President Bush is finally getting over his version of the Vietnam syndrome.

"If you're 60 years old, you tend to be a product of the Vietnam era," Bush told me and other journalists in the Oval Office a few months ago when asked if we needed more troops in Iraq. "I remember the tactical decisions being made out of the White House during that period of time. I thought it was a mistake then, and I think it's a mistake now."

Bush will eat these words if he orders the troop "surge" into Baghdad that is considered skeptically by some of his top generals. He thought he was avoiding a mistake of the Vietnam War by deferring to his generals on troop levels, but he has only internalized an erroneous conservative belief about that conflict. Conservatives falsely think that it was the civilian leadership that lost the Vietnam War by restraining the military.

The true lesson of Vietnam is that the civilian leadership should exercise close supervision of the military and ensure that, when fighting an insurgency, it acts in ways that don't come naturally to a U.S. Army that is most comfortable when smashing a conventional enemy.

As Andrew F. Krepinevich Jr. recounts in his classic book on the military's failures in the war, "The Army and Vietnam," it was a civilian, President John F. Kennedy, who was prescient about the coming era of guerrilla warfare. He pushed the Army to learn counterinsurgency warfare, but it ignored him.

The civilian who bears the brunt of conservatives' ire is President Lyndon B. Johnson. He once bragged that "they can't bomb an outhouse without my approval" and imposed political constraints on the use of force. But in a limited war, such constraints are inevitable. The question is whether they make sense or not. Some of LBJ's limits were for sound reasons. We understandably feared provoking the Chinese by a too-wide-ranging bombing campaign in the North.

If LBJ meddled on the air campaign, he didn't meddle enough on the ground. When Gen. William C. Westmoreland wanted 200,000 troops in 1965, LBJ quickly ponied them up.

The problem was that the military didn't know how to win the war. It was clueless about counterinsurgency, which typically requires careful discrimination in applying firepower, light infantry undertaking intensive patrolling, and political action to undermine the basis of the insurgents' support in the population. Instead, it dreamed of replicating the conventional clashes of World War II.

Westmoreland wanted to attrit the Communists, but the Communists wanted to attrit us, and they had a much better understanding of whose will would be broken. So the military did a perfectly fine job of losing Vietnam all on its own. "Westmoreland himself,"historian Eliot Cohen writes, "operated under remarkably little civilian oversight."

Too late, Gen. Creighton Abrams, Westmoreland's replacement, emphasized pacification of populated areas and other classic counterinsurgency tactics. Together with more bombing in the North, they met with some success. "By 1970," historian Max Boot writes, "more than 90 percent of the South's population was under Saigon's control." But, by then, the U.S. was ready to quit the war.

In Iraq, Bush has been deferring to generals of widely varying quality. Some deserved deference, others didn't. The question of troop levels might seem a mere tactical issue, but it has vast strategic implications -- without enough troops, it is impossible to provide the security to the population that is one of the foundations of a sound counterinsurgency strategy. As it became clear that the military strategy in Iraq wasn't working, Bush stuck with it, partly on grounds that he didn't want to gainsay his generals, when he should have been firing them.

Now that he might order a surge, Bush will have to backtrack on his conviction that generals are best left alone. As he does, he should go back and understand the source of his mistake -- a misinterpretation of Vietnam.