in participation reºected
suspicion and unrest, as well as the lack of a level playing ªeld. In 1990, only
19 percent of Salvadorans were conªdent that there would not be electoral
fraud; 41 percent thought there would be; and 40 percent were not sure.167
Popular support for a negotiated peace, as opposed to a government military
victory, rose. In 1983, 51 percent of the electorate supported talks to end the
war, whereas 10 percent wanted a military victory over the insurgency.168
A
June 1989 poll found that 76 percent of Salvadorans favored negotiations.169
summary
The Malayan Emergency little resembled a campaign implementing the good
governance theory of counterinsurgency. The government’s indiscriminate use
of force against civilians played a major role in British success, as the coercion
theory predicts. Military targeting had left the insurgents unable to mount coordinated
attacks by 1949, several years before the accommodation of civilians
held in prison camps, also as the coercion theory predicts. The British did not
undertake liberalizing reforms and did not gain public support. Yet they
won, as the coercion theory predicts. These ªndings in the case most widely
known as a model counterinsurgency success suggest that the good goverThe
“Hearts and Minds” Fallacy 111
162. Consul General D.G. Crawford to Wright, Bahrain, April 18, 1971, FCO 8/1667, TNA; and Sir
Geoffrey Arthur, Political Resident, Bahrain, to Muscat, June 11, 1971, FCO 8/1667, TNA.
163. Graham diary, November 14, 1971, folder 1, box 4, Graham Collection, OA.
164. Chief of Defence Staff, Notes on a Visit to Oman, January 19 to January 22, 1976; and Chiefs of
Staff Committee, Defence Operational Planning Staff, Progress Report on Oman, draft.
165. Karl, “Imposing Consent?” p. 34.
166. Ibid.
167. McClintock, Revolutionary Movements in Latin America, p. 112.
168. Karl, “Imposing Consent?” p. 27, citing Segundo Montes, “Condicionamientos SocioPolítícos
del Proceso Electoral” [Socio-political conditioning in the electoral process], Estudios
Centroamericanos, April/May 1983, pp. 187–196, at p. 188.
169. Tommie Sue Montgomery, Revolution in El Salvador: From Civil Strife to Civil Peace (1982; repr.
Boulder, Colo.: Westview, 1995), p. 215.
nance theory has limited explanatory power. In contrast, the coercion theory of
counterinsurgency success has signiªcant explanatory power.
Findings in the Dhofar and El Salvador cases offer further support for the
coercion theory. In Dhofar, the military defeated the insurgency on the battleªeld,
forcefully controlled civilians to cut the ºow of resources to insurgents,
and accommodated rival elites. The sultan implemented no reforms and
gained no popular support. The El Salvador campaign also shows the absence
of reforms and the indiscriminate use of force against civilians. In El Salvador,
as in Malaya and Dhofar, the government’s primary tool in achieving success
was military force. The differences in El Salvador emphasize the importance of
the coercion theory’s predictions regarding the treatment of elites and civilians:
the Salvadoran government did not accommodate rival elites; it did not
cut ties between civilians and insurgents; and it did not attain military victory.
Conclusion
I have argued that my theory of coercion offers a more robust explanation for
successful counterinsurgencies than does the good governance theory. According
to the coercion theory, counterinsurgent governments must use force to
control civilians, and thus cut the ºow of resources to insurgents, as well as
provide accommodations to rival elites to obtain information and enhance
their military power. The coercion theory better explains counterinsurgency
success in cases that proponents of the good governance approach present
as model campaigns—cases that should present strong evidence of the effectiveness
of good governance in defeating insurgencies. If the predictions of
the good governance approach are accurate, the government should limit its
use of force in countering the insurgents; it should propose and implement reforms
to reduce popular grievances; and popular support for the government
should rise. The empirical evidence demonstrates that none of these factors is
present in these three cases considered by supporters of the good governance
theory as models of counterinsurgency success: Malaya; Dhofar, Oman;
and El Salvador. The evidence did, however, yield strong support for the coercive
theory.
In the Malaya case, the government used considerable force against civilians;
it did not implement reforms; and it did not gain popular support for its
counterinsurgent campaign. A brief review of the Dhofar and El Salvador
cases yields similar ªndings. The El Salvador case, in particular, underscores
the importance of the use of force against civilians and accommodation of elite
actors because the government used brute force against civilians and still succeeded,
yet did not accommodate rivals and did not defeat the FMLN.
International Security 42:1 112
Moving beyond theory to consider the implementation of a counterinsurgency
campaign by a speciªc government facing a speciªc insurgency, the
governance theory confronts several difªculties. First, governments that can
make reforms do so, leaving only the most challenging cases of full-blown insurgency.
Second, governments willing to make reforms may not have the
ability to execute them, even with patron support. Third, violence hardens actors’
positions and changes their calculations. It can exacerbate actors’ sense of
playing a zero-sum game and make resolving the conºict harder.170
Fourth, violence
exacerbates the difªculties associated with establishing democracy in
a non-democratic state, and democratization efforts may even intensify vio-
lence.171
Fifth, there is an inherent tension between the strategic goals of the
governance approach and the strategic necessities of defeating an insurgency.
The government’s need to accommodate warlords or other elites to gain information
and ªghting power and to control the populace undercuts the good
governance goal of empowering and serving the populace.172
These ªndings raise important questions for policymakers, including questions
about moral hazard. Liberal great powers may try to do good by fostering
reforms in states confronting an insurgency and fail at high human and
moral cost. Alternatively, they may achieve their goal of long-term political
stability through elite domestic political accommodation, also at some moral
cost. Support for an illiberal client means direct or indirect support for its illiberal,
even brutal, policies. These difªculties suggest the need for policymakers
to assess the value of keeping client elites in power compared to the costs
of trying to preserve their rule against an insurgent threat. Acceding to corruption
and warlord rule or supporting it to gain political stability is normatively
unpalatable. There is also a moral argument, however, for ending or reducing
the suffering engendered by civil war and for supporting postwar humanitarian,
infrastructure, economic, and political construction whatever the regime
type. Western policymakers should consider just how high the moral and human
costs of a successful counterinsurgency campaign are likely to reach before
choosing the path of intervention.
The “Hearts and Minds” Fallacy 113
170. Robert J. Art, A Grand Strategy for America (Ithaca, N.Y.: Cornell University Press, 2003), p. 6.
171. Edward Mansªeld and Jack Snyder, “Democratization and War,” Foreign Affairs, Vol. 74, No. 3
(May/June 1995), https://www.foreignaffairs.com/articles/china/1995-05-01/democratizationand-war;
and Paris, At War’s End.
172. Maj. Niel Smith and Col. Sean MacFarland, “Anbar Awakens: The Tipping Point,” Military
Review, March/April 2008, pp. 41–52; and Rod Nordland, “Some Police Recruits Impose ‘Islamic
Tax’ on Afghans,” New York Times, June 12, 2011, http://www.nytimes.com/2011/06/13/world/
asia/13police.html?scpϭ1&sqϭafghanistan%20%22afghan%20local%20police%22&stϭcse.