One problem with state security, and the meaning behind the adage, “those who would give up liberty for security deserve neither,” is that it's a one-size-fits-all solution. The cost-benefit analysis is completely out of the hands of the consumer, and, as the analysis is conducted by the “provider” of the benefit, biased towards those special interests.

Whereas, the analysis should start from the default position of full privacy and add benefits/subtract costs from there, in practice the privacy invasions are lost in the assumptions, and only touted security benefits are considered. Predictably, the result of the analysis is, then, the maximum violation of personal privacy allowed by law (and, then a little) for the least amount of security.

Even if we accept the reality that this flawed methodology will always be used, we must recognize that even the security benefits considered are claims, not independent results of sound testing. Security expert Bruce Schneier points us to an example of how a full-body scanner fails even the most rudimentary test on this German television show:

Of note, the test environment is not the same as one the scanner will see in practice (jacket on, only the one instrument is used, etc.) and the scanner used is an infra-red scanner and not the millimeter-wave scanners being implemented in the United States, but neither was the test double-blind. The scanner operator found exactly the items the subject said he had on him before the test began.

When confronted with the Thermite and detonator the subject also carried, the scanner operator launches into a barrage of defensive excuses eerily similar to those of psychics, mediums, and dowsers when subjected to double-blind trials.

Does this mean the millimeter-wave scanners have similar efficacy problems? Certainly not, although there have been indications that the PETN carried by the Christmas day crotch-bomber would not have been reliably detected by these scanners.

The take-away from all this is that, before any cost-benefit analysis (even a flawed one) can be conducted, the security benefits of the considered technology must be independently tested, verified, and published.

-------------------------------------

On a related note, the depositions taken by Marc J. Victor in preparation for the defense of Steven Anderson (the pastor roughed-up by Arizona DPS and Border Patrol) motivated me to search for the results of double-blind trials of drug and bomb-sniffing dogs. The dog handler in that case stated the “alerts” he looks for in the dog are changes in respiration or hesitation. That sounds suspiciously like dowsing.

Even I am not cynical enough to believe that these working dogs have not been subjected to independent trials (if nothing else, as competitive sport). However, I was only able to find scant references to trials, and those ranged from damning to dubious. If anyone knows of such a trial, please pass that information along.