Von at Obsidian Wings revisits two points on Iraq that come up now and again. The first is the "we should have invaded with more troops" bit. Kevin Drum says we just didn't have the troops. On this, I think Von's rejoinder is technically correct: the U.S. never had the manpower to sustain 400,000 troops in Iraq for three years, given the need for rotations and the like, true, but in theory we had enough to go in with 400,000 or so initially, stabilize things in the early days, secure Baghdad, and quell the insurgency before drawing down to more sustainable troop levels by late 2003. Von also says, as many others have, that disbanding Saddam's Iraqi Army was a big mistake, since it decimated the country's native security force, and all those soldiers went to go join the Sunni insurgency. In general, hawks who think Iraq was a decent idea but got badly botched in the implementation point to these two mistakes.

I tend to think, though, that even if the United States had done both of those things, Iraq still would have been a gamble, and hardly destined to succeed. The U.N. plastered Kosovo with troops, mind you, and even that region is in danger of imploding even today; see the ICG's report on this. Moreover, history is ambiguous on whether keeping the repressive old order in place—all those Baathist soldiers, for instance—"works" or not. It proved effective in Japan and Germany, sure, but the relevant comparison here might be the postwar occupation of South Korea, where Lt. Gen. John Hodge relieved the Japanese, but kept the old Imperial ministries in place, staffed with former Korean collaborators. (As with Iraq, the U.S. administrators in South Korea had precisely zero knowledge of the local culture.) Due to the ensuing resentment, many Koreans ended up revolting and proved moderately receptive to Communist insurgents filtering in from the North, and pretty soon you had violence and then a war in which millions died. Whoops. It's not hard to imagine that had the U.S. kept the Baathist Army in place in Iraq, something similar might have happened, with the Shiites creating an Iranian-backed insurgency.

I don't know how useful these sorts of historical analogies are; usually they make sense only in retrospect. (Looking at it now, the occupation of Iraq does resemble the bumbling postwar occupation of South Korea, but no one would have offered up that comparison beforehand.) Prior to the war, the only clear lesson from America's past was that nation-building was always an extremely iffy enterprise. As far back as 1865 the United States had deposed one of the more sordid, illiberal regimes around, occupied the region, and tried to do a bit of nation-building. But a decade later the occupation ended in the face of an armed insurgency and dwindling domestic support, and the occupiers left a corrupt one-party state in place that didn't get around to respecting minority rights until 100 years later, and to this day still exports militant fundamentalism abroad that threatens world peace. Nation-building failed, and that was here at home. From the very start, we've never been much good at this.

Ultimately, it seems, that's because these sorts of nation-building operations tend to live or die on the internal social and political forces of the occupied country in question, rather than the best-laid plans of the occupiers. Good preparation helps, of course, but I don't think it's ever decisive. The Truman administration didn't get a working plan for Germany until 1947; the occupation of the rest of Europe—Austria, especially—was nothing short of disastrously incompetent, but things turned out okay. Elsewhere, the immediate post-WWII liberation of Vietnam went smoothly, unlike Malaysia and Indonesia, but decades later, Vietnam was imploding and Indonesia stable. Whether or not competent people are running things, it's very much a crapshoot. The best thing to do, then, is not to roll the dice unless you absolutely have to.