While some are discussing whether the U.S. presence in Afghanistan should be maintained, the Obama administration does not appear to be seriously considering withdrawal. Sen. Carl Levin (D-Mich.) and others have instead proposed expanding the Afghan National Security Forces (ANSF) first and raising U.S. force levels only if that approach is unsuccessful. This option holds out hope of success without the need to send more U.S. troops, but we believe it is illusory.
Withdrawal now would allow Afghanistan to again become a haven for terrorists. It would destabilize Pakistan by giving refuge to terrorist and insurgent groups attacking Islamabad and by strengthening the forces within the Pakistani government and security forces that continue to support the Taliban as a hedging strategy against precisely such an American retreat. Pursuing an offshore strategy of surgical strikes using aircraft and Special Forces units would destabilize Pakistan for the same reasons. Further, if such a strategy could work against al-Qaeda, the commander in Afghanistan, Gen. Stanley McChrystal -- the most accomplished practitioner of Special Forces counterterrorism campaigns -- would be advocating it. Instead, he is advocating counterinsurgency.
Failing to act to reduce the insurgent threat while simply trying to expand the ANSF is precisely what the United States did in Iraq in 2005 and 2006, with well-known results. It will not succeed any better in Afghanistan.
Inherent constraints on our ability to expand the Afghan army quickly make it extremely likely that the enemy will succeed before the necessary Afghan forces can take the field. We should commit to the most rapid possible growth of Afghanistan's national forces, and we can do more to speed up expansion. But an increase in U.S. forces will be needed in the interim.
Afghanistan's security force capabilities are complex. The ANSF includes the Afghan National Army, the Afghan National Police, the Afghan Border Police, the Afghan National Civil Order Police and other, smaller organizations. There will be roughly 107,000 army soldiers and 96,000 police officers on the rolls by the end of September. There are perhaps 13,000 border police and 5,000 civil order police. Most troops in the army and the National Civil Order Police are moderately effective; some are very effective. A large portion of the national and border police are ineffective or part of the problem. Reforming these institutions and removing their malign actors, as we did with similar organizations in Iraq, must be key to the counterinsurgency strategy. Meanwhile, we must not assume that every police officer in Afghanistan aids security.
Under current parameters, Afghanistan could add about 50,000 soldiers and police officers annually. By altering the training schedule, increasing the number of locations at which training takes place, accelerating the fielding of equipment, and other adjustments, Afghanistan could produce as many as 80,000 new police and soldiers per year. But accelerating that growth will take time. Realistically, we can hope to get the Afghan army up to its expected ceiling, 134,000 troops, by October 2010. The need to vet, replace and retrain members of the national police will probably make only a fraction of that organization's effective personnel available within the same time frame.
Doctrine suggests that at least 360,000 troops and police are required to fight a counterinsurgency among Afghanistan's 16 million Pashtuns and to police the 16 million Tajiks, Uzbeks and Hazaras. Even under the most aggressive expansion programs, Afghan force levels could not reach 300,000 before the end of 2011. Still, warfare rules of thumb are not guarantees of success. Estimates of population in Afghanistan are imprecise, as are estimates of the numbers of effective Afghan National Police. None of our figures accounts for AWOL rates, ghost soldiers or police officers, and other factors that significantly reduce the number of troops available to function as counterinsurgents. Given the national caveats that confine thousands of coalition forces to their bases, even those troops cannot be firmly counted as counterinsurgents.
These rough calculations lead to two conclusions: First, relying solely on growth in Afghanistan's national forces to address the deteriorating situation is very likely to fail. Second, accelerating the expansion of Afghanistan's national forces offers a meaningful strategy to allow international forces to begin withdrawing gradually in a few years without compromising security.
The Taliban maintains significant sanctuaries in Afghanistan that neither coalition nor ANSF forces have been able to contest, especially in the districts around Kandahar City; in parts of Helmand, Ghazni, Uruzgan, Zabol, Paktika and Farah provinces; and in some areas in the north (especially Kunduz province). The districts around Kandahar City are strategically vital because continued Taliban control of those areas threatens Kandahar City -- the loss of which could radically destabilize the country. Retaking those districts will be difficult. Raw Afghan recruits or recently formed units without adequate coalition partners are unlikely to be able to do so alone. It is almost impossible to imagine a scenario in which accelerating the expansion of Afghan national forces would create conditions for the elimination of Taliban sanctuaries in strategically critical districts in time to prevent a significant Taliban success.
Coalition forces will have to go into these and other districts to clear them, in partnership with such Afghan forces as are available, and will then have to hold them against insurgent attempts to retake them for many months. The aim is to reduce the strength and capability of the insurgency while increasing the size and effectiveness of the ANSF, ultimately allowing Afghans to take responsibility for maintaining order in a significantly less challenging environment. This strategy has worked in Iraq. It has the highest probability of working in Afghanistan.
The good news is that an accelerated growth plan for Afghan national forces is very likely to allow coalition forces, including American reinforcements, to begin reducing their footprint within two or three years without compromising their gains. Withdrawing completely from Afghanistan now would put America's vital national interests in serious danger.
Building Afghan forces dramatically is part of a strategy for succeeding in Afghanistan and permitting the reduction of foreign forces. It cannot, however, be the whole strategy.
Frederick Kagan, director of the Critical Threats Project at the American Enterprise Institute; Kimberly Kagan, president of the Institute for the Study of War, where retired Lt. Gen and James Dubik, a former commander of Multi-National Security Transition Command-Iraq, is a senior fellow. All three advised Gen. McChrystal in Afghanistan, but the views expressed here are their own.