“How many troops did the United States need in Iraq?” is probably destined to become an iconic foreign policy question, like “Who lost China?” or “Could we have prevented the 9/11 attacks?” Like most such questions, its meaning exceeds mere words; political agendas, future elections and personal reputations are all bound up with it.

Today’s conventional wisdom holds that civilian officials rejected requests from uniformed officers for additional forces and that the resulting troop shortage led to the current insurgency. This is a parsimonious storyline, as such lines go, but the issue is really much more complex. To understand what actually happened, one must recall how specific officials responded, and what motivated them, at specific points in time.

In truth, the “how many troops” issue evolved in three distinct phases—the run-up to the war, combat, and the postwar insurgency—and often took separate tracks in the military, political and public arenas. The issue is important because the question about troop levels in Iraq is not just about what happened in the past; it’s about the future, too. It concerns how the United States can respond to conflicts in the post-Cold War, post-9/11, post-Iraq world, and the U.S. strategy that would organize that response.

Phase One: What Will It Cost?

By early 2003 most of Washington assumed that the United States would probably launch military operations against Iraq. Congress had passed its resolution authorizing military force in October 2002. President Bush had said he would not wait indefinitely for Iraq to comply with UN resolutions requiring it to disarm. U.S. Central Command had conducted exercises in the theater, and troops were assembling for deployment.

During this period, public discussions about force levels were mainly linked to the cost of the impending war. The reason for this was that opponents of the war were working in a difficult political environment. Most Americans favored military intervention; a February 2003 Gallup Poll put public support at 73 percent. Opponents needed to reframe the issue to gain political traction, and one way to do that was to highlight the potential costs, which were directly linked to troop requirements.

Senator Carl Levin (D-MI) raised the cost issue in a hearing of the Senate Armed Services Committee on February 25, 2003. Levin asked General Eric Shinseki, the Army Chief of Staff, how many troops he thought would be needed for the impending operation. Shinseki answered, “Something on the order of several hundred thousand soldiers.”¹1.
Eric Schmitt, “Pentagon Contradicts General On Iraq Occupation Force’s Size”, New York Times, February 28, 2003.
Critics of the war have since enshrined this estimate as the number of forces that should have been provided for the war and occupation.

But to really understand this exchange, it is important to remember who was asking and who was answering it. Levin was one of 23 senators who had voted against the resolution authorizing the use of military force. Also, as Army Chief of Staff, Shinseki had no direct responsibility for planning the pending operation; that belonged to General Tommy Franks, head of Central Command. Levin’s question was designed to produce an estimate of war costs from a military expert that would raise concerns about the operation, not to give the Defense Department an opportunity to ask for more troops.

The Levin-Shinseki exchange triggered a debate that played out largely in the media. The debate was fueled by the reluctance of Bush Administration officials to provide their own estimate for the cost of the impending war. Any figure was likely to be used to portray a worst-case scenario just as they were trying to muster support. Two days later, a reporter asked Defense Secretary Donald Rumsfeld about Shinseki’s comment. “There are so many variables”, he said, “that it is not knowable. However, I will say this. What is, I think, reasonably certain is, the idea that it would take several hundred thousand U.S. forces, I think, is far from the mark.”²2.
Defense Department transcript, February 27, 2003.

Except for a few cases, Administration officials refrained from engaging opponents by hazarding an estimate. The most notable exception was Lawrence Lindsey, head of the White House’s National Economic Council. Lindsey had told a Wall Street Journal reporter that projected war costs had an “upper bound” of 1 to 2 percent of U.S. GDP, which the reporter translated into $100–200 billion.³3.
Bob Davis, “Bush Economic Aid Says the Cost of Iraq War May Top $100 Billion”, Wall Street Journal, September 16, 2002.
Even this rough, implied estimate was reported to have displeased senior Administration officials.⁴4.
Six months later, soon after Lindsey resigned, a reporter at a White House briefing asked Press Secretary Ari Fleischer to give an estimate of the war’s costs. The reporter said, “The recently departed Larry Lindsey put forward an estimate back in December [sic] based on a percentage of GDP, which was in line with the spending…” Fleischer quickly shot back, “Are you asking me to follow the example and be recently departed?” See “Press Briefing by Ari Fleischer”, Office of the Press Secretary, February 26, 2003.

Opponents of the war made clear their frustration with not being able to get a more specific cost estimate from the Administration. Representative James Moran (D-VA) told Deputy Secretary of Defense Paul Wolfowitz at a House Budget Committee hearing, “I think you’re deliberately keeping us in the dark.” He continued by questioning Wolfowitz’s candor, saying, “We’re not so naive as to think that you don’t know more than you’re revealing.”⁵5.
Schmitt, New York Times, February 28, 2003.

For his part, Shinseki appears to have understood the link between troop levels and costs. A few days before, when he was preparing to testify before the Senate Appropriations Committee, he reportedly cut off an aide who tried to remind him of the Administration’s position that costs were uncertain. “We know how many troops are there now, and the projected numbers”, he said. “We know how much it costs to feed them every day. We know how much it cost to send the force there. We know what we have spent already to prepare the force and how much it would cost to bring them back.”⁶6.
James Fallows, “Blind Into Baghdad”, The Atlantic (January/February 2004).

In sum, opponents of the war with Iraq needed an issue that emphasized how costly the war might be, and Administration officials, naturally enough, refused to help them. The resulting debate centered on costs, costs that depended on how many troops were needed. That was how the issue was first raised in the public.

Phase Two: Defeating Iraq

Once the war began, the troop-level question shifted focus to whether the United States had enough forces to defeat the Iraqi army. It is often forgotten today, but in March 2003 many experts worried that U.S. force levels were inadequate for the coming battle for Baghdad. The Iraqi army was supposedly among the world’s best in building prepared defenses (Iran had discovered this in the 1980s). Also, Saddam Hussein had saved his best troops for defending his capital, and U.S. military planners still believed that Iraq had chemical or biological weapons.

Criticism peaked at the end of March when a dust storm slowed the ground offensive and Iraqi irregulars attacked the Coalition’s supply lines. Reporters embedded with the troops covered these developments first-hand. Media experts and retired military officers soon began to fault the war plan in newspapers and on television, claiming that U.S. planners had not provided enough troops.

Retired General Barry McCaffrey was one of the most visible and influential of these critics. McCaffrey, a successful commander in the 1991 Gulf War and greatly respected in the Army, told a reporter that “there should have been a minimum of two heavy divisions and an armored cavalry regiment on the ground—that’s how our doctrine reads.”⁷7.
Vernon Loeb and Thomas E. Ricks, “Questions Raised About Invasion Force”, Washington Post, March 25, 2003.
He added in an op-ed a few days later: “We face a war of maneuver in the coming days to destroy five Iraqi armor divisions with only one U.S. armored unit, supported by the modest armor forces of the First Marine Division and the Apache attack helicopters of the 101st Airborne.” He also warned that U.S. forces had “inadequate tube and rocket artillery to provide needed suppressive fires . . . .”⁸8.
McCaffrey, “A Time to Fight”, Wall Street Journal, April 1, 2003.

In the event, of course, there was no war of maneuver for Baghdad. The Iraqi army collapsed, U.S. casualties were light, and most of the media quickly forgot about the controversy.

In reality, this phase of the “how many troops” controversy was just the latest round in a debate between advocates of speed and agility and advocates of mass and firepower that is almost as old as land warfare itself. Franks used a “rolling start”, sending units into battle as they arrived in theater, rather than wait until he could concentrate them. One can point to cases in which both approaches have worked. But as McCaffrey implied, Army doctrine has generally favored mass and firepower, so Franks was operating contrary to doctrine.

But to fully understand the criticism of the U.S. strategy at the time (and criticism of Donald Rumsfeld by several retired generals since), one must consider the larger picture. Rumsfeld’s desire to “transform” the U.S. military into a faster, more agile force—at the expense of mass and firepower—was so strong that more orthodox officers had been galvanized to resist these reforms from the time he entered office. This resistance was so intense that some press reports in the summer of 2001 speculated Rumsfeld would leave office. Criticism abated after 9/11 and the seemingly successful campaign in Afghanistan, but underlying opposition remained. It resurfaced when the Iraq offensive slowed in March and subsided after Coalition forces captured Baghdad. It resurfaced again as the insurgency became a major problem. In each case, anyone favoring larger forces claimed vindication whenever U.S. forces appeared inadequate, which is to say, whenever they ran into difficulty—whatever the actual reasons for the difficulty may have been.

Critics have also argued that civilian leaders pressured Franks to use fewer troops than he wanted. The evidence does not support this. Franks’ war plan was consistent with Rumsfeld’s desire for transformation, but in his memoir, American Soldier, Franks claims this strategy as his own; this was how he wanted to carry out the operation. And according to Franks, there were arguments for an even smaller force than he proposed, as well as for a larger one. Franks recalls that some Pentagon civilians floated the idea of taking Iraq with just a few special operations forces, air strikes and Iraqi opponents of Saddam. Retired General Wayne Downing, deputy national security advisor during 2001 and 2002 and former head of Special Operations Command, also reportedly favored this approach.⁹9.
Michael Dobbs, “Old Strategy on Iraq Sparks New Debate”, Washington Post, December 27, 2001; and “Battle Plans for Iraq”, New York Times, July 6, 2002 [unsigned editorial]. But Franks says he rejected their proposals—just as he rejected proposals from others for a larger force.

Franks recalls that Secretary of State Colin Powell, himself a retired four-star general who has favored large forces, argued for more troops. Franks says that Powell made his case to the President, but the President supported Franks. Three years after the war Powell recalled in a television interview, “I made the case to General Franks and Secretary Rumsfeld before the President that I was not sure we had enough troops.” He added, though, “A judgment was made by those responsible that the troop strength was adequate.”¹⁰10.
Libby Quaid, “Powell Forces Rice to Defend Iraq Planning”, Associated Press, April 30, 2006.
Loyal public servant that he was, and is, Powell saluted and went back to work.

It was not simply that Franks did not want more troops; the problem was that adding troops would have compromised his basic concept of attack, which was to strike quickly, prevent damage to the oil fields (which were believed essential to financing Iraq’s recovery), and minimize losses to both Coalition forces and Iraqi civilians. Adding troops would have slowed the attack, especially if they were provided a typical support tail of proportionate size. In other words, the desire for speed dictated leanness, and this leanness was reflected in the overall troop levels. As General William Wallace, who had commanded the U.S. Army’s V Corps, explained to a reporter, “I don’t know any commander who would refuse additional troops. But I didn’t request more troops because the support structure might have been inadequate.”¹¹11.
“General William S. Wallace reflects on invasion of Iraq”, TRADOC News Service, March 14, 2006.

In their history of the Iraq war, Michael Gordon and Bernard Trainor make much of the fact that Lieutenant General David McKiernan added more troops to Central Command’s initial plans after joining Franks’ staff in September 2002.¹²12.
Michael R. Gordon and General Bernard E. Trainor, Cobra II: The Inside Story of the Invasion and Occupation of Iraq (Pantheon Books, 2006), pp. 74–94.
But this is really no surprise, nor is it necessarily as significant as they suggest. Plans always change as they get “chopped” in coordination. The more important point is that Franks oversaw the plan’s evolution as his staff made additions and modification, and the final plan remained true to the initial “commander’s concept” of using a light force to strike rapidly and then keep the enemy off balance by attacking continuously.

A question sometimes raised is whether defeating the Iraqi army more decisively might have averted the insurgency that took hold later. Some experts—including McCaffrey—had warned of a guerrilla war and that additional conventional forces were therefore needed for a decisive “takedown.” But additional combat forces could not have eliminated the insurgents because, for the most part, the insurgents did not yet exist. According to analysts who interviewed former Iraqi leaders for the Iraqi Perspectives Project (a study commissioned by Joint Forces Command after the war),

Since the end of the war one question that has come up regularly is whether the regime made plans to continue the conflict through the insurgency the United States is currently combating. As far as can be determined through interviews . . . and the tens of thousands of records reviewed so far, there were no national plans to transition to a guerilla war in the event of defeat. . . . Nor, as their world crumbled around them, did the regime appear to cobble together such plans.13

Iraqi paramilitary forces, like the al-Quds Army and Fedayeen Saddam, were established to protect the regime from internal enemies, namely the Shi‘a and Kurds, who had rebelled after the 1991 war. When pressed into battle against Coalition forces, these groups were sometimes fanatical, but almost entirely ineffective. According to Franks, when U.S. forces encountered these paramilitaries, they tracked them back to their staging points (for example, a regional Ba‘ath Party headquarters) and destroyed them en masse. But if the paramilitaries (and regular troops who shed their uniforms) decided to blend into the population, additional conventional forces would not have helped target them. Making the rubble bounce with more tanks, rockets and tube artillery would not have destroyed them, and such tactics would have killed more civilians, making a counterinsurgency campaign even harder to win. It is important to recall that McCaffrey specifically recommended armored and mechanized forces, not dismounted soldiers—the kind one needs most in counterinsurgency operations.

In sum, much of the criticism of the numbers of U.S. troops used in the assault on Iraq was based on a disagreement over military doctrine and a misunderstanding of how Central Command developed its war plan. Although that plan succeeded, it was ineffective against the insurgency that was about to develop. But the additional forces critics proposed at the time would not have eliminated the insurgency, and could easily have made conditions even worse.

Phase Three: Occupying Iraq

It is clear now that the United States was inadequately prepared for the occupation. But the main problem was not just or mainly inadequate troop numbers; it was more fundamental than that. There was simply a stunning lack of realistic planning for postwar stabilization, and without such a plan, it was impossible to know how many troops were needed. Without a plan, too, any difference that additional troops might have made is pure speculation.

Some critics claim that plans for a potential occupation were available, but that the Defense Department ignored them. The evidence does not support this. To the contrary: Not only did the Defense Department fail to plan systematically for an occupation, no one else did either.

The State Department’s “Future of Iraq Project”—often cited as a blueprint for reconstruction—said nothing about how many troops would be needed to secure the country. The project report consisted mainly of policy papers outlining broad goals for the postwar period, like eliminating the Iraqi army as a threat to neighbors and democratizing the Iraqi government. It did not define specific operational requirements, plans for achieving discrete objectives—or how many troops would be needed to do all this.

Another study sometimes cited as a “plan not followed” was prepared by a team headed by Conrad Crane and W. Andrew Terrill at the Strategic Studies Institute of the U.S. Army War College. This report, “Reconstructing Iraq”, anticipated the situation we now face, including a detailed analysis of the major groups and factions likely to compete for power after Saddam fell. It also provided a checklist of activities needed for an occupation. But it, too, stopped short of translating objectives (securing borders, protecting infrastructure and so forth) into hard plans and requirements for troop levels.¹⁴14.
Crane and Terrill, Reconstructing Iraq: Insights, Challenges, and Missions for Military Forces in a Post-Conflict Scenario (U.S. Army War College Strategic Studies Institute, February 2003).

The occupation needed the kind of planning that the combat phases of the operation received, where goals were used to identify missions, missions were analyzed to define requirements, and requirements were matched to assignments for specific units. Any such planning process, if properly done, soon makes clear if there is a mismatch between objectives and the resources provided to achieve them. It was this kind of systematic planning for the postwar environment that was never done—and the main reason is fairly obvious. No one planned for an occupation because no one—the Office of the Secretary of Defense, the uniformed services, the National Security Council, the State Department, the CIA—wanted an occupation. This had a ripple effect. For effective planning to have occurred, someone would have had to fight for the money, the people and the equipment needed for an occupation. With no advocate, serious planning for a long occupation simply did not happen. And with no planning, there was no detailed set of requirements for troops.

Before the war, some experts not actually involved in policy or operational planning did offer estimates of how many troops would be needed to occupy Iraq. But these were all top-level, seat-of-the-pants guesses based on experience and instinct, not rigorous analysis. The irony, however, is that many, if not most, experts who hazarded an opinion in early 2003 gave estimates for post-combat troop requirements quite close to the number of troops that were in fact deployed in the occupation.

For example, Crane and Terrill wrote in February 2003 that past U.S. military occupations suggested that “the number of troops that may be needed for an occupation of Iraq are somewhere around 100,000.” Philip Gordon and Michael O’Hanlon of the Brookings Institution wrote in April 2003 that “about 100,000 to 200,000 troops may be needed to police the country, find and destroy weapons of mass destruction, prevent remnants of Hussein’s loyalists from regrouping, maintain a northern presence to deter conflict among Kurds, Arabs and Turks, and train a new Iraqi military.” Scott Feil, a retired Army colonel experienced in peacekeeping operations, advised the Senate Foreign Relations Committee that 75,000 troops would be needed. The Army History Center estimated that 100,000 troops would be needed. The non-partisan Center for Strategic and Budgetary Assessments said that 20,000–90,000 troops was a “reasonable estimate” for occupation and peacekeeping duties. Others offered higher or lower estimates, but they were in the minority.¹⁵15.
Gordon and O’Hanlon, “No Easy Victory”, Washington Post, April 12, 2003; “Prepared Statement of Colonel Scott R. Feil (Ret.)”, Senate Foreign Relations Committee, February 11, 2003; for the Army History Center study, see Vernon Loeb, “100,000 Soldiers Needed to Rebuild Iraq after Invasion”, Washington Post, September 23, 2002; Steven M. Kosiak, “Potential Cost of a War with Iraq and Its Post-War Occupation”, Center for Strategic and Budgetary Assessments, February 25, 2003.

Why did the experts disagree? At least three factors were at work.

First, as noted, no one did a systematic analysis of requirements for troops in a postwar occupation. According to Lieutenant General Jay Garner, individual agencies started developing ideas for an occupation in October 2002, but no one tried to integrate them into an overall plan (which would have identified requirements for troops) until January 2003.¹⁶16.
Interview with Garner, July 27, 2003, part of the October 9, 2003 Frontline broadcast, “Truth, War, and Consequences.”
With no hard data, opinion and gut instinct filled the vacuum.

Second, different people drew on different personal experiences. Shinseki, for example, had served in the Balkans peacekeeping operations during the 1990s, and was extrapolating from his experience there. Officials such as Wolfowitz, on the other hand, had just seen the Taliban regime fall in Afghanistan in 2002, and that occupation, though incurring casualties, seemed at the time to be achieving its objectives with a very small force.

Third, the history of military occupations was ambiguous. James Dobbins of the RAND Corporation directed a study of past occupations that was completed just before the Iraq war, and drafts of the report circulated at the time the occupation was being considered. Dobbins himself had taken part in the recent occupations of Haiti, Somalia, Bosnia, Kosovo and Afghanistan.¹⁷17.
Dobbins, John G. McGinn, Keith Crane, Seth G. Jones, Rollie Lal, Andrew Rathmell, Rachel M. Swanger and Anga Timilsina, America’s Role in Nation Building: From Germany to Iraq (RAND Corporation, 2003).
Gordon and Trainor cite this report, saying that the message of the Dobbins study was that “large peacekeeping forces were better than smaller ones.”¹⁸18.
Gordon and Trainor, p. 477.
They also say that if the United States had used the Balkans as a baseline—another occupation of a multi-ethnic region with a Muslim population—it would have deployed a peacekeeping force of 450,000 troops (about the number Shinseki proposed). In his memoirs, U.S. Ambassador L. Paul Bremer says he sent an executive summary of this study to Rumsfeld, but never received a response.¹⁹19.
Bremer with Malcolm McConnell, My Year in Iraq: The Struggle to Build a Future of Hope (Simon & Schuster, 2006), pp. 9–10.

However, a look at the RAND study suggests that the lessons of past occupations are not at all clear-cut. The success of a peacekeeping operation and the casualties incurred depend on factors whose interplay, to say the least, is complex. In addition to the size of the occupying force, one needs to acknowledge many variables to capture all of the differences among the cases:

• Was the adversary defeated, or was there a continuing insurgency?
• Was the occupied territory ethnically homogeneous or diverse?
• Was the goal simply displacing a dictator, or creating a functioning democracy?
• Was partition an option? A federation? Or is the desired goal a unitary state with a strong central authority?
• Was the environment rural or urban? Distant or close to the United States?
• How large was the population? How large was the occupied territory, how long were the borders, and were the neighboring countries hostile or cooperative?
• Could the United States exit quickly if necessary, or did circumstances oblige it to stay, casualties notwithstanding?

Most of these factors did not favor an easy occupation in Iraq (multi-ethnic, large territory and population, hostile neighbors, distant from the United States, and not presenting an opportunity for an early exit if things went poorly). But it was difficult to extrapolate from that experience to a hard number of required troops. The range of uncertainty is significant.

For example, U.S. forces had a hundred soldiers for each thousand Germans in the post-World War II occupation, but just five soldiers for each thousand Japanese. Both occupations were successful. Replicating the German ratio in Iraq would require 2.6 million soldiers—an impossible goal—but replicating the Japanese ratio would require only 150,000 soldiers, about the number of troops deployed in Iraq today.

Similarly, both the Balkans and Afghanistan are ethnically diverse areas with Muslim populations, and at the time Dobbins published his study, the occupations of both seemed to be proceeding reasonably well. The ratio of troops to population in the Balkans operations was 20:1,000; in Afghanistan it was less than 1:1,000. What is the resulting rule of thumb? To further complicate matters, Iraq resembles Bosnia and Kosovo in the degree to which each is urbanized, but it more closely resembles Afghanistan in size of territory and population.

The most important factor in U.S. planning seems to have been that the responsible officials anticipated that the Iraqi regime and army would capitulate, and the government bureaucracy and population would then cooperate in managing affairs during the occupation. They did not. Gordon and Trainor rightly observe that U.S. postwar planning hinged on this key premise, and remark—also rightly—that “rarely has a military plan depended on such a bold assumption.”²⁰20.
Gordon and Trainor, p. 105.

But again, there was conflicting experience and evidence. During the 1991 Gulf War, the Iraqi army capitulated in droves, with 85,000 Iraqi soldiers surrendering to become prisoners of war. Large numbers of both Shi‘a and Kurds rebelled against the Ba‘athi regime, but the regime did not disappear. It was at least plausible to believe that events would follow a similar course in 2003. Indeed, it wasn’t just American officials who thought the Iraqi regime would capitulate instead of evaporate; many Iraqis did, too. Iraqi officers told American interviewers that they had believed the war would be short and the Coalition would quickly win. Their war plan consisted mainly of avoiding death. According to one Iraqi colonel, “We wanted the Americans to come quickly and finish the war rapidly.”²¹21.
Woods et al., p. 97.

Even Saddam Hussein thought collapse was possible. According to the Iraqi Perspective Project, he feared his regime coming apart more than he feared a military defeat by Coalition forces—and with good reason based on earlier experience. That was why he focused so much on internal controls—he feared Shi‘a and Kurdish rebellion, or a plot by the army.

In effect, everyone agreed: Iraqi capitulation, not evaporation, was plausible and even likely. U.S. planners were counting on it, Saddam was afraid of it, and Iraqi officers just wanted to get it over with as quickly as possible. Alas, it did not happen. Instead of capitulating, the regime dissolved. U.S. policy then compounded the problem by disbanding the Iraqi army and banning Ba‘athis from the civil service, thus eliminating whatever structures might have held the remnants of the government together.

Moreover—and this has been pointed out before—Germans and Japanese citizens cooperated with Allied occupiers because their countries were utterly devastated. Recall photographs of Berlin or Tokyo after the war. It did not matter in 2003 if the United States deployed 100,000 troops or 500,000 troops; no U.S. leader was going to allow American forces to inflict that kind of death and destruction on Iraq.

Finally on this point, Germany and Japan each had a regime that, evil or not, commanded the allegiance of most citizens. After the bombing of Hiroshima and Nagasaki, Emperor Hirohito instructed his subjects “to endure the unendurable and bear the unbearable.” No Iraqi leader could have delivered a similar message to Iraqis, and few Iraqis would have paid the slightest attention in any case, especially those who saw themselves oppressed by a regime not of their choosing.

Sustainable Capacity

Better planning might have revealed the crucial assumptions that the capitulation scenario depended on. It might also have raised the most important question of all: Even if U.S. officials had believed they needed more troops on the ground, where did they think they would find them?

Disagreements over troop numbers have almost always missed the crux of the problem. It is inherently difficult, even for a country as powerful as the United States, to remove a government and occupy a politically unstable, hostile territory the size of Iraq located on the other side of world. Hard numbers make this clear. Consider the situation facing the Army, which, along with the Marines, is bearing the brunt of the occupation. The Army has just over a million soldiers. Of these, 340,000 are in combat units; the other 700,000 are in support and administration units, or are in training or between assignments. Half of the combat units—170,000 soldiers—are active Army divisions and brigades. Under Army guidelines, they should deploy no more than one year out of three. The other 170,000 soldiers assigned to combat units are National Guard that are supposed to deploy no more than one year out of six.²²22.
Adam Talaber, Options for Restructuring the Army, Congressional Budget Office (May 2005).

This rotation is important to give the soldiers time to “reset”—that is, rest, recover and train. Active duty soldiers can deploy more frequently because they are expected to be available, and deployments round out their experience, which is important to their proficiency. Reserve and National Guard soldiers are expected to deploy less often because they have commitments to their civilian jobs.

Simple math says that if the Army keeps to its desired rotation rates, it can on average maintain a force of 85,000 soldiers in combat units abroad indefinitely as the soldiers cycle through their normal deployments.²³23.
The United States has maintained about 130,000 personnel in Iraq. This has included, in addition to 17 to 18 Army brigades, 26,000–27,000 Marines.
If this seems like a poor tooth-to-tail ratio for an organization with more than a million soldiers, keep in mind that no other country in the world has such a capability to take and hold territory far from its borders.

To deploy more soldiers, only three variables matter: stretching deployments, increasing their frequency or reducing time to reset. None alters the basic mathematics, and each carries a cost. Longer deployments and shorter time at home wear out both the equipment and the soldiers. Recruitment and reenlistments will suffer. One can issue “stop loss” orders and offer reenlistment bonuses (as we are today), but eventually it will be harder to attract people to an Army career if it means spending years overseas separated from family and, in the case of reservists and Guards, their jobs.

Alternatively, one can try to squeeze more efficiency out of the system—which the Army is doing. But according to the Congressional Budget Office, even the new “modular force” will increase the number of soldiers available for combat by only 5 percent.²⁴24.
Talaber, p. xii.
As for enlarging the Army, as some officials and experts propose, the CBO says that adding just 18,000–23,000 troops to sustain the current operations in Iraq would cost $18–19 billion up front and $23–29 billion annually—a sum about equal to a 5 percent increase in the total defense budget.²⁵25.
Statement of Douglas Holtz-Eakin before the House Armed Services Committee, “The Ability of the U.S. Military to Sustain an Occupation in Iraq”, Congressional Budget Office, November 5, 2003. The problem is that each deployed combat soldier in the modern Army requires three to four soldiers for support and to accommodate rotation, so there is a multiplier effect and costs rise quickly.

In short, it is hard to imagine how the United States could have deployed several hundred thousand troops in an occupation for any length of time. Critics who argue that the United States deployed 550,000 troops in Operation Desert Storm—often observing that the Army was one-third larger at the time—overlook the more important fact: These forces were mobilized for an operation that lasted only seven months. The limited length of deployment went hand-in-hand with the nation’s limited goal: defeating (but not eliminating) the Iraqi army and restoring the previously existing government of Kuwait.

It is also worth noting that Coalition partners provided 400,000 troops in Desert Storm. Most were from Muslim nations (118,000 Saudis, 40,000 Egyptians, 40,000 Emiratis and 25,000 Omanis, plus about 100,000 Turks poised along Iraq’s northern border). But in 2003 the mission was different—enforcing UN resolutions and regime change rather than liberation. U.S. policy also added to the challenge of finding partners for the occupation by barring countries that did not support the combat phase of the war from bidding on contracts for the reconstruction.

One last complication is that postwar stabilization operations require special skills: policing, civil affairs, public outreach and special operations against hardcore insurgents. Currently, about 80 percent of the soldiers with these specializations are reservists. This is a legacy of the Cold War. NATO planned to repel a Soviet invasion, push the attackers back into Eastern Europe, and then restore order in the Low Countries, Germany and perhaps parts of France—friendly, familiar territory. Then everyone would go home. Now we face almost exactly the opposite situation. Combat can require only a week or three, but an occupation can go on for years. It’s hard for the police officers, utility repairmen and local officials who serve in the Reserves to be away from their main jobs for long periods—long enough to put a self-employed professional or small-business owner out of business.

Partly because of the experience of Iraq, in November 2005 the Defense Department adopted a new policy to improve its capabilities for postwar stabilization.²⁶26.
Department of Defense Directive 3000.05, Military Support for Stability, Security, Transition, and Reconstruction (SSTR) Operations, November 28, 2005.
Under this policy, stability operations are to be “given priority comparable to combat operations.” But this does not resolve the larger issue: Where will the soldiers come from?

The plan for changing the regime in Iraq through military force was based on plausible, but crucial, assumptions. Despite concerns by many about the combat phases of the war, that part of the operation went extraordinarily well. The issue was always the “war after the war”, and that boiled down to a single question: Would the Iraqi government capitulate, and would its army and civil service cooperate in an occupation? If so, fine; if not, U.S. forces were entering a risky situation where time would not be on their side. It was sometimes said that U.S. leaders decided not to proceed to Baghdad during Operation Desert Storm because occupying Iraq would be like trying to occupy a country the size of California with the population of Canada.²⁷27.
Bruce Berkowitz, “Gulf War II? There’s a Better Way”, Wall Street Journal, February 5, 1998. Events have borne this out, and now we must deal with the consequences.

U.S. armed forces have a unique capability to carry out military operations throughout the world, and both the United States and the world as a whole benefit on balance from this capability. It is the key to American predominance, and American predominance is, on the whole, a good thing, especially given the possible alternatives. But this capability is not limitless. That is why U.S. leaders need to be prudent, realistic and candid with each other and the American people when deciding to use it.

1.
Eric Schmitt, “Pentagon Contradicts General On Iraq Occupation Force’s Size”, New York Times, February 28, 2003.

2.
Defense Department transcript, February 27, 2003.

3.
Bob Davis, “Bush Economic Aid Says the Cost of Iraq War May Top $100 Billion”, Wall Street Journal, September 16, 2002.

4.
Six months later, soon after Lindsey resigned, a reporter at a White House briefing asked Press Secretary Ari Fleischer to give an estimate of the war’s costs. The reporter said, “The recently departed Larry Lindsey put forward an estimate back in December [sic] based on a percentage of GDP, which was in line with the spending…” Fleischer quickly shot back, “Are you asking me to follow the example and be recently departed?” See “Press Briefing by Ari Fleischer”, Office of the Press Secretary, February 26, 2003.

5.
Schmitt, New York Times, February 28, 2003.

6.
James Fallows, “Blind Into Baghdad”, The Atlantic (January/February 2004).
7.
Vernon Loeb and Thomas E. Ricks, “Questions Raised About Invasion Force”, Washington Post, March 25, 2003.

8.
McCaffrey, “A Time to Fight”, Wall Street Journal, April 1, 2003.
9.
Michael Dobbs, “Old Strategy on Iraq Sparks New Debate”, Washington Post, December 27, 2001; and “Battle Plans for Iraq”, New York Times, July 6, 2002 [unsigned editorial].
10.
Libby Quaid, “Powell Forces Rice to Defend Iraq Planning”, Associated Press, April 30, 2006.

11.
“General William S. Wallace reflects on invasion of Iraq”, TRADOC News Service, March 14, 2006.

12.
Michael R. Gordon and General Bernard E. Trainor, Cobra II: The Inside Story of the Invasion and Occupation of Iraq (Pantheon Books, 2006), pp. 74–94.

13.
Kevin M. Woods with Michael R. Pease, Mark E. Stout, Williamson Murray and James G. Lacey, Iraqi Perspectives Project: A View of Operation Iraqi Freedom from Saddam’s Senior Leadership: Study by the Joint Center for Operational Analysis (Institute for Defense Analysis, 2006), p. 149. Some journalists reported that Saddam planned the insurgency in some significant way; see, for example, Joe Klein, “Saddam’s Revenge”, Time, September 25, 2005. The Iraqi Perspectives Project appears to refute this.
14.
Crane and Terrill, Reconstructing Iraq: Insights, Challenges, and Missions for Military Forces in a Post-Conflict Scenario (U.S. Army War College Strategic Studies Institute, February 2003).
15.
Gordon and O’Hanlon, “No Easy Victory”, Washington Post, April 12, 2003; “Prepared Statement of Colonel Scott R. Feil (Ret.)”, Senate Foreign Relations Committee, February 11, 2003; for the Army History Center study, see Vernon Loeb, “100,000 Soldiers Needed to Rebuild Iraq after Invasion”, Washington Post, September 23, 2002; Steven M. Kosiak, “Potential Cost of a War with Iraq and Its Post-War Occupation”, Center for Strategic and Budgetary Assessments, February 25, 2003.

16.
Interview with Garner, July 27, 2003, part of the October 9, 2003 Frontline broadcast, “Truth, War, and Consequences.”

17.
Dobbins, John G. McGinn, Keith Crane, Seth G. Jones, Rollie Lal, Andrew Rathmell, Rachel M. Swanger and Anga Timilsina, America’s Role in Nation Building: From Germany to Iraq (RAND Corporation, 2003).

18.
Gordon and Trainor, p. 477.

19.
Bremer with Malcolm McConnell, My Year in Iraq: The Struggle to Build a Future of Hope (Simon & Schuster, 2006), pp. 9–10.
20.
Gordon and Trainor, p. 105.

21.
Woods et al., p. 97.
22.
Adam Talaber, Options for Restructuring the Army, Congressional Budget Office (May 2005).

23.
The United States has maintained about 130,000 personnel in Iraq. This has included, in addition to 17 to 18 Army brigades, 26,000–27,000 Marines.

24.
Talaber, p. xii.

25.
Statement of Douglas Holtz-Eakin before the House Armed Services Committee, “The Ability of the U.S. Military to Sustain an Occupation in Iraq”, Congressional Budget Office, November 5, 2003.
26.
Department of Defense Directive 3000.05, Military Support for Stability, Security, Transition, and Reconstruction (SSTR) Operations, November 28, 2005.

27.
Bruce Berkowitz, “Gulf War II? There’s a Better Way”, Wall Street Journal, February 5, 1998.