From all I saw in Afghanistan last month, everything indicates that the "clear, hold, build" strategy now employed by the military is not the solution. In the past two months, 20,000 troops, mobilized in the most important operation ever undertaken in Afghanistan, have failed in their attempt to clear the central part of the Helmand province. How could more resources turn the tide, when the plan is so clearly not working? Astronomical costs aside, it is important to reaffirm that even 200,000 troops would not be sufficient to take the rural districts back from the Taliban and seal the border with Pakistan.

On Monday, reporters sparked a fury when they seized on remarks from the top US general in Afghanistan, claiming that the war there would be lost without more troops. In his sixty-six-page assessment of the war, Gen. Stanley McChrystal claimed the United States would need more boots on the ground in Afghanistan to pursue a concerted counterinsurgency strategy. But watch the hands, not the mouth: only a day before, President Obama appeared on several major news shows, expressing skepticism that sending more soldiers into the country would accomplish any concrete objectives or make Americans safer.

"I'm not interested in just being in Afghanistan for the sake of being in Afghanistan or saving face or, in some way - you know, sending a message that America is here for the duration," the president said on Meet the Press. "We're not going to put the cart before the horse and just think by sending more troops we're automatically going to make Americans safe," he remarked on Face the Nation. And on CNN's State of the Union, he added, "The first question is, are we doing the right thing? Are we pursuing the right strategy?"

So what exactly is the problem? One key element of the new thinking is the relationship between the Taliban and the population. The standard opinion is that the insurgents are terrorizing the population and that their level of support is marginal. Consequently, the objective is to "separate the Taliban from the population" or "protect the population" from the Taliban. Yet, at this stage of the war in the Pashtun belt, there is no practical way to separate the insurgency from the population.

In fact, the U.S. approach reflects a misunderstanding about who the Taliban are: Afghan natives. Even if it were possible to find cases where the Taliban were mostly foreigners and were undeniably oppressing the villagers, the overall situation in the Pashtun belt is much more complex.

The Taliban have successfully exploited local grievances against corrupt national officials and the behavior of the foreign forces, framing them as a jihad. Moreover, the Taliban are generally careful not to antagonize the population. They are much more tolerant of music and beardless men than they were before 2001, and Mullah Omar has repeatedly made clear that fighters should show respect for the people, paying for the food they take, for instance. In a land of mixed religious and nationalist feelings, local solidarities tend to work in favor of the Taliban and against foreigners, who remain extremely unpopular in the Pashtun belt, especially when fighting occurs. This political context is a key driver of the insurgency.

How does the coalition control the (supposedly) cleared areas? There is no trust between the coalition and the Afghan population-especially the Pashtuns-and after eight years in the country, it has definitely lost the battle for hearts and minds. The coalition forces simply don't know how to be accepted locally: patrolling the villages is useless, and the linguistic and cultural barriers are de facto insurmountable when the average soldier's stay in the country is no more than six months. The behavior of the coalition forces has also not been beyond reproach, and has included cultural insensitivity, heavy-handed searching of houses, aggressive behavior on the roads, arbitrary imprisonment, beating of prisoners and of course the inadvertent bombing of civilians.

In addition, there is no state structure to speak of in the Pashtun belt. The military operations there are foreign alone, including no more than token Afghan National Army forces. No Afghan forces can effectively take charge of secured areas after the "clear" phase, as they are nowhere near numerous or well-trained enough, and the police are often corrupt or inefficient. In addition, the pro-government tribes or communities that are present in a few districts cannot venture outside their areas without great difficulty.

The supposed "ink spot" strategy-whereby the coalition establishes control in a key part of a province and security radiates outward-is not working, because of the social and ethnic fragmentation. Stability in one district doesn't necessarily bleed over into the neighboring one, since groups and villages are often antagonistic to one another, and compete for the resources provided by the war economy. In this context, to secure an area means essentially to stay there indefinitely, under constant attack by the insurgency. Even if only 20 percent of a village sympathizes with the insurgents, "clearing" cannot work.

As long as the coalition persists in its current strategy, increasing the number of troops in country will not only be inefficient, it will be dangerously counterproductive. As Secretary of Defense Robert Gates said not so long ago, more troops would fuel opposition amongst the Afghan population. Considering the growing illegitimacy of the Karzai regime, more foreign troops will be resented as a military occupation. To this end, the coalition's communiqués stating that the foreign presence in Afghanistan will go on for two generations-which were intended to reassure the Afghan partners-are staggering diplomatic blunders, especially in a country where feelings towards outsiders are at best ambiguous.

The more foreign troops fight to take territory back from the Taliban, the more the population rejects them, because it sees them as the major provider of insecurity. In addition, more troops mean more casualties, leaving the coalition less time to do its work before public opinion turns too far against the war. Yet it is unrealistic to expect quick results, especially in training the Afghan National Army. And at the same time, it is more and more difficult to argue in support of the discredited Karzai regime.

What now are realistic objectives for the coalition? The defeat of the Taliban is not feasible. The overall objective should be to build an Afghan partner that can fight for itself and contain the insurgency. That doesn't mean perfect security on the border, or the destruction of the Taliban. In fact, the best the coalition can do in the Pashtun belt is to protect the cities and the major roads against the mounting attacks-a difficult task in itself. The priority is now to stop the Taliban in the north, where some areas are spiraling rapidly out of control. These cities are where the state-building can work and where the foreigners are more easily accepted.

Are the core U.S. interests preserved with this alternative strategy? Yes. The Taliban do not threaten transnational attacks against Western countries, and al-Qaeda is based in Pakistan. Nor does the war make the United States safer. On the contrary, the current conflict fuels the creation of radical networks with global agendas. One can even argue that the current state of war in Afghanistan serves al-Qaeda's interests even more than a Taliban victory, because the jihad acts as a magnet for radicals all around the world.

The only logical link between fighting al-Qaeda and fighting the Taliban is that, if the insurgency takes the cities, al-Qaeda would gain a sanctuary and shift part of its operations from Pakistan to Afghanistan. To avoid that, however, the coalition does not need to defeat the Taliban, but to secure the cities of the east and south and diminish the level of violence. If the Afghan government can accomplish that, al-Qaeda will have little interest in Afghanistan, because it will have fewer international troops in its crosshairs, and no easy access to the cities it would need to launch international actions.

The United States faces a grave and growing problem in Afghanistan, but so far, public debate continues to validate the mistaken assumptions of a failed strategy. Congress is asking the wrong question: It's not how many troops the United States is sending to Afghanistan, it's what they're being asked to do.

Gilles Dorronsoro is a visiting scholar at the Carnegie Endowment for International Peace.