Cordesman's Contradictions

By Dave Anderson:

Anthony Cordesman of both the McCrystal insider strategy review session and the Center for Strategic and International Studies is arguing in the Washington Post for a surge into Afghanistan as a prelude to another five to ten years of sustained combat as well as hundreds of billions in additional expenditures. He asks for just a couple more Friedman units to avoid defeat (although he poorly defines desired end-states so what is defeat or even victory) that can only be reveresed by applying the latest in modern COIN doctrine to.

The United States cannot win the war in Afghanistan in the next three months -- any form of even limited victory will take years of further effort. It can, however, easily lose the war....

it [Bush Administration] did not react to the growing corruption of Hamid Karzai's government or the major problems created by national caveats and restrictions on the use of allied forces and aid. It treated Pakistan as an ally when it was clear to U.S. experts on the scene that the Pakistani military and intelligence service did (and do) tolerate al-Qaeda and Afghan sanctuaries and still try to manipulate Afghan Pashtuns to Pakistan's advantage.

Okay, this is a decent diagnosis of some of the problems --- massive lack of local legitimacy by the supposed central government, massive corruption, Pakistani interests in creating and maintaining strategic depth vis a vis India (especially since the Northern Alliance which supplies most of the warlords backed by the US, were supplied and armed by India for most of the 1990s), and NATO forces that are fighting for countries that are not too sure what the point of being in NATO is any more. Not sure how tractable any of these problems are given US time, resources, troop strength and political constraints, but these are not a bad set of symptons listed.

The appointments this summer of Karl Eikenberry as ambassador to Afghanistan and McChrystal as commander of U.S. and allied forces have created a team that can reverse this situation. In fact, given the rising unpopularity of the war and Taliban successes, they are our last hope of victory. Yet they can win only if they are allowed to manage both the civil and military sides of the conflict without constant micromanagement from Washington or traveling envoys. They must be given both the time to act and the resources and authority they feel they need. No other path offers a chance of a secure and stable Afghanistan free of terrorist and jihadist control and sanctuaries.

Hopefully Cordesmann means "terrorists and jihadists" of "the far-enemy/deep strike" groups. If that is the unstated but embedded assumption, that is a reasonably stated victory condition --- remove the threat of long-strike terrorism based in Afghanistan.

However there are two problems with this reading. The first is that most of the planning for the most notable far enemy strike (9/11) took place in Karachi, Kuala Lumpor and Hamburg, so Afghanistan is neither neccessary nor sufficient for deep-strike planning. The second is that I am not sure if this actual distinction between deep strike/far enemy jihadis and the "accidental guerrillas" is actually being made here. The United States has minimal interest in Pashtun politics and inter-ethnic group relationships near the Durand line except on the question of deep strike terrorism. Cordesman may be elevating such an interest in making the Pashtuns as a group suck on it.

almost every expert on the scene has talked about figures equivalent to three to eight more brigade combat teams -- with nominal manning levels that could range from 2,300 to 5,000 personnel each -- although much of that manpower will go to developing Afghan forces that must nearly double in size, become full partners rather than tools, and slowly take over from U.S. and NATO forces. Similarly, a significant number of such U.S. reinforcements will have to assist in providing a mix of capabilities in security, governance, rule of law and aid. U.S. forces need to "hold" and keep the Afghan population secure, and "build" enough secure local governance and economic activity to give Afghans reason to trust their government and allied forces. They must build the provincial, district and local government capabilities that the Kabul government cannot and will not build for them. No outcome of the recent presidential election can make up for the critical flaws in a grossly overcentralized government that is corrupt, is often a tool of power brokers and narco-traffickers, and lacks basic capacity in virtually every ministry.

Here is where the op-ed falls apart on several grounds. The first is on US sustainability. Next problem is the Afghan Army and National Police sustainability problem. Finally, the op-ed is arguing for a deux ex machina of counterinsurgency and legitimacy enhancement in that the greatly expanded US/ISAF presence, combined with a greatly expanded ANA and ANP will overwhelm the structural political problems that promote illegitimacy of the Karzin government and the current governing structure.

Three to eight combat brigades is between 15,000 and 45,000 more troops once you count on the logistic enablers. As our good blog friend, Dave Schuler likes to point out, the logistics of supplying combat forces in Afghanistan suck --- each individual US soldier costs about three times as much to support in Afghanistan as he did in Iraq. The biggest difference is that US supply ships could haul heavy cargo to modern ports (Aquaba, Ceyhan, and Kuwait City) and then onto trucks using modern highways on mostly direct routes to the major supply hubs in Iraq. Air transport was reserved for either low-weight, high value items or for limited logistics support in no-go areas. The basic opposites apply in Afghanistan once a ship reaches Karachi --- the roads are worse, the terrain is almost impassible, and the through-put is restricted to a few narrow and easily targetted passes (and all of that is on the Pakistani side of the border). And once supplies reach a major US logistics hub, the deployed combat arms units are either Fobbits and operationally useless or out in the boonies with almost no road network back to the logistics hub. Air delivery of basic and frequently consumed supplies is used far more often in Afghanistan than in Iraq.

Secondly, as Steve, myself and many others have pointed out, the current Afghan Army and National Police are at a size that is unsustainable by the Afghan government's local resources. Increasing the Afghan security forces means increasing the security budget which creates a few long term problems. The first problem is the operational problem of recruiting and training new units while still trying to hold onto the quality of the existing units. That is solvable. The second problem is a strategic problem of what to do about the Kabul's government legitimacy --- either the forces are paid for by foreigners, which is a massive hit to legitimacy in general, or the forces sustainably de-volve into either quasi-warlordism (which is a reasonable component of Afghan national politics) or rent seeking coup initiators on any day that ends in "y" as the Kabul government can not pay for its army or its loyalty from local resources.

This leads into the greatest problem with Cordesman's argument --- it is what he wants those additional forces to do. He wants the US military, in conjunction with ISAF forces and nominally sovereign Afghan military units to build a modern nation state from the ground up despite the following description of the state and its institutions:

"grossly overcentralized government that is corrupt, is often a tool of power brokers and narco-traffickers, and lacks basic capacity in virtually every ministry.

WTF

The 10 second description of US COIN doctrine is legitimacy based:

Enhance the host government's legitimacy and capability while denying legitimacy and capacity to the insurgent(s).

Cordesman wants to build local, decentralized institutions that have minimal relationships to the Kabul government. That may reduce the level of violence to "acceptable" levels but every local success delegitimatizes Kabul's authority. I hope I am missing something here as the argument being made for state construction from local actions goes against the legitimacy of the host government and seems to be counter-productive in actually achieving the deliverables that US COIN doctrine promises.