Pascal’s Wager

In yesterday's post I described an argument that reminded people of something called Pascal's Wager.

In a nutshell, Pascal was a dude who argued you should consider Christianity because if it's true, the downside of not believing is eternal Hell. But if you become a Christian and there's no God, all you've lost is your Sunday mornings. (Here I am simplifying.)

Many of my blog readers left comments alluding to the well-known "flaws" in Pascal's argument. Here’s a handy list of them.

http://en.wikipedia.org/wiki/Pascal's_Wager#Rebuttals

Chief among the alleged flaws in Pascal’s argument is that you still have to pick the correct religion among many, or else you go to Hell anyway.

Sure. But picking any religion that promises salvation slightly improves your odds over picking an option that doesn’t. You're still probably doomed, given your bad religion-picking skills, but a one-in-a-million chance of reducing the risk of eternal Hell is a move worth taking, mathmatically speaking.

Another noted "flaw" in Pascal's wager is that you can't rule out the possibility that only skeptics are spared from Hell. Perhaps, it is argued, God loves the spunky fact-loving personality of skeptics and saves them alone, or saves them in the greatest percentage.

That argument passes the math test, but does it pass the sniff test? It’s a viewpoint that exists only as a debate tool. While we can't rule it out, surely it is the worst bet if you must pick a theory of God. No rational person on earth, including skeptics, has concluded that God prefers skeptics over believers.

Personally, if I were more rational, and less focused on immediate gratification, I would become a moderate, peace-loving Muslim. My reasoning is that Islam has the best chance of becoming the dominant world religion in the future, and therefore probably has God’s backing, if he exists. The Muslim belief that death is sometimes a good thing is a huge advantage in a future where weapons are improving, and the only thing keeping people from using them is fear of death.

If you believe God exists, the smart money says he’s backing the team with the best strategy and long term viability. Based on what I see today, I’m betting on Islam being the only religion in a thousand years. Once you can build your own nuke from stuff you buy online, don’t be betting on the Buddhists.

I realize it's unscientific to try and compare one absurdity to another. But if you assume our perceptions are often flawed, you have to allow the possibility that some apparent absurdities are due to our limited powers of perception. So, for example, while the notion of a loving God who allows eternal damnation seems absurd, it is less absurd than assuming the world is run by invisible unicorns, or that God discriminates against those who believe in him.

The God theory has built into it the assumption we are not bright enough to understand the mind of an omnipotent being. That sounds reasonable. Hey, if God exists, and he does things different that I would, just maybe the problem is on my end. If you believe in God, the apparent absurdities have a reasonable explanation, even if wrong.

But what’s the reasonable explanation for God preferring skeptics? If God appreciates reasoning skills, he can’t be too impressed by the fact he created the entire Universe and skeptics still can’t find any good clues he exists. God would only be impressed by skeptics if God did NOT exist. You can’t top that for absurdity.

Back to Islam. It also has the most satisfying answer to the multiple prophet issue. If we assume God speaks through prophets, as all God-oriented religions do, then how can you be sure the last prophet finished the job? Islam gives us Mohammed, the "seal of the prophets," and promises that God intends him to be the last one. That’s a tidy package.

All the other religions seem to leave open the possibility that God has a few more prophets up his sleeve. If you bet on one of those other religions, you can't know for sure if you're living by God's first draft or his finished manuscript.

Picking the "right" religion is a long shot no matter how hard you try. But if rational thought has any value at all, it's in narrowing down options and improving our odds of making good choices. Rational thought hasn't led anyone to conclude that there's a God who only saves people who don't believe he exists. We can't rule it out, but can't we rate its likelihood compared to a God who prefers that his lumps of clay hold him in higher esteem than their own eye crud?

I’d prefer to make all of my decisions on the basis of peer reviewed science. But I don’t have that option when considering the great beyond. So I settle for looking at the competing absurdities and picking the one that seems relatively least absurd.

There are other arguments against Pascal’s wager, but none of them looked any stronger than the ones I mentioned here.

Pascal's wager was designed to make people consider the importance of grappling with the question of a Christian God. That's not my argument. I simply borrowed the math part of his argument and followed it to its logical conclusion: The most rational worldview is moderate Islam.

And since I am not a Muslim, I must conclude that I am not rational. I sure hope God likes moist robots.

I wrote this at 3 AM. I won’t even pretend it makes sense.

[Update: Early commenters are noting that someone who only pretends to believe, to get into heaven, won't fool God. But the science strongly supports the notion that going through the motions of believing will, in most normal people, turn them into believers in fact. It's the same principle as why people who join a political party start to believe most of the positions of the party. -- Scott]

Comments

[seeker]

Scott,

I have posted the first of two commentaries on this post on my site - see
- Scott Adams continues to question atheism http://www.twoorthree.net/2007/08/scott-adams-con.html

The appalling fact that you use one point of reason to pick moderate Islam as the most "reasonable" faith requires a much longer response - but you can get a head start on my thinking with
- What are the main differences between Islam and Christianity? http://www.twoorthree.net/2006/08/what_are_the_ma.html
- There are no moderate Muslims http://www.twoorthree.net/2007/02/there_are_no_mo.html (not what you might think).

Even moderate Islam is a weak intellectual world view, with nary a valid or useful ethical, moral, intellectual, or soteriological framework.

(sorry if this is a duplicate)

[seeker]

Scott,

I have posted the first of two commentaries on this post on my site - see
- Scott Adams continues to question atheism http://www.twoorthree.net/2007/08/scott-adams-con.html

The appalling fact that you use one point of reason to pick moderate Islam as the most "reasonable" faith requires a much longer response - but you can get a head start on my thinking with
- What are the main differences between Islam and Christianity? http://www.twoorthree.net/2006/08/what_are_the_ma.html
- There are no moderate Muslims http://www.twoorthree.net/2007/02/there_are_no_mo.html (not what you might think).

Even moderate Islam is a weak intellectual world view, with nary a valid or useful ethical, moral, intellectual, or soteriological framework.

[Adrian D.]

Stomper:

The biggest problem with books like the bible is not that they are materially false, but rather that they endorse harmful actions. There are too many people willing to do evil for the sake of "god".

[Robert]

I note that majority of you posters are those that hold with anostic or atheistic beliefs. I hope most readers do not get the opinion that such are the majority. Most likely, those of us who are religious and have a firm belief system know that on anything but the simplest topics you can not convince someone else with a brief resposne.
The truth (not just relgious truth) tends to be complicated a lot more then easy sound bites and spurious remarks. There are quite logical ways to argue FOR God, but they are no so simple and simple minded as talking points againt Him.

[Stomper]

Dunno about Catholics, but most of the Protestants I know have indeed read large parts of the Bible. Many Christians and Jews are not Biblical literalists. Many of us are content to read the Bible for useful lessons, without insisting that every word be taken as literally true.

At my church, some creationists showed up, and a whole roomful of adults just laughed at them. Not to be cruel -- we just didn't realize they were serious. We thought we were laughing WITH them, until we realized they weren't laughing. We were all embarassed, and we never saw them again.

So I agree the Bible is not literally true, I generally believe that Evolution is a useful theory (and probably mostly right, even though incomplete). I don't see science and faith as being at odds, and disproving the literal truth of the Bible doesn't affect my faith (or the faith of the substantial minority like me).

So again, please don't stereotype us. The fundamentalists are louder and may get more press, but that doesn't mean they speak for all of us. There are many different, nuanced approaches to faith. As soon as you make a generalization about what the faithful believe, you are wrong.

--Stomper

[adora]

Nothing makes you give up Christianity faster than reading the bible. I grew up a Catholic, and I'm convinced that 99% of Christians don't read the bible. They're just skeptics doing it as an insurance policy.
There is a major problem for using bible as a moral guide:
- It's OK to sell your daughters to slavery.
- We should stone disobedient children to death.
- Natural birth control. (i.e. Starvation)
They aren't exactly as printed because I read the Chinese version. There are many more of these, I feel embarrassed for people who try to prove the stories are true. Look! Fossils!
Here's the catch: You *might* go to heaven for killing your devil-possessed talk-back children, but you're definitely going on death row.
I don't usually argue. Just like you wouldn't tell people that you can't see their imaginary friend.

[Stomper]

Pascal's Wager is flawed by its assumption that the only (or at least primary) function of faith is to avoid Hell and gain entry to Heaven. Then, Scott and numerous posters extend that flaw by addressing the quasi straw-man argument that you can only win that wager by picking the single, precisely correct approach to faith.

A just God won't reward or punish people based on whether they guessed right or wrong. If there was a specific set of rules God wanted us to follow, then a just God certainly could have been much clearer about which set of rules applies -- i.e., by striking down false prophets, by correcting transcription and translation errors, and by clarifying apparent ambiguities.

The fact that none of this has happened tells me very clearly that God is not concerned about which version of faith we adopt. Rather, I believe (no, I can't prove it -- that's why it is called "faith") that we are invited to appreciate the mystery of a life in which some things can never be proven, and then we are judged by how we respond to that invitation.

I call it a "quasi" straw-man argument, because there are actually many believer of all faiths who DO believe that theirs is the only "true" way (as though the empirical and logical concept of "truth" can even be applied to faith). However, that majority does not justify non-believers in stereotyping all of us who have faith.

Are we "required" to "waste" our our time by (i) praying 5 times a day, (ii) spending Saturday or Sunday mornings going to church, or (iii) otherwise engaging in formal worship? No, but here's a news flash: some of us find these activities rewarding, whether spiritually, intellectually, musically, or otherwise. We engage in formal worship because we want to, not because we are compelled to. For us, worship is no more a waste of time than a workout at the gym. Formal worship helps us stay spiritually healthy, but there are other ways to maintain spiritual health.

So there are several silly arguments implied by Scott's post, and/or appearing in the comments (in fact, these arguments appear many times). It would be nice if people would expressly recognize their assumptions, and then acknowledge that those assumptions are based on generalizations. I'm not holding my breath, though.

--Stomper

[extabgrad]

All you've done is assert that the odds of a non-believer going to hell if a god exists are greater than the odds of you picking the wrong way of worshipping god. Without any analysis of the mathematics. You've also skipped carelessly over the negative aspect of choosing belief in the event that there is /no/ god. In actual fact it is trivial to stick in numbers that show that one is /far/ better off betting on there being no god, because the odds of choosing the right way to worship god are so vanishingly small that the /guaranteed/ benefits of non-belief become significant in comparison.

Plus a major problem with Pascal's wager is that it completely avoids any analysis of the odds of gods existing. The actual odds of there being a god that throws non-believers in hell, when one notes that the universe is clearly completely indifferent to mankind's existence, is infinitessimal. Add that to the tiny odds of choosing the right way to worship god and non-belief wins hands down.

[Charlie Wood]

I haven't read all the comments so forgive me if this has already been mentioned.
I'd like to playing devil's advocate (har har har) on the subject of skeptics being rewarded. I'm a pretty strong believer in the argument that as flawed finite beings we can't perceive the mind of an omnipotent, infinite one, almost to the point where applying any rationality is useless. A mouse has a much better chance figuring out why a scientist is experimenting on him than we do when pondering why God might reward or punish skeptics since both mice and people are have finite knowledge and ability.

That being said and even though my attempt at reason is fruitless consider the situation between a white lab mouse in the maze and the experimenting scientist. To the mouse the scientist looks like a God. He's omnipresent, reaching anywhere in the maze around the walls. He gives the mouse food, water and shelter. Now the mouse in the maze runs around looking for the cheese at the end. If the mouse by some chance develops the false belief that some random activity (sitting still, walking repeatedly down dead ends, praying every day facing Mecca, etc) will be rewarded by this god-scientist, the mouse is going to starve since it's never going to get to the cheese. Now if the mouse attacks the maze with a systematic, scientific approach, with the attitude that it's not going to be rewarded unless it works out the maze by itself, then it's bound to find the cheese sooner or later.

We've all been raised in the presence of the idea of a Judeo-christian-muslim god as some loving god that cares for his creation so naturally a scientific god who tests and rewards skepticism and logic seems strange. There are a number of reasons for this behavior. Maybe the god (or whatever being set things up, not necessarily omnipotent or omnipresent, just far enough beyond us to seem so) is simply running a behavioral experiment that we have as much hope of understanding as the average lab rat does. Maybe a god that loves order (having created the cosmos and all it's intricate components out of chaos) prizes intelligence and logic above blind faith as more orderly systems. Whatever its reason (although I think any infinite being would have a much better system on which to make decisions than reason, which is simply a tool human beings evolved over the millennia to help us survive), I think the idea that a god rewards skeptics is at least as plausible as one that rewards blind faith. Even if it isn't plausible, as I mentioned before why should we think that our reason has the ability to make any headway in the analysis of the behavior of a being so far beyond our abilities.

[jesse]

It seems that if there is (a) a God (proper noun) and (b) no conclusive evidence either way as to God's existence, then the logical place to find out would not be from conclusive evidence at all, but to ask God.

There are, of course, obstacles to asking God if he exists. #1 is the problem of ordinary social pressure. It feels silly. #2 is the problem of address; if you have already made up your mind that God does not exist, then you wouldn't really try to ask. Saying "God, do you exist?" in a sarcastic tone wouldn't really draw an answer. If you were subconsciously scared, etc, that could ruin the meaning of your question. Only if you genuinely wanted to know would you find out.

This kind of evidence, which would be spiritual and personal, would of course be unable to convince anyone else if obtained. It could only draw others to ask the same question. That's what scriptures are really for. That's why the Bible contains testaments. If they believed it, it might be true. If it is true, you can ask. If you can, do.

[Mark Robinson, Eindhoven]

Hi Egg - I'm back - it's all true...

[David MacMillan III]

I don't know if anyone else said this yet, but....

Pascal's Wager is, at face value, pretty much bunk. To say "You should be a Christian because the risk/benefit arrangement is better than atheism" is a false dichotomy.

But Pascal's Wager is excellent if taken differently - instead of being used as an argument to BECOME a Christian, or a Muslim, or a Buddhist, it should be used as an argument to EXPLORE Christianity, Islam, or Buddhism. Actually, not Buddhism; they see conscious existence as a curse. But that's beside the point.

In other words, since the risk/benefit of this particular traditional religion is a heckuvalot better than traditional atheism when the two are compared, maybe you better see how likely that traditional religion is to be true.

Which is what Scott does quite often....

[Kthulu]

Had a though. Maybe I am god. Odds are low, but greater than zero (as in infinite universe nothing is 0% or 100%, we are usually rational and assume close to 0 can be taken as actual zero for convenience, but this is discussion on religion so what has being rational got to do with it.) So there is a non zero chance I am god. If I am god then in the afterlife I will sentence all who don't believe in me to an eternity of hell (seems to be what gods do, I wouldn't want to let the side down.) So if you do not believe I am god the risk you are taking is a finite non zero chance multiplied by infinite penalty of eternity in hell. Finite number greater than zero times infinity gives infinity.

So that is infinite risk if you don't worship me. Get on your knees and praise me now!.

[sam gates]

i think the logical conclusion of pascal's wager is to cover as many bases as possible. right now i'm good for buddhism, hinduism, relativism, mormonism, most moderate islam, most judaism, and last but not least, Christianity. most religions would put me somewhere in second or third tier paradise, or purgatory like Dante's Virgil. nice post though, it seemed reasonable to me.

[TallDave]

"My reasoning is that Islam has the best chance of becoming the dominant world religion in the future, and therefore probably has God’s backing, if he exists. "

Not even close. Christianity is already the numerically dominant religion and is still gaining adherents in Africa and Asia. Islam is not gaining adherents anywhere (except, marginally, among disaffected Westerners such as in the prison system); they are only growing through immigration, driven by those fleeing the poverty which stems from Islam's generally repressive practice.

This is not a recipe for success.

You have to realize, Islam has generally been spread by the sword, not proselytization. That's why it's called "Submission" while Christianity is the "good news" that God so loves us that he sent His only son to die for our sins. I'm a Deist and not sure if I buy all that, but the meme is both uplifting and infectious. I have met recent converts in Asia and this idea of a God that loves them so much makes them very very happy.

Islam peaked around the Second Siege of Vienna, the last time military expansion of their religion was viable, and has been losing ground ever since. Much of their early success in expansion was driven by Christianity's essential pacificism, which so enervated the military power of Rome that by the time Muslim invaders came, the former Imperial provinces were only fielding a tenth of the troops they had several hundred years earlier.

Here's a video of the spread of the major religions.

http://video.google.com/videoplay?docid=1032421163698055013&q=spread+of+religion+video&total=1027&start=0&num=10&so=0&type=search&plindex=1

[Ivy]

God prefers skeptics. He is a jealous God and takes it personally when his creation claims to know His devine truths. The religious people are damned...and the rest of us standing here scratching our heads will be rewarded for our humility. But I don't actually KNOW this....it's just a theory. Honest, I'm not claiming to know anything.

[ipsissimus_clay]

We get that Scott's post was funny Maggie - that's probably why most of us started to read his blog in the first place. But it's quite possible [if not entirely common] to be both funny and thought-provoking at the same time - a particular skill which is what, personally, keeps me coming back here. And Scott, in his funny and thought-provoking way, even coined a term for this: philosotainment.

[Paul C]

"How can you even consider being a Muslim as the most likely religion? They do not even have a football team."

Yes they bloody well do, the Iraqui national side just won the Asian Cup, or didn't American TV report all the (non-violent and this time not anti-American)jubilation in the streets? Iraq should also do well in the World Cup proper and might give the top European and South American national sides a few shocks, whch is all for the good!

[ronbus]

Basically what you're saying is that the "religion" that comes up with the worst outcome of not believing in it, wins. Similarly, the "religion" that comes up with the best outcome of believing in it, wins. Since it's possible for two different "religions" to come up with one of the outcomes above, and both can't win, one needs to decide which is better: avoiding the worst outcome, or going for the best outcome. That's a little like choosing to have the most wonderful sex everyday of your life (if you enjoy sex) and dying a miserable death at the age of 90; or not every having sex ever but never feeling pain.

Of course, my main point is that if you choose a "religion" based on the downside of not believing, or upside of believing, you will always run into problems. Someone can always come up with a new "religion" that promises worse outcomes of not believing, or better ones of believing. You would be caught in eternal religious band-wagoning.

You would also be susceptible to being taken advantage of by those looking to profit from your belief. Take the Heaven's Gate people for instance. Granted, that's an extreme case, but if you would be tortured forever in hell if you didn't kill yourself, it would be better to kill yourself. Most religions take less extreme trade offs: believe in the lord (and pay me some money) and you'll be saved; much easier trade.

I myself am agnostic since anything is possible, but most things are unlikely.

[Kevin Kunreuther]

My last comment on subject, which is actually Scott's, when I think about it (BTW Scott, what happened to my first comment, not the reply to Nathaniel, the one recommending we pray or give thanks to the sun, the earth and ourselves?)
Everyone download and read (or reread) God's Debris, Scott's lovely little thought experiment. It's on the Dilbert website.

[Air Phloo]

Another consideration is the size of the reward/penalty. I think most religions have a fairly equal probability of being true (or at least true enough). So, why not choose the religion with the best reward. That is Mormonism - not many other religions offer you the chance to become a god. I figure a .0001% chance of omnipotence is better than the .0001% chance of being Allah's eternal servant.

[Kevin Barnes]

I am going to start a Primitive Greek Mythologist movement. The odds suck of anything good coming of it. My buddies love when I come over for poker night.

[Joey]

Two things I have against this refutation of the refutation of Pascal's Wager:

1) As they say, a bird in hand is worth two in the bush. If I go to church every sunday, that's a certain inconvenience, which I'm trading for the mere chance of reward. In other words, the negative value of a Christian-in-a-Godless universe's life is certain even if trivial, while the negative value of an atheist-in-a-godfull-universe's, or even the negative value of a christian-in-a-different-godful-universe's are uncertain and of minuscule probability.

2)How about non-eternal punishments/rewards. What if in the afterlife you get to have sex one more time, then you're gone for good? Would that be worth a lifetime of devotion and cost? Would that be worth killing yourself at the age of 20? Rationally, non-eternal afterlife is far less absurd than any sort of eternal afterlife. How cant he human mind possibly stay intact for eternity? I guarantee you'd go insane by a billion years, and that's 0% of eternity. No eternity, even one of bliss, would be happy or even not destroy your mind. It's completely absurd, I would even say impossible for eternal afterlives to exist.

Therefore, the probability of Pascal's wager being statistically worth taking drops infinitely.

And btw, when it comes to religions that claim to have non-eternal afterlives, your choices are very limited, meaning your true best bet is to make one up, or believe in something with a completely different type of post-death experience.

[Hank]

"If we arose here through random events that we're planned by nobody and that are being watched by nobody,,, and if when we die we simply cease to exist,,, then human life has zero true value"

Why exactly is the psychological belief in a controlling, observing god offering non-physical life after death the ONLY THING IN THE ENTIRE WORLD that gives human life any value to you?

If you sit and think for two seconds, you can probably come up with some better, more realistic reasons to assign value to human life.

[mick]

Religion is to civilization as gravity is to earth, a basic construct to keep things orderly and predictable. However, man designed religion and used crude tools such as fear and hate so we should not be surprised that it fails in its mission so completely. Read the essays of Bertrand Russell for a well reasoned viewpoint on god and religion