Bureaucratic frictions can paralyze strategic decision-making. To manage national security affairs, the US government has developed powerful institutions. The National Security Council consists of the president of the United States, the vice president, the secretary of state, and the secretary of defense. They are supported by national security staff at the White House and by their departments. The Department of Defense manages the military, the Department of State runs diplomatic efforts, and the US Agency for International Development (technically part of the State Department) coordinates international aid and development. Various agencies provide intelligence (the Central Intelligence Agency, National Security Agency, and Defense Intelligence Agency, among others). A National Security Council meeting includes typically around twenty officials, fewer than half of whom have a voice. This national security structure, primarily constructed in 1947, was organized for conventional war. Diplomats aim to avoid conflict or build a coalition to fight the war.1 Once a war is declared, military forces fight to win, lose, or draw. Diplomats return to the fore to negotiate peace. Then aid agencies move out to repair the damage.
The United States has amassed greater global reach and obligations since 1947, stretching this structure. The same small group of people manages nearly every national security crisis across the globe, in addition to persistent matters, such as space, cyber, climate change, nuclear weapons and materials, and the rise of China. They also have a domestic policy to manage and departments and agencies to run. The load is crushing. Even such incredibly capable people as US cabinet officials lack the bandwidth to attain expert knowledge on every issue, and their agencies may lack procedures for nonstandard crises. In their landmark study of the 1962 Cuban Missile Crisis, Graham Allison and Philip Zelikow explain how organizational processes and bureaucratic politics can result in frictions that impede decision-making and execution.2 The organizational process model explains how bureaucratic procedures can place limits on a policymaker’s freedom of action.3 Instead of examining a crisis holistically, governments tend to break down the problem along organizational lines. Each bureaucracy addresses its portion of the issue. Government agencies use their existing procedures to execute assigned tasks and peacetime routines. When applied during a crisis, they can result in delays, major oversights, or rigidity.
The bureaucratic politics model, on the other hand, explains decision-making as a product of politicking and negotiations among the government’s top leaders. These leaders have varying levels of power based upon their charisma, relationship to the president, and their interpersonal and persuasive skills. The intense discussions, often filled with miscommunication and misunderstandings, can result in a consensus that differs significantly from individual preferences. Conversely, individuals may take actions that the group would not condone.4
The organizational process and bureaucratic politics models explain why the US government may develop suboptimal intervention strategies that overemphasize the military instrument and make implicit assumptions about decisive victory. They also illustrate the reasons why losing or ineffective approaches could be difficult to change—the same decision-makers who agreed on the strategy have a vested interest in making it work. Leaders prefer to tinker on the margins rather than to overturn the existing strategy. New governing coalitions are often necessary to change an approach. Interest misalignment with the host nation and moral hazard can paralyze decision-making.
Political and bureaucratic frictions within the US government, the lack of a body of expert knowledge for wartime negotiations, and resistance from Karzai undermined reconciliation even as exploratory talks began. In late 2010, US officials started meeting periodically with Tayyab Agha. After a few sessions, the discussions centered on confidence-building measures. The Taliban wanted several prisoners released, particularly five senior leaders detained in the US military prison at Guantanamo (GTMO). They also wanted UN sanctions on them lifted and recognition of their political office in Doha. In turn, the United States wanted the Taliban to release Sergeant Bowe Bergdahl, a US soldier they captured in 2009, and for the Taliban to denounce international terrorism, announce support for a peace process, and begin meeting with the Afghan government.5 The Taliban said they would talk with the Afghan government at the end of the confidence-building process.6 The talks continued throughout 2011 and into 2012. US diplomats informed Karzai of the main points after each meeting.7
Grossman and I joined the talks in mid-2011 as the confidence-building measures were gaining definition. The London Conference in early 2011 had solidified Karzai’s call for December 2014 to be the end of the ISAF combat mission. Obama’s speech in June 2011 specified that the transition to ANSF-led security was to be complete by the end of 2014.8 The timeline announcements likely solidified the Taliban’s negotiating strategy. Having withstood the surge, they had no incentive to negotiate an end to the conflict until they could take on the Afghan government after international forces had left. They also had little incentive to make compromises that might cause tension within their ranks, particularly any actions that might confer legitimacy on the Afghan government.
Grossman played the hand he was dealt as well as he could. He focused on confidence-building measures to gain the Taliban’s consent to meet with the Afghan government. These steps included the opening of a Taliban political office in Doha and Taliban statements denouncing international terrorism and supporting a peace process, detainee releases (the GTMO-5 and Bergdahl), and a meeting with the Afghan government.9 A couple of obstacles remained. First, the Taliban needed to agree to terms of reference for the detainees sent to Doha, which included limitations on activities and a travel ban until the end of 2014. Second, they needed to agree to rules for the political office. The most important restriction was that the office not appear as an embassy or use the name “Islamic Emirate of Afghanistan.”
Tayyab Agha was soft-spoken, even-tempered, and highly pragmatic. His closely cropped hair and beard little resembled the 2001 pictures of him. He gained Mullah Omar’s trust as his secretary and, in 2009, became the head of the political commission. He described the Taliban’s views of the conflict in two dimensions: external and internal. The Taliban wanted to deal first with the external conflict (particularly with the United States), and then address the internal conflict.10 The external conflict meant the withdrawal of all foreign forces—the international coalition as well as al Qaeda and other foreign militant organizations that were fighting the government. The lifting of international sanctions, the establishment of the political office, and the transfer of the GTMO-5 were the confidence builders he said the Taliban needed to agree to have the Afghan government in talks. GTMO was a symbol of injustice to them.11 The five detainees they wanted to be released had either surrendered to the Northern Alliance or turned themselves in based on agreements that they would be free to live in peace. The GTMO transfers were their biggest test of American sincerity.
Some administration officials, including ones at State and Defense, viewed the Taliban participation in talks as insincere. They believed that if the Taliban wanted peace, they would first stop fighting. High-profile or large-scale attacks were cited as proof of Taliban deception, even as the United States continued night raids and operations against them. There was a mismatch in the view that the United States could fight and talk, but the Taliban could not. Some exchanges with very senior officials became intense over these issues. Such challenges illustrated the importance of taking the effort slowly and step-by-step to build political space on all sides. If US officials found the notion of talks with the Taliban distasteful, many Afghans had far stronger reactions. Skeptical officials demanded significant unilateral signals from the Taliban as proof of sincerity.
Intelligence officials and Taliban experts often described the Taliban as a decentralized, pragmatic, consensus-based organization.12 If they have an ideology, it is unity and the prevention of discord. Although Mullah Omar was the iconic leader of the movement, he did not rule by diktat. The Taliban use councils (shuras and jirgas) to discuss issues, examine ideas, and come to a consensus.13 In a traditional jirga system, decisions require unanimous approval. This approach lowers the risk of dispute but makes decision-making slow and conservative. The status quo bias in an organization like the Taliban tends to be quite high, so major unilateral concessions from the Taliban at that point were extremely unlikely.
Confidence-building measures consumed most of the US-Taliban meetings, which took place once every four to six weeks through the summer and fall of 2011. Unfortunately, no one took official minutes. Tayyab Agha and the Qatari intermediaries spoke excellent English, but it was clear to me from Agha’s body language that he would sometimes miss parts of the conversation. I would note to Grossman when I detected this, and he reiterated vital points, but the absence of an agreed record heightened the risk of misunderstandings. Nonetheless, texts containing the terms of reference for the Taliban political office went back and forth. These were decided by November 2011, except for some ambiguity about the office’s name. The Americans sent the terms of reference to the Afghan government in advance of the hoped-for office opening in early 2012.14
Tensions with Karzai over reconciliation were growing. The US officials provided Karzai the points of each meeting with the Taliban, but this was not creating ownership and buy-in.15 Unlike the carefully coordinated security transition effort, the United States had not developed an agreed approach with the Afghan government on reconciliation.16 Defense officials expressed concerns, but no changes occurred.17 “Karzai is irrelevant,” a senior SRAP official told me at the time. “His interests are so different from ours that there is no point in trying to discuss it.”18
Karzai grew increasingly worried that the United States was attempting to make a deal with the Taliban as a cover for withdrawal, just as America did in Vietnam.19 A Foreign Affairs article by former US diplomat Robert Blackwill argued for the soft partition of Afghanistan. Karzai viewed this article as reflecting an option being considered seriously by the Obama administration—ceding the south to the Taliban (and Pakistan) in exchange for a cease-fire and an end to the conflict.20 He perceived the efforts by the United States on reconciliation to be dangerously naive and potentially catastrophic.21 The lack of serious engagement and coordination with Karzai on an effort so central to the political order and future of Afghanistan would have grave consequences.
These problems exploded into controversy just before the beginning of the Bonn II Conference on December 5, 2011. The atmosphere in Kabul was tense, particularly after the High Peace Council chairman, Afghanistan’s former President Rabbani, was killed in September by a suicide bomber posing as a Taliban representative. Karzai’s political rival, Abdullah Abdullah, warned, “This is a lesson for all of us that we shouldn’t fool ourselves that this group, who has carried out so many crimes against the people of Afghanistan, are willing to make peace.”22 A concession to the Taliban was likely to create a backlash in Kabul.
The way the Taliban office discussions were unfolding amplified these concerns. After receiving the unsigned agreement on the Taliban office, Karzai requested a meeting with Secretary Clinton. Participants with knowledge of the exchange recalled that Karzai was outraged at being blindsided with a fait accompli.23 Knowing that the US military command and embassy in Kabul backed Karzai’s views on reconciliation and opposed SRAP’s probably gave Karzai the confidence to press the matter hard. He demanded suspension of the effort until he had the opportunity to review and comment on the document. He recalled his ambassador from Qatar a few days later.24 The United States reluctantly acceded to his demands.
Frictions increased in Washington as well. Sensitive to how negotiating with the Taliban would be perceived by Congress, the White House arranged meetings with House and Senate leaders in the late fall of 2011 and January 2012.25 An interagency team supported the briefings. Grossman outlined the different measures under consideration and explained that the entire focus of the effort was to arrange a meeting between the Afghan government and the Taliban. He noted that the chances of success were low and surmised that the assassination of Rabbani and the September 13, 2011, attack on the US embassy in Kabul might have reflected the Taliban’s true intentions about peace.26
His explanation came across to House and Senate leaders as a high-risk, low-reward proposition. The administration, they perceived, was having Grossman offer significant concessions to get the Taliban to agree to a meeting with the Afghan government. Why, they questioned, was the administration even considering such a bad deal? These concessions, they argued, would improve the Taliban’s legitimacy and capabilities while placing American soldiers at higher risk for no meaningful return. Those present for the briefings were many of the same congressional leaders that received testimony from Defense and State officials that everything was on track in Afghanistan and that the risks were manageable.27 The discussion was leaked to the press immediately. Members voiced strong opposition to the talks.28
Still, the administration sought to lay the foundations for a potential detainee exchange. Some observers mistakenly believe the Defense Department opposed the effort and tried to derail reconciliation talks.29 Defense officials did want to take smaller steps first because transferring detainees from GTMO had become highly politicized. Congress passed legislation in 2009 that the Secretary of Defense had to certify in writing to Congress that he had taken all measures necessary to ensure the transferred individual would no longer pose a national security threat. Congress maintained these requirements through 2014.30 In short, the Secretary of Defense would be personally accountable if a transferred detainee returned to the battlefield. The Obama administration often complained that such provisions made GTMO transfers virtually impossible.31 Prematurely moving forward on such a potentially explosive issue could create more obstacles, heighten suspicion and cynicism, and potentially undermine the entire effort.
The Principals Committee determined that the United States needed assurances from the Qatari government that the detainees would be monitored, not allowed to engage in acts against the United States or its allies, and not permitted to leave the country. A senior Defense official and I worked closely with the Qatari attorney general on the provisions, capturing them in writing over a series of meetings. After a few months, we agreed on the terms of reference and gained the approvals of Secretary of Defense Leon Panetta and Secretary of State Hillary Clinton. The Qataris, however, would enforce the provisions only if the Taliban agreed to them. They understandably wanted the Taliban to share the blame for any transgression. The Taliban, however, refused to accept the travel ban. When the Principals Committee discussed the matter in late 2011, their decision was unanimous and unequivocal—the transfers could not take place until the Taliban agreed to the travel ban.32
The lack of clear, authoritative language about negotiated outcomes, strategic incoherence, and poor coordination within the US government and with the Afghan government sorely undermined confidence in reconciliation. The damage was about to expand.