Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Pascals wager is an extremely flawed argument. It boils down to the idea that whether there is a god or not, you may as well believe because if it turns out that if there is a god you win infinite reward, and if there's not then what's the loss?

However it presupposes that the Christian ideal of a god holds true, one that will reward you for belief and not reward you (or punish you for disbelief). There is a multidimensional spectrum of other options available. What if there is a god but (s)he created a mechanistic universe deliberately left behind no evidence, and so rewards those that use their critical thinking and form no belief? What if there's a benevolent god that rewards everyone regardless? What if you picked the wrong god and the real one hates wrong-believers more than atheists?

So it really only works in a situation where you have a binary choice, and that doesn't model the reality. In other words it's an exercise in apologetics written by someone who had already decided to believe, based on other criteria (or no criteria), trying to persuade others through a pseudo-rational argument.

I don't mean any of this as an attack on you, or as justification to look down on others' beliefs, but Pascal's Wager kinda grinds my gears :)



Here's my version of Pascal's wager:

Even if there is/are god(s), she/he/they obviously don't want to be known/found as she/he/they have left no obvious clues from which we might deduce such a being or beings. So, we should all be atheists/agnostics because once we die, if it turns out that there is/are god(s), atheists/agnostics are the ones who will end up in heaven. Why? Because, they alone are following the will of a possible creator.

"Did I not give you a brain," such a deity would say to believers of various religions. "Did I not hide all indications of my existence, so that you try to make sense of the universe with the data available to you? Why, then, did you go and make up random fantastic beings to worship without a shred of evidence? You have ignored my will. How dare you presume that you are capable of knowing an almighty being that obviously doesn't want to be known? Look at these atheists, this small band of the faithful, who have stayed true to the faculties of reasoning I endowed them with and reached the logical conclusion that they could to the best of their ability. These alone have stayed true to their creator's purpose, and these alone shall party with us in Valhalla."

So there. :-)


Gods are all about the past; I'm more interested in the gods that are yet to be. It would be a shame if I were to piss off Roko's Basilisk!


Pascal's wager works as long as you think there is a reasonable chance of the Christian god existing. This would have been true in Pascal's time when everyone around you believed, and much of the knowledge we take for granted didn't exist. E.g. the theory of evolution, or any understanding of physics, etc.

The stronger version of the argument also is true provided you only care about maximizing expected utility. That is any finitely small probability that god exists times the infinite negative or positive utility of believing/not believing, and you the expected utility of believing is infinite.

The counter is "maybe there is a god but he punishes believers and rewards atheists." As long as the probabilities are exactly equal, they perfectly cancel each other out. However if they aren't exactly equal, if the probability of the Christian god is even slightly higher, then it will vastly overwhelm the expected utility calculation.

The consequence of this is you can go up to someone who believes in maximizing expected utility and say "I am a god. If you don't do what I say I will torture you for an infinite amount of time. If you think there as an even tiny probability that what I'm saying is correct, you must obey me." It doesn't just apply to the Christian god or even religion.

Thus expected utility maximization may not actually be the best way to make decisions.


If you have multiple possible gods, your utility maximizing option is to choose the god with the highest probability of being the correct one.


How do you calculate that probability?


Start with a prior based on your subjective view of the world and perform Bayesian updates based on evidence (e.g., the bible, koran, or currently unexplained miracles), same as you make other decision.

I just find this cheap dismissal of Pascal's wager to be mathematically nonsensical. It's almost surely not correct that the probabilities of all gods (real and honeypot) exactly cancel and therefore atheism is the correct choice.

Even if at some time those positive and negative utilities did cancel, even a small amount of evidence would shift the balance of probabilities pushing you back into Pascal's wager territory.


Since no data has ever been collected that would indicate existance of any god the only thing you can assume (thhough you shouldn't) is that they are all equally probable and since their number raises in time I'd advise you to pick the one that offers infinite reward for believing in him. I you don't know about such god just make one up. Of course that does not guarantee you infinite expected utility of your choice because someone could make up infinite number of gods. I'm pretty sure at least one person already did. Thus driving probablility of your god existing to exact zero and making expected utility of your choice indeterminate. Still you're better off than believeing in a god that offers finite rewards.

Of course you may just correctly reason that in absence of data you can't assume anything about gods and save yourself all the thinking about gods as devoid of any predictive utility.


While I agree with your argument, it doesn't really apply to Niels Bohr's horseshoe.

The cost of having a horseshoe over your door is zero, and if there is any negligible impact of it's positive luck, it's a worthy wager.


Unless the actual god really hates horseshoes.


This! Horseshoes might be the work of the devil! :-)


It might fall on your head...


Yes, but religions of the orient, do not actually punish you for not believing in God. There were atheists, agnostics, ... in India, long before they were tolerated in the West.

My interpretation, and Niels Bohr's, is that it doesn't matter if you believe or not, all that matters is that it brings you happiness, and does not cause harm to others. You don't actually have to believe in it.

Be that as it may, if you do interpolate it to less constricting religions, Pascal's wager does breaks down; but as you do say, it is an argument for an apologetic. I'm happier being an apologetic than to assume I'm rational.

I'm infact an agnostic, if you're inclined to believe me; but all this bashing of scientists for being religious is annoying. There are, things like, the world being created 6k years back, or that there lived a Super-monkey (Ramayana ?), which is plain silly. Superstitions like the one Sanal Edamaruku, also fall into this latter category. Anything that can be disproved by Science and reason, is not to be believed in.

Yes, the Middle ground is vague and ill-defined, but it's the only sane place to be.




Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: