Scrubcutters experiment is a move in the right direction. A lot of dowsers refuse to submit to scientific testing.
However, you would absolutely, definitely, have to have two friends - one to operate the switch and one to record the response of the dowsing rods. It is fundamental that the person observing and recording the response is unaware of the switch state.
The switch operator would also have to toss a coin or use some other random method to decide on whether the switch should be up or down.
You would also have to do the test many times - say one hundred. Of course, if dowsing does not work, you would expect the dowser to be right half the time by chance. If dowsing works as well as its supporters claim, then I would be expecting close to 100% accuracy. In fact, it would be worth doing a pre-test run where the dowser KNOWS the switch position. If he is 100% accurate in the pre-test run, then it should be equally accurate in the actual experiment, if dowsing does indeed work.
Jodies point about headphones is a good one. It is crucial that the dowser is unaware of the switch position.
As bushscout says, an ac current is detectable by the surrounding magnetic field. However, in the experiment above, there is no current flowing at any time, because the equipment itself is switched off. Which makes me wonder what the dowser is actually detecting. Presumably the potential difference between the live cable and earth. Just to complicate matters, if your house has protective multiple earthing, the earth cable may be at a different potential to real earth. But that should not affect the experiment.
I would love to hear of any results. I am very sceptical of dowsing, because the results of controlled trials are very unimpressive, but like James Randi, I am struck with the genuiness of dowsers. They genuinely, really, truely believe in it. However, we humans do have an extraordinary capacity for self-deception.
One problem in science is that people are very keen to publish positive results, but often reluctant to publish negative results (especially if they were expecting a positive). So if you do the test, undertake to post the result regardless of what it is.
However, you would absolutely, definitely, have to have two friends - one to operate the switch and one to record the response of the dowsing rods. It is fundamental that the person observing and recording the response is unaware of the switch state.
The switch operator would also have to toss a coin or use some other random method to decide on whether the switch should be up or down.
You would also have to do the test many times - say one hundred. Of course, if dowsing does not work, you would expect the dowser to be right half the time by chance. If dowsing works as well as its supporters claim, then I would be expecting close to 100% accuracy. In fact, it would be worth doing a pre-test run where the dowser KNOWS the switch position. If he is 100% accurate in the pre-test run, then it should be equally accurate in the actual experiment, if dowsing does indeed work.
Jodies point about headphones is a good one. It is crucial that the dowser is unaware of the switch position.
As bushscout says, an ac current is detectable by the surrounding magnetic field. However, in the experiment above, there is no current flowing at any time, because the equipment itself is switched off. Which makes me wonder what the dowser is actually detecting. Presumably the potential difference between the live cable and earth. Just to complicate matters, if your house has protective multiple earthing, the earth cable may be at a different potential to real earth. But that should not affect the experiment.
I would love to hear of any results. I am very sceptical of dowsing, because the results of controlled trials are very unimpressive, but like James Randi, I am struck with the genuiness of dowsers. They genuinely, really, truely believe in it. However, we humans do have an extraordinary capacity for self-deception.
One problem in science is that people are very keen to publish positive results, but often reluctant to publish negative results (especially if they were expecting a positive). So if you do the test, undertake to post the result regardless of what it is.