Quotes of the Day

President Barack Obama will on Wednesday announce the size of the troop withdrawal from Afghanistan he will order in July in keeping with the symbolic drawdown he promised to begin this summer. His top military men appear to want to keep most combat troops in the field for at least another two years; other advisers want the withdrawal that begins this summer to involve substantial numbers of troops and on an accelerated schedule. But the number of U.S. troops deployed in Afghanistan appears increasingly unlikely to decisively determine the outcome of the war.

Soon after he arrived in office, President Obama buckled to pressure from the Pentagon and doubled down on the U.S. commitment of blood and treasure in Afghanistan, sending an extra 30,000 soldiers as part of the Afghan "surge". Whatever tactical gains have been made in that time — and are likely to be made in the next two years even if current force levels were maintained — Obama will be well aware that a decade of warfare in Afghanistan has failed to qualitatively transform the strategic equation. The key argument from the generals and Administration figures like Defense Secretary Robert Gates in support of keeping the full complement of fighting men in Afghanistan is that the gains made in recent years are "fragile". Of course they are: Experience has shown that there's no reason to expect any area "cleared" by NATO forces will remain that way once they reduce their force levels there. It was ever thus, and there's little reason to expect any different in a conflict that pits a foreign armies against an indigenous insurgency backed by a significant section of the population and with sanctuary in a neighboring country. But who'd want to be the President who has to call time on a failed expeditionary war?

Obama's track record suggests the withdrawal he announces for this year will involve a desultory figure of less than 5,000 of the 100,000 troops the U.S. currently has deployed in Afghanistan. But expect, also, to see the President define downward the benchmark of success, focusing it narrowly on protecting the American mainland from another transnational terror attack by al-Qaeda, and using the killing of Osama bin Laden to underscore the fact that the job is almost done. The terror network, much reduced by a decade of pummeling, today operates more from Pakistan than Afghanistan.

The Afghanistan mission's more ambitious nation-building goals originally proclaimed by the Bush Administration, however, are plainly beyond reach for the foreseeable future. The U.S. finds itself propping up a corrupt and ambivalent Karzai government knowing it has limited ability to win hearts and minds in the war zone, but also that it has no better alternative. The argument that the U.S. needs to be in Afghanistan to stop Pakistan falling to the Taliban has always been specious, for the simple reason that those currently in power in Pakistan see the Afghan Taliban as an ally that, unlike the U.S., shares Islamabad's strategic interests in Afghanistan. And Pakistan's generals act accordingly. Pakistan has been the headquarters of the Afghan Taliban's leadership shura ever since U.S. forces and their Afghan allies swept the movement from power in late 2001. The generals who make Pakistan's decisions didn't stop seeing Afghanistan as a key front in their existential strategic rivalry with India just because it became the initial geo-strategic focus of the U.S. "war on terror".

Afghanistan on June 7 superseded Vietnam as the longest war in U.S. history, having dragged on twice as long as American involvement in World War II. The more relevant comparison, however irksome it may be to American exceptionalists, may be to Soviet Union's nine-year debacle in Afghanistan — a milestone the U.S. passed last November.

Sure, there are plenty of differences between the two cases — Moscow hadn't suffered terror attacks originating in Afghanistan; it faced insurgent opposition in all of Afghanistan's major ethnic groups whereas the Western presence is opposed from within the Pashtun majority, but has found willing allies the Tajik, Uzbek and Hazara communities whose leaders were traditionally grouped in the Northern Alliance that had fought Taliban rule. And, of course, U.S. counterinsurgency warfare is far smarter and more effective than the Red Army's.

Still, like the Soviets, the U.S.-led operation controls only part of the country; the writ of the feckless government it has propped up doesn't extend much beyond the capital; Pakistan continues to function as a rear base for the insurgents; and the long-term trend lines offer no reason to expect that the insurgents will be eliminated.

The U.S. is not going to be routed in Afghanistan, like some 19th Century British expeditionary force. But nor is it able to vanquish a foe that, in traditional guerrilla style, simply melts away to form up elsewhere whenever the foreign army concentrates its forces. The Taliban doesn't need to storm Kabul to win. As Henry Kissinger famously noted, in an insurgency of this type, the guerrilla army wins by not losing; the conventional army loses by not winning.

And the U.S.-led NATO operation is clearly not winning, even if it's managing to tread water in areas where it has concentrated its forces. Polls show that a majority of Americans no longer believe the war is worth fighting. And that's an opinion long shared by the majority of Afghans in the Pashtun war zone.

The comforting prospect offered to Americans looking for an end to a war that is costing upward of $100 billion a year and a slow but steady drip of U.S. casualties is that the U.S. is training and equipping Afghan security forces projected to number 300,000 by October. But few Western military observers have much confidence that this force will be either able or willing to continue NATO's fight once Western troops leave, ostensibly in 2014 — although the rump Northern Alliance elements will certainly fight tooth and nail to protect their turf from any Taliban return. Right now, Kabul doesn't exactly control much of the territory of which it is the sovereign capital, and that's unlikely to change no matter who's in power there.

Washington appears to have recognized that the only way to relieve itself of an open-ended burden in Afghanistan is to negotiate a peace agreement with the Taliban enemy it had originally hoped to destroy. Defense Secretary Robert Gates acknowledged last weekend that preliminary negotiations are already under way, although with little expectation of near-term progress.

The purpose of the NATO war effort, over the past two years, has been to set the negotiating table more favorably to the Western side, hoping to forcefully "demonstrate" to the Taliban that it can't prevail on the battlefield and therefore that it needs to be more amenable to U.S. terms for peace. And, of course, the Taliban's purpose has been the exact opposite — to demonstrate the futility of the NATO effort, so as to give it a better chance of imposing its own terms at the peace table.

Despite the "surge", the Taliban doesn't appear to be feeling squeezed to accept U.S. peace terms. On the contrary, right now it seems to playing hard to get, believing that time and circumstance work in its favor.

So outcome of the Afghanistan war is unlikely to be determined by the troop numbers to which President Obama commits on Wednesday. Instead, it will be shaped by what the U.S., the Northern Alliance, and Afghanistan's key neighbors, most importantly Pakistan, are willing to accept by way of a political compromise.

• Tony Karon