The Little Bomb-Detecting Device That Couldn't

theodp writes "Widely deployed in Iraq and promoted by military leaders, BusinessWeek reports the ADE 651 bomb-detecting device had one little problem: it wouldn't detect explosives (earlier Slashdot story). 'The ADE 651,' reports Adam Higginbotham, 'was modeled on a novelty trinket conceived decades before by a former used-car salesman from South Carolina, which was purported to detect golf balls. It wasn't even good at that.' One thing the ADE 651 did excel at, however, was making money — estimates suggest that the authorities in Baghdad bought more than 6,000 useless bomb detectors, at a cost of at least $38 million. Even though ADE 651 manufacturer James McCormick was found guilty of three counts of fraud and sentenced to 10 years in prison in May, the ADE 651 is still being used at thousands of checkpoints across Baghdad. Elsewhere, authorities have never stopped believing in the detectors. Why? According to Sandia Labs' Dale Murray, the ideomotor effect is so persuasive that for anyone who wants or needs to believe in it, even conclusive scientific evidence undermining the technology it exploits has little power."

Comments

It has a deep tradition it seems

[Sycraft-fu]

Some people don't even think it is special powers, just a thing you do. My grandpa did the dowsing thing to decide where to put the various wells on his property. Not because he thought he had special powers, it was just how he'd learned you select your well spot. Anyone could do it. He figured it worked since every time he'd drill that spot, and before long have a functional well.

For him it wasn't magical or special powers, it was just the standard process. Get Y shaped stick, walk around, it signals where the well goes, put it there.

Re:It has a deep tradition it seems

[master_kaos]

I was a skeptic to, but my grandfather was one (they called it water witching around here). There were a few locally, but my grandfather was the most known and best, everyone within a half hour radius would call my grandfather when they needed a well dug out (and this was 30+ years ago). He would use any standard stick that was laying around. He never charged the people money (people were a lot more neighborly back then instead of just looking out for themselves), so wasn't like he was out scamming them, very religious so not a liar.

Not once did he screw it up, he hit water every time. I was a skeptic to before I seen it, and it didn't seem like it would work for just anyone. I tried it along with a lot of my relatives, and it wouldn't do a thing, my one uncle it did it a bit.

It was funny one time this guy tried digging a well on his property twice kept getting dry, my grandfather went out and did the dowsing told him here this is where you got to dig(they guy didn't like my grandfather for some reason, and was a major skeptic), the guy ignored my grandfathers advice, dug up 3 more spots in the following 2 years, kept hitting dry again. Finally fed up he called my grandfather back to confirm the location, grandfather goes back, exact same spot detects water, guy digs there and sure enough hits water.

Re:It has a deep tradition it seems

[black3d]

Right, which is the exact ideometer effect that's being discussed here. There are other (subconcious) cues at work which lead him to believe where the water will be - or just pure coincidence. Aside from the obvious fact there's no actual mechanism at work, it can be easily disproven. Take a dowser out until they find a spot "with water", then blindfold them and drive them around to re-test various random spots including this one. The vast majority of the time, they'll get it wrong - suddenly not able to detect water at the spot they previously said it was at, or detect water in places they previously said it wasn't. Also fun is taking them to an area known to be entirely over a natural aquifier and watch them wander around until they "find" water in some exclusive spot.
 
Map-based dowsing is even easier to disprove - again, aside from the obvious lack of any mechanism (ie, it doesn't really need proof, but just to satisfy the idiots out there we have to go through it). Give a map-dowser a map without orientation or contour lines and suddenly their "abilities" go away. Give them a fully-detailed map but blind-fold them, and similarly, they're no longer able to "detect" where the water is.
 
In all cases, it's either fraudulent, subconcious, or simply luck. Likewise, stories about "other people" are steeped in grandeur. A guy who gets it right "a couple of times" is suddenly a legendary dowser, and every re-telling by both others and himself get better and better each time.

Re:It has a deep tradition it seems

[Sycraft-fu]

In plenty of places, you can put a well wherever you like and it'll work. I'm quite sure that's the case on my grandfather's property. There's a lot of homes there with their own wells, there's presumably a big aquifer or the like underneath (I've never bothered to check to see what). So the reason dowsing worked was that any spot was fine.

He did it just because he believed it was how it was done. Of course each time it 'worked' and as such he kept doing it.

What I found interesting about the thing was that it was a 'common man' kind of thing for him and others. He wasn't a huckster that went around dowsing for people, he did it himself, for his wells, and just using whatever Y shaped stick he'd come across. To him, it wasn't mystical, it was just a process one did like so much else in farming and ranching and it was something anyone could, and would, do.

I think that might have something to do with why dowsers keep believing in it. There seems to be a real strong cultural thing that dowsing just works, and so they believe that must be the case.

Re:It has a deep tradition it seems

[TheCarp]

> There is something to this. If you have an approved mechanism for intuitively detecting bombs they
> you have probable cause to terminate a prospective bomber, and if your intuition is right more than
> half the time on average, you're a hero. Since some few are more accurate with intuition and some
> less, and the metrics are classified, you are free to open fire indiscriminately anywhere anywhen.

Half the time? Nah, I think you are overestimating how accurate you need to be, because, even if you find nothing, you can, like the scammer of these did, claim that it hit on some residue and you just got them at the wrong time.

In fact, this is very much the way drug dogs are used. Dogs, it turns out, have great snouts and can detect all manner of things, and do great in really blind trials. However, they are even better at playing "Clever Hans" ( http://en.wikipedia.org/wiki/Clever_Hans ), that is, if the handler is given any clues as to where their might be hits, the animals false positive rate goes through the roof...in exactly the places where the handler expects to find something.

So.... dogs are great at finding bombs or drugs in random packages... but in one of their most common use cases, sniffing a suspects car, they are just about guaranteed a hit, because their handler is expecting one. A hit, that can be explained away and dismissed when nothing is found, so their real hit rate can be far lower than chance without bringing them into question.

One study used no drugs or explosives at all, but flagged several points with information for the handler indicating the type of hit expected to set his expectations. If the dogs were 100% effective, there would have been not a single hit amongst any of the trials...the results?

from http://www.economist.com/blogs/babbage/2011/02/animal_behaviour

The findings, which Dr Lit reports in Animal Cognition, reveal that of 144 searches, only 21 were clean (no alerts). All the others raised one alert or more. In total, the teams raised 225 alerts, all of them false. While the sheer number of false alerts struck Dr Lit as fascinating, it was where they took place that was of greatest interest.

Effectiveness of fake bomb detectors

[Firethorn]

Now, while they're technologically incapable of their purpose, I wonder if they might actually be somewhat effective in real life? IE a different type of placebo?

It says that they're being used at a number of checkpoints. Now, one of the things I know about is that the insurgents/terrorists tend to observe such places before they target them. Often at some distance, but eh.

The ones doing the observing are often no more educated than those working the checkpoint, often less. So they see the operators using their 'bomb detector' in all seriousness. They think 'crap! They'd find our bomb, time to figure out a different plan!' and either delay or go elsewhere. So the end result is that they still have fewer attacks against that checkpoint.

And in other news...

[sirwired]

Shocker: In the face of conclusive evidence understandable to anybody with an IQ higher than a kumquat, people still believe in:

Ponzi schemes
Homeopathy
Dowsing
Young-earth creationism
Psychics

Never underestimate the stubbornness of otherwise-rational people.

Re:I haven't played golf in several years

[Megane]

It would probably be simpler to put an old-school chirp transponder like on wildlife tracking collars. You could probably use inductive charging to avoid the need to open the ball. If the golf course was next to the transmitter for a radio station, you could even get away without needing a battery.