- [mysterious music]

- So what if I told you that I could use a pendulum to mark the spot on a map where you should drill for oil? Would you walk out of the room, or hand me a check? If you find the question ridiculous, you'll be surprised to learn that it's actually a service people pay for. It just involves finding water and not oil. It's called dowsing, and its practitioners use metal rods, pendulums and Y-shaped branches. These objects supposedly move when a dowser finds water underground. 

Some dowsers claim to locate buried utility lines, valuable ores, unmapped property lines, or even spiritual auras. This should tip you off that this isn't exactly physics here. How can people charge money for this nonsense? [soft music] Dowsing is also commonly referred to as divining or witching.

- Sounds like a man.

- So this medieval superstition can hardly be accused of false advertising. In Georgius Agricola's 1556 book, De Re Metallica, [lightning bolt blasting] a passage describing dowsing says, wizards who also make use of rings, mirrors, and crystals, seek for veins with a divining rod shaped like a fork, but its shape makes no difference, for it is not the form of the twig that matters, but the wizard's incantations which it would not become me to repeat. So why, centuries later, are people still relying on a wizard's incantations when they drill water wells? Basically, groundwater's just a mystery to most people. 

The average person imagines that groundwater exists as underground rivers. With that model in your head, you'll be worried you might drill an expensive well somewhere that misses the river and produces no water. The thing is, groundwater almost never looks like that. Instead, imagine filling a bucket with dry sand, and then dumping in a smaller volume of water. It doesn't matter where'd you jab a straw into that sand, you're gonna be able to suck some water out. This is the dowser's trick, whether they know it or not. It doesn't matter what hocus-pocus you do before you drill the well, because just about any spot is as good as any other. They simply cannot lose. It's like paying someone to tell you which window in an office you should open to get some air. They find air no matter what. And you agree to hire them again for next Tuesday based on their track record of excellent air finding. 

What would the world be like if this were right? There are reports claiming to prove that dowsing works, but most are just collections of anecdotes. Stuff like, we drilled 40 wells with a dowser, and a lot of them were good. These reports rarely make comparisons with groups of wells drilled without a dowser. In the 1980s, there was actually one large-scale experiment in Germany that worked with some 500 self-proclaimed dowsers. A pipe with water flowing through it was moved back and forth across the floor of a large barn. Up on the second floor of the barn, dowsers would be brought in to detect the position of the pipe below. The organizers of the study claimed that it showed dowsing is real, but the data actually showed exactly the opposite. The organizers systematically removed any dowsers who guessed poorly, claiming that they weren't the real deal. They picked out just the few who had gotten the luckiest. Surprise, imagine, they were all skillful. 

This is people acting out what's called the Texas sharpshooter fallacy. It's like firing 100 bullets randomly at the side of a barn, finding the tightest cluster of five holes, and then painting a bull's eye around them. In the German experiment, the cream of the crop dowsers were chosen based on the results in the final round of testing, but if you looked at their earlier tests, their results sucked, proving that their performance in the final round was just a fluke. So why do people believe this? Both the US Geological Survey and the National Groundwater Association will tell you that dowsing is complete nonsense, and a total waste of money. But as we mentioned earlier, if there is groundwater present, then it's hard to miss. 

Dowsers are right often enough that people keep using them, but that doesn't mean that finding groundwater is all a matter of luck. There are people called hydrogeologists who can plan wells using information on the layers of sediment and rock beneath an area. Experience at other wells shows which layers provide the largest volumes of clean water. Wells can then be drilled down until they reach the best aquifer. Known geology beats walking around with a stick any day. Even in unpermeable bedrock like granite, where most of the water will be pulled from fractures in the rock rather than through the rock itself, and geological maps and aerial photos can be used to identify the best locations to drill. So even if you're dead set on ignoring science, at least save yourself some money. Just throw a dart at the map and declare yourself a wizard. [soft music]