Also, He’s Always Happy and Affectionate When the Defense Secretary Comes Home, Even if He Spent the Day Lying to the Country About Why We Went to War

Friday, October 22nd, 2010

It was under their nose all along.

Picking up the chemical signature of those bombs should be relatively straightforward — just a matter of picking up the stray molecules that float away from unstable explosive material. In practice, it hasn’t been so easy. In 1997, a young program manager at Darpa launched the “Dog’s Nose” progam, to develop a bomb-sniffer as good as a canine’s. Today, that program manager, Regina Dugan, runs the entire agency.

It’s now 2010. What have they found?

Drones, metal detectors, chemical sniffers, and super spycams — forget ‘em. The leader of the Pentagon’s multibillion military task force to stop improvised bombs says there’s nothing in the U.S. arsenal for bomb detection more powerful than a dog’s nose.

Despite a slew of bomb-finding gagdets, the American military only locates about 50 percent of the improvised explosives planted in Afghanistan and Iraq. But that number jumps to 80 percent when U.S. and Afghan patrols take dogs along for a sniff-heavy walk. “Dogs are the best detectors,” Lieutenant General Michael Oates, the commander of the Joint Improvised Explosive Device Defeat Organization, told a conference yesterday, National Defense reports. That’s not the greatest admission for a well-funded organization — nearly $19 billion since 2004, according to a congressional committee — tasked with solving one of the military’s wickedest problems.

Seems like $19 billion would have bought and trained a lot of dogs.

CORRECTION: The $19 billion figure is DARPA’s entire budget since 2004, not the amount of money allocated to Dog’s Nose.

Digg it |  reddit | |  Fark

18 Responses to “Also, He’s Always Happy and Affectionate When the Defense Secretary Comes Home, Even if He Spent the Day Lying to the Country About Why We Went to War”

  1. #1 |  zendingo | 

    wow, i want a dog…..

  2. #2 |  Cynical in CA | 

    Interestingly, the idea of eliminating the exploitation of dogs as bomb squads is a noble one (with the standard disclaimer of being done voluntarily without stolen money).

    So, now that the project is a failure, is the plan to go back to exploiting dogs? That’s good news?

    Here’s a clue — if it were possible and profitable to make a bomb-sniffing robot, some private company somewhere would or would have developed it. Or will develop it — if it is possible and profitable.

  3. #3 |  delurking | 

    Clueless article author leads to clueless blog post. Sorry:
    “Picking up the chemical signature of those bombs should be relatively straightforward — just a matter of picking up the stray molecules that float away from unstable explosive material. ”
    Explosive material has an incredibly low vapor pressure, often 10^-13 torr (10^-18 atmospheres) or so. The word “unstable” is inapposite. Very few stray molecules float away from explosives. Those molecules can be pretty similar to other molecules that exist in much larger quantities in the air (like vehicle exhaust), so designing a sensor for them is very difficult. No one ever thought it would be easy, they just thought it would be worth trying, given the threat.

    Not even close to $19B have been spent on direct molecular explosives detection research. It is almost certainly less than $100M.

  4. #4 |  Brandon | 


    “With Democrats increasingly outraged over the Supreme Court’s Citizen United decision that allowed unlimited corporate spending in elections — a change conservatives have been more successful at taking advantage of — a Democratic congressman is raising the prospect of impeaching the Supreme Court’s chief justice over the issue.”

    Democrats being sore pre-losers?

  5. #5 |  Brad | 


    Not to mention the money spent on these detectors is basic science research that also has applications in biofuels, fuel cells, hydrogen storage and many other areas.

  6. #6 |  JS | 

    I don’t see what’s the big deal about the military wasting money, they can always just print more of it.

  7. #7 |  JS | 

    Radley that puppy in the picture looks so sad. Did you yell at him or something?

  8. #8 |  Pinandpuller | 

    PBS recently had a program about a guy who, during WWII, proposed training dogs to specifically attack Japanese. The idea was to train a bunch of them then turn them a’ loose on the Japanese homeland. They tried it out for about 3 months on an island off Alabama. It didn’t turn out that great and I believe that the guy was investigated by congress. The Japanese American GI’s that were used as bait weren’t all that unhappy about the program ending.

  9. #9 |  Pete | 

    One of the real unrecognized ridiculous elements here is that’s it’s not at all unreasonable to initially accept that a government agency could spend nineteen-some-odd billion dollars on this.

    It happens to not be true, but it easily could be, you know?

  10. #10 |  luvzbob | 

    1) we don’t know what the cost of the program actually is, the only number quoted is the DARPA budget. The author COULD have found out, but that wouldn’t have fit the general anti-government narrative.

    2) We don’t know how much it costs to train and keep dog – it is a lot, but without this information we have no way of comparing to the cost of this program.

    3) the machines are 50% effective while dogs are 80% effective – that means the machines are getting with in the ball park of the dogs, and any one with the vaguest notion of the technological development knows that this new technology will improve rapidly with each generation and will be better than the dogs in no time. 0 to 50% in a few years – that’s pretty good progress- a dog’s nose is a pretty sophisticated instrument.

    THis is the usual libertarian anti-government hit piece, it sounds good, but when you look just a little deeper you see that it is baseless.

    But what the hell, lets defund DARPA anyway, nothing but government welfare for scientists. All they have given us is the internet, and we would be better off with it.

  11. #11 |  SJE | 

    What happened to the bomb-sniffing rats and other critters they were investigating? I heard that rats were very good, are more mobile and, if they were killed, less mourned than Fido.

    My other thought: how is it that the US military can use dogs to save lives, while cops can only see them as something to shoot.

  12. #12 |  MacGregory | 

    Here is the bottom line: we, as libertarians, just wanna be left alone. Tired of your bullshit.

  13. #13 |  Chance | 

    I like DARPA – I think we get a pretty good return on our investment from them (but since I’m not libertarian, take one collectivist’s opinion for what it’s worth), but like any investment, sometimes it doesn’t pay off – at least immediately or in the ways expected. I suppose you could leave this type of research to private sector, but a lot of basic science probably wouldn’t happen, or would take a lot longer to develop and spread. Would a company or companies have ever built the Large Hadron Collider I wonder?

  14. #14 |  Gerald A | 

    DARPA deals in the high rate of failure, high rate of return. They take the risk a normal business wouldn’t take because of the money involved, long lead times, and unknown returns.

    Besides, I enjoy watching the DARPA Challenge when it’s on.

  15. #15 |  Gritsforbreakfast | 

    I heard somebody discussing this recently at the Texas Forensic Science Seminar whose colleague had received a research grant on the subject. After all this time and however many millions were spent so far, they haven’t even figured out exactly what the dog is reacting to when it alerts.

  16. #16 |  David Chesler | 

    FWIW, I’ve recently finished a contract with a company making a bug sniffer (bio-agent detection, not chemical detection.) We were trying to make it be cheap, and work fast, and be broad-spectrum. It’s not easy (which means it was lots of fun as an engineer.) Stateside I’m more worried about bio-agents than explosives.

  17. #17 |  delurking | 

    “CORRECTION: The $19 billion figure is DARPA’s entire budget since 2004, not the amount of money allocated to Dog’s Nose.”

    No, $19B is JIEDDO’s entire budget. JIEDDO is a separate organization that deals with shorter-term solutions than DARPA. That paragraph about the “dog’s nose” program is just there to give a bad impression. If you knew nothing about it, you would assume that spending on “dog’s nose”-type programs continued from 1997 to the present, growing all the time as Dr. Duggan rose through the ranks to become director. But, you would be wrong. The “dog’s nose” program ended, she went off to found an investment company which supported another company that commercialized explosives detection technology (among other ventures). It was only recently that she came back to DARPA to take over as director.

  18. #18 |  ALowe | 

    I wonder how the false-positive rates compare between dogs and machines. That’s the real test, and I’m not convinced that dogs are particularly reliable in that regard.