who used the radio waves for classified information. In addition, the Iraqis made heavy use of buried telephone lines and motorcycle couriers.

          It was also difficult because Washington, concerned with avoiding any situation that could lead prematurely to war, decided not to fly any reconnaissance aircraft over Iraq or occupied Kuwait during Operation Desert Shield.

          CENTCOM's second reason for keeping the focus away from people counts was a fear of reliving the preoccupation with statistics on enemy strengths and casualties that developed during the Vietnam War. The commander of Operation Desert Storm had no intention of beginning another war defined by body counts. Still, there was a public demand for this kind of information in Operations Desert Shield / Desert Storm. To arrive at a figure for the number of Iraqi troops in the KTO, the Pentagon public affairs office did some simple arithmetic. The number of divisions in the theater -- 42 -- was known with some certainty. This number was multiplied by the number of men thought by intelligence analysts to comprise a division.

          The estimated personnel complement of an Iraqi division had been revised downward shortly before the estimate was made public. The revision had reduced the number of support troops thought to be associated with a division. When the multiplication was done, it produced an estimate of approximately 547,000 Iraqi troops in the KTO. This prompted some confusion because the earlier, higher estimate of men per division had leaked unofficially. The best estimate, however, was the figure of 547,000, which was not very good in any case. In fact, the one certainty is that there never really were 547,000 Iraqi troops in theater because - and this was not known until after the war - many units were sent to the theater substantially understrength.

          While CENTCOM rightly felt troop counts were not necessary, solid post - war information is very useful. Knowing how many of the enemy were killed is politically important; knowing how many Iraqi troops were in theater when the ground attack began is militarily important for future contingencies. A credible figure would help answer more definitively questions about the role of air power in the war. We would know how air power devastated the ground forces, either by killing them or by so affecting their morale that they deserted their posts and abandoned their equipment. We would know how strong an army the coalition ground forces swept aside.

          At this point, no one knows -- not even Saddam. Interviews with captured Iraqi officers revealed that many of them lied about their daily strength so their superiors would not know how miserably they had failed in keeping their units intact. Captured documents offer another source of information, but little analysis had been done at the time research for this analysis was conducted. Eventually, there may be a reasonably hard figure on the number of Iraqis killed and the number of soldiers in theater when the ground campaign started. In the meantime, an attempt can be made to make a closer estimate of opposing troop strengths and casualties.

          Table I below lists seven relevant categories for assessing Iraqi strength. Some of the estimates emerge from hard data, others are mere guesses. But the key point is that all seven categories must be considered together. Estimates of Iraqi dead cannot be intelligently dealt with in isolation from the estimates for deserters or escapees. Because we are starting with the figure of 547,000 as the notional number of Iraqi troops

<29>