When the war on terrorism began, Secretary of Defense Donald Rumsfeld promised that it would be a war like no other. The first battles in Afghanistan seemed to confirm this. An ingenious and now famous combination of modern technology and traditional tribal skills—a cavalry charge with global communications and precision guided munitions—brought the Taliban regime to an end.

Although these hybrid tactics and the organization that implemented them were new, in another sense, the war in Afghanistan was just like every other war the U.S. military has fought. It was a question of finding the enemy and going at him. This was a good strategy for dealing with the Taliban, which was a government and had an organized military force defending it. We had superior force to engage and destroy that military organization, and its destruction would lead to the fall of the Taliban.

Since the fall of the Taliban, the strategy of finding the enemy and killing him has been less successful. The enemy is hiding and does not distinguish himself, as an organized military does, from the civilian population. In this situation, a strategy of finding the enemy and killing him often leads to the deaths of civilians. This in turn builds resentment among the civilian population, makes them less likely to help us, and makes our presence more difficult to maintain. This has been what has been happening in Afghanistan over the last several months. Blowing up a wedding party from the air, because the revelers are unexpectedly firing guns, may be legitimate self-defense but whether legitimate or not, such incidents do not improve our position in Afghanistan nor do they bring us closer to the goal of destroying those who threaten the new government there.

What is the alternative? Instead of driving for head-on conflicts with an elusive enemy, we could try to outmaneuver him. The Special Forces who took high-tech weaponry into battle on horseback train to live and work among civilian populations, gaining their trust and cooperation. This is precisely what they did in northern Afghanistan early in the war, before launching their attacks. The tribes would not have joined them in the fight, providing critical local knowledge and risking their lives, if they had not come to trust the Special Forces. Today in Afghanistan, these forces could operate in the same way, gaining the trust of local populations. In doing so, they would gain intelligence on the enemy, allowing our targeting to be more discriminate. More important, cut off from information and supplies, the enemy would find it harder and harder to operate. In time, he would find himself so outmaneuvered that he would no longer be able to operate effectively.

Certainly, a strategy of maneuver would be difficult to implement in Afghanistan. But over time, it would be more likely to lead to success there than the strategy we are currently following.

The same can be said for the wider, global war on terrorism. In this case, too, we often appear to be following a strategy of locating the enemy and killing or capturing him. We have had some success with this strategy. Almost every week brings news of some arrest of suspected al Qaeda operatives or those who sympathize with them. Recently, such reports have come from the Netherlands, Germany and Detroit. In the long run, however, we should wonder whether this strategy will prevail. There are at least two reasons to have doubts. First, a significant number of these arrests are of people previously suspected by the police, whose arrest now has been made possible apparently because of increased urgency following the attack on September 11 or because of information from the interrogation of prisoners from Afghanistan or other perishable sources. Over time, all the already suspected will have been picked up and the detainees will have given us all the information they have. Where will the new intelligence come from? Second, and more important, the strategy of finding and killing or arresting the enemy is a strategy of attrition. Its success depends on our ability to find and kill or arrest the enemy faster than the enemy can put new troops or operatives into the field. Even in Afghanistan, our direct head-on confrontational approach may be or may soon be creating enemies faster than we can dispose of them. In the global war on terrorism, this is even more likely to be the case. We are seeing cases, such as recently in Germany, of people not formally affiliated with al Qaeda or trained by this organization adopting its war aims and preparing to take action.

As maneuver is the alternative to attrition in Afghanistan, so is it in the wider war on terrorism. What would such a strategy of global maneuver look like? On analogy with the situation in Afghanistan, in the global war on terrorism, the local populations we must work with, gaining their trust and cooperation, are a number of the other nations of the world and their populations. Particularly important in this regard are the Muslim nations of the world and their populations. Without their assistance, we cannot effectively fight the global war on terrorism. With it, we will over time cut off our enemies from their intelligence and resources, outmaneuvering them and, ultimately, making their operations ineffective.

In some cases, an important basis of cooperation with other governments will be shared principles. In all cases, appealing to national interest will be important. Also, we should not forget that although there are some with whom cooperation will not be possible, there are others favorably disposed to us, whom we should try to reach. One recent survey found that in nine Muslim countries stretching from the Middle East to Southeast Asia, 68% of urban, relatively well-educated Muslims aged 15 to 24 viewed the United States favorably, a higher rating than any other country in the world.

When devising a strategy of maneuver for the global war on terrorism, we will have to take into account the peculiar character of each nation and our relations with it and the secondary and tertiary effects of our actions on other nations, our allies and ourselves. This more complex and comprehensive strategy might contain such elements as the following:

• We should announce, as part of our effort to overthrow Saddam, that we will withdraw our troops from Saudi Arabia when he and his weapons of mass destruction are gone, assuming that he is not replaced by someone as bad or worse. We kept the troops in Saudi Arabia after the Gulf War only because of the threat Saddam posed. When the cause disappears, so should the effect.
• We import relatively little oil from the Middle East but the world depends on this supply and to the extent that we want to deal with the problem of terrorism with a global reach, we cannot isolate ourselves from Middle East oil. We might, on the other hand, consider isolating parts of the world from Middle East oil. For example, we might work with the other industrialized countries to develop a long-term strategy to replace oil as much as possible as our fuel of choice. Perhaps we could start this with just Japan and Europe. This would help take care of our Kyoto problem—resentment over what the Europeans and Japanese consider our indifference to global warming—and improve our relations with Japan and Europe.
• Al Qaeda uses lawless or outlaw areas of the globe to hide in and operate from. For example, having been thrashed in Afghanistan, some of its members have apparently moved into Kashmir, control of which is disputed by Pakistan and India and held effectively by neither. Resolving this problem, if it can be resolved, will require work with both India and Pakistan and not just when they are threatening to launch their nuclear missiles at each other. Similarly, handling potential terrorist safe havens in central Asia (the former Soviet Union) will require concerted cooperation with Russia and China.