With all due respect (and I sincerely mean this, not meant just as the requisite boiler-plate), I find Peter a tad too cock-sure in how he portrays more boots on the grounds as a total no-brainer ("well, do the math"). I understand the importance of boots on the ground for stability operations, indeed in the pages of my blog urged for supplementing our forces in Iraq back in the day, before the decision was belatedly made on the surge (once finally implemented after the myriad criminal ineptitudes of the Rumsfeld era, I disagreed with the wisdom incidentally, as it was not accompanied by a serious regional diplomatic strategy, so that we were merely forging tactical, localized security improvements but missing the wider strategic lens the situation demanded, and indeed still does today).
The question about more boots on the ground is a relatively easy one to answer.
None, or few, of those new boots will come from NATO allies and if they do come they will come so freighted with national caveats and domestic political considerations that will make them largely ineffective. So they will have to come from the U.S.
Why are more needed? Well do the math: Afghanistan is a country ideally suited to guerilla warfare with its high mountain ranges and it is a third larger than Iraq and its population is some 6 million or so greater, yet the numbers of soldiers and policemen in Iraq are more than three times larger than in Afghanistan.
I agree with almost everything Peter writes below (particularly his "second" mistake, the clever subtlety he flags in his "third mistake", and then too his last paragraph-to which I'd add the need to effectively engage with Syria too).
A small quibble however.
I also wanted to respond to the idea that somehow we are making the same mistake in Afghanistan that the Soviets did. This is a real misreading of history. The Soviets killed at least 1.5 million Afghans and they turned a third of the population into refugees; some 6 million fled to Iran and Pakistan.
Our policies in Afghanistan are failing and require a complete rethink but no matter how many problems we have encountered there (and in Pakistan) it is not because we are repeating the same mistakes as the Soviets who imposed a brutal, totalitarian war on a population who, in the main, loathed them with a passionate intensity.
As a lawyer, I will adjudge this at best circumstantial evidence, but take your point.
No leaders of al Qaeda have been killed or captured in urban areas since 2004. Before that many were-Khalid Sheik Mohammed, al Shibh, Abu Zubayda, etc. Since then Al Qaeda leaders who have been killed or narrowly avoided being killed have been on the receiving end of Hellfire missiles in the tribal areas.
How can you be sure of this?
For what its worth none of the AQ leadership are in urban areas. That was a very costly mistake for them. Since 2004 they are all in the Federally Administered Tribal Areas [of Pakistan.]
In response to Greg's query to me: Briefly, I don't assume that Bin Ladin and al-Zawahiri are hiding in urban areas (but, of course, I have no intelligence upon which to base this assumption). So your question underscores my point precisely: the military is not the lead counterterrorism agency in urban areas, where so many segments of the jihadist movement thrive. And that is why the pre-9/11 mindset is a canard.