If the first step to making war a rational instrument is to determine the political goals, the second is to understand the character of the conflict.1 This assessment includes estimating the aims of the different belligerents; evaluating the importance of those goals to them (the “value of the object” for Clausewitz2); assessing the strengths and weaknesses, capabilities, and limitations of each actor; calculating comparative advantages; and weighing potential risks and opportunities.3 This process, political scientist Richard K. Betts argues, has a high risk of error as the policymakers struggle to use intelligence reports and assessments that can be vague, incomplete, misleading, or contradictory to make strategic decisions.4 At the same time, he notes, policymakers can ignore, discount, or misinterpret accurate intelligence.5 Intelligence producers and consumers are imperfect.
External powers face the additional problem of competing risks. Interventions do not occur in isolation of world events and other national interests. When interests clash, countries generally prioritize the more important ones, which may be to the detriment of an ongoing war. This risk calculus helps to explain why the Soviet Union did not invade Western Europe or start a nuclear war with the United States over its tangible support to Afghan insurgents in the 1980s. Likewise, the United States did not invade Iran over its support to Shi’a militants in Iraq and has not conducted a ground invasion of Pakistan to eliminate Taliban sanctuaries. Such actions could have increased the costs of supporting an insurgency. But their consequences to broader national security interests were far too high. Careful management of risk and uncertainty by the intervening power across an array of national security concerns may result in choices that reduce the prospects of decisive victory even further. The new Obama administration grappled with these tough choices as it faced a wide array of domestic and international challenges.
As is often the case, a change in administration was necessary for a change in strategy. Obama’s changes to the Afghanistan strategy, however, were more significant in scale than approach. Cognitive bias, bureaucratic frictions, and patron-client problems continued to impede the US government’s ability to recognize the limited prospects for a successful transition. Obama campaigned that Afghanistan was the “good” war—a war of necessity—while Iraq was the war of choice.6 One of his first acts as president was to order an interagency review of the war in Afghanistan, led by former CIA official and South Asia expert Bruce Reidel. The process served to concentrate the minds of senior administration officials and bring a shared appreciation of the challenges.
Reidel’s report outlined that the situation in Afghanistan was worse than expected. Afghanistan and Pakistan, it argued, should be a single integrated theater. The United States needed a stronger relationship with Pakistan to change its strategic calculus so they would stop using militant groups to advance their interests. The report called for a more effectively resourced counterinsurgency campaign in Afghanistan, to include holding President Karzai accountable for dealing with corruption. It also supported an Afghan-led reconciliation effort. President Obama outlined these findings in a March 27, 2009, speech and approved ISAF commander General David McKiernan’s request for 17,000 more American troops.7
Obama also made personnel decisions. He directed Secretary of State Hillary Clinton to create a special representative for Afghanistan and Pakistan (SRAP) and selected veteran diplomat Richard Holbrooke for the position. Holbrooke’s mission was regional diplomacy and prospective talks with the Taliban. Soon, other international partners created similar posts. The director of the Joint Staff, Lieutenant General Stanley A. McChrystal, created the Pakistan-Afghanistan Coordination Cell (PACC) to improve the effectiveness of the Joint Staff’s efforts. Robert Gates, who Obama had kept on as secretary of defense, picked a former deputy chief of mission to Afghanistan, David Sedney, to lead an upgraded Afghanistan-Pakistan-Central Asia office. He also appointed a former CIA operative during the Soviet-Afghan War, Michael Vickers, as the assistant secretary for special operations. Not satisfied with how McKiernan was responding to the new direction, Gates relieved him in favor of McChrystal. As the new commander was leaving for Kabul, Gates directed him to provide, within 60 days, an assessment of the war considering the president’s new direction and to inform him of any additional resources required.8
McChrystal’s assessment described the situation as “serious and deteriorating.”9 The Taliban and other insurgent groups had sanctuary in Pakistan and were threatening Kandahar and several critical locations across the country.10 They were tightening their grip in the provinces and districts around Kabul. The Afghan government was weak and corrupt, while the international effort was disjointed, ineffective, and creating animosity among the population. ISAF, the assessment argued, needed to address two critical threats: a growing insurgency and a crisis in confidence in the coalition and the Afghan government. Success was achievable, McChrystal noted, but not by “doubling down” on the same ways of doing business. A fundamentally different approach was needed, which included protecting the civilian population from harm by the Taliban and the Afghan government. He emphasized that responsible and accountable governance should be on par in priority with security. Without the former, success was not possible. He called for much greater unity of effort, both within the military coalition and with the international civilian efforts, and he recommended conflict resolution initiatives, such as reintegration and reconciliation.11
The assessment outlined five strategic risks: (1) the loss of coalition will and support; (2) lack of Afghan government political will to enact needed reforms; (3) failure by ISAF partners to provide adequate civilian capabilities to support good governance and economic development; (4) significant adaptations by insurgent groups; and (5) external malign activity from Pakistan and Iran.12 All five were clear and present problems; the magnitude of risks 2, 3, and 5 was becoming apparent.13 These are issues that tend to fall in the seams between bureaucratic silos. At no time was any American official held responsible or accountable to address them. The assessment did not address the question of war termination directly. Still, it argued that helping the government win the battle of legitimacy in the eyes of Afghans would reduce the Taliban’s ability to recruit local fighters and control population, thus forcing them to seek an end to the war.14
McChrystal provided periodic updates on the assessment to Gates, Chairman of the Joint Chiefs of Staff Admiral Michael Mullen, and Central Command’s General David Petraeus. Recognizing that any discussions over troop increases would be contentious, Gates directed McChrystal not to provide a resource recommendation until the NSC reviewed the assessment.15
The assessment, however, was leaked to the Washington Post soon after being sent to Gates.16 The issue of “more resources” was already being discussed by pundits in Washington, including some civilian members of the assessment team after they returned home.17 The White House, already suspicious of the military, believed the Pentagon was using leaks to force the president to approve a troop surge. The leak, which was not by McChrystal or his staff, exacerbated the suspicion into mistrust.18 These internal bureaucratic and principal-agent problems set the stage for a civil-military crisis in June 2010 when a Rolling Stone article reported disparaging remarks by McChrystal’s staff about the president.19
As the NSC reviewed the assessment, Gates directed McChrystal to develop resourcing options. McChrystal offered low-, medium-, and high-risk options: 80,000, 40,000, and 20,000 troops, respectively.20 Equally important, he urged, was a significant increase in intelligence capabilities, civilian expertise, development assistance, and a doubling of the size and capability of the Afghan National Security Forces. The administration deliberated from August until late November.21
These discussions took place in the wake of the August 20, 2009, presidential elections in Afghanistan. Convinced that good governance was impossible under another Karzai term, SRAP Holbrooke actively promoted Karzai’s rivals, particularly former Foreign Minister Abdullah Abdullah.22 Karzai suspected this and began accusing Holbrooke of interference.23 Karzai won 55 percent of the initial tally to Abdullah’s 28 percent.24 Abdullah refused to accept the results and accused Karzai of widespread fraud.25 Holbrooke and UNAMA deputy Peter Galbraith pushed hard for the internationally staffed Electoral Complaints Commission to investigate. Galbraith was fired after accusing UNAMA chief Kai Eide of a cover-up.26 The Electoral Complaints Commission declared roughly 1 million ballots to be fraudulent, just enough to put Karzai’s percentage below 50 percent and trigger a run-off between the top two candidates. Karzai refused Abdullah’s entreaties and American suggestions for a power-sharing deal.27 By November, a frustrated Abdullah declined to participate in the run-off, ceding the election to Karzai.28 Obama reportedly called Karzai afterward and lectured him about corruption.29 In his inaugural speech, Karzai called for a transition to Afghan-led security to begin within two years, for Afghan forces to take over security responsibility by the end of 2014, and for international forces by then to be reduced and limited to training and support roles.30 Criticized for getting nothing out of the elections dispute in terms of lucrative jobs, Abdullah and his supporters would increase their brinksmanship in the disputed election of 2014.31
The mess surrounding the elections intensified the debate in Washington over McChrystal’s requests for more resources.32 The US ambassador in Kabul, Karl Eikenberry, cabled that a surge would be a mistake because Karzai was an “unreliable partner” who lacked the political will to reform.33 Vice President Joe Biden continued to advocate for a smaller-footprint counterterrorism mission (dubbed “CT-plus”).34 Obama was reportedly upset that the military only offered options in terms of troop numbers.35 He believed the Pentagon was trying to box him in to approve a significant troop surge. Only the military, though, was asked to provide options.36 No one sought—and no agency provided—strategies that placed diplomatic efforts (such as reconciliation) or political efforts (such as addressing governance and corruption) as the top priority with the military in support. America’s bureaucratic way of war offered only military-centric options to the president.
Afghanistan, moreover, was not being considered in isolation. Obama wanted to wind down wars in Iraq and Afghanistan so that he could fund his domestic agenda, which included a major national health care initiative, addressing massive budget deficits, and bringing the economy out of recession after the 2008 banking crisis. Expanding the war in Afghanistan even further than he already had could inhibit those priorities.
Seeking to limit the US commitment to Afghanistan and to send a signal to Karzai to get serious about reform, the Obama administration began debating a timeline to withdraw troops. The administration did not address conditionality or other options to tackle corruption. The military deferred to its recent experience in Iraq as a gauge for when the campaign would show results.37 Based on advice from the uniformed military, Gates suggested that areas cleared of the Taliban could transition to the Afghan government within two years.38 The White House thus decided to withdraw the surge forces beginning in July 2011.
The NSC did not discuss methods besides transition-and-withdraw for achieving a favorable and durable outcome, nor did they examine the feasibility of transition.39 The questions for debate were limited to the scale of the surge and its timeline. The military was to attrite the Taliban and expand the ANSF while the civilians were to build government capacity. Diplomats were to convince Pakistan to pressure and eventually shut down insurgent sanctuaries.40 The Obama administration believed these efforts would reduce the Taliban to a residual insurgency by the end of 2014. At that point, the ANSF could defeat them. The NSC discussions did not address the probability of tackling government legitimacy and insurgent sustainability. Beyond the Iraq example, the NSC did not examine readily available studies to assess the validity of the transition theory and the likely amount of time required.41 The White House reportedly rebuffed suggestions to review comparative examples.42 “History,” former Under Secretary of Defense for Policy Michèle A. Flournoy recalled, “never had a seat at the table.”43