Maybe to some extent or another - but the figures cited there out of proportion. the government is not always frank about commando casualties, and sometimes the government will hush up or perhaps give an erroneous reason for deaths. And incident during the second world war comes to mind, where a boat carrying a battallion or so of US GIs sank during a training exercise for D-Day. About 900 Americans died, but most of the British crew survived. Not good for allied unity, and so it wasn't reported on. In fact, the truth widely known until the 1970s.

The actual U.S. military losses in Iraq may have reached 2,000 personnel, more than twice the official figure of 900, as Washington badly understates its casualty statistics, a military diplomatic source told the Itar-Tass news agency.

"Official statistics do not include casualties among non-U.S. nationals who sign up to serve in the American armed forces in order to get a U.S. `green card.' According to reliable information the share of non-Americans in the U.S. force in Iraq may be as high as 60 per cent," the source said. "The real number of U.S. losses may be as high as 2,000 casualties and up to 12,000 wounded," the military diplomat said.

60 percent non US nationals? That simply isn't so, especially considering the ever-higher proportion of national guard units being deployed.

In this age of the internet and easily searchable databases, etc. I find it hard to believe that over a thousand US troops could just disappear.

Are there some deaths that won't be acknowledged? I'll wager, yes, but in the dozens not the hundreds. Has there been some messing around with the figures? Probably -- during the Fallujah/Sadr flare-up earlier in the year I think there was big spike of US casualties that was flattened out somewhat in the official accounts. But the full number was eventually reported, spread out over some days.

The 12,000 wounded figure is probably correct, but I think 1/2 of those qualified for purple hearts but were back in action in a week or less.