There have been conflicting reports about the timing of American troop withdrawals from Iraq. Gen. George Casey, commander of U.S. forces there, has announced that the United States intends to begin a "fairly substantial" withdrawal of U.S. forces after the projected December elections establish a constitutional government. Other sources have indicated that this will involve 30,000 troops, or some 22 percent of U.S. forces in Iraq. Some high-level statements from Baghdad have indicated that the beginning of withdrawals may be delayed until next summer. On either schedule, progress is dependent upon improvements in the security situation and in the training of Iraqi forces.

A review of withdrawal strategy therefore seems in order. For one thing, how are the terms "progress" and "improvement" to be defined? In a war without front lines, does a lull indicate success or a strategic decision by the adversary? Is a decline in enemy attacks due to attrition or to a deliberate enemy strategy of conserving forces to encourage American withdrawal? Or are we in a phase similar to the aftermath of the Tet Offensive in Vietnam in 1968, which at the time was widely perceived as an American setback but is now understood as a major defeat for Hanoi?

For someone like me, who observed firsthand the anguish of the original involvement in Vietnam during the Kennedy and Johnson administrations, and who later participated in the decisions to withdraw during the Nixon administration, Casey's announcement revived poignant memories. For a decision to withdraw substantial U.S. forces while the war continues is a potentially fateful event. It affects the calculations of insurgents and government forces alike, so that the definition of progress becomes nearly as much a psychological as a military judgment. Every soldier withdrawn represents a larger percentage of the remaining total. The capacity for offensive action of the remaining forces shrinks. Once the process is started, it runs the risk of operating by momentum rather than by strategic analysis, and that process is increasingly difficult to reverse.

Despite such handicaps, the decision to replace U.S. forces with local armies during the Vietnam War -- labeled "Vietnamization" -- was, from the security viewpoint, successful on the whole. Between 1969 and the end of 1972, more than 500,000 U.S. troops were withdrawn. American involvement in ground combat ended in early 1971. U.S. casualties were reduced from an average of 400 a week in 1968 and early 1969 to an average of 20 a week in 1972.

These measures were possible because, after the failure of Hanoi's Tet Offensive, the guerrilla threat was substantially eliminated. Saigon and all other urban centers were far safer than major cities in Iraq are today. Saigon controlled perhaps 80 percent of the country with relatively well-established front lines. Vietnamese army units were increasingly able to repel offensives from the regular forces of Hanoi.

When the Vietnamese army, with substantial U.S. air support, broke the back of the North Vietnamese all-out offensive in 1972, Vietnamization could be judged a success. Shortly afterward the North Vietnamese accepted terms that they had rejected for four years. (That they did, however, does not settle the debate over whether a different withdrawal rate -- slower, faster or none at all until after a settlement -- could have speeded that day.) Three years later, these results were reversed, not because of internal violence but because of an external attack by Hanoi's conventional military force, in violation of every provision of the Paris agreement.

America's emotional exhaustion with the war and the domestic travail of Watergate had reduced economic and military aid to Vietnam by two-thirds, and Congress prohibited military support, even via airpower, to the besieged ally. None of the countries that had served as guarantors of the agreement was prepared to lift a diplomatic finger.

All this demonstrated two principles applicable to Iraq: Military success is difficult to sustain unless buttressed by domestic support. And an international framework within which the new Iraq can find its place needs to be fostered

History, of course, never repeats itself precisely. Vietnam was a battle of the Cold War; Iraq is an episode in the struggle against radical Islam. The stake in the Cold War was perceived to be the political survival of independent nation-states allied with the United States around the Soviet periphery. The war in Iraq is less about geopolitics than about the clash of ideologies, cultures and religious beliefs. Because of the long reach of the Islamist challenge, the outcome in Iraq will have an even deeper significance than that in Vietnam. If a Taliban-type government or a fundamentalist radical state were to emerge in Baghdad or any part of Iraq, shock waves would ripple through the Islamic world. Radical forces in Islamic countries or Islamic minorities in non-Islamic states would be emboldened in their attacks on existing governments. The safety and internal stability of all societies within reach of militant Islam would be imperiled.

This is why many opponents of the decision to start the war agree with the proposition that a catastrophic outcome would have grave global consequences -- a fundamental difference from the Vietnam debate. On the other hand, the military challenge in Iraq is more elusive. Local Iraqi forces are being trained for a form of combat entirely different from the traditional land battles of the last phase of the Vietnam War. There are no front lines; the battlefield is everywhere. We face a shadowy enemy pursuing four principal objectives: (1) to expel foreigners from Iraq; (2) to penalize Iraqis cooperating with the occupation; (3) to create a chaos out of which a government of their Islamist persuasion will emerge as a model for other Islamic states; and (4) to turn Iraq into a training base for the next round of fighting, probably in moderate Arab states such as Egypt, Saudi Arabia and Jordan.

North Vietnamese forces possessed heavy weapons, had sanctuaries in adjoining countries and numbered at least a half-million trained troops. Iraqi insurgents number in the tens of thousands and are lightly armed. Their most effective weapon is a homemade explosive, their most effective delivery system the suicide bomber and their most frequent targets unarmed civilians.

The Iraqi population has shown extraordinary equanimity in the face of this deliberate and systematic slaughter. In the end, its perception will determine the outcome as much as the military situation does. It will know how secure it is; it will determine the sacrifices it is prepared to make.

In essence, the Iraq war is a contest over which side's assessment turns out to be correct. The insurgents are betting that by exacting a toll among supporters of the government and collaborators with America, they can frighten an increasing number of civilians into, at a minimum, staying on the sidelines, thereby undermining the government and helping the insurgents by default. The Iraqi government and the United States are counting on a different kind of attrition: that possibly the insurgents' concentration on civilian carnage is due to the relatively small number of insurgents, which obliges them to conserve manpower and to shrink from attacking hard targets; hence, the insurgency can gradually be worn down.

Because of the axiom that guerrillas win if they do not lose, stalemate is unacceptable. American strategy, including a withdrawal process, will stand or fall not on whether it maintains the existing security situation but on whether the capacity to improve it is enhanced. Victory over the insurgency is the only meaningful exit strategy.

The quality of intelligence will be crucial. Specifically, these issues require attention: How do we assess the fighting capacity of the insurgents and their strategy? To what level must attacks on civilians be reduced, and over what period, before a province can be described as pacified? What is the real combat effectiveness of Iraqi security forces, and against what kind of dangers? To what extent are the Iraqi forces penetrated by insurgents? How will Iraqi forces react to insurgent blackmail -- for example, if a general's son is kidnapped? What is the role of infiltration from neighboring countries? How can it be defeated?

Experience in Vietnam suggests that the effectiveness of local forces is profoundly affected by the political framework. South Vietnam had about 11 divisions, two in each of the four corps areas and three others constituting a reserve. In practice, only the reserve forces could be used throughout the country. The divisions defending the provinces in which they were stationed and from which they were recruited were often quite effective. They helped defeat the North Vietnamese offensive in 1972. When moved into a different and unfamiliar corps area, however, they proved far less steady. This was one of the reasons for the disasters of 1975.

The Iraqi equivalent may well be the ethnic and religious antagonisms between Sunnis, Shiites and Kurds. In Vietnam, the effectiveness of forces depended on geographic ties, but the provinces did not perceive themselves in conflict with each other. In Iraq, each of the various ethnic and religious groupings sees itself in an irreconcilable, perhaps mortal, confrontation with the others. Each group has what amounts to its own geographically concentrated militia. In the Kurdish area, for example, internal security is maintained by Kurdish forces, and the presence of the national army is kept to a minimum, if not totally prevented. The same holds true to a substantial extent in the Shiite region.

Is it then possible to speak of a national army at all? Today the Iraqi forces are in their majority composed of Shiites, and the insurrection is mostly in traditional Sunni areas. It thus foreshadows a return to the traditional Sunni-Shiite conflict, only with reversed capabilities. These forces may cooperate in quelling the Sunni insurrection. But will they, even when adequately trained, be willing to quell Shiite militias in the name of the nation? Do they obey the ayatollahs, especially Grand Ayatollah Ali Sistani, or the national government in Baghdad?

And if these two entities are functionally the same, can the national army make its writ run in non-Shiite areas except as an instrument of repression? And is it then still possible to maintain a democratic state?

The ultimate test of progress will therefore be the extent to which the Iraqi armed forces reflect -- at least to some degree -- the ethnic diversity of the country and are accepted by the population at large as an expression of the nation. Drawing Sunni leaders into the political process is an important part of an anti-insurgent strategy. Failing that, the process of building security forces may become the prelude to a civil war.

Can a genuine nation emerge in Iraq through constitutional means?

The answer to that question will determine whether Iraq becomes a signpost for a reformed Middle East or the pit of an ever-spreading conflict. For these reasons, a withdrawal schedule should be accompanied by some political initiative inviting an international framework for Iraq's future. Some of our allies may prefer to act as bystanders, but reality will not permit this for their own safety. Their cooperation is needed, not so much for the military as for the political task, which will test, above all, the West's statesmanship in shaping a global system relevant to its necessities.

The writer, a former secretary of state, is chairman of Kissinger Associates.

Commentary:
I counted seventeen question marks in Kissinger's letter. It would have been better for the pro war side to have thought of these questions before the war started. Now, it's too late.

Most wars lose public support after two years but Bush and the GOP needed the lies from the war (a connection to terrorism) to get them reelected.

While there are many options, the most important is that no option can succeed without public support and that's never going to happen again. This leaves us with only two major options; put in more troops so we can win (which the public is strongly against) or getting the hell out of there. Logic dicates the latter.