On his second full day in office, President Barack Obama signed an executive order committing to close the military detention facility at Guantánamo Bay, Cuba, by January 2010. The attorney general-led review team is now underway, reviewing the detainees' files and determining what to do with the 245 men who remain there.

As an initial matter, the team needs to consider who the United States should be seeking to detain long-term, far from any battlefield, and on what grounds.

In 2001 and 2002, the Bush administration answered this question without any clearly articulated criteria. In addition to the foot soldiers fighting to defend the Taliban's Afghanistan, it brought to Guantanamo those who may have once attended an al-Qaeda training camp, had some affiliation with an al-Qaeda member, or merely appeared to dislike the United States. Hundreds were held for years without charge before ultimately being released. Others remain in Guantanamo - even though the United States has concluded that they should be released - because they cannot be returned to their home country for fear of ill-treatment and the United States cannot find a third country to accept them.

As Guantanamo has proven, it's not good counterterrorism policy to bring every potential terrorism supporter into a U.S.-run system of detention without charge. Mistakes are inevitably made and innocent people detained, fueling anger toward the United States and becoming part of the terrorist recruiters' job pitch. Moreover, it's simply impractical. Assume, for example, that the United States wanted to detain all potential al-Qaeda members. The U.S. military could march through the streets of Riyadh or Islamabad, arrest and detain just about any "dangerous-looking" male between the ages of 20 and 35. After all, at least some portion of them might one day join forces with anti-U.S. forces, or want to. But no prison is large enough to hold all of the angry young men in the world.

A smart counterterrorism policy would instead focus on incapacitating those who provide value-added to the al-Qaeda network - the leaders, financiers, and technological experts who cannot be easily replaced, and who can be prosecuted for conspiracy to commit terrorism, if not more.

The United States Army's new Field Manual on Counterinsurgency Operations provides the rationale for a more pragmatic - and effective - approach. It is simply not possible to kill or capture every enemy in a battle with a non-traditional enemy like al-Qaeda. Nor is it necessarily a good idea. "Dynamic insurgencies can replace losses quickly," explains the Counterinsurgency Field Manual. The only way to win, therefore, is to "cut off the sources of that recuperative power" by diminishing the enemy's legitimacy and appeal while increasing one's own. The manual cautions that the United States loses its legitimacy, and therefore its ability to win the fight against al-Qaeda, if it engages in illegitimate actions. "Unlawful detention" and "punishment without trial" are cited as illegitimate actions to be avoided.
To establish legitimacy," the manual continues, "commanders transition security activities from combat operations to law enforcement as quickly as feasible. When insurgents are seen as criminals, they lose public support."

Some have warned that the U.S. criminal justice system is not up to the task. But since 9/11, federal courts have successfully convicted over 145 terrorism suspects. The Classified Information Procedures Act, as well as other case-by-case means of handling complicated evidentiary issues, have been relied on to successfully protect both a suspect's due process rights and sensitive national security information, including information gathered overseas. Moreover, the evidence needed to prove a suspect guilty of the crimes of conspiracy to commit terrorism, or providing material support for terrorism, is surprisingly little. To obtain a conspiracy conviction, for example, prosecutors need only establish a criminal agreement between two or more people and a single step, no matter how minor, in furtherance of the agreement.

To be sure, by narrowing the definition of who should be detained by the United States, and by relying on the criminal justice system to justify the detentions, the United States will incur some risk. Some of the detainees who will be released may ultimately turn out to be dangerous. Indeed, the Pentagon claims that some 61 detainees released from Guantanamo have "returned to the fight," although that number - which has never been supported by a full list of names and other identifying information - reportedly includes those who engaged in "propaganda warfare" by speaking out about their experiences in Guantanamo.

But detaining without charge a few dozen potentially dangerous men also involves risk. As the counterinsurgency manual explains, a terrorist network like al Qaeda is not static, but dynamic and fluid, and the world is full of angry young men who are potential recruits to al Qaeda's cause. Locking up a few dozen without charge does little to make the United States safe. To the contrary, these men become glorified as martyrs, and facilitate the terrorists' recuperative power in the future.