In May 2006 a unit of American soldiers in Afghanistan's Uruzgan valley were engulfed in a ferocious fire fight with the Taliban. Only after six hours, and supporting airstrikes, could they extricate themselves from the valley. But what was most revealing about the battle was the fact that many local farmers spontaneously joined in, rushing home to get their weapons. Asked later why they'd done so, the villagers claimed they didn't support the Taliban's ideological agenda, nor were they particularly hostile toward the Americans. But this battle was the most momentous thing that had happened in their valley for years. If as virile young men they had stood by and just watched, they would have been dishonored in their communities. And, of course, if they were going to fight, they could not fight alongside the foreigners.
In describing this battle, the Australian counterinsurgency expert David Kilcullen coins a term, "accidental guerilla," to describe the villagers. They had no grand transnational agenda, no dreams of global jihad. If anything, those young men were defending their local ways and customs from encroachment from outside. But a global terrorist group—with local ties—can find ways to turn these villagers into allies of a kind. And foreign forces, if they are not very careful, can easily turn them into enemies.
Reduced to its simplest level, the goal of American policy in Afghanistan should be to stop creating accidental guerrillas. It should make those villagers see U.S. forces as acting in their interests. That would mark a fundamental turnaround.
Let's be clear. The war in Afghanistan is not going well; almost all trends are moving in the wrong direction. But I don't believe it is a quagmire—yet. We still have time to focus our goals, improve our strategy, calibrate our means. The two men in charge now, Gen. David Petraeus and Ambassador Richard Holbrooke, are extraordinarily talented. But what should they do? We need to overhaul U.S. policy in four steps, each more complicated than the last.
Do counterinsurgency right. Despite Petraeus's demonstrable success in Iraq, U.S. forces have to this point largely relied on more old-fashioned tactics—raids, search-and-destroy missions, air attacks. Partly this is because the U.S. military has deployed too few troops to hold territory that's been cleared. "In Afghanistan, we do what we can. In Iraq we do what we must," Joint Chiefs of Staff Chairman Adm. Mike Mullen explained in 2007. It is also because many American troops believe that they are chasing global terrorists who must be captured or killed urgently.
Instead of aggressive and punitive—in military parlance, "kinetic"—operations, Petraeus's counterinsurgency approach emphasizes the need to make local populations feel secure. Troops are meant to live among the people, use less force, gain trust, not overreact to every provocation and be seen as a positive force within the community. Above all, the priority is to get local forces—in this case, the Afghan National Army and the police—to do as much as possible, even when the job might not be done as well as by foreign troops.
The number of additional U.S. troops needed is not large. Afghanistan is predominantly rural, and the large population centers that truly need protection are limited. U.S. forces would also need to control the key roads and transit points. In fact, the commander in Afghanistan, Gen. David McKiernan, has begun to focus his efforts on this approach. Between the addition of two to four more American brigades and a ramp-up of the Afghan Army, there should be enough troops to execute the strategy.
Strengthen the Afghan government. The central government is widely seen as weak, dysfunctional and utterly corrupt. Disgust with its performance has reopened the door for the Taliban, who are unpopular almost everywhere but who promise justice—albeit very rough justice—rather than the chaos of the Karzai reign. The international community should have considerable influence on this matter because the Kabul government, unlike in Iraq, has virtually no revenue sources other than foreign aid. Unfortunately, so far many of the most corrupt elements in government are allies of the West and have gained a kind of immunity as a result.
The most immediate way to enhance the legitimacy of the Afghan government would be to ensure that both presidential and local elections take place this year without disruption, and that viable alternative candidates are free to campaign. But elections are only one form of political legitimacy in a country like Afghanistan. There should be a much more broad-based effort to reach out to tribal leaders, hold local councils and build a more-diverse base of support. The goal in Afghanistan should not be a strong central government—the country is decentralized in its DNA—but a legitimate government with credibility and local allies throughout the country. This is how Afghanistan was ruled before the wars that have consumed it since the 1980s.
Talk to the Taliban. The single most important consequence of the surge in Iraq was the fact that large parts of the Sunni community—including insurgents who had been attacking U.S. troops for years—reconciled with America and, provisionally, the Baghdad government. "The challenge in Afghanistan," Petraeus said in a recent interview with Foreign Policy, "is figuring out how to create conditions that enable reconciliation, recognizing that these will likely be different somewhat from those created in Iraq."
Timing is important. Petraeus argues that in Iraq, reconciliation became easier once the United States had regained a position of strength, having killed or captured many Sunni fighters. (And after many more were savagely killed by Shiite militias.) But the basic idea is obvious—to divide the enemy and thereby reduce the number of diehard opponents arrayed against you. The process of political bargaining goes on in every society during such conflicts. The goal in Afghanistan must be to separate, as often as possible, the global jihadist from the accidental guerrilla.
In America, this has turned into a somewhat ideological debate about "talking to the Taliban." Critics rage that this would be doing business with evil people. But in a country like Afghanistan—one of the poorest in the world—politics is often less about ideology and more about a share of the spoils. While some members of the Taliban are hard-core Islamic extremists, others are concerned with gaining a measure of local power—of access to money and clout.
The most important departure from current thinking would be to make a distinction between Al Qaeda and the Taliban. The United States is properly and unalterably opposed to Al Qaeda—on strategic, political and moral grounds—because its raison d'être is to inflict brutality on the civilized world. We have significant differences with the Taliban on many issues—democracy and the treatment of women being the most serious. But we do not wage war on other Islamist groups with which we similarly disagree (the Saudi monarchy, for example). Were elements of the Taliban to abandon Al Qaeda, we would not have a pressing national-security interest in waging war against them.
In fact, there is a powerful military advantage to moving in this direction. Al Qaeda is a stateless organization that controls no territory of its own. It can survive and thrive only with a host community. Our objective should be to cut off Al Qaeda, as far as possible, from its allies in Afghanistan and Pakistan. Deprived of local support, Al Qaeda would be a much diminished threat. Now, it is certainly true that some elements of the Taliban might be closely wedded to Al Qaeda. But others are not. Even the most hard-line Taliban—the so-called Quetta Shura led by Mullah Omar—have at various points made overtures to the Afghan government, always asking that they be distinguished from Al Qaeda. In Guantánamo, for example, Afghans who had played minor roles as drivers and servants for Qaeda officials have been treated just the same as 9/11 mastermind Khalid Sheikh Mohammed.
Given that the United States is in its seventh year of war in Afghanistan, it might surprise many Americans to recognize that not one Afghan was involved at any significant level in the 9/11 attacks. Barnett Rubin, who has studied the region for decades and is chairing an Asia Society report on Afghanistan, makes the point more forcefully: "Afghans have played no significant role in any major terrorist attack before or after 9/11." This is true. All the plots that have been traced back to the region lead not to Afghanistan but to Pakistan, where U.S. officials acknowledge the top leadership of both the Taliban and Al Qaeda now reside.
Pressure Pakistan. When the United States invaded Afghanistan, it did not defeat Al Qaeda and its supporters among the Taliban. They simply fled to Pakistan, their original home. The story is by now familiar. During the 1980s, the Pakistani military—through its Inter-Services Intelligence agency—helped form militant Islamic groups to wage asymmetrical war against Soviet-occupied Afghanistan and against India in Kashmir. This strategy was financed by Saudi Arabia and the United States. It gave birth to the Taliban and helped provide Al Qaeda with a home when Osama bin Laden was expelled from Sudan.
It is crucial to recognize that the Pakistani military achieved substantial success with these militias. They bled India at very low cost, neutralizing New Delhi's much larger army, and chased the Soviet Union out of Afghanistan. These represent the only two significant strategic successes for the Pakistani military in decades, perhaps in its history.
The American debate on the need to "press" Pakistan to dismantle these militias misses this point. Pakistan has long viewed its clients as having given the country "strategic depth"—keeping its historic foes, India and Afghanistan, off balance. For Islamabad to genuinely renounce these groups would require a fundamental strategic rethinking within the Pakistani military.
This is hard but not impossible. The civilian government in Pakistan, weak and ineffective though it may be, is allied with the international community on these issues. It too wants a Pakistani military that knows its boundaries, does not run militant groups and conceives of the country's national interests in less-confrontational terms. The United States has enormous influence with the Pakistani Army, though it has not always used it well. (When we cut off military-to-military relations in the 1990s, because of congressional sanctions against Pakistan's nuclear tests, we lost a generation of officers who felt betrayed by America.) If the military agrees to dismantle these jihadist networks—demonstrably—Afghanistan and India should respond with concessions to ease regional tensions. I don't want to make this sound easy. Of all the tasks that Petraeus and Holbrooke have, this one is the hardest. And yet, if the problem with Pakistan cannot be solved, the war in Afghanistan cannot be won.
Afghanistan is a complex problem, and progress will be slow and limited. But we need to stabilize the situation, not magically transform one of the poorest, most war-torn countries in the world in the next few years. It will help immeasurably if we keep in mind the basic objective of U.S. policy there. "My own personal view is that our primary goal is to prevent Afghanistan from being used as a base for terrorists and extremists to attack the United States and its allies," said Secretary of Defense Robert Gates last week. That is an admirably clear statement.
It is not that we don't have other goals—education, female literacy, centralized control of government services, drug eradication, liberal democracy. But many of them are objectives that will be realized over very long stretches of time, and should not be measured as part of military campaigns or political cycles. They are also goals that are not best achieved by military force. The U.S. Army is being asked to do enough as it is in Afghanistan. Helping it stay focused on a core mission is neither cramped nor defeatist. It is a realistic plan for success.