In testimony last week before Congress, the American ambassador to Afghanistan, Karl Eikenberry, insisted that President Obama's revised war strategy will "build support for the Afghan government," while Gen. Stanley McChrystal, the top American commander there, vowed that it will "absolutely" succeed in disrupting and degrading the Taliban.
Confidence is important, but we also have to recognize that the decision to commit 30,000 more troops to a counterinsurgency effort against a good segment of the Afghan population, with the focus on converting a deeply unpopular and corrupt regime into a unified, centralized state for the first time in that country's history, is far from a slam dunk. In the worst case, the surge may push General McChrystal's "core goal of defeating Al Qaeda" further away.
Al Qaeda is already on the ropes globally, with ever-dwindling financial and popular support, and a drastically diminished ability to work with other extremists worldwide, much less command them in major operations. Its lethal agents are being systematically hunted down, while those Muslims whose souls it seeks to save are increasingly revolted by its methods.
Unfortunately, this weakening viral movement may have a new lease on life in Afghanistan and Pakistan because we are pushing the Taliban into its arms. By overestimating the threat from Al Qaeda in Afghanistan, we are making it a greater threat to Pakistan and the world. Afghanistan and the tribal areas of Pakistan are unlike Iraq, the ancient birthplace of central government, or 1960s Vietnam, where a strong state was backing the Communist insurgents. Afghanistan and Pakistan must be dealt with on their own terms.
We're winning against Al Qaeda and its kin in places where antiterrorism efforts are local and built on an understanding that the ties binding terrorist networks today are more cultural and familial than political. Consider recent events in Southeast Asia.
In September, Indonesian security forces killed Noordin Muhammad Top, then on the F.B.I.'s most-wanted terrorist list. Implicated in the region's worst suicide bombings -- including the JW Marriott and Ritz-Carlton bombings in Jakarta last July 17 -- Noordin Top headed a splinter group of the extremist religious organization Jemaah Islamiyah (he called it Al Qaeda for the Malaysian Archipelago). Research by my colleagues and me, supported by the National Science Foundation and the Defense Department, reveals three critical factors in such groups inspired by Al Qaeda, all of which local security forces implicitly grasp but American counterintelligence workers seem to underestimate.
What binds these groups together? First is friendship forged through fighting: the Indonesian volunteers who fought the Soviet Union in Afghanistan styled themselves the Afghan Alumni, and many kept in contact when they returned home after the war. The second is school ties and discipleship: many leading operatives in Southeast Asia come from a handful of religious schools affiliated with Jemaah Islamiyah. Out of some 30,000 religious schools in Indonesia, only about 50 have a deadly legacy of producing violent extremists. Third is family ties; as anyone who has watched the opening scene from "The Godfather" knows, weddings can be terrific opportunities for networking and plotting.
Understanding these three aspects of terrorist networking has given law enforcement a leg up on the jihadists. Gen. Tito Karnavian, the leader of the strike team that tracked down Noordin Top, told me that "knowledge of the interconnected networks of Afghan Alumni, kinship and marriage groups was very crucial to uncovering the inner circle of Noordin."
Consider Noordin Top's third marriage, which cemented ties to key suspects in the lead-up to the recent hotel bombings. His father-in-law, who founded a Jemaah Islamiyah-related boarding school, stashed explosives in his garden with the aid of another teacher at the school. Using electronic intercepts and tracing family, school and alumni ties, police officers found the cache in late June 2009. That discovery may have prompted Noordin Top to initiate the hotel attacks ahead of a planned simultaneous attack on the residence of President Susilo Bambang Yudhoyono.
In addition, an Afghan Alumnus and nephew of Noordin Top's father-in-law was being pursued by the police for his role in a failed plot to blow up a tourist cafe on Sumatra. Unfortunately, Noordin Top struck the hotels before the Indonesian police could penetrate the entire network, in part because another family group was still operating under the police radar. This group included a florist who smuggled the bombs into the hotels and a man whose eventual arrest led to discovery of the plot against the president. Both terrorists were married to sisters of a Yemeni-trained imam who recruited the hotel suicide bombers, and of another brother who had infiltrated Indonesia's national airline.
Had the police pulled harder on the pieces of social yarn they had in hand, they might have unraveled the hotel plot earlier. Still, their work thwarted attacks planned for the future, including that on the president.
Similarly, security officials in the Philippines have combined intelligence from American and Australian sources with similar tracking efforts to crack down on their terrorist networks, and as a result most extremist groups are either seeking reconciliation with the government -- including the deadly Moro Islamic Liberation Front on the island of Mindanao -- or have devolved into kidnapping-and-extortion gangs with no ideological focus. The separatist Abu Sayyaf Group, once the most feared force in the region, now has no overall spiritual or military leaders, few weapons and only a hundred or so fighters.
So, how does this relate to a strategy against Al Qaeda in the West and in Afghanistan and Pakistan? Al Qaeda's main focus is harming the United States and Europe, but there hasn't been a successful attack in these places directly commanded by Osama bin Laden and company since 9/11. The American invasion of Afghanistan devastated Al Qaeda's core of top personnel and its training camps. In a recent briefing to the Senate Foreign Relations Committee, Marc Sageman, a former C.I.A. case officer, said that recent history "refutes claims by some heads of the intelligence community that all Islamist plots in the West can be traced back to the Afghan-Pakistani border." The real threat is homegrown youths who gain inspiration from Osama bin Laden but little else beyond an occasional self-financed spell at a degraded Qaeda-linked training facility.
The 2003 invasion of Iraq encouraged many of these local plots, including the train bombings in Madrid in 2004 and London in 2005. In their aftermaths, European law and security forces stopped plots from coming to fruition by stepping up coordination and tracking links among local extremists, their friends and friends of friends, while also improving relations with young Muslim immigrants through community outreach. Morocco, Saudi Arabia and Turkey have taken similar steps.
Now we need to bring this perspective to Afghanistan and Pakistan -- one that is smart about cultures, customs and connections. The present policy of focusing on troop strength and drones, and trying to win over people by improving their lives with Western-style aid programs, only continues a long history of foreign involvement and failure. Reading a thousand years of Arab and Muslim history would show little in the way of patterns that would have helped to predict 9/11, but our predicament in Afghanistan rhymes with the past like a limerick.
A key factor helping the Taliban is the moral outrage of the Pashtun tribes against those who deny them autonomy, including a right to bear arms to defend their tribal code, known as Pashtunwali. Its sacred tenets include protecting women's purity (namus), the right to personal revenge (badal), the sanctity of the guest (melmastia) and sanctuary (nanawateh). Among all Pashtun tribes, inheritance, wealth, social prestige and political status accrue through the father's line.
This social structure means that there can be no suspicion that the male pedigree (often traceable in lineages spanning centuries) is "corrupted" by doubtful paternity. Thus, revenge for sexual misbehavior (rape, adultery, abduction) warrants killing seven members of the offending group and often the "offending" woman. Yet hospitality trumps vengeance: if a group accepts a guest, all must honor him, even if prior grounds justify revenge. That's one reason American offers of millions for betraying Osama bin Laden fail.
Afghan hill societies have withstood centuries of would-be conquests by keeping order with Pashtunwali in the absence of central authority. When seemingly intractable conflicts arise, rival parties convene councils, or jirgas, of elders and third parties to seek solutions through consensus.
After 9/11, the Taliban leader, Mullah Omar, assembled a council of clerics to judge his claim that Mr. bin Laden was the country's guest and could not be surrendered. The clerics countered that because a guest should not cause his host problems, Mr. bin Laden should leave. But instead of keeping pressure on the Taliban to resolve the issue in ways they could live with, the United States ridiculed their deliberation and bombed them into a closer alliance with Al Qaeda. Pakistani Pashtuns then offered to help out their Afghan brethren.
American-sponsored "reconciliation" efforts between the Afghan government and the Taliban may be fatally flawed if they include demands that Pashtun hill tribes give up their arms and support a Constitution that values Western-inspired rights and judicial institutions over traditions that have sustained the tribes against all enemies.
The secretary of state, Hillary Clinton, and the special envoy to the region, Richard Holbrooke, suggest that victory in Afghanistan is possible if the Taliban who pursue self-interest rather than ideology can be co-opted with material incentives. But as the veteran war reporter Jason Burke of The Observer of London told me: "Today, the logical thing for the Pashtun conservatives is to stop fighting and get rich through narcotics or Western aid, the latter being much lower risk. But many won't sell out."
Why? In part because outsiders who ignore local group dynamics tend to ride roughshod over values they don't grasp. My research with colleagues on group conflict in India, Indonesia, Iran, Morocco, Pakistan and the Palestinian territories found that helping to improve lives materially does little to reduce support for violence, and can even increase it if people feel such help compromises their most cherished values.
The original alliance between the Taliban and Al Qaeda was largely one of convenience between a poverty-stricken national movement and a transnational cause that brought it material help. American pressure on Pakistan to attack the Taliban and Al Qaeda in their sanctuary gave birth to the Pakistani Taliban, who forged their own ties to Al Qaeda to fight the Pakistani state.
While some Taliban groups use the rhetoric of global jihad to inspire ranks or enlist foreign fighters, the Pakistani Taliban show no inclination to go after Western interests abroad. Their attacks, which have included at least three assaults near nuclear facilities, warrant concerted action -- but in Pakistan, not in Afghanistan. As Mr. Sageman, the former C.I.A. officer, puts it: "There's no Qaeda in Afghanistan and no Afghans in Qaeda."
Pakistan has long preferred a policy of "respect for the independence and sentiment of the tribes" that was advised in 1908 by Lord Curzon, the British viceroy of India who established the North-West Frontier Province as a buffer zone to "conciliate and contain" the Pashtun hill tribes. In 1948, Pakistan's founder, Muhammad Ali Jinnah, removed all troops from brigade level up in Waziristan and other tribal areas in a plan aptly called Operation Curzon.
The problem today is that Al Qaeda is prodding the Pakistani Taliban to hit state institutions in the hopes of provoking a full-scale invasion of the tribal areas by the Pakistani Army; the idea is that such an assault would rally the tribes to Al Qaeda's cause and threaten the state. The United States has been pushing for exactly that sort of potentially disastrous action by Islamabad. But holding to Curzon's line may still be Pakistan's best bet. The key in the Afghan-Pakistani area, as in Southeast Asia, is to use local customs and networks to our advantage. Of course, counterterrorism measures are only as effective as local governments that execute them. Afghanistan's government is corrupt, unpopular and inept.
Besides, there's really no Taliban central authority to talk to. To be Taliban today means little more than to be a Pashtun tribesman who believes that his fundamental beliefs and customary way of life are threatened. Although most Taliban claim loyalty to Afghanistan's Mullah Omar, this allegiance varies greatly. Many Pakistani Taliban leaders -- including Baitullah Mehsud, who was killed by an American drone in August, and his successor, Hakimullah Mehsud -- rejected Mullah Omar's call to forgo suicide bombings against Pakistani civilians.
In fact, it is the United States that holds today's Taliban together. Without us, their deeply divided coalition could well fragment. Taliban resurgence depends on support from those notoriously unruly hill tribes in Pakistan's border regions, who are unsympathetic to the original Taliban program of homogenizing tribal custom and politics under one rule.
It wouldn't be surprising if the Taliban were to sever ties to Mr. bin Laden if he became a bigger headache to them than America. Al Qaeda may have close relations to the network of Jalaluddin Haqqani, an Afghan Taliban leader living in Pakistan, and the Shabi Khel branch of the Mehsud tribe in Waziristan, but it isn't wildly popular with many other Taliban factions and forces.
Unlike Al Qaeda, the Taliban are interested in their homeland, not ours. Things are different now than before 9/11. The Taliban know how costly Osama bin Laden's friendship can be. There's a good chance that enough factions in the loose Taliban coalition would opt to disinvite their troublesome guest if we forget about trying to subdue them or hold their territory. This would unwind the Taliban coalition into a lot of straggling, loosely networked groups that could be eliminated or contained using the lessons learned in Indonesia and elsewhere. This means tracking down family and tribal networks, gaining a better understanding of family ties and intervening only when we see actions by Taliban and other groups to aid Al Qaeda or act outside their region.
To defeat violent extremism in Afghanistan, less may be more -- just as it has been elsewhere in Asia.
This article originally appeared in the New York Times.
Scott Atran, an anthropologist at the National Center for Scientific Research in Paris, John Jay College and the University of Michigan, is the author of the forthcoming Listen to the Devil.