If you give a fuck at all that hundreds of thousands of Iraqi children are dead, if you give a fuck that tens of thousands more are about to die, if you give a fuck that tons of depleted uranium from weapons used against Iraq could (see edit) cause up to a half million cancer deaths and untold severe birth defects (and that's what's already thereGo
(
Read more... )
I find it interesting that your two primary articles dispelling the "myths" about depleted uranium both involve sources that have a vested interest in the continued deployment of weapons systems incorporating DU. In fact, one comes straight from an Army web site, and is based entirely on the claims of an Army "DU consultant". The US and other militaries want to use DU because of its superior characteristics as armor and antitank munitions. "'It's tactically a significant advantage,' to use DU instead of tungsten in armor-piercing rounds, [the Army consultant] said. M-1 Abrams rounds with DU can effectively engage targets at 3,000 meters, he said, adding that Tungsten rounds fired by the Iraqis in the Gulf War were only effective at about 2,000 meters."
You yourself state (and a report from the Army Environmental Policy Institute agrees) that uranium is a heavy metal and is toxic as such. It is ALSO radioactive. Your seeming implication that it's not really any worse than using lead because they're both toxic anyway and it's only a little bit radioactive glosses over the fact that the two negative effects are added together. Chemical toxicity PLUS radiation. Radiation exposure is defined by the Army's own report as a "stochastic health effect". Stochastic health effects are those defined as follows, quote:
* An increase in exposure raises the incidence of a naturally occurring disease but the severity of the disease is not related to the level of exposure.
* No exposure threshold.
* All exposures, regardless of the source or strength, contribute to the risk.
* In a single person it is medically impossible to determine the true cause of a stochastic health effect.
The report also states: "[T]he demography of large human populations is required to detect stochastic health effects with any significant degree of confidence." Is anyone doing a well-organized demographic study on the illness trends among people of Iraq? I don't know. Should we be doing one, considering we've exposed many of them to SOME dose of DU? After all, we're better than the man with the poison gas... right? Yet it seems that even when questions have been brought up regarding DU and our own troops, the standard policy has been Deny, Deny, Deny.
All this conveniently adds up to a difficult time for those attempting to investigate causal links between DU exposure and human illness. It's often next to impossible for people living in housing developments next to big industry spills to "prove" a "causal link" between the chemicals in their soil or water or air and the fact that they're all getting sick. Without all that demography of large human populations, lawyers argue, where's the "proof"? It could be something else. Apply the same principle to Iraq and your task gets worse, due to the large area of exposure (and no way to do much more than guess the average dose for most of it), the mobility of people involved, the lack of studies being done, and so on.
Reply
Leave a comment