Alexander Downes makes a strong and smart argument about the perils of regime change. He lists several of those perils, and I will expand on one, which further explains why regime change is such a risky business.

That is the role of the intervener’s violence, specifically in Iraq. Downes notes that regime change can generate grievances that inspire segments of the population to fight the new government and interveners. He also points out that occupiers don’t like to provide law and order, and “these conditions create an opening for rebellion.” These are important insights, too rarely articulated, that tell us much about the chaos of post-invasion Iraq.

The most popular accounts of the growth of insurgencies in Iraq tend to ignore U.S. violence. Academics contend that Sunnis suffered an intolerable loss of status; journalists’ favorite has been the more prosaic “ancient hatreds” argument; and Washington emphasizes the destabilizing roles of al Qaeda and Iran. But we have ample evidence showing that U.S. violence is essential to any explanation.

The people of Iraq, when asked, have said that violence and insurgency were directly attributable to U.S. military actions. Even including Kurds living in relatively tranquil provinces—20 percent of the population—a September 2006 poll by the University of Maryland’s Program on International Policy Attitudes found, “78 percent of Iraqis believe the U.S. military presence causes more conflict than it prevents.” A year later 81 percent of Iraqis said that to protect themselves they stayed away from U.S. and coalition forces. Other polls confirm these views.

Recall the arguments about troop levels in the lead-up to war, arguments that persisted when the war went badly. Were there enough “boots on the ground,” we wondered, to carry out the mission? One consequence of having too few U.S. troops was that firepower was often substituted for manpower, and firepower—including a panoply of tactics, detention practices, intimidation, etc.—stirred resistance. The dense social and kinship networks of Iraq were quickly and easily activated to resist U.S. forces when trauma was visited upon one of their number.

Violence begets violence, undermining the prospects for new regimes to thrive.

In this light the scale of casualties takes on particular military (as well as moral) significance. The higher the number of casualties, the fiercer the resistance. The insurgency in Iraq was many-headed, essentially leaderless, and without specific political objectives or ideology—precisely what one would expect in a reactive phenomenon. Many, if not most, insurgents believed they were defending their communities against the occupier—the source of the threatening violence.

Occupation forces are legally obligated to provide order, yet the United States failed to do this, and one outcome was a volatile space of contention between Sunni and Shia militias, and between other hostile groups. This internecine conflict took many lives, probably more than those lost to U.S. actions alone, as each act of violence stirred reprisals in an ever-more violent cycle. Altogether, according to household surveys and other indicators, probably 750,000 or more Iraqis died as a result of the war.

In this, then, regime change was a bloody affair. It often is. Regime change during the Vietnam War and throughout Southeast Asia in the 1960s and ’70s was accompanied by colossal carnage. Unstable regimes, themselves prone to repression and violence, typically followed.

In the simplest terms, violence begets violence, and this undermines the prospects for new regimes to thrive. But large-scale, organized violence as an instrument of regime change is particularly destructive. Such violence devastates infrastructure and creates deep-seated resentment, social disintegration, widespread population displacement, and new openings for political extremism. In this atmosphere, it is not surprising that, as Downes concludes, “The record of foreign-imposed regime change over the last two centuries is not a happy one.”