was a concession to anti-American sentiment within Pakistan, including fears that CIA agents were slipping into the country en masse. The cost to civilian assistance efforts was considerable. Often, State Department officials simply couldn’t get into the country. In one case, a day before I was scheduled to depart for a trip to Islamabad, I learned that my months-old visa request was still languishing. As was usually the case with Pakistan, the answer didn’t lie with the civilians. Instead, I secured a meeting with the military attaché, Lieutenant General Nazir Ahmed Butt. We met in his large office on the fourth floor of Pakistan’s embassy, with a view of China’s across the street. Butt, in full uniform with three stars on his collar, was distinguished looking, with a graying chevron moustache and, unusual for a Pakistani, electric blue eyes. He leaned back and listened intently as an assistant poured tea out of a china pot dusted with pink flowers and I spoke about the importance of working with Pakistani civil society, trying my best. An hour later, I walked out of the embassy with a multiple entry visa, good for a year. Not everyone was so lucky. At any given time, hundreds of applications were pending, requiring direct clearance by the Pakistani military or intelligence operatives.
The situation finally became so problematic that Hillary Clinton raised it with Pakistani prime minister Yousuf Raza Gilani. Gilani quietly authorized Husain Haqqani to begin approving visas without going through Islamabad—making him, as Haqqani put it, “visa czar.” Over the course of the following year, he approved a wave of American visa requests, keeping the relationship from tilting into hostilities. He was, in his view, “papering over a lot of problems between Pakistan and the US.” Haqqani knew his efforts would draw suspicion from the Pakistani political establishment. For him, as was the case with Robin Raphel’s misunderstood diplomatic endeavors, talking to the other side was about to become a dangerous game.
UNABLE TO MOVE THE NEEDLE away from what he called “mil-think,” Richard Holbrooke set to work around its margins. He still felt that any hope for success depended on broadening America’s role, in both Afghanistan and Pakistan, beyond tactics.
On the Afghan side of the border, he proposed a flood of new civilian-led assistance. Prompted by his agitation, the Obama administration requested from Congress $800 million more for reconstruction in 2009 than the Bush administration had the year before. Holbrooke commandeered control of USAID projects, insisting on signing off on many of them personally. He was able to secure that control because USAID reports to the State Department. His outsize influence was a source of bureaucratic rancor—especially when Holbrooke, always a whirling dervish of activity, would leave projects awaiting approval for months, unwilling to relinquish control. But he considered the move necessary. Afghanistan was full of expensive, embarrassing USAID boondoggles—from cobblestone roads that Afghans considered unusable as they hurt camels’ feet, to farming projects on land with groundwater too salty to sustain crops, to fertilizer handouts that inadvertently enhanced poppy cultivation and, in turn, Afghanistan’s drug economy. When Holbrooke was in Vietnam, USAID had a robust corps of technical specialists in areas like agriculture. By the Obama administration, decades of budget cuts had shrunk the size of the workforce and robbed it of such expertise. The funding USAID did receive was often mismanaged and misspent, with projects going to American mega-contractors with high overhead and little understanding of circumstances on the ground. This was one of the symptoms of the imbalance that had bedeviled Holbrooke throughout his career. By the modern war on terror, almost all of the capacity and resources lay instead on the military side.
Holbrooke was convinced that the key was agriculture. The US military, which led many of the counternarcotics efforts in the region, had long contended that lucrative poppy cultivation for heroin sustained the Taliban. So the Bush administration had focused on crop eradication, slashing and burning its way through Afghanistan’s fields. Holbrooke was incensed by this. He pointed to intelligence assessments that showed support from Pakistan and the Gulf States was far more central to the Taliban’s livelihood. He argued that eradication pushed penniless farmers into the arms of the Taliban—often their only source of employment after their crops were wiped out.
He set out to refocus the United States on supporting Afghan farmers. “They need the kind of soup-to-nuts agricultural support that Roosevelt gave farmers during the great depression,” he said. He was a man possessed. Pomegranates, once a lucrative export for the Afghans, were a particular obsession. At Holbrooke’s request, I organized dozens of meetings focused on the fruit. Sometimes he’d cut me off in the middle of an unrelated sentence and say, from a faraway place, “Where are we on the pomegranates?” By the end of his first year on the job, Richard Holbrooke, a man who as far as I was aware had never so much as kept a potted cactus alive, could explain the pomegranate’s required levels of moisture, favorable types of soil, and ideal timeline for harvest. Hillary Clinton took to calling him Farmer Holbrooke.
But, despite Holbrooke’s efforts, civilian reconstruction remained dwarfed by an order of magnitude by the Pentagon’s programs in the same space. In the early years of George W. Bush’s administration, State reconstruction spending sometimes outweighed the Pentagon’s by a ratio of more than ten to one. By the time Holbrooke came to State, the situation had nearly reversed. The trend lines were hard to miss: From 2008 to 2010, State spending on reconstruction in Afghanistan jumped from $2.2 billion to $4.2 billion, while the Pentagon’s budget for similar efforts more than tripled from $3.4 billion to $10.4 billion. This included a sea of development projects conventionally associated with State and USAID, ranging from counternarcotics programs, to education, to the catchall Commander’s Emergency Response Program, which was primarily used for road building and repair. The Army Corps of Engineers, likewise, worked on infrastructure projects around the country, and USAID was sometimes the last to know.
Even for the projects underwritten through the new USAID and State Department funding, Holbrooke had trouble disentangling development and military objectives. Counterinsurgency strategy was typically described in three steps: “clear, hold, build”—as in, clear the area of the enemy, hold it with our forces, and begin to build capacity. As the first year of the Obama administration wore on, that COIN language, ripped from Petraeus’s counterinsurgency manual, began appearing in USAID development contracts. One request for applications for a community-based development initiative called for USAID’s partner charities to “enable COIN-focused, unstable communities to directly implement small-scale community level projects” and “support military … efforts in communities by helping to ‘hold’ areas after they are cleared.”
Security and development objectives in an active theater of war are never completely separable, but there had, historically, been an acknowledgment that development should be driven by technical expertise and long term goals, not shackled to tactics. Explicitly militarizing the contract language was new—and, it turned out, tone-deaf. The nongovernmental organizations applying for the contracts revolted. The head of one charity told me its staffers had been targeted for attacks based on their visible identification with the military. Others said it was destroying trust with communities of Afghans who were comfortable with American reconstruction, but not American military might.
Holbrooke correctly judged that the years of narrow, military-driven engagement under the Bush administration had also atrophied relationships with civil society, especially at a local level. Large American companies took equally large commissions, then subcontracted to other groups, which in turn sometimes subcontracted yet again. The result was obvious: