Recently, an acquaintance asked me whether I believed in destiny. I told her I didn’t, and she told me a long story boiling down to this: someone she knows was prevented, by a series of improbable accidents, from getting on a plane. The plane then crashed, killing this person’s entire family.
“How do you explain that?” she asked.
“I don’t really see the need for an explanation,” I said.
I relayed the cached wisdom that there are billions of people on Earth, all living eventful lives, and therefore we can expect one-in-a-billion experiences to occur daily.
It just occurred to me that people in this sort of situation are subject to something very similar to the red/green paradox first discussed here. Suppose 10^9 people have a commonly agreed-on model of the universe (each person assigns it a probability of about 10^-3 of being false). This model says that, each time you wake up, there is precisely a 10^-9 chance of discovering you have been transformed into a giant insect, with the chance being independent for each person for each day. If we read in the news that someone was transformed, we don’t update against the model being true—the model predicts that about one person should be transformed each day.
On the other hand, if you yourself wake up to find yourself transformed into a giant insect, it is tempting to say that you should update against the model, since it is more likely that the model underestimates the chances of this happening than that you have experienced a 1 in 10^9 event. Indeed, if someone within 2 degrees of separation from you is transformed, it seems you should update.
Such a population could experience a long period of statistical adherence to the model, yet contain a growing population of skeptics who believe that a lot of transformations are unreported or covered up.
Is this, generalized, the situation we actually find ourselves in with respect to what’s usually called “belief in the supernatural?”
Recently, an acquaintance asked me whether I believed in destiny. I told her I didn’t, and she told me a long story boiling down to this: someone she knows was prevented, by a series of improbable accidents, from getting on a plane. The plane then crashed, killing this person’s entire family.
“How do you explain that?” she asked. “I don’t really see the need for an explanation,” I said.
I relayed the cached wisdom that there are billions of people on Earth, all living eventful lives, and therefore we can expect one-in-a-billion experiences to occur daily.
It just occurred to me that people in this sort of situation are subject to something very similar to the red/green paradox first discussed here. Suppose 10^9 people have a commonly agreed-on model of the universe (each person assigns it a probability of about 10^-3 of being false). This model says that, each time you wake up, there is precisely a 10^-9 chance of discovering you have been transformed into a giant insect, with the chance being independent for each person for each day. If we read in the news that someone was transformed, we don’t update against the model being true—the model predicts that about one person should be transformed each day.
On the other hand, if you yourself wake up to find yourself transformed into a giant insect, it is tempting to say that you should update against the model, since it is more likely that the model underestimates the chances of this happening than that you have experienced a 1 in 10^9 event. Indeed, if someone within 2 degrees of separation from you is transformed, it seems you should update.
Such a population could experience a long period of statistical adherence to the model, yet contain a growing population of skeptics who believe that a lot of transformations are unreported or covered up.
Is this, generalized, the situation we actually find ourselves in with respect to what’s usually called “belief in the supernatural?”