REPORTS ON the Bush administration's efforts to craft a new strategy in Iraq often use the term "surge" but rarely define it. Estimates of the number of troops to be added in Baghdad range from fewer than 10,000 to more than 30,000. Some "surges" would last a few months, others a few years.
We need to cut through the confusion. Bringing security to Baghdad--the essential precondition for political compromise, national reconciliation and economic development--is possible only with a surge of at least 30,000 combat troops lasting 18 months or so. Any other option is likely to fail.
The key to the success is to change the military mission--instead of preparing for transition to Iraqi control, that mission should be to bring security to the Iraqi population. Surges aimed at accelerating the training of Iraqi forces will fail, because rising sectarian violence will destroy Iraq before the new forces can bring it under control.
Any military strategy must of course be accompanied by a range of diplomatic, political, economic and reconciliation initiatives, but those alone will not contain the violence either. Success in Iraq today requires a well-thought-out military operation aimed at bringing security to the people of Baghdad as quickly as possible--a traditional counterinsurgency mission.
Of all the "surge" options out there, short ones are the most dangerous. Increasing troop levels in Baghdad for three or six months would virtually ensure defeat. It takes that long for newly arrived soldiers to begin to understand the areas where they operate. Short surges would redeploy them just as they began to be effective.
In addition, a short surge would play into the enemy's hands. Both Sunni insurgents and Shiite militias expect the U.S. presence to fade away over the course of 2007, and they expect any surge to be brief. They will naturally go to ground in the face of a short surge and wait until we have left. They will then attack the civilian population and whatever Iraqi security forces remain, knowing them to be easier targets than U.S. soldiers and Marines. They will work hard to raise the level of sectarian violence in order to prove that our efforts have failed.
We have seen this pattern so many times before that we can be virtually certain the enemy will follow it in the face of a short surge. The only cure is to maintain our presence long enough either to root out the hiding enemy or to defeat him when he becomes impatient. A surge that lasted at least 18 months would achieve that aim. It would also provide time to bring Iraqi forces up to the level needed to fight whatever enemy remains.
The size of the surge matters as much as the length. Baghdad is a large city. Any sound military plan will break the problem of bringing security to the Iraqi capital into manageable parts. But there remains a minimum level of force necessary to make adequate progress in a reasonable time.
U.S. forces working with Iraqi troops can clear neighborhoods fairly quickly. Unfortunately, past endeavors such as Operation Together Forward relied too much on that ability. We sent forces into the city that were large enough to clear a few neighborhoods at a time but not large enough to maintain the security they had established. Any plan for bringing security to Baghdad must include forces for the "hold" phase as well as the "clear" phase.
Clearing and holding the Sunni and mixed Sunni-Shiite neighborhoods in the center of Baghdad, which are the keys to getting the overall levels of violence down, will require around nine American combat brigades (27 battalions, in partnership with Iraqi forces, divided among some 23 districts). Since there are about five brigades in Baghdad now, achieving this level would require a surge of at least four additional combat brigades--some 20,000 combat troops. Moreover, it would be foolhardy to send precisely as many troops as we think we need. Sound planning requires a reserve of at least one brigade (5,000 soldiers) to respond to unexpected developments. The insurgents have bases beyond Baghdad, especially in Anbar province. Securing Baghdad requires addressing these bases--a task that would necessitate at least two more Marine regiments (around 7,000 Marines). It is difficult to imagine a responsible plan for getting the violence in and around Baghdad under control that could succeed with fewer than 30,000 combat troops beyond the forces already in Iraq.
It is tempting to imagine that greater use of Iraqi forces could reduce the number of U.S. troops needed for this operation. The temptation must be resisted. We should of course work with the Iraqi government to get as many trained and reliable Iraqi troops as possible into Baghdad, and we should pair our soldiers and Marines with Iraqis as much as we can. But reducing the violence in the Sunni and mixed neighborhoods in Baghdad is the most critical military task the U.S. armed forces face anywhere in the world. We cannot allow that mission to fail simply because some Iraqi units don't show up, aren't at full strength or are less reliable than we had hoped.
The United States faces a dire situation in Iraq because of a history of half-measures. We have always sent "just enough" force to succeed if everything went according to plan. So far nothing has, and there's no reason to believe that it will. Sound military planning doesn't work this way. The only "surge" option that makes sense is both long and large.
Jack Keane is a retired Army general. Frederick W. Kagan is a resident scholar at the American Enterprise Institute and a contributing editor to THE WEEKLY STANDARD. This piece originally appeared in the December 27, 2006 edition of the Washington Post.