Afghanistan’s Legacy: Emerging Lessons of an Ongoing War

June 01, 2014

by Stephen Biddle

Download the print version P D F icon

The war in Afghanistan is not over. Nor is it ending any time soon. The U.S. role may end in 2016, in whole or in part, but the war will continue—and its ultimate outcome is very much in doubt. The conflict is now stalemated militarily, and will likely stay that way as long as outsiders pay the large bills needed to keep the Afghan National Security Forces (ANSF) in the field and fighting. The war will thus grind onward until this funding dries up or the two sides negotiate a compromise settlement, neither of which is imminent. Depending on how any talks unfold, historians in 2050 could thus look back on this war as a costly but tolerable outcome for the West, as a wasteful disaster, or as something in between; for now, all we know for sure is that it continues.

Yet, for many Americans the war is already receding in the rearview mirror. President Obama, for example, often suggests that by as early as 2014 “this long war will come to a responsible end.”1 It will not, in fact, but Americans could be excused for thinking otherwise, given such framing. Nor is the President alone in this thinking, though many of his compatriots are taking a notably less optimistic tone in their retrospectives.

Among the most important of these pre-postmortems is an emerging debate on the war’s lessons. Perhaps the most common position is that Afghanistan shows the futility of counterinsurgency (COIN). By contrast with the President’s view, many now believe the war can already be deemed a failure in spite of massive investment. For critics such as Gian Gentile or Andrew Bacevich, this failure in Afghanistan shows how counterinsurgency’s demands cannot be met, and how the United States should therefore avoid such wars at all costs in the future.2 For many in this camp, Iraq, too, was a failure; others see Iraq as an eleventh-hour turnaround brought about by good luck in the idiosyncratic Sunni realignment of the Anbar Awakening.3 Even the latter group, however, usually sees Afghanistan as evidence that COIN cannot succeed without the happy accident of the enemy changing sides.

Within the military, debate on the future of COIN revolves as much around how the United States should wage counterinsurgency in the future as around whether it should. Formal studies of the lessons of the Afghan and Iraq wars are still very thin on the ground, perhaps unsurprisingly—with the Afghan war still ongoing, there is yet no equivalent to the post-World War II Strategic Bombing Survey or the post-1991 Gulf War Airpower Survey. But an emerging informal argument heard among many veterans of Afghanistan and Iraq holds that prevailing U.S. COIN doctrine now undervalues military security and combat relative to “softer” governance reform and economic development. Many view the doctrinal manual co-authored by then-Lieutenant Generals David Petraeus and James Amos in 2006 (FM 3-24, Counterinsurgency),4 as advocating an approach centered on winning civilians’ hearts and minds via superior public services. Yet, the actual conduct of both the Iraq and Afghanistan wars has proven far more violent than many readers of this manual have expected; even officers who see Iraq or Afghanistan as successes now often attribute this largely to the “kinetic” security campaign and the hard fighting it produced in both theaters. For them, the wars show that FM 3-24’s “hearts and minds” provisions are overemphasized, and a central lesson is thus that future COIN doctrine should subordinate governance to security, and reemphasize the combat function.5

In an important sense, these debates are premature: because the Afghan war continues with results as yet unknown, there is no observable final outcome to explain, or from which to draw lessons. But the debates matter all the same. Wartime experience always shapes post-war policy, and the impressions formed first are often the most persistent and influential. For Afghanistan, ideas and interpretations are already taking shape and could easily become established well before the war is actually over. Hence, it makes sense to give them an early look, and to evaluate how well or badly they correspond with what can actually be observed in the war’s conduct to date—premature as any such findings must necessarily be.

What we know so far suggests a more nuanced view than what one often finds in the emerging debate. There are important elements of truth in both of these early accounts (i.e. the “never again” school and the combat-emphasis camp within the military). But the first is overstated, and the second overlooks important flaws in the way governance reform has been undertaken in Afghanistan (and also Iraq, though my focus here is on the former).

COIN is obviously hard and slow. But the Afghan experience shows that current U.S. methods can return threatened districts to government control, when conducted with the necessary time and resources. This certainly does require combat and hard fighting. Counterinsurgency is not social work, and its purpose is not to make local civilians like Americans. But combat and security alone will have difficulty sustaining control if all they do is allow a predatory government to exploit the population for the benefit of unrepresentative elites. Yet, the governance reform approach in FM 3-24 misunderstands the issue and tends to make the problem worse rather than better, as I will explain later.


Hard? Yes. Impossible? No.

For most of the post-Vietnam era, the conventional wisdom among Americans held that COIN was a hopeless project, especially for democracies like the United States. Defeat in Vietnam preconditioned many Americans to see insurgencies as intractable quagmires that would inevitably overwhelm any occupier’s patience. Soviet defeat in Afghanistan just reinforced this view. Even the Bush administration shared it: its refusal to acknowledge a mounting insurgency in Iraq in 2003–04 and its reluctance to commit to a large-scale presence in Afghanistan in 2001 both stemmed in part from this widespread view that insurgencies could not be defeated, hence avoiding them was the only sensible approach.6

The Iraq surge led to a brief reversal of this orthodoxy. After reaching murderous heights in early 2007, violence in Iraq then plunged by a factor of about ten in a period of just a few months.7 This nose-dive in violence followed the deployment of about 30,000 U.S. reinforcements and the appointment of FM 3-24’s co-author, David Petraeus, to command in Baghdad. Many saw this reduction in violence as validation of the new COIN doctrine and a demonstration that one could tame insurgencies in brief campaigns with the right methods.

COIN critics see this reassessment as a wave of unwarranted enthusiasm which led the Obama administration down a primrose path to a misbegotten repeat in the Afghan surge of 2009. Speaking only for the analysts who wrote General Stanley McChrystal’s assessment report in spring 2009 (I served as a member of the analysis team), I can say that none saw Afghanistan as a meaningful analog to Iraq, or thought COIN was a cheap and easy ticket to success in the war. I doubt anyone in the Obama administration thought so, either. But the Iraq surge did suggest that COIN was not impossible, and that if the United States was willing to invest the time and resources, then it could eventually defeat an insurgency.

The ensuing 2009–12 surge in Afghanistan certainly looked different than Iraq’s in 2007. No Iraq-like sudden turnaround occurred. Five years after surge forces arrived in Afghanistan, the Taliban remain militarily viable and continue to control strategically important terrain, especially in the east but also in parts of the south and southwest. U.S. casualties were lower in 2013 than in 2009,8 but by nothing like Iraq’s 2007 factor-of-ten plummet, and many now expect at least a temporary increase in Taliban attacks as they test the ANSF to see whether it can function after Western combat forces withdraw. Anyone who thought the Iraq surge had created a new template for rapid success in COIN must surely be disabused of this notion by now.

But does this warrant a return to the general COIN pessimism of the Vietnam-to-2006 era? Does the Afghan surge show that COIN is as hopeless as many now believe? In fact, what the Afghan experience shows is that counterinsurgents can indeed re-establish government control of threatened areas, hold them against insurgent counterattacks, and hand that control off to indigenous security forces. What COIN cannot do, though, is to accomplish this quickly or cheaply.

Among the Afghan surge’s earliest targets were the districts and subdistricts of Nawa, Marjah, Garmser, and Nad Ali in the central Helmand River Valley.9 By mid-2009, these had become Taliban strongholds with little to no government presence, populations who complied with Taliban edicts, and dense networks of improvised explosive devices and ambush sites confronting any government effort to patrol or penetrate the area. By early 2010, the arrival of U.S. surge forces enabled systematic offensives to clear these districts of insurgents. Taliban defenders resisted, and the result was a spike in violence with heavy U.S. and insurgent casualties. Following FM 3-24-prescribed methods of dismounted patrolling, forward basing among the population, and intensive efforts to build intelligence networks for guiding operations, clearance forces that met or exceeded the doctrinal norm of 20 troops per thousand members of the population succeeded in driving the Taliban from their strongholds, defeating a series of counterattacks, and establishing government control with substantial freedom of movement.10

As a result, violence fell: by May 2011, Marines in Nawa had gone over a year without a serious firefight; in Nad Ali, reported violence fell by 85 percent by June 2010; in Garmser, attacks fell by 90 percent by March 2011; in Marjah, Taliban-initiated attacks fell from almost 1,600 in 2011 to 782 in 2012, and have continued to decline since.11 By early 2012, shops had reopened, displaced populations were returning, and visiting American civilians were able to walk through marketplaces without body armor.12 By late 2013, lead security responsibility was transferred to the Afghan Army, U.S. forces had largely withdrawn, violence was continuing to decrease, and all four areas remain in government control today despite repeated Taliban efforts to retake them following the U.S. handoff to the ANSF.13 Similar experiences accompanied offensives in Arghendab, Zari, Panjway, and many other districts in which large U.S. or other Western forces were introduced into Taliban-held areas, displaced insurgents, held the results against counterattack, and eventually handed security over to Afghan forces.14

None of these offensives were cheap. Not only did they require large forces relative to the size of the population to be secured, but clearing the Taliban and establishing government control cost many lives: in Marjah alone, 28 U.S. Marines were killed in the first twelve months.15 Nor were any of these operations quick. The Allied command initially hoped that it could mostly re-establish government control in such places within a single fighting season, after which the ANSF could take over; that rule of thumb then stretched into a year, then eighteen months.16 The Taliban have proven themselves as tough opponents capable of sustained resistance against better-equipped, numerically superior forces.

The results have also fallen a good deal short of creating Eden in the Helmand River Valley. The Taliban there are not annihilated; counterattacks never disappeared altogether, and will probably increase now to test the ANSF. Many Afghans continue to see insecurity as a significant problem in their daily lives.

But the Taliban could not prevent properly trained forces from seizing and holding contested areas. Nor have Taliban counterattacks made much net headway to date against ANSF defenders in regaining districts the Taliban lost to the surge. Modern counterinsurgency methods have shown in Afghanistan that, when properly resourced, they can indeed drive insurgents out, and mostly keep them out, even after transition to indigenous control. And this means that U.S. forces can progressively expand the government’s zone of control by using the pattern of clearing districts, holding them against counterattack, handing off to indigenous forces, then moving on to repeat the process elsewhere while indigenous forces hold the cleared areas left behind. This suggests that such “ink spot” processes can eventually clear enough of a country to compel insurgents to end the war on terms the government can accept.

Nor have the Taliban been able simply to move from cleared areas to others, increasing violence elsewhere in a balloon-squeezing phenomenon. After an initial spike, violence nationwide fell in Afghanistan during the surge. Moreover, in areas where violence increased after 2009 (especially the country’s east), there were new U.S. clearance operations ongoing as forces swung from the initially reinforced south into regions that had not previously received surge forces.17 Offensives for clearance always increase local violence before stabilizing an area under government control; this means that an ink-spot strategy will rarely show a large drop in nationwide violence until all the local offensives are complete everywhere (in the meantime, drops in violence in one locality where the process is complete will be counterbalanced by local increases in violence elsewhere where the process is only beginning). In fact, there is little evidence that the Taliban made compensating gains elsewhere to offset its losses in the surge.

The commonplace narrative that Afghanistan shows how COIN is impossible is thus overstated. What experience to date suggests is that it can work—but only where counterinsurgents invest the lives, forces, and time needed. Neither FM 3-24 nor any other contemporary COIN technique can produce quick or inexpensive results. Anyone who employs it expecting otherwise will be deeply disappointed—and would be better advised not to start the process at all.


Hearts, Minds, and Firepower

Many felt the early U.S. military response to Iraq’s insurgency was heavy-handed and overly violent.18 This built on a common impression of U.S. tactics in Vietnam, which have been criticized for applying conventional-style firepower to destroy an elusive enemy, rather than protecting and winning over the civilian population.19 When FM 3-24 appeared in 2006, its emphasis on the latter was widely seen as a response to this criticism, and the manual itself goes to great lengths to contrast the less violent methods needed for COIN with the kind of firepower dominance emphasized in conventional warfare. In fairness, the manual does not present COIN as armed humanitarianism. The manual clearly intends a balance that includes a major combat function which will have to be violent, but the contrast with conventional doctrine is striking by design. And given this, it is perhaps predictable that many would feel the manual swung the pendulum too far.

When General McChrystal took command in Kabul in 2009, one of his first major initiatives was to reduce civilian casualties in consonance with the doctrine’s emphasis on protecting the population. He never meant this to suggest that killing the enemy was unimportant, or that troops in contact could not use firepower to protect themselves. But communicating a nuanced message of balance and change-by-degree across a sprawling organization of more than 100,000 people, many nations, and multiple military services proved very difficult in practice. The result was an unhappy combination of persistent heavy-handedness in some places, and widespread resentment in many others among troops who felt hamstrung by the new restrictions.

This frustration interacted with widespread awareness among the troops that COIN was continuing to involve bitter combat with sometimes heavy U.S. firepower. Certainly the Iraq surge was anything but soft or non-violent: in a single day on January 10, 2008, for example, U.S. aircraft dropped 47,000 pounds of bombs in support of Operation Marne Thunderbolt, south of Baghdad.20 And when General Petraeus replaced McChrystal in Kabul in 2010, he both re-emphasized the need to kill the enemy and de-emphasized governance reform, in an effort to prioritize near-term security progress of the kind he thought necessary to secure support for the war in Washington.

Taken together, these things have promoted a view among many within the military that recent experience shows how FM 3-24 swung the pendulum too far toward an almost humanitarian conception of victory through popularity and away from traditional military concerns with violent combat, combat that was continuing in any case. FM 3-24 is now under review for revision; one widespread and influential view of Afghanistan’s lessons thus holds that this revision should swing the pendulum back toward combat and away from governance.

Yet, this view risks over-shooting the target in the opposite direction. Combat and violence are clearly essential for successful COIN. But the biggest problem with FM 3-24 is not an underemphasis on combat and an overemphasis on governance—it is the manual’s prescription for the wrong kind of governance effort. Whereas FM 3-24 advocates apolitical capacity building, what Afghanistan actually shows is that governance problems are about political interests at least as much as administrative capacity. Merely improving the capacity of government actors who seek to prey on their population makes things worse, not better.

The underlying problem here is a tacit assumption throughout FM 3-24 that U.S. political interests align with the host’s. The manual implicitly assumes that the host government wants to be an effective, disinterested, broadly representative defender of its people’s well-being. The host thus theoretically shares with the United States a mutual ambition for what the manual refers to as legitimacy—the host just lacks the means to achieve it. Hence, the manual focuses on providing the means: equipping would-be Jeffersonian democrats in the host government with schools, clinics, judges and prosecutors, trained administrators and police, well-equipped soldiers. The doctrine also emphasizes connecting this more capable government to the people, so the public will see their stake in the government’s success and sacrifice on its behalf.

In Afghanistan, this assumption has proven badly mistaken. Much of the Afghan government operates as a collection of patronage networks that systematically extract resources from the population for the benefit of those networks and the power brokers who run them. This is hardly a disinterested attempt to improve the lives of average Afghans. And it happens not because the government lacks an adequate supply of trained administrators or roads or schools—it happens because the elites who run the system profit from it and want it to function this way. Hence, this problem will not diminish merely by the West increasing Afghanistan’s material capacity to govern. On the contrary, poorly monitored “capacity building” often makes the problem worse by supplying patronage networks with the money and resources needed to reward their allies, punish their rivals, and extract resources from the public more effectively. Worse still, U.S. attempts to connect the public to local government organs that the public sees as predatory implicate the United States in official predation and suggest to prospective victims that they have nowhere to turn for succor but the insurgency.

Nor is this problem limited to Afghanistan. In Iraq, the biggest challenge was sectarianism rather than profiteering (though plenty of the latter also existed), but in Iraq, too, U.S. interests and the host government’s were badly misaligned.

In fact, U.S. interests will rarely align naturally with the host’s in COIN. Local governments which actively promote a fair distribution of resources rarely face major insurgencies in the first place—hence, Americans are rarely asked to wage counterinsurgency at all in such places. It is no accident that the U.S. military does little COIN in Switzerland or Canada. If there is an insurgency to counter, then the odds are good that the local government is ruling in the interest of elites who do not want to redistribute power toward groups they distrust or exploit. The basic premise underlying governance reform doctrine in FM 3-24 is thus almost always false. Afghanistan shines a spotlight on the problem, but it is anything but unique to Kabul.

This does not make governance reform impossible for counterinsurgents. Among the reasons Iraq’s violence fell in 2007 is that Petraeus and U.S. Ambassador Ryan Crocker forced Iraqi Prime Minister Nouri al-Maliki to replace known sectarians in the Iraqi security forces, and to moderate Iraqi government behavior toward Sunnis (for example, by tolerating the mostly-Sunni Sons of Iraq groups that actualized the Anbar Awakening). They did this by coercing Maliki via conditional provision of support: if Maliki refused to act, Petraeus and Crocker would deny essential U.S. logistical support to Iraqi army or police units, or they would withhold other U.S. support Maliki valued and needed.21 Making COIN work here was not a process of benign, apolitical capacity building, a la FM 3-24. On the contrary, it was profoundly political and often highly coercive.

In Afghanistan, by contrast, the West has been systematically unwilling to use conditionality as leverage for reform. For some, especially in the military, the reason has been the absence of doctrinal guidance for this kind of coercive political bargaining. For others, the chief barrier has been fear that Afghans would call the bluff, assistance would have to be withheld, and this would delay the creation of a higher-capacity Afghan civil and military administration— which is usually seen as the key enabler for Western withdrawal. The net result is that the problem of government predation has grown worse, not better. Foreign presence has mostly served to aggravate the problem by fueling the corruption rather than solving it.

This in turn has contributed to the difficulty of truly eliminating the Taliban presence in Afghan population centers. Through combat, counterinsurgents can establish predominant control in which they regain freedom of movement and enable something like normal commerce. But even in major cities such as Kabul and Kandahar, resentment of government predation by some of its victims facilitates a shadow presence of Taliban fighters and sympathizers who can mount occasional attacks which keep the violence from truly disappearing. If government predation continues to worsen, the troop levels the ANSF will need to maintain control in such areas will leave them without the strength they need to expand the government’s writ over time. This will consign the war to a future of grinding stalemate for as long as the ANSF’s foreign funding holds up.

What Afghanistan suggests, then, is that if the U.S. military actually does wage counterinsurgency in the future, such apolitical “capacity building” is unlikely to work. Effective governance reform during insurgency is inherently political, typically coercive, and must proceed in a manner for which FM 3-24 offers little guidance. Yet, it cannot be ignored in favor of a unidimensional focus on security via combat. Without tolerable governance, the force levels and violence needed to sustain control of a threatened population will prove extreme even by doctrinal COIN standards, and residual instability is likely to remain unsatisfyingly high. It is not wooly-headed humanitarianism to suggest that voluntary public compliance with government edicts reduces security force requirements—or that public resentment increases them. Counterinsurgency is inherently slow, costly, and difficult; to conduct it amidst a population that views the government as a threat is to add substantially to an already substantial bill.


Never Again?

Of course, it is too early to offer conclusive lessons from an ongoing war. But in the spirit of presenting hypotheses for definitive analysis later, and in the interest of spurring debate before conventional wisdom consolidates, two observations are worth considering now.

First, Afghanistan to date suggests that COIN can indeed restore government control of threatened areas, but not cheaply or quickly. Where counterinsurgent forces were committed in doctrinal numbers and stayed long enough, the Taliban have been driven out, and kept out—this is not mission impossible. But neither is it some kind of silver bullet that can win wars before democratic publics take notice.

Worse, it is not a method that adapts well to partial application. A year before the Obama administration’s surge, there were already about 70,000 Western troops in Afghanistan; this did not yield half the stability for half the price of the 140,000-soldier peak Western presence.22 On the contrary, it was the 70,000 soldiers’ inability to prevent Taliban expansion that persuaded the administration to reinforce, lest the Taliban win the war outright. This makes it very hard to triangulate and serve limited interests with limited versions of COIN—the limited version attempted in Afghanistan before 2009 was well on its way to complete failure before the surge. Partial application is less likely to yield limited success than it is to produce a more expensive version of defeat.

This turns COIN into a thorny dilemma. In Afghanistan, the Obama administration has seen real but limited U.S. interests;23 its instincts have been to seek some real but limited means to secure them at a cost that fits the stakes. There is no such means. COIN can secure the stakes, but at a cost which many find greater than the benefit. Withdrawal would be cheaper (certainly in the short run), but would sacrifice the U.S. security stakes in the conflict. If COIN were impossible, then sacrifice would be the only logical approach, and no dilemma would exist. If COIN were quick and cheap, it would be worth the stakes in Afghanistan, and again no dilemma would exist. Yet, COIN is neither.

This in turn makes answering the question of COIN’s future more difficult than a simple verdict of “impossible, so never again.” As Gentile and others rightly note, COIN is a means, not an end. To judge the means, one must therefore assess the ends, which are the U.S. interests at stake where the United States would wage counterinsurgency. This is challenging but tractable for specific wars and theaters. In Afghanistan, the Obama administration saw a close call, deciding in fall 2009 that the interests at stake there were worth the cost of counterinsurgency, if only barely so, then wrestled with buyer’s remorse afterwards. For projections into uncertain futures in unknown theaters, the task of balancing interest and cost is harder still. The United States has interests of some kind in many places, the stakes vary in scale, and many conflicts affecting interests will probably not be anticipated before events bring them into focus.

Many COIN critics nevertheless think the stakes will never prove worth it; hence large, expensive, labor-intensive ground forces for COIN should shrink. This is not an unreasonable position. With dwindling budgets and pressing needs in the Pacific and elsewhere, the United States will surely have to reduce real capability of some kind; the least-damaging capability to divest might be that for large-scale COIN. But if so, this will not be a simple, cost-free call made easy by COIN’s manifest impossibility as demonstrated in Afghanistan—the war offers no simple escape from the dilemmas posed by the need to accept expensive means if one wants to secure real but limited interests.

Nor does Afghanistan suggest that future counterinsurgencies can be waged more cheaply by dispensing with difficult governance programs and refocusing on security and combat. This brings in the second tentative observation of the war to date: the United States needs a different approach to governance in counterinsurgency. COIN can reinstate government control without meaningful reform, as the Helmand and Kandahar campaigns suggest. But to sustain control under predatory misgovernance adds great expense to an already expensive form of warfare. Afghanistan shows not that governance is unimportant, but that apolitical capacity building is the wrong approach.

Of course, COIN will remain expensive even with astute governance policies. Given this, Americans may choose to avoid it in the future. But this will probably not avoid the debates that hard decisions spur, and it is plausible that future U.S. administrations might resolve the dilemmas posed by real but limited interests via COIN as the Obama Administration did. So it is important to preserve, as much as possible, the skills and knowledge in COIN’s conduct developed at such cost in Afghanistan and Iraq. This includes the knowledge of how to balance security with a more political understanding of governance reform. Overlooking this crucial balance under the influence of some popular early interpretations of the war’s lessons would represent a great and unnecessary loss.



1. See, e.g., “Obama: Afghan war to end in 2014,” United Press International, January 11, 2013,; Office of the Press Secretary, The White House, “Remarks by the President on the Way Forward in Afghanistan,” June 22, 2011,

2. For illustrative examples, see Gian P. Gentile, “A Requiem for American Counterinsurgency,” Orbis 57, no. 4 (Autumn 2013), pp. 549–558; Andrew Bacevich, Breach of Trust: How Americans Failed their Soldiers and their Country (New York: Metropolitan Books, Henry Holt and Company, 2013), pp. 13, 109, 149; “The War in Afghanistan: Year Ten,” Salon, October 7, 2010; also Karl Eikenberry, “The Limits of Counterinsurgency Doctrine in Afghanistan,” Foreign Affairs (September/October 2013); Douglas Porch, Counterinsurgency: Exposing the Myths of the New Way of War (New York: Cambridge University Press, 2013).

3. On the Anbar Awakening, see, e.g., Austin Long, “The Anbar Awakening,” Survival 50, no. 2 (April/May 2008), pp. 67–94; and Daniel R. Green, “The Fallujah Awakening: A Case Study in Counter-Insurgency,” Small Wars and Insurgencies 21, no. 4 (December 2010), pp. 591–609, both of which attribute Iraq’s 2007 violence reduction chiefly to the Awakening. For a counterargument, see Stephen Biddle, Jeffrey A. Friedman and Jacob N. Shapiro, “Testing the Surge: Why did Violence Decline in Iraq in 2007?” International Security 37, no. 1 (Summer 2012), pp. 7-40.

4. Headquarters, Department of the Army, FM 3-24: Counterinsurgency (Washington, DC: U.S. Government Printing Office, 2006), reprinted as United States Department of the Army, The U.S. Army/Marine Corps Counterinsurgency Field Manual (Chicago: University of Chicago Press, 2007).

5. Though this position is to date more widely held than published, for an illustrative articulation see Bing West, The Wrong War: Grit, Strategy, and the Way Out of Afghanistan (New York: Random House, 2011); also the views in U.S. Army Military History Institute, Iraq Surge Collection, esp. audio file 69.

6. See, e.g., Bob Woodward, Bush at War (New York: Simon and Schuster, 2002): pp. 220, 237, 241, 252-263, 310; Terry H. Anderson, Bush’s Wars (NY: Oxford University Press, 2011): pp. 88, 159-67.

7. See “Operation Iraqi Freedom,” Iraq Coalition Casualty Count,

8. U.S. fatalities in Afghanistan totaled 317 in 2009 and 127 in 2013: “Operation Enduring Freedom,”,

9. For general accounts of the post-2009 fighting in Helmand, see, e.g., Jeffrey Dressler, Counterinsurgency in Helmand (Washington, DC: Institute for the Study of War, 2011); Rajiv Chandrasekaran, Little America: The War within the War for Afghanistan (New York: Knopf, 2012); Carter Malkasian, War Comes to Garmser (New York: Oxford University Press, 2013).

10. For the doctrinal troop density norm for COIN, see FM 3-24, para 1-67.

11. Rajiv Chandrasekaran, “Nawa turns into proving ground for U.S. strategy in Afghan War,” Washington Post, December 12, 2010; Matt Millham, “Despite Lingering Reliance on US Aid, Afghanistan’s Nawa District a Model of ‘Transition,’” Stars and Stripes, May 28, 2011; Theo Farrell, Operation Moshtarak: An Analysis of the Campaign in Nad-e-Ali District, Helmand, Redacted version (British Army Land Warfare Centre, June 7, 2010), p. 16; Malkasian, War Comes to Garmser, figure 2, Op. cit.

12. See Tyrone C. Marshall Jr., “Carter’s Walk Through Marjah Market Shows Progress,” American Forces Press Service, February 24, 2012:

13. UK House of Commons, Defense Committee, Securing the Future of Afghanistan, Vol. 1 (London: The Stationery Office Limited, 10 April 2013), p. EV-92.

14. See, e.g., Carl Forsberg, Counterinsurgency in Kandahar (Washington, DC: Institute for the Study of War, December 2010).

15. Nancy Montgomery, “One Year After Offensive: Signs of Progress in Marjah,” Stars and Stripes, February 14, 2011,

16. Author’s observations, ISAF Headquarters, Kabul, June 2009, January 2010, June-July 2010, November 2010, March 2011, and November 2011.

17. For nationwide and provincial violence statistics, see the annual series Report to the Congress on Progress Toward Security and Stability in Afghanistan (Washington, DC: U.S. Department of Defense, various years).

18. See, e.g., Tom Ricks, Fiasco: The American Military Adventure in Iraq, 2003 to 2005 (New York: Penguin, 2007); Michael Gordon and Bernard Trainor, Endgame: The Inside Story of the Struggle for Iraq, from George W. Bush to Barack Obama (New York: Pantheon, 2012).

19. See, e.g., Andrew Krepinevich, The Army and Vietnam (Baltimore: Johns Hopkins University Press, 1988); Jeffrey Record, The Wrong War (Annapolis: Naval Institute Press, 1998).

20. Dale Andrade, Surging South of Baghdad: The Third Infantry Division and Task Force Marne in Iraq, 2007-2008 (Washington, DC: U.S. Army Center of Military History, 2010), p. 318.

21. See, e.g., Fred Kaplan, The Insurgents: David Petraeus and the Plot to Change the American Way of War (New York: Simon and Schuster, 2013), pp. 263-4, 341; Linda Robinson, Tell Me How This Ends: General David Petraeus and the Search for a Way Out of Iraq (New York: Public Affairs, 2008), pp. 81, 156, 261, 331.

22. For troop counts, see Ian S. Livingston and Michael O’Hanlon, Afghanistan Index (Washington, DC: Brookings, August 2013), figures 1.1 and 1.2; figures above are for March 2008.

23. Specifically, the administration has identified two chief U.S. interests in Afghanistan: that the country not again become a base for al-Qaeda to attack the West, and that it not become a base for militants to destabilize Afghanistan’s neighbors. On the Administration’s interest calculus in Afghanistan, see Stephen Biddle, “Is It Worth It? The Difficult Case for War in Afghanistan,” The American Interest 4, no. 6 (July/August 2009), pp. 4-11.