Is there a sense in which this argument proves too much? Like, traditionally, evangelism doesn’t involve murder because God said that murder is bad. But if we’re going full-bore infinite-utilitarian about all this, is there a sense in which any extra time in heaven is infinitely valuable, and outweighs the nastiness of killing?
Presumably for the same reason we should prefer a 99% chance of infinite value to a 1% chance, we should prefer infinitely-awesome experience for 99 days to 1 day. I don’t know how to make that math work out, but if it does, then it seems like a truly effective missionary would find himself committed to murdering those he’s converted, no?
Interesting point! I think that will give hinge on a lot of things, including not least your eschatology.
One thing is that I doubt whether you can have infinitely good experiences in a finite amount of time. I sort of doubt that, but who knows. Even so, any day on earth is presumably worse than any day in heaven.
In any case it seems like the conclusion should be that you should kill someone when you have the highest credence that they're saved. This is complicated by "once saved always saved" but you should presumably still give some credence to the negation of this.
Another question is whether killing someone might reduce your chance of being saved drastically. If so, then it seems like you should be very sure you save roughly an expected person by doing so.
As to whether this "prooves too much," I don't think that's straightforward. A decision theory that allows this result is strange, but it's just a standard counterexample to fanaticism, and denying fanaticism also has very counterintuitive implications.
But I'm probably not gonna go out and murder people for this cause so maybe it is too much.
Imagine Gorgia..er...George accepts some aspect of Pascal's Wager, and furthermore is confident that some Armenian Protestant Christianity is the ticket to heaven, i.e., infinite expected positive utility. It's also the case that George really loves the never-ending parties at the Temple of Dionysus (best sex, wine, and entertainment in the world, as far as he's concerned). And we are already assuming doxastic voluntarism; therefore, George can freely choose to commit to the creed of Protestant Christianity at any point, and secure his spot in heaven. In fact, as long as George commits to this creed before he dies and remains in the faith, he will go to heaven, regardless of what he did and believed before committing himself to the Protestant faith.
In other words, he has a finite but very large number of opportunities to commit to the correct faith that secures himself a spot in heaven, for he will surely die at some point that he cannot predict. George is a reasonable person and would like to maximize utility. His strategy is to enjoy the heck out of the parties at the Temple of Dionysus as long as he can, and then repent and commit to the Protestant faith. So each day when George wakes up, hung over at the Temple of Dionysus, he reasons as follows: "No matter what I chose today, my expected utility will be the same whether I repent now or wait until tomorrow, since I have good reason to believe there will be a tomorrow. So I am indifferent, at this moment, between choosing sin and choosing salvation, or maybe slightly more in favor of more parties." And so George carries on, living a life of sin at the Temple of Dionysus, every day, until his mate Alexander trips on lute, knocking a marble plinth onto a passed out George, killing him. George never repents and therefore never goes to heaven.
Has George exhibited a failure of rationality? It depends on George's ability to set an intention and stick to it, which in turn depends on whether George presumes that the norms of rationality will be constant for his infinite future self. If George accepts irreducibly global norms of rationality, then his choice to squeeze in some more parties is irrational, even though any individual decision to stay at the never-ending party is seemingly rational. What this illustrates is that to set an intention for an infinite future requires a belief that the norms of rationality will remain constant into the infinite future, and in every setting, and that his future self will value heaven just as much today as it will in a trillion millennia. But is it rational to believe that? It seems under-determined, to me.
Assuming George is a fanatic and all that, it seems like he should just use the heroes me heuristic of maximizing the probability of an infinite positive outcome: for every day, he assigns a small probability that he'll die. And so the optimal chance of the infinite outcome is to stop parroting me partying today.
No matter what probability George assigns to his confidence that he will die imminently, when multiplied by a non-convergent utility sum, George should repent immediately. Doing anything else is irrational. But under this utilitarian eschatology, George is still stuck under infinitarian paralysis: all other actions are sort of meaningless. Imagine George is already the world's expert in molecular genetics, and he's pretty convinced he can create, test-tube babies that have the xtian-25 gene, which predisposes them to believing in classical theism. Should George spend the rest of his life making sure he's first in heaven, or should he forgo his chances of getting into heaven and instead focus on producing and infinite amount of god-fearing test tube babies, each of which have a good chance of getting into heaven? Assume there's a non-zero probability that this technology will be used ad infinitum, thereby creating an infinite amount of potential Christians. Obviously we're getting deeper and deeper into the absurd, but I think it highlights the problem of intuiting probability spaces when we're dealing with infinities.
Is there a sense in which this argument proves too much? Like, traditionally, evangelism doesn’t involve murder because God said that murder is bad. But if we’re going full-bore infinite-utilitarian about all this, is there a sense in which any extra time in heaven is infinitely valuable, and outweighs the nastiness of killing?
Presumably for the same reason we should prefer a 99% chance of infinite value to a 1% chance, we should prefer infinitely-awesome experience for 99 days to 1 day. I don’t know how to make that math work out, but if it does, then it seems like a truly effective missionary would find himself committed to murdering those he’s converted, no?
Interesting point! I think that will give hinge on a lot of things, including not least your eschatology.
One thing is that I doubt whether you can have infinitely good experiences in a finite amount of time. I sort of doubt that, but who knows. Even so, any day on earth is presumably worse than any day in heaven.
In any case it seems like the conclusion should be that you should kill someone when you have the highest credence that they're saved. This is complicated by "once saved always saved" but you should presumably still give some credence to the negation of this.
Another question is whether killing someone might reduce your chance of being saved drastically. If so, then it seems like you should be very sure you save roughly an expected person by doing so.
As to whether this "prooves too much," I don't think that's straightforward. A decision theory that allows this result is strange, but it's just a standard counterexample to fanaticism, and denying fanaticism also has very counterintuitive implications.
But I'm probably not gonna go out and murder people for this cause so maybe it is too much.
(not *your* argument necessarily, to clarify—the one that Gordon and that guy in that paper I’ll admit I didn’t read are making)
Imagine Gorgia..er...George accepts some aspect of Pascal's Wager, and furthermore is confident that some Armenian Protestant Christianity is the ticket to heaven, i.e., infinite expected positive utility. It's also the case that George really loves the never-ending parties at the Temple of Dionysus (best sex, wine, and entertainment in the world, as far as he's concerned). And we are already assuming doxastic voluntarism; therefore, George can freely choose to commit to the creed of Protestant Christianity at any point, and secure his spot in heaven. In fact, as long as George commits to this creed before he dies and remains in the faith, he will go to heaven, regardless of what he did and believed before committing himself to the Protestant faith.
In other words, he has a finite but very large number of opportunities to commit to the correct faith that secures himself a spot in heaven, for he will surely die at some point that he cannot predict. George is a reasonable person and would like to maximize utility. His strategy is to enjoy the heck out of the parties at the Temple of Dionysus as long as he can, and then repent and commit to the Protestant faith. So each day when George wakes up, hung over at the Temple of Dionysus, he reasons as follows: "No matter what I chose today, my expected utility will be the same whether I repent now or wait until tomorrow, since I have good reason to believe there will be a tomorrow. So I am indifferent, at this moment, between choosing sin and choosing salvation, or maybe slightly more in favor of more parties." And so George carries on, living a life of sin at the Temple of Dionysus, every day, until his mate Alexander trips on lute, knocking a marble plinth onto a passed out George, killing him. George never repents and therefore never goes to heaven.
Has George exhibited a failure of rationality? It depends on George's ability to set an intention and stick to it, which in turn depends on whether George presumes that the norms of rationality will be constant for his infinite future self. If George accepts irreducibly global norms of rationality, then his choice to squeeze in some more parties is irrational, even though any individual decision to stay at the never-ending party is seemingly rational. What this illustrates is that to set an intention for an infinite future requires a belief that the norms of rationality will remain constant into the infinite future, and in every setting, and that his future self will value heaven just as much today as it will in a trillion millennia. But is it rational to believe that? It seems under-determined, to me.
Assuming George is a fanatic and all that, it seems like he should just use the heroes me heuristic of maximizing the probability of an infinite positive outcome: for every day, he assigns a small probability that he'll die. And so the optimal chance of the infinite outcome is to stop parroting me partying today.
I don't know what happened with "heroes me" lol
No matter what probability George assigns to his confidence that he will die imminently, when multiplied by a non-convergent utility sum, George should repent immediately. Doing anything else is irrational. But under this utilitarian eschatology, George is still stuck under infinitarian paralysis: all other actions are sort of meaningless. Imagine George is already the world's expert in molecular genetics, and he's pretty convinced he can create, test-tube babies that have the xtian-25 gene, which predisposes them to believing in classical theism. Should George spend the rest of his life making sure he's first in heaven, or should he forgo his chances of getting into heaven and instead focus on producing and infinite amount of god-fearing test tube babies, each of which have a good chance of getting into heaven? Assume there's a non-zero probability that this technology will be used ad infinitum, thereby creating an infinite amount of potential Christians. Obviously we're getting deeper and deeper into the absurd, but I think it highlights the problem of intuiting probability spaces when we're dealing with infinities.