As long as there has been war, as long as one nation has sought to impose its military will on another, there have been insurgents… and there have also been counterinsurgents.

It was the ancient Romans who popularized what has become an oft-imitated means of dealing with recalcitrant and subjugated peoples—the so-called Roman Method of repression to quell the noncompliant. Since those long-ago days, numerous countries and empires (from autocracies to democracies) have used similar counterinsurgency (COIN) tactics. Indeed, the history of COIN is a depressing and unremitting tale of coercion and violence generally aimed not just at armed insurgents but at civilian populations as well.

The United States has not been immune from such conflicts—or their brutality. Counterinsurgent fights were waged at home against Native American tribes and in Central America, the Philippines and Vietnam. These small wars are in some manner the defining element of the American Way of War.

In recent years, however, the US military has not only rediscovered counterinsurgency but reinvented it, first in Iraq and now in Afghanistan. To listen to the American military or, better, to read the Army and Marine Corps counterinsurgency manual, FM 3-24, is to hear a tale of COIN’s bloody legacy of coercion and violence magically transformed.

According to this new telling, the key to winning COIN conflicts is in protecting civilians and providing a better future for them—that is, winning hearts and minds via civic action rather than military force. As written in FM 3-24, counterinsurgency fights are population-centric, not enemy-centric, and are defined by an extraordinarily broad array of capabilities, "political, economic, military, paramilitary, psychological, and civic actions."

On the most basic level, the key to COIN victory rests in protecting the population, providing good governance, extending the legitimacy of the host government and gaining the trust of the people. This notion of warfighting has become so internalized in US military thinking that even the past has been rewritten. New assessments of conflicts in Malaya and Vietnam, conducted by COIN advocates, confidently assert that civic action, not coercion—much less forced separation of insurgents from the population—turned the tide of those conflicts. In the case of Vietnam, COIN advocates argue that once the United States turned to counterinsurgency, the direction of the war shifted; and that had the United States had the resilience to see the fight through, the embrace of population-centric COIN would have led the way to success.

Much of this new take on COIN is a result of the US military’s experience in Iraq, where, as the argument goes, a focus on protecting civilians and standing up the Iraqi government led to a decrease in civilian casualties. The supposed success of COIN tactics in Iraq, twinned with the 2007 surge of 30,000 US troops, led to the claim that these successes could be replicated in Afghanistan. Indeed, by the fall of 2009 Gen. Stanley McChrystal offered a strategic review of the situation in Afghanistan that concluded that a population-centric counterinsurgency—and only that strategy—could grasp victory from the jaws of defeat.

It is this approach that the United States began using in 2009 and the first half of 2010. Upon being named Afghanistan commander, McChrystal declared in Congressional testimony that it was more important to protect civilians than to kill insurgents. NATO officials even discussed the possibility of creating a new medal for "courageous restraint" for soldiers who take measures to avoid civilian casualties.

Indeed, COIN has become increasingly fetishized in US military thinking, training and guidance. Col. Gian Gentile, a frequent critic of population-centric counterinsurgency, recounts a recent experience listening to a three-star Army general telling West Point graduates that "what they needed to be good at when they went out into the field army—was establishing ‘trusting relationships’ with local populations." As Gentile acidly notes, "One would have liked soon-to-be-lieutenants told that they must be proficient in their basic branch skills: infantry and armor, basic fire and maneuver with their platoon as part of a maneuver company/team; artillery, fire support; logistics, logistical support; and so on."

The military’s focus on COIN has been reflected in more than just rhetoric. When troops were sent into the town of Marja in Helmand province in February, the element of surprise was purposely given away. Military commanders made clear their intention to take the town and their hope that Taliban forces would flee, thus protecting civilians from being caught in the cross-fire.

Under the logic of COIN, such efforts to protect civilians are not only the right thing to do but are essential to military success, because safe and secure civilians will throw in their lot with the government and provide crucial intelligence for use in targeting insurgents. By securing the people, counterinsurgents can win the competition for the loyalty of the population.

This method of placing civilians front and center in military deliberations may on the surface seem quite progressive (indeed, COIN advocates have repeatedly made this argument). But it hasn’t worked out that way. The United Nations recorded this past summer that the number of Afghan civilians killed in the first six months of the year jumped by a depressing 31 percent; in southern Afghanistan, where the war’s heaviest fighting is taking place, civilian casualties (based on hospital admissions) have increased dramatically. While military officials are quick to argue that these increased deaths are largely the result of insurgent actions (and they are correct), this doesn’t mean that local Afghans don’t blame the United States and NATO. Recent opinion polling of Afghans suggests they do.

In fact, efforts to protect the Afghan population are failing. According to recent press accounts, insurgents in Kandahar province have carried out a successful campaign of intimidation and violence against local citizens. Writing recently in the Christian Science Monitor, Julius Cavendish reported that Taliban assassinations of officials there have become so pervasive that more than 600 government jobs remain unfilled. The situation in Marja—ten months after US and Afghan troops entered the town in what was billed as a major element of the military’s counterinsurgency strategy—also remains challenging.

Beyond the obvious difficulty of making protecting lives a priority in an active war zone, it has become evident over the past year that Afghanistan is a spectacularly poor choice for a population-centric COIN campaign. An effective counterinsurgency relies, in large measure, on a competent and legitimate host country government. Afghanistan has neither. Corruption is so bad that the UN Office on Drugs and Crime has estimated that "drugs and bribes are the two largest income generators in Afghanistan."

Afghanistan’s security services are unable to hold areas cleared by US forces, and Afghan police are even less effective and lack the support of a functioning legal system. Making matters more difficult is the continued presence of undisturbed Afghan Taliban safe havens across the border in Pakistan. Finally, there are the Taliban, who have demonstrated a brutal willingness to use violence to cow the civilian populace into not cooperating with the United States, NATO or their own government.

A successful COIN campaign in Afghanistan is predicated on an extraordinary confluence of events: first, militarily reversing the Taliban’s momentum without causing significant civilian casualties; second, standing up the Afghan government and security forces so they are able to hold and build on the gains made by Western troops; third, minimizing the impact of Taliban sanctuaries across the border, in part by convincing the Pakistani government to reduce its support for them. And all this must be done while operating under Obama’s July 2011 deadline for commencing the withdrawal of US troops.

Achieving these goals demands an enormous political will and outlay of resources. Even by the measure of the military’s counterinsurgency doctrine, at approximately 130,000 US and NATO troops and a US outlay of an estimated $100 billion a year, the current effort is underresourced. Achieving these goals would require not only tens (perhaps hundreds) of thousands of more troops; it would require a stated commitment to maintain the fight for years to come. After nine years of war, these are elements that seem increasingly in short supply.

To be sure, many of these arguments have been made for some time by COIN skeptics. But what is most revelatory is that the US military seems to be finally getting the message. In the media uproar over the Rolling Stone article that sank the career of General McChrystal, often overlooked was the recounting of his conversation with US soldiers in southern Afghanistan who complained that restrictive rules of engagement made it nearly impossible for them to do the job they were trained for—killing the enemy. With no sense of irony, McChrystal complained, "This is the philosophical part that works with think tanks…but it doesn’t get the same reception from infantry companies."

McChrystal’s replacement, Gen. David Petraeus, has adopted a far different approach to the war. While efforts at nation-building continue hesitantly, the biggest change has been in the use of direct military action against the insurgents. Air power, which for a while was minimized because of the risk of civilian casualties, has increased dramatically. According to the blog Danger Room, "The U.S. and its allies have unleashed a massive air campaign in Afghanistan, launching missiles and bombs from the sky at a rate rarely seen since the war’s earliest days. In November alone, NATO aircraft launched 850 missions—three and a half times more than the same period in November 2009. Petraeus is also using more Special Operations forces in targeting Taliban commanders; and night raids, which have particularly inflamed the Karzai government, have recently tripled. In the three-month period ending October 21, Special Forces units conducted more than 1,500 operations, killing or capturing 339 insurgent leaders and more than 3,400 foot soldiers. NATO officials now hand out daily updates boasting about the number of insurgents killed or captured.

News reports leaked by military officials to reporters today do not speak of shuras convened, schools opened or corruption battled but instead boast of insurgents eliminated. The shift in tactical approach was perhaps an effort to show "progress" before the White House’s planned December review of Afghan policy. But it should also be seen as something else: a more accurate reflection of how counterinsurgency conflicts—even modern ones—are waged. As a COIN theorist and military official said to me, "COIN is a form of warfare and thus involves violence. Don’t be fooled by the fact that Petraeus found some useful idiots to make it sound more palatable and humanitarian."

The best evidence of this comes, ironically, from the US experience in Iraq. What led to the decrease in violence there was not an enhanced focus on protecting civilians; it was a confluence of factors: the extraordinary bloodletting in and around Baghdad that led to the forced separation of Sunnis and Shiites into ethnic enclaves; the flight of, by some estimates, 5 million Iraqis who have been internally and externally displaced; the paying off of, and support for, Sunni tribes who took on Al Qaeda in Iraq. And it wasn’t just Iraqi-on-Iraqi violence—the number of civilians killed by US troops rose dramatically after the surge and the adoption of supposedly civilian-friendly measures. Four times as many Iraqis were killed by US airstrikes, which rose sevenfold after Petraeus took command in Iraq. While certainly there were US efforts to capture Iraqi hearts and minds, no one can reasonably argue that those measures were decisive.

According to Jeff Michaels, a COIN expert and research associate at King’s College in London, "Killing lots of people was a key element of the Iraq surge, and so this [war in Afghanistan] is not too dissimilar. Indeed, too much emphasis has been placed on FM 3-24 rather than the real script they are operating from, which is to say that the Iraq case illustrated a considerable divergence from the theory."

What is happening in Afghanistan is an embrace of the aggressive approach to counterinsurgency once publicly dismissed by FM 3-24 advocates. This is not to suggest that US and NATO forces in Afghanistan have given up on trying to reach hearts and minds. But their embrace of techniques they once argued against is an implicit acknowledgment that the population-centric tactics of FM 3-24 have only marginal effectiveness in a nonpermissive environment like that of Afghanistan today. Like so many counterinsurgents before them, US generals are finding that the carrot is far less effective than the stick.

Their actual approach bears startling resemblance to the smaller-military-footprint counterterrorism strategy outlined by Vice President Biden during last year’s strategic review debates. Put aside for now are dreams of state-building in the Hindu Kush or the belief that only by turning the people away from the insurgents can America secure its interests. Instead, military planners have shifted their focus to an end-game strategy of using lethal force to drive the Taliban to the negotiating table.

The shift in emphasis toward a more traditional conflict is compelling evidence of the disconnect between the theory of population-centric COIN and actual US capabilities—and an unstated recognition that FM 3-24 has so far not succeeded. This hasn’t stopped COIN advocates from arguing that the shift in military emphasis is all part of the larger COIN effort; after all, they claim, direct military action is a crucial element of counterinsurgency. But these are self-serving and deceptive arguments, intended in part to mask the failure of the military to capture Afghan hearts and minds.

All of this matters for the future of US national security strategy. Much of the rationale for escalation in Afghanistan was based on the story of "success" in Iraq, and in particular on the supposed effectiveness of more population-sensitive counterinsurgency strategies executed there. But what should really be taken away from the US military’s experience over the past ten years is not that the United States understands how to fight and win population-centric counterinsurgencies but that counterinsurgencies are as violent and inconclusive as any other conflicts, and that the United States should avoid such wars at all costs.