7.

The Second Washington Fight

Afghanistan

Counterinsurgency is messy and slow under the best of circumstances, and Afghanistan was a long way behind Iraq in almost every way that mattered. The human capital in Iraq was far better; by 2009, the Afghan people had been shattered by thirty continuous years of war. The Soviet invasion of Afghanistan in 1979 was followed by a bitter guerrilla war in which mujahedeen, supported by the CIA, fought to expel the invaders, an almost unbelievable true story well chronicled in George Criles’s book Charlie Wilson’s War. After the Soviets’ departure came a civil war in which the Taliban eventually seized control of most of the country, although the central government survived as long as funds from the Soviet Union kept coming. The Afghan government fell only when the Soviet Union collapsed, some three years after the last Soviet troops crossed the Termez Bridge out of the country. The Taliban made the error of providing a home for Al Qaeda and its leader, Osama bin Laden. That didn’t work out well for the Taliban, as the American invasion in 2001 drove the remnants of both organizations into the frontier regions of Pakistan, where they almost immediately resorted to plotting an insurgency inside Afghanistan against the American forces there.

As a result of the decades of constant fighting, good leaders and administrators were even harder to find in Afghanistan than they had been in Iraq. Iraq also had the advantage of substantial and increasing oil revenues, meaning that it could largely pay for its own security forces and development, and impressive infrastructure. Afghanistan was going to be a ward of the international community for the foreseeable future, and those in power had much more temptation to try to grab aid money to line their own pockets while the getting was good. They knew that the wartime funds were not going to last forever, unlike the prospect that Iraq’s leaders enjoyed: a share of substantial long-term oil revenues if they governed that country reasonably well.

Afghanistan faced other challenges as well. It was a rural rather than an urban insurgency. The success rate of rural insurgencies is far higher than those fought largely in urban areas because the population in an urban insurgency is already concentrated in cities and easier to separate from insurgents, using expedients like the T-walls that divided ethnic factions in Baghdad. Furthermore, the population of Afghanistan was largely illiterate. I had explained the impact when briefing my classes of future advisers at Fort Riley: “In Iraq, they knew how to read; we had to teach them how to fight. In Afghanistan they know how to fight; we have to teach them to read. Unfortunately, it takes a lot longer to teach someone how to read than it does to teach them how to fight.”

But the biggest problem in Afghanistan was, and remains, Pakistan. The initial U.S. mistake of providing too few American troops to pin down Osama bin Laden in the Tora Bora Mountains allowed Al Qaeda and much of the Taliban’s leadership to escape from Afghanistan. They settled in Pakistan, which either willingly or through lack of control over its own territory provided bases for both organizations to lick their wounds and regenerate. This they did with a vengeance. The Taliban regained strength in Pakistan and began reinfiltrating into Afghanistan in the mid-2000s while the attention of the United States was focused on the escalating disaster in Iraq. By 2008, when I had been shocked at the positive progress of the war in Iraq during my summer visit but disappointed to see the opposite in our other war, the Taliban clearly had the upper hand in Afghanistan. I reported this fact to Admiral Michael Mullen, who had replaced General Peter Pace as chairman of the Joint Chiefs of Staff, in an office call upon my return from Afghanistan in the fall of 2008. Not long afterward he correctly told the Senate Armed Services Committee that “In Iraq, we do what we must; in Afghanistan, we do what we can,” with whatever resources we had left over after we had sent to Iraq what was needed there to prevent full-scale civil war in the heart of the Arab world.

I had visited Afghanistan for the first time in early 2007, while training U.S. Army, Navy, and Air Force teams at Fort Riley to serve as advisers to the Afghan Army and Police. It seemed like a good idea to have actually been to Afghanistan at least once before standing on a podium telling American soldiers, airmen, and sailors what their war would be like. I found a critically underresourced war, starved of the soldiers, dollars, and attention required to beat back a resurgent Taliban. It was difficult to return to Fort Riley and send advisory teams to a war that I knew was an economy-of-force effort.

Although central Kansas was a long way from the Pentagon, there were still a few things I could do to help. I was astounded during my 2007 visit to find that many of the counterinsurgency lessons that had been painfully relearned in Iraq had not been transferred to the Afghan theater. Chief among them was the need for an in-country counterinsurgency academy to teach tactics, techniques, and procedures to those arriving for a tour of duty. General Casey had established one in Taji, Iraq, back in 2005, but two years later there was no comparable effort in Afghanistan, where knowledge of Afghan culture and the principles of counterinsurgency were sadly lacking among the troops we deployed there.

I got in touch with a friend from my Pentagon days who was now working as a strategic adviser to the head of the Afghan training mission and suggested that he establish a COIN academy in Afghanistan, explaining how effective General Casey’s effort in Iraq had been at improving the counterinsurgency performance of U.S. units sent there, at relatively low cost. I was pleased to find a friendly ear. The idea was quickly adopted, and I was invited to head over to Afghanistan to set up the Counterinsurgency Academy myself, but Fort Riley and the First Infantry Division, in which I was then serving, denied permission. As a battalion commander training advisers for two wars, I couldn’t be released for this mission.

Fortunately, I knew just the man for the job. Captain Dan Helmer had just completed the makeshift adviser training at Fort Riley and been deployed to Afghanistan as a combat adviser. Dan was a West Pointer and Rhodes scholar who had taught himself Arabic during a previous Iraq rotation, had read and internalized the counterinsurgency literature and philosophy, and was completely impervious to conventional views on the proper role of a captain in the U.S. Army. On my recommendation, he was given the mission of standing up the COIN Academy–Afghanistan. Working on a shoestring, Dan lied and stole to accomplish the mission, acquiring officers senior to him as meat puppets to put more authority behind his pronouncements and generally making miracles happen. When he left Kabul after a year, the COIN Academy–Afghanistan was fully up and running. At my encouragement and to the detriment of his Army career, Dan then came back to Fort Riley to train future advisers for Afghanistan, bringing with him a boxed flag for me that had flown over the COIN Academy in Kabul. He inscribed the flag box “To the architect of the counterinsurgency insurgency—the wheels are in motion.”

Captain Dan Helmer teaching at the COIN Academy–Afghanistan he helped establish.

I visited Afghanistan again as a civilian senior fellow working at CNAS in late 2008 and again found an underresourced theater of war, although General David McKiernan, who was now commanding the effort there, at least appeared to have a good handle on what it was he was supposed to accomplish. One of the highlights of my visit was a short session at the Counterinsurgency Academy that Dan Helmer had set up outside Kabul. The stage was set for an increase in American attention to the Afghan theater of war.

Senator Barack Obama, who had campaigned in part on refocusing attention from Iraq to the war in Afghanistan, was elected to the presidency a few days after my return to the United States from my Afghan visit. Upon assuming office, President Obama recognized that the situation in Afghanistan was even more dire than he had thought as a candidate. When he took office, he found waiting for him a request for tens of thousands more troops to secure the Afghan presidential election scheduled for the summer of 2009. The request was already several months old, but the George W. Bush White House had not acted on it. Sending the additional troops to Afghanistan as General McKiernan had requested would have revealed the truth of then-candidate Obama’s argument that “the good war” in Afghanistan was being neglected. If General Casey had been correct that “President Bush has given me a load of shit” in Iraq, President Obama could perhaps be forgiven for feeling the same way about Afghanistan.

President Obama faced an immediate decision about Afghanistan even before the Senate had confirmed his whole defense team. Important voices in the debate included Michèle Flournoy, recently confirmed as the top policy official in the Pentagon, and my former West Point student Craig Mullaney, who after fighting in Afghanistan and leaving the Army had handled Afghanistan for Senator Obama during the campaign and been given a job in the Pentagon with responsibility for that war. With Michèle and Craig’s encouragement, President Obama immediately established a review panel headed by Brookings Scholar and South Asia expert Bruce Reidel, and within three months ordered the deployment of 30,000 additional U.S. forces to the fight there. Nate Fick and I attended the troop increase announcement at the White House and were heartened by the new president’s commitment to counterinsurgency principles and to success in Afghanistan.

Changes in strategy often require changes in personnel to implement them. General David McKiernan understood the war in Afghanistan far better than any of his predecessors save Lieutenant General Dave Barno, who had run the war there in 2003 and 2004. Barno had worked to implement counterinsurgency principles despite being poorly resourced—he had only one Army brigade at his disposal to cover the requirements in all of Afghanistan. General Barno would later tell me that he’d kept his hardback copy of Learning to Eat Soup with a Knife on his desk during his command in Kabul, and he had me sign it in General Helmick’s backyard after the Army ten-miler upon Barno’s return from theater in 2005. In a sign of how little the Pentagon cared about the Afghanistan conflict, Lieutenant General Barno was given the job of running Army installation management after commanding the U.S. effort in Afghanistan. He retired from the Army to teach at National Defense University, and I later hired him to run the CNAS Afghanistan program.

McKiernan was judged to be too conventional and not energetic enough by Secretary Gates, who relieved him in favor of Stanley McChrystal in June 2009. General McChrystal had impressed the Pentagon through his leadership of Joint Special Operations Command for the past five years, making significant contributions to the kill/capture part of the counterinsurgency effort in Iraq and Afghanistan, and through his focus on supporting our efforts in those wars during his subsequent appointment as director of the Joint Staff in the Pentagon.

While he was serving in the Pentagon in 2009, I helped General McChrystal implement the long-overdue creation of “AfPak hands,” developing officers and sergeants specially trained and educated in counterinsurgency, regional languages, and local culture. Admiral Mullen dedicated the AfPak hands to repeated assignments to the region, either in theater or back in the United States in between deployments. That it took until 2009 to stand up a small corps of a few hundred military experts on the region was another indication of the Iraq War stealing all the oxygen from the Afghanistan effort. Despite significant flaws in execution—many of the people selected as AfPak hands had been “voluntold” that they would be participating in it, and the education they received has varied widely in quality—the program at the very least requires repeated assignments to the same region to build enduring relationships between Afghan and American leaders, a lesson that took far too long to learn.

Upon being given command in Afghanistan, General McChrystal set up a strategy review panel and invited me to be a part of it. Although I dearly wanted to accept, I had just taken the position of president of CNAS and couldn’t afford to take the time away from my new responsibilities of hiring talent and raising money. For the first time I understood fellow defense policy analyst Mike O’Hanlon’s incredulity that I had accepted the promotion to head CNAS; he told me that I would no longer be able to serve as a thinker and opinion leader because of the administrative burdens of my new role. I wasn’t smart enough to figure out that he was right until I had to tell McChrystal no. In my stead Nate and I sent Andrew Exum to serve on the Afghanistan strategy review panel. Ex was a former Army Ranger with time in both Iraq and Afghanistan who was finishing his doctoral dissertation on the strategy of Hezbollah; he was one of the first people Nate and I had hired.

General McChrystal’s strategy review panel included not just Ex but also Fred Kagan, who had been an important thinker behind the decision to “surge” troops to Iraq, and Steve Biddle, with whom I’d traveled to Afghanistan a few months earlier, visiting the COIN Academy in Kabul for the first time. The review panel found a situation that was “deteriorating,” as the first sentence of their assessment noted, and recommended “an integrated civil-military counterinsurgency strategy” to turn things around, which would require another significant troop increase on top of the one that had just been decided upon.

This caused real consternation in the White House, which had only a few months earlier signed off on the troop increase for Afghanistan that the Reidel study group had recommended. McChrystal’s ultimate request for an additional 40,000 troops began a long period of deliberations in the White House that became quite public. I had the chance to influence the debate when I testified before the Senate Foreign Relations Committee on the prospects for fighting “a better war in Afghanistan” on September 16, 2009. I argued that the United States was trying to accomplish the ends of keeping continued pressure on terrorist groups in the region, preventing a sanctuary for terrorists (as was the case before 9/11), and ensuring that there was no regional meltdown—code for Pakistan losing control of its nuclear weapons to radical Islamists. The ways I suggested were the same clear-hold-build strategy that had worked in Iraq, and the means required to do it was a U.S. troop surge to create space in which to build and professionalize more Afghan troops and police. I also suggested a renewed effort to conduct more effective information operations—in Afghanistan, in the region, and here at home.

In a helicopter in Afghanistan with Mike O’Hanlon.

My fellow expert witnesses included Rory Stewart, who had walked across Afghanistan in 2002, soon after the fall of the Taliban, and written about the journey in a book titled The Places in Between. He had later set up a nonprofit, Turquoise Mountain, to encourage economic development in Afghanistan. The other was Steve Biddle, freshly back from McChrystal’s review group; Biddle noted correctly that the answer to the question of whether Afghanistan was worth the cost of a fully resourced counterinsurgency campaign was “a close call on the merits.” Nonetheless, he came down in favor of more civilian and military resources for Afghanistan, applied in the same manner as they had been in Iraq, where Biddle’s support for the surge had played some role in influencing President Bush’s decision to implement the strategy.

In addition to rather publicly advocating a counterinsurgency strategy for Afghanistan before the Senate Foreign Relations Committee, I also had the chance to make the argument before a smaller but more influential audience. Vice President Joe Biden played a major role in the decision over troop levels for Afghanistan that occupied much of 2009. He was not an advocate of a fully resourced counterinsurgency strategy, believing that a smaller footprint of American troops focused on counterterrorism operations would be sufficient. I was invited to the White House to try to change his mind, spending ninety minutes in his office explaining that counterterrorism by itself would do nothing to change the dynamics that led terrorists to decide to wage war against the Afghan government and its American supporters. A counterterrorism strategy by itself was a recipe for endless war; it was necessary to resource improvements in governance, economic development, and the provision of services to the population to persuade Afghans not to support a Taliban insurgency that promised it would do all those things if it regained power. The other problem with a counterterrorism strategy was that it offered little to the Afghans themselves in return for basing rights; in order to get them to permit us to conduct such operations from their territory, I believed that we had to offer them something in return—economic development, greater security, and the prospect of eventual peace through the defeat of the Taliban. The vice president was unconvinced, but I appreciated the chance to make the argument in person.

The White House debate over a second increment of troops for Afghanistan in 2009, in which the vice president played a major role, took place against the background of the summer Afghan presidential election that President Karzai won in an election aided by significant ballot stuffing on his behalf. He would have won without cheating, so the effort cost him in international credibility for no real gain. It also made it harder for American advocates to argue that the fragile democracy in Afghanistan was worth an increased investment of our blood and treasure.

The Obama administration had inherited the summer elections in Afghanistan from the Bush administration, which pursued a “freedom agenda” throughout both of its terms in office. This was one of the critical errors of President Bush’s foreign policy and was ironically based on an overly simplistic understanding of international relations theory. The Democratic Peace Theorem notes that mature democracies do not wage war with each other very often. The key word is mature. Countries going through the process of democratization are actually more likely to wage war than are autocracies or established democracies. The essence of mature democracies is not honest voting to determine the will of the majority; it is effective institutions to protect the rights of the minority, even if the majority doesn’t like them. These institutions play a major role in establishing peaceful relations both internally and externally, but building them can take decades.

The Bush administration’s misreading of international relations theory had real-world implications in many countries, including Iraq and Afghanistan. By moving too quickly to create democracies without doing the hard work to create institutions to protect the rights of minority groups—work that can literally take generations, as it did during the development of strong democracies in Taiwan and South Korea—the Bush administration simply made it more likely that the most ruthless and best-organized thugs would seize power in the Palestinian territories, in Iraq, and in Afghanistan. We have since seen similar results in Libya, Egypt, and in the insurgency in Syria.

The Bush administration’s “freedom agenda” mistake was even more consequential than it might appear at first glance. In Malaya, Gerald Templer had essentially embodied political and military power. He had no need to negotiate with the government of Malaya—he was the government of Malaya for the duration of the Emergency, during which Britain suspended the normal political order. In both Iraq and Afghanistan, the United States and the international community granted sovereignty to regimes that didn’t yet have the capacity to exercise it and encouraged democratic elections before founding and nurturing the political parties and other institutions that underlay all functioning democracies worldwide. In both Iraq and Afghanistan, American commanders and ambassadors would have to cajole actions from the local political leaders that General Sir Gerald Templer could have simply ordered to happen. It is a crippling distinction in the history of counterinsurgency campaigns, and it was a self-inflicted wound for our efforts in both Iraq and Afghanistan—imposed at the national command level, well outside the control of military leaders.

However, in 2009 it was far too late to undo the Afghan constitution, and so after the flawed elections of the summer, the Obama administration was stuck with President Karzai for another five years. Rather than making political decisions unilaterally, or even in consultation with junior Afghan partners, American ambassadors and generals were forced to plead with Karzai to fire corrupt governors and replace them with more capable successors. Improving governance and encouraging economic development would have been a challenge in Afghanistan in any case, given the limits of human capital in the country after three decades of war. Accomplishing these tasks when it was impossible for American leaders to independently fire or prosecute Afghan leaders who stole development funding (just one of many examples of corruption and malfeasance) added a degree of difficulty to the task that would prove almost insurmountable.

The political decision to allow democratic elections so early in the lifespan of the new Afghanistan was far from the only error of U.S. policy there. The lack of significant, well-resourced U.S. effort in Afghanistan over the eight years preceding President Obama’s two decisions in 2009 to increase troop strength there had given the Taliban a second chance. They took the opportunity to regain strength both in the south around Kandahar, where their movement had begun, and in the east, close to their sanctuaries in Pakistan. General McChrystal’s request for 40,000 additional troops provided sufficient resources to take on both problem areas nearly simultaneously, but it would prove too big a bill for the president to swallow in the fall of 2009 after already doubling down on troops for Afghanistan earlier in the year.

Along with several other think-tank denizens, I was invited to the White House Situation Room to hear the results of the months of Afghan policy deliberations on the afternoon of December 1, 2009, as Marine One was preparing to ferry the president to West Point for the announcement. The troop numbers were about 10,000 short of what I had recommended, which seems like a small “commander in chief tax,” as some in the White House called it, but was actually hugely important. The president’s decision to send just 30,000 American troops—and to pledge to get the rest from coalition countries—meant that the effort to defeat the Taliban could not be carried out simultaneously Afghanistan-wide but would instead have to be sequential—focusing effort first on the south and swinging to the east later.

Although America’s allies in the Afghan effort did, over time, promise an additional 10,000 troops, they came in packages of varying size, capability, and thus utility. And the sad truth was that only American soldiers were consistently effective enough in an offensive role to clear the Taliban from their sanctuaries in the south and east; the additional allied troops might be able to hold areas that Americans cleared, but that was about it. The International Security Assistance Force (ISAF) consisted of troops from nearly forty countries, and they varied widely in degree of commitment to the fight. While some were excellent—the Canadians for many years paid a heavier toll in blood per capita than did the United States, while the Brits, Australians, and French were also great fighters—the many jokes about the acronym ISAF, such as “I Suck at Fighting” and “I Saw Americans Fight,” hint at a bitter truth about which nations were willing to pay the high cost of doing the hard work of clearing the Taliban.

The Afghan “surge,” to be announced that evening at West Point, suffered from another problem. In Iraq, the fight to build host-nation security forces began in earnest not long after the Iraqi Army had been disbanded. Real assets were devoted to the effort to rebuild it, including, already in June 2004, putting Lieutenant General Petraeus in charge of the effort, although it took some six months to assemble the organization he needed, and many of the troops under his command who were assigned advisory and training duties were not the first tier. But unlike in Iraq, in Afghanistan the effort to build an Afghan Army and Police force was cursory at best. I had discovered this at Fort Riley in my last assignment in the Army, when I could tell advisers assigned to Iraq exactly where they would be assigned even before they arrived in Kansas, while their counterparts assigned to Afghanistan literally got their assignments when they landed in Kabul. As a result, the American soldiers who cleared Baghdad during the Iraq surge had capable, well-advised Iraqi forces to help hold what they had cleared with U.S. forces, while the effort to build a decent Afghan force began roughly simultaneously with the commitment of American forces to clear southern Afghanistan. There would be no Afghan troops trained and ready to hold what Americans cleared, much less to build a better country in the newly purchased free territory.

But the biggest flaw in the president’s decision to commit more forces to Afghanistan was not in the sequencing of building Afghan forces simultaneously with the commitment of American surge forces (a decision that had effectively been made for him by the neglect of the advisory effort under the previous administration) or the commitment of insufficient U.S. forces to clear both the south and the east simultaneously. Instead, the critical error was the decision to announce that the “surge” troops for Afghanistan would begin their withdrawal in just eighteen months, during the summer of 2011.

When this fact was briefed to the think-tankers in the White House Situation Room on the afternoon of the president’s West Point speech, our reaction was immediate and unanimous: this was a disaster of the first order. In Afghanistan, and just as important in Pakistan, the announcement of an increased American troop commitment to the war would be instantly overshadowed by the announcement of the time limit on the American troop commitment. The only words that our allies, and our enemies, would hear from the president’s speech would be the withdrawal date. The president was cutting his own policy off at the knees before it had even gotten up on its feet.

The cause for this self-defeating policy decision was domestic politics. The president paid a heavy price with his Democratic base for the increased troop commitment to Afghanistan. They apparently hadn’t believed his repeated promises on the campaign trail to increase the U.S. troop commitment to Afghanistan. I explained to Rachel Maddow that night on her program on MSNBC that the president had done exactly what he had promised to do in Afghanistan, at some cost to his popularity among Democrats. Announcing that the Afghan surge would be limited in duration was in part at least a commitment to his political base to start ending the war in his first term, before the 2012 election.

But nothing is free. The president undercut the value of the increased U.S. troop commitment by announcing its withdrawal date, giving both the Taliban and Al Qaeda a finish line in sight: if they could hold on for eighteen months, the American tide would recede. The Afghans were listening as well, and the lesson they took was similar: The Americans are going to abandon us again, just as they did after we defeated the Soviet Union in the 1980s, so get everything you can now while the getting is good. Announcing the withdrawal date in advance was a grievous self-inflicted error, or “own goal” in the British parlance.

The White House would claim that Secretary Gates, Admiral Mullen, and Generals Petraeus and McChrystal agreed to or at least accepted the president’s decision to announce the date to begin drawing down the surge at the same time as he announced the additional forces. However, some of those involved stated later that there was no true discussion; rather, they were confronted with a “take it or leave it” moment, and they decided to take it, knowing that leaving it would have given no chance to achieve our objectives. General McChrystal had no choice but to comply with the president’s directives, and he set to work attempting to implement a fully resourced counterinsurgency strategy for the first time during our decade of war in Afghanistan—at least in the south, as he didn’t have the resources to simultaneously get started with a fully resourced COIN campaign in the east.

The process began in Helmand province, which had been the responsibility of the British Army, who called it “Helmandshire,” but was now being buried under the weight of Marines. They were no longer responsible for Al Anbar Province in Iraq, my old stomping grounds. The first test of counterinsurgency in Afghanistan was slated to be the town of Marjah, where on February 13, 2010, Marines poured in to clear out the Taliban. General McChrystal made the mistake of overpromising, telling The New York Times’s Dexter Filkins, “We’ve got a government in a box, ready to roll in.”

We didn’t. Afghan forces were not yet developed enough to hold what the Marines had cleared, and there were insufficient political and development specialists in Afghanistan to roll into Marjah for the build phase. The fight in Marjah, which had been scheduled to take just a few weeks, dragged out for the rest of the year, consuming vast resources and attracting far more attention than it deserved. It became an early indicator that Afghanistan was not going to “flip” as quickly as had Iraq under General Petraeus. Afghanistan was harder because of the insurgent sanctuaries in Pakistan, insufficient Afghan human capital, the lack of adequate governance and institutions, and limited infrastructure and government revenues.

McChrystal was deeply concerned about civilian casualties in Afghanistan, correctly understanding that counterinsurgents who killed the innocent created recruits for the enemy. This was a primary tenet of the Counterinsurgency Field Manual, and the war in Afghanistan was one of the cases we were thinking about when we wrote it. John Paul Vann, a famous counterinsurgent during the Vietnam War, had noted of that conflict, “This is a political war and it calls for discrimination in killing. The best weapon for killing would be a knife, but I’m afraid we can’t do it that way. The worst is an airplane. The next worst is artillery. Barring a knife, the best is a rifle—you know who you’re killing.” Unfortunately, given how underresourced the fight in Afghanistan had been, there were too few rifles available, and when troops got in trouble, often the only option available to bail them out was airpower, which sometimes went astray and killed the innocent. McChrystal imposed firm restrictions on the use of airpower and artillery in his attempt to win over the Afghan population, dramatically reducing civilian casualties—but perhaps overcorrecting, at least in the eyes of subordinate commanders.

General Petraeus had done a good job of underpromising and overdelivering in Iraq, but McChrystal took the opposite approach. He also misread Petraeus’s skill in managing members of the media. Although Petraeus worked the media constantly, answering press e-mails at all hours of the day or night, he was very careful about how he talked with members of the press and how much he revealed to them. Between his time at grad school at Princeton and in the Social Sciences department at West Point, he had gotten to know a large number of writers, and he worked assiduously to show them only the sides of him that he wanted revealed on the front page of The New York Times.

McChrystal hadn’t had the same developmental experiences during his long career in the special forces and didn’t have the same natural caution. When, encouraged by his staff, he allowed Michael Hastings, a stringer with Rolling Stone magazine, to embed with him for a few days in early 2010, he let Hastings inside the circle of trust. A veteran of many years of service inside the close-knit special operations community, knowing almost no one who did not adhere to his personal code of loyalty to mission and command, it was beyond Stan McChrystal’s ability to imagine that a fellow American would report things that would hurt him.

He was wrong. The Rolling Stone article that was published that summer under the title “Runaway General” featured comments by members of McChrystal’s staff about Vice President Biden that were deeply disrespectful and, according to Hastings, encouraged by the general. McChrystal was recalled to Washington and, on June 23, tendered his resignation to the president, who accepted it and announced on the same day that Petraeus would step down from his position at Central Command in order to take responsibility for the war in Afghanistan. Petraeus hadn’t had time to tell his wife, Holly, that he was going back to war.

He did so on July 4, almost immediately overriding his predecessor’s overly restrictive guidance on the employment of lethal force in the counterinsurgency campaign. Although counterinsurgency is mostly political, there are committed insurgents who cannot be persuaded to quit and who have to be captured or killed. Even those who are persuadable are more likely to abandon their objections if they face a real threat of being killed or captured; one of the indices of progress we tracked in Khalidiyah in 2004 was the rising price the insurgent leaders had to pay to get a poor subordinate to fire a rocket-propelled grenade at one of our tanks. The cost went up in consonance with the likelihood that the RPG firer wouldn’t survive to enjoy his earnings.

McChrystal had overinternalized the guidance in the Counterinsurgency Field Manual. Only some of the best weapons for COIN don’t shoot bullets, and although dollars are weapons in this kind of fight, bullets work pretty well in a lot of circumstances. McChrystal’s guidance had limited the ability of American units to call in air support or artillery fire even when they were under enemy fire, but Petraeus restored firepower to its proper place in a counterinsurgency fight, noting that “protecting the Afghan people does require killing, capturing, or turning the insurgents.”

This change was enormously popular in theater, and the results were almost immediate, as the employment of air strikes and of joint special operations raids increased rapidly. Petraeus and his team implemented a number of other changes, including the commencement of the Afghan Local Police program, support for an insurgent reintegration program modeled on the one that had paid off in Al Anbar, an anticorruption task force led by newly promoted Brigadier General H. R. McMaster, a huge increase in the program to train and equip Afghan forces, and an effort to build capability and capacity to expand the rule of law led by Brigadier General Mark Martins. All made progress in key areas, but time was always the limiting factor.

Petraeus succeeded in putting more pressure on the Taliban and in rapidly increasing the size and capability of the Afghan security forces, but he was fighting an opponent as implacable as Afghan government corruption: the Washington clock. He would later tell me, “We didn’t get the inputs even close to right in Afghanistan until the late fall of 2010. . . . And, in the end, we only had the benefit of all the forces for some six months or so, with the drawdown commencing in the summer of 2011.” By the time he left his position in Afghanistan in July 2011, the drawdown of U.S. surge forces from Afghanistan had already begun. President Obama was true to his word; he announced the end of the Afghan surge in June 2011, exactly eighteen months after it had begun, and began bringing the troops home in July.

General Jack Keane, longtime mentor of Petraeus and intellectual godfather of the Iraq Surge, urged Petraeus to resign in June 2011 when President Obama announced the drawdown would be carried out at a more rapid pace than Petraeus had recommended. General Petraeus, already nominated to become the CIA director, felt that he had been given every opportunity to make his case to the president and chose to continue to serve despite his disagreement with the decision to end the Afghan surge as originally promised.

In fact, Afghanistan was harder than Iraq in some key respects, even though it was never anywhere near as violent as Iraq was in 2006 and the first half of 2007, and the outcome of the campaign was very much in doubt when he departed. The results on the ground were as mixed as we think-tankers would have predicted that afternoon of December 1, 2009, in the White House Situation Room. On the glass-half-full side of the equation, the U.S. troop increase had been substantial, and it did have a huge impact in the south and the west, where the main effort was led by the Marines and an Army division. The clear-hold-build transfer strategy proved its effectiveness on the ground, not for the first time in the history of counterinsurgency campaigns. After years of neglect, the Afghan National Army grew in both size and capability and was able to hold areas that had been cleared by U.S. forces with only a minimal advisory presence. The Afghan Police also improved, albeit from a lower base, and the Afghan Local Police program initiated by Petraeus was scaled across Afghanistan in an example of a historical COIN best practice that was eventually adopted across the campaign.

Kill/capture operations to take down high-value targets proved successful as a result of improvements in the links between operations and intelligence gathering and some very innovative intelligence fusion operations. With our forces and those of our NATO and Afghan allies, we came tantalizingly close to the 1:50 ratio of counterinsurgency forces to the local population that the Counterinsurgency Field Manual noted had proved a tipping point in past COIN campaigns. The numbers were staggering: 600,000 troops would be indicated to protect the 30 million in Afghanistan’s population. At the peak in 2011, NATO had as many as 150,000 troops, with Afghan security forces still building to an expected total of 354,000 and an Afghan Local Police force of some 40,000. Unfortunately, Afghan forces were still being built up even as the NATO drawdown began, long before the Taliban had been defeated. Afghan forces will have to confront a challenge that may be more than they can handle.

Five years after President Obama doubled down on troops to Afghanistan and the AfPak hands program was stood up, the results are mixed. The greatest successes have come in the military field, but that is not the most important battlefield in this kind of war. David Galula said that COIN is 20 percent military and 80 percent everything else—political, economic, and information operations. It is possible to get everything right in the military domain and still lose the war, and there is some danger of just that outcome.

Counterinsurgency is as much political as military, of course, and the political shortfalls in Afghanistan are also significant. The most pressing is the continued disappointment of Afghan governance, which remains marred by corruption that is excessive even by Afghan standards. Brigadier General H. R. McMaster was given the task to reduce corruption, and the fact that even he wasn’t able to put much of a dent in the problem shows what a significant—and potentially decisive—obstacle Afghan governance is to achieving our objectives there. Promoting democracy in the short term can perversely make it harder to achieve stability in the short term and human rights in the long term, and there are many indications that this will be the result of American and international decisions to move directly to democracy in Afghanistan.

And if Afghan governance is troubling, Pakistan’s is worse. The most dangerous country in the world for the United States continues to allow the Taliban to operate from its territory, although it is unclear whether this failing results from a lack of ability to control its own territory or from a lack of will to do so; most likely it is a mix of both. The Pakistani deterrence triad has been described as nuclear weapons, the conventional Pakistani Army, and irregular forces including the Pakistani Taliban, which has turned into something of a Frankenstein’s monster that now threatens its own creator. Pakistan is the core of the problem in the region, the main driver of U.S. interests in Southwest Asia, and the keystone to solving it—but not for many years to come. It is worth underlining the fact that the success or failure of the Afghan security forces remains perhaps the most powerful lever to influence Pakistan’s strategic calculus.

The long-term presence of Osama bin Laden inside Pakistan for some five years prior to his dispatch by a SEAL team underscores the problems that the country poses. And the killing of Bin Laden, ironically, increased the challenge for the United States of staying the course in Afghanistan. U.S. public opinion, never much engaged in the Afghan War, turned even more negative toward the effort after the killing of Bin Laden and the effective dismantling of Al Qaeda by drone strikes inside Pakistan. Public opinion could be summarized as noting that Bin Laden is dead, Al Qaeda is no longer a threat to the United States, and it’s time to go home.

The war in Afghanistan is likely to end with a negotiated settlement that allows the Taliban some role in the Afghan government in return for verifiable Taliban commitments to renounce violence and Al Qaeda (and that will be easier to do as Al Qaeda fades more deeply into irrelevance) and to adhere to the terms of the Afghan constitution. As long as the United States remains committed to the security of the Afghan government over the long term, we are likely to accomplish our core national security objectives in the region, although at a much higher price than would have been necessary had we not taken our eye off the Afghan War for eight years to focus on an unnecessary one in Iraq.

America will leave a force of some thousands of troops to conduct counterterrorism operations, to support the Afghan government, and to advise Afghan security forces, particularly with intelligence assets, airpower, medical evacuation and assistance, and logistical support. These forces will be sufficient to prevent the Afghan government from falling to the Taliban, largely by making it too expensive for the Taliban to mass forces for a frontal assault on Kabul. Although the United States will no longer conduct counterinsurgency directly in Afghanistan, the Afghan government in Kabul will, with American help, continue to counter the Taliban’s insurgency. The remaining U.S. troops will focus on counterterror operations—part of counterinsurgency, and a part that can continue only if the Afghan government remains in control of its capital and airspace.

In 1975 Saigon fell not to guerrillas but to conventional forces from North Vietnam; a tank, the very icon of conventional war, broke down the gate at the U.S. embassy in Saigon. There is no danger of a conventional Taliban invasion of Afghanistan from its bases in Pakistan, only of continued infiltration and subversion—serious threats, to be certain, but ones that the Afghan forces, if enabled by American support, will likely be able to contain. The outcome will not be pretty; however, it should be marginally acceptable if we continue to provide sufficient support.

Great powers lose small wars for only one reason: they run out of will to continue the fight. After more than twelve years in Afghanistan, it is understandable that America is tired of war. But the commitment of a very small number of American troops and a relatively small investment of dollars may make the difference between the survival of a representative government in Afghanistan post-2014 and the return of the Taliban. At the end of the movie Charlie Wilson’s War, after American-backed mujahedeen have expelled the Soviet Union from Afghanistan, the Texas congressman and Naval Academy graduate pleads for a small investment in the future of the country. He failed. America chose to ignore Afghanistan after the war in that country had broken the Soviet Union, and in the power vacuum we left behind, the Taliban came to power and provided Al Qaeda with a base of operations that enabled it to attack us. It would be a shame to repeat that particular crime of neglect in the same place just a generation later.

This question of American support for Afghanistan over the long term may be the longest-lasting impact of General Petraeus’s affair with Paula Broadwell. I met Paula at a CNAS event sometime in 2009 and saw her around Washington at other events, including Dave Kilcullen’s wedding to Janine Davidson and a party at her brother’s house; I found her both impressive and ambitious.

I was apparently not the only one. Paula was writing her doctoral dissertation on leadership at King’s College, London, with General Petraeus as a case study; when he was unexpectedly assigned to take command in Afghanistan, she adroitly decided to turn her dissertation research into a book. Most of us do it in the opposite order, although I have been accused of similarly writing my book and then doing the research afterward, in Al Anbar. Petraeus developed a closer relationship with Broadwell than was proper over the course of her writing what became the biography All In and resigned from the CIA when it came to light.

The affair with Paula surprised many of us who had known Petraeus for decades and who believed him to be invulnerable. If an M16 round in the chest during a training accident and a collapsed parachute that resulted in a shattered pelvis couldn’t slow him down, what could? I was shocked to discover that Petraeus was human; I knew my own faults, but had imagined none in him.

And so one of the great Americans of our time was brought down by a sin that appears to have been more notable by its absence than by its presence in the great men of all eras past. Petraeus’s reputation will recover as, like President Clinton, he continues to serve the public interest. My friend and fellow baseball fan Paul Yingling has described Petraeus as having the most “wins above replacement” (the acronym WAR is particularly appropriate here) of any general officer in his generation. It is hard to imagine anyone else buying the time and space to turn Iraq around, or performing better than he did in the Afghan theater.

The real cost of Petraeus’s infidelity is the loss of a compelling advocate for the people of Afghanistan at a critical time in America’s involvement in that nation’s history. Over the course of 2014, the Obama administration decided how many American troops to commit to the security of Afghanistan after the formal end of combat operations in December 2014. The need is clear; without American help to check them, the Taliban are likely to regain control of much of rural Afghanistan and even threaten Kabul. Those concerned about stability in the region, the security of Pakistan’s growing nuclear arsenal, and the residual elements of Al Qaeda central want Afghan bases from which to conduct counterterrorism operations and intelligence gathering. The cost of those bases will be enough American advisers and airpower to give Afghanistan’s own security forces a fighting chance against the Taliban.

Petraeus would have been a powerful advocate for a strong and enduring American presence of perhaps some 15,000 troops; if the absence of his voice results in a considerably smaller force, the price of his private sins will be heavy indeed—and much of it is likely to be paid by the innocent people of Afghanistan, who have already suffered so much.