The military uses some of the most technologically sophisticated machinery and innovative medical techniques in history. But a disturbing current of pseudoscience in the military is wasting money, perpetuating myths, and putting our troops in danger. I am a retired U.S. Air Force colonel, so this hits close to home. An organization I was once proud to belong to has become a source of embarrassment.

The most appalling episode was the bomb detector scandal. Our military was trying to protect its soldiers and civilians with dowsing rods—very expensive, deceptively high-tech-looking electronic wands that worked no better at detecting bombs than would a forked stick or a coin toss. Aside from being supremely unintelligent, the whole episode was tainted with conflicts of interest. Mother Jones reported that a high-powered Pentagon official and her family had money invested in these bogus devices.

Another ongoing DoD failure is the infiltration of quackery into military medicine. It's not as dangerous to our troops as a bomb detector that can't detect bombs, but it's wasting tax dollars and medical resources on unscientific mumbo-jumbo that "works" only as a placebo. In some cases, it is demonstrably harmful.

Acupuncture is based on a mythical, nebulous energy called qi that has never been detected, even though scientific instruments are capable of measuring quantum energies at the subatomic level. It is said to flow through hypothetical meridians and to be altered by sticking needles into hypothetical acupuncture points. Originally, there were 360 acupuncture points, corresponding to the days of the year, which is not surprising since the idea grew out of astrology. Now so many acupoints have been described that one wag suggested there was no place left on the skin that wasn't an acupuncture point in someone's system.

Many proponents of acupuncture reject the esoteric explanations but believe acupuncture has a real physiological effect. Various mechanisms have been proposed, but none is convincing. Needling can release pain-killing endorphins in the brain, but that's a nonspecific effect: Placebo pills do the same thing, and just throwing a stick for a dog releases endorphins in the dog's brain.

We don't need to know how it works to know if it works. Acupuncture has been tested repeatedly and found wanting. Studies have shown that it doesn't matter where you stick the needles, and it doesn't matter whether you pierce the skin. Stimulating intact skin with toothpicks or electricity works just as well. The crucial factor seems to be whether patients believe they are getting acupuncture.

The claimed benefits of acupuncture range from treating infertility to aiding smoking cessation, but the evidence argues against its usefulness for anything but easing pain and possibly nausea. A recent comprehensive review of the literature by Edzard Ernst found little evidence that acupuncture is even truly effective for pain. He also found 95 published cases of serious adverse effects, including death. There is a double standard here: The quality of evidence offered to support acupuncture would not pass muster for a proposed prescription drug.

Some acupuncturists have accepted that the evidence is lacking and are now saying: "Maybe it's just a placebo, but let's use it anyway. Placebos are good." But placebos amount to lying to the patient. Surely our troops deserve better.