As the president and his war council meet to finalize the latest most updated new new new strategy review in Afghanistan, ex-CIA man Paul Pillar gave the House Armed Services Committee the right question to ask:

[I]s the difference between the terrorist threat Americans would face if we wage a counterinsurgency in Afghanistan, and the threat we would face if we do not wage it, sufficiently large—and in the right direction—to justify the costs and risks of the counterinsurgency itself?

The short answer is, “No.”

One of the urban legends about the birth of the Internet is that it was conceived in response to a Pentagon desire for a communications system that could survive a nuclear strike. As a network, the Internet routes around clusters of lost nodes. So, if, say, a chunk of the nation was destroyed by a Russian nuclear weapon, the military would still be able to communicate by routing command and control through still-functioning nodes outside of the blast zone. (Nobody, though, seems to ponder how that Internet would work alongside the massive disruption to the national power grid that would be sparked by such an attack. Details details…) That’s what we mean when we say certain networks as a structure are resilient. Taking out a random mass of local nodes doesn’t cause it to fail. It routes around failure. It’s what’s known as a scale free network, which, to skip over a bunch of nerdy talk about growth and preferential attachment, means that you can take out contiguous chunks of the network all day long, but, because you’ll be predominantly destroying small nodes versus network hubs, the hubs outside of the lost chunk keep all the surviving nodes connected, and the network functioning.

Al-Qaida is not a conventional military force. It’s a scale-free network. When attacking a such a network, seizing a particular piece of geography is a laughable way to proceed unless you know that the region you target contains a non-random concentration of network hubs. Further, AQ’s network displays small world behavior, meaning that unless you silence a sufficient number of hubs at once, the pieces of the network remain functional as mini-networks.

But Petraeus and Mullen have both flat out told us that there are no al-Qaida “hubs” in Afghanistan.

In other words, Afghanistan holds no hubs, and thus mucking around there has no chance of inducing network failure.

So, why on earth would we continue to spend $1 million per year, per soldier, to seize and hang onto a piece of territory empty of the hubs we’re seeking?

Counterinsurgency represents the conventional military’s bid to stay relevant in current conflicts against unconventional enemies. Conventional forces are a weapon designed to fight other conventional forces. Thus, to stay relevant, commanders of conventional forces (and those with interests in making sure said forces remain relevant, like defense contractors and politicians of a certain stripe) must impose conventional attributes onto the public perception of the unconventional enemy. This is often done by euphemism: “bases” become “safe havens,” for example. When you hear people talking about “safe havens,” you should recognize the projection of the conventional force structure model onto the terrorist group by a person used to thinking about a conventional opponent. What they mean by “safe haven” is essentially a “military base.” But terrorists don’t need military bases. As Pillar explained:

Most important activities that transnational terrorist groups have performed in safe havens also can be performed, often with comparable ease, elsewhere. Terrorist attacks can be conceived anywhere. Operations can be, and often have been, planned and prepared in Western cities. Most recruitment and radicalization takes place outside anything that could be called a safe haven, and again often in the West.

Terrorists’ exploitation of information technology, and of the ease of movement that goes under the label of globalization, have made them less dependent than ever before on any physical haven. Terrorist leaders exert command and control through cyberspace at least as effectively as through older methods. As 9/11 demonstrated, the planning, training, and other preparation for a major operation need not be centered in any one location but instead can span several continents.

…The terrorist threat to U.S. interests—even just the Sunni, Salafi, jihadist portion of that threat—does not all emanate from Afghanistan or from any other single place. It is decentralized, more so now than it was at the time of 9/11. It involves a diverse movement with different groups, cells, and individuals having different mixtures of local and global objectives. The movement does not have a central command, even though bin Ladin’s al-Qa’ida continues to be the most recognizable part of it. Even some groups that have found it advantageous to adopt the al-Qa’ida brand name, such as the one in the Maghreb of North Africa, act more on local initiative than in response to any orders from South Asia. One hears frequent mention of “links” back to Afghanistan or Pakistan, but links do not necessarily mean direction or instigation. They can mean nothing more—as in what has become publicly known so far about the Zasi case—than having once passed through a training camp. They may mean even less.

And, even if al-Qaida did need a base, it’s not clear that they’d want Afghanistan, or that a Taliban-controlled Afghanistan would want them:

[T]he conditions of any al-Qa’ida return to Taliban-controlled territory would be a source of strain between the groups. This in turn would affect al-Qa’ida’s perceptions of the relative attractiveness of Afghanistan and the current haven in northwest Pakistan. It is hard to discern much that the former would offer over the latter. In any event, it is not apparent how a move of al-Qa’ida or parts of it from one side of the Durand line to the other would substantially affect the threat the group poses to U.S. interests. Any such threat should be no less from Waziristan than it would from Nuristan.

The sad joke is that even if President Obama implements General McChrystal’s called-for troop increase, he still won’t have eliminated the ability of terrorists to create a safe-haven within the borders of Afghanistan:

Regardless of whether a renewed haven inside Afghanistan were attractive and useful to al-Qa’ida or any other terrorist group, there is the question of whether a counterinsurgency would preclude it. A haven would not require a patron with control over all of Afghanistan, which has an area of 647,000 square kilometers, but instead only a small slice of it. As described in General McChrystal’s assessment, a “properly resourced” strategy would leave substantial portions of the country—those portions not deemed essential to the survival of the Afghan government—outside the control of that government or of U.S. forces. In short, even a counterinsurgency that was successful, in the sense of accomplishing the mission of bolstering the government in Kabul and stabilizing the portions of the country where most Afghans live, still would leave ample room for a terrorist haven inside Afghanistan should a group seek to establish one.

General McChrystal’s commander-in-chief has defined the war in Afghanistan as a “war of necessity,” and defined the objectives thus: “to disrupt, dismantle, and eventually defeat at Qaida and prevent their return to Afghanistan.” McChrystal’s mandate is restricted to Afghanistan. The problem is that, as Mullen and Petraeus have admitted, there is no AQ in Afghanistan. The task given to McChrystal, then, is nonsense, so his answer is nonsensical. The McChrystal strategy posits no end to al-Qaida. Nothing in it bears directly on the question of “how do we destroy al-Qaida.” The only thing a counterinsurgency campaign could do even if it were successful beyond our wildest dreams is deny AQ a safe haven within the areas under the control of the Kabul government. But even if McChrystal miraculously manages to deny safe haven in all of Afghanistan, well, congratulations. You’ve managed to deny al-Qaida the use of 652,230 square miles out of a total 57,511,026.002 square miles of the Earth’s surface, the roughly 1 percent of the planet’s landmass which, according to U.S. commanders, contains none of the terrorist network hubs they’re after. Good job.

Al-Qaida’s safe haven is a globalized, Internet-connected world. You’re probably not going to defeat it by taking and defending territory. McChrystal’s request comes straight out of a conventional-thinking military accustomed to political deference and unlimited resourcing, striving (successfully, mind you) to stay relevant versus today’s threats, and shaped by a presidential goal that makes no sense given the public statements of U.S. generals and admirals.

COIN is a waste of resources if the goal of the U.S. is to defeat the al-Qaida network.

Note: Derrick Crowe is the Afghanistan blog fellow for Brave New Foundation / The Seminal. Learn how the war in Afghanistan undermines U.S. security: watch Rethink Afghanistan (Part Six), & visit http://rethinkafghanistan.com/blog.