The Mystery of the Haunted Rationalist

Followup to: Simultaneously Right and Wrong

“The most merciful thing in the world, I think, is the inability of the human mind to correlate all its contents.”

- H.P. Lovecraft, The Call of Cthulhu

There is an old yarn about two skeptics who stayed overnight in a supposedly haunted mansion, just to prove they weren’t superstitious. At first, they laughed and joked with each other in the well-lit master bedroom. But around eleven, there was a thunderstorm—hardly a rare event in those parts—and all the lights went off. As it got later and later, the skeptics grew more and more nervous, until finally around midnight, the stairs leading up to their room started to creak. The two of them shot out of there and didn’t stop running until they were in their car and driving away.

So the skeptics’ emotions overwhelmed their rationality. That happens all the time. Is there any reason to think this story proves anything more interesting than that some skeptics are cowards?

The Koreans have a superstition called “fan death”: if you sleep in a closed room with a fan on all night, you will die. Something about the fan blades shredding the oxygen molecules or something. It all sounds pretty far-fetched, but in Korea it’s endorsed by everyone from doctors to the government’s official consumer safety board.

I don’t believe in ghosts, and I don’t believe in fan death. But my reactions to spending the night in a haunted mansion and spending the night with a fan are completely different. Put me in a haunted mansion, and I’ll probably run out screaming the first time something goes bump in the night1. Put me in a closed room with a fan and I’ll shrug and sleep like a baby. Not because my superior rationality has conquered my fear. Because fans just plain don’t kill people by chopping up oxygen, and to think otherwise is just stupid.

So although it’s correct to say that the skeptics’ emotions overwhelmed their rationality, they wouldn’t have those emotions unless they thought on some level that ghosts were worth getting scared about.

A psychologist armed with the theory of belief-profession versus anticipation-control would conclude that I profess disbelief in ghosts to fit in with my rationalist friends, but that I anticipate being killed by a ghost if I remain in the haunted mansion. He’d dismiss my skepticism about ghosts as exactly the same sort of belief in belief afflicting the man who thinks his dragon is permeable to flour.

If this psychologist were really interested in investigating my beliefs, he might offer me X dollars to stay in the haunted mansion. This is all a thought experiment, so I can’t say for certain what I would do. But when I imagine the scenario, I visualize myself still running away when X = 10, but fighting my fear and staying around when X = 1000000.

This looks suspiciously like I’m making an expected utility calculation. Probability of being killed by ghost * value of my life, compared to a million dollars. It also looks like I’m using a rather high number for (probability of being killed by ghost): certainly still less than .5, but much greater than the <.001 I would consciously assign it. Is my mind haunted by an invisible probability of ghosts, ready to jump out and terrify me into making irrational decisions?

How can I defend myself against the psychologist’s accusation that I merely profess a disbelief in ghosts? Well, while I am running in terror out of the mansion, a bookie runs up beside me. He offers me a bet: he will go back in and check to see if there is a ghost. If there isn’t, he owes me $100. If there is, I owe him $10,000 (payable to his next of kin). Do I take the bet?

Thought experiments don’t always work, but I imagine myself taking the bet. I assign a less than 1100 chance to the existence of ghosts, so it’s probably a good deal. The fact that I am running away from a ghost as I do the calculation changes the odds not at all.

But if that’s true, we’re now up to three different levels of belief. The one I profess to my friends, the one that controls my anticipation, and the one that influences my emotions.

There are no ghosts, profess skepticism.
There are no ghosts, take the bet.
There are ghosts, run for your life!


1: I worry when writing this that I may be alone among Less Wrong community members, and that the rest of the community would remain in the mansion with minimal discomfort. If “run screaming out of the mansion” is too dramatic for you, will you agree that you might, after the floorboards get especially creaky, feel a tiny urge to light a candle or turn on a flashlight? Even that is enough to preserve the point I am trying to make here.