Backgammon, Pascal, Sklanskyanity

[David Sklansky]

There is actually a connection between those three subjects.

Pascal's wager, I believe, is basically the idea that you should believe in the Christian God because even if there is only a small chance that he exists, you are getting essentially "infinite odds" since it will get you into heaven. Flaws in this idea include the fact that other religions might believe that such a belief guarantees that you DON'T get into heaven (since you are blasphemous) and the concept that if your belief is based on Pascal's argument, you don't truly believe. You can't make yourself believe if you don't.

But there is a somewhat altered version of Pascal's wager (The Sklansky-Pascal Wager) that doesn't have these flaws. The idea comes up all the time if you play bridge or backgammon. In those games you are frequently faced with situations where you cannot possibly win UNLESS something else is true. That something else may be quite unlikely. Your opponent will throw double sixes on his next roll. East is holding no spade. But those facts alone are not enough to win. It is also necessary that you play the situation correctly. Which means play the UNDER THE ASSUMPTION that he will roll double sixes or that East has no spade.

If you don't play that strategy you have no chance at all. However that doesn't mean that you should somehow persuade yourself that your assumptions are likely to be correct. It does no good to be inaccurate with your probability assessments if it flies in the face of evidence and math. Play optimistally while realizing that you were lucky if things turn out OK.

When I defined the basics of Sklanskyanity I mentioned that God wants his conscious evolved creations to "be happy". (Of course that was an oversimplification. I really should have said that he was concerned with their overall well being as a group.) Someone questioned the logical basis for this. I didn't answer at the time but my answer is simply that I assumed it to be true. Not a farfetched assumption if there is God. And without it there is no religion. Meanwhile if God wants to improve Man's overall well being, one way to do this is to be the third party in Prisoner Dillema situations who punishes non cooperaters. In other words make it worthwile to follow the Golden Rule.

So lets's change Pascal's wager thusly:

It is plus EV to ASSUME that there is a God who in some way will make it infinitely better if you follow the Golden Rule than if you don't. (Only if you believe that the chances of such a God existing is zero does this not work.)
Notice that this revised Sklansky- Pascal Wager eliminates the two earlier flaws mentioned. No reasonable person believes that DISOBEYING the Golden Rule may actually be a way into heaven. (As opposed to some religions that might believe that disbelieving in Christ is neccesary for salvation.) Secondly, anybody can ASSUME that God exists without actually believing that there is a high probability that he does. Any non zero probability in their mind suffices.

[Joe826]

But your new interpretation still doesn't escape the criticisms of Pascal's wager as it was. The obvious thing to do would be to say that perhaps God is actually an evil God, and he rewards us in the after life for doing everything we can to break the Golden Rule. Obviously such a scenario is far-fetched, but it's at least conceivable, so there is a slight chance that this is actually the case (no matter how infintesimal).

If we are arguing on the merits of Pascal's wager alone, we can't know which one of these two antithetical beliefs to accept, since they both have the same after-life implications. Of course, we could extend this example and make it even more hairy by talking about those other religions that could very well cast you into eternal damnation even if you do following the Golden Rule.

I just don't see how you escape this.

[David Sklansky]

The difference between your criticisms and the original ones have to do with reasonableness. Your technically correct argument is unreasonable. The original ones aren't.

[Joe826]

As far as I understand, reasonability isn't a part of Pascal's wager, and you've made it a part of yours. Ok.

I'm still not sure I totally understand, though. You've said that in order for there to be a God, it MUST be the case that he would have us follow the Golden Rule, so we ought to follow the Golden Rule JUST IN CASE God actually exists, so we'll be saved or what have you.

You've done nothing to defend your basic premise regarding the Golden Rule, other then to say that most normal people would think that should be the case. That's not really a sound argument.

Additionally, if we're going to make wagers on what's reasonable, it seems that accepting one of the world's religions with at least some historical evidence (like we know Christ did exist, etc.) would be much more reasonable than simply believing that we should follow the Golden Rule and we'll be ok. We have no evidence for this, outside of what you think God should be.

[jjacky]

[ QUOTE ]
You've done nothing to defend your basic premise regarding the Golden Rule, other then to say that most normal people would think that should be the case. That's not really a sound argument.

[/ QUOTE ]

i think thats the major flaw in the theory.

[West]

[ QUOTE ]
Additionally, if we're going to make wagers on what's reasonable, it seems that accepting one of the world's religions with at least some historical evidence (like we know Christ did exist, etc.) would be much more reasonable than simply believing that we should follow the Golden Rule and we'll be ok.

[/ QUOTE ]

I definitely disagree with this...

[West]

[ QUOTE ]
I'm still not sure I totally understand, though. You've said that in order for there to be a God, it MUST be the case that he would have us follow the Golden Rule, so we ought to follow the Golden Rule JUST IN CASE God actually exists, so we'll be saved or what have you.


[/ QUOTE ]

I don't think he is saying that it "MUST" be the case that he would have us follow the golden rule. He is saying that it is far more reasonable to assume that he would have us follow the golden rule than otherwise. And since following the golden rule ultimately results in things being "infinitely better" for you if such a god exists, then it is "+EV" to do so.

[PairTheBoard]

The belief that if there is a diety he must be a good diety rather than an evil one, is more a matter of faith than of reasonableness. You could cite a lot of evidence in making the case for an Evil Diety. Human beings don't seem to do this though. They consistently choose to ignore such evidence in favor of faith that God must be Good.

PairTheBoard

[David Sklansky]

"The belief that if there is a diety he must be a good diety rather than an evil one, is more a matter of faith than of reasonableness. You could cite a lot of evidence in making the case for an Evil Diety. Human beings don't seem to do this though. They consistently choose to ignore such evidence in favor of faith that God must be Good."

PairTheBoard

People sometimes take my posts too literally. What I was really tying to say had little to do with Pascal's wager. The main point was that for those who think there might be a God who punishes misdeeds, (especially if nothing good is done to make up for those misdeeds) there need not be something sinful about living your life as if there is such a god, even while you doubt his existence.

The fact that such an attitude would fix most of the problems with Pascal's wager was really mainly an afterthought.

[PairTheBoard]

David Sklansky -
"The main point was that for those who think there might be a God who punishes misdeeds, (especially if nothing good is done to make up for those misdeeds) there need not be something sinful about living your life as if there is such a god, even while you doubt his existence."

Ok. I see your point. To apply to Pascal's wager I assume the punishment and reward are super big, like hell and heaven? The problem with this for me is that even if I Knew such a God existed I would refuse to believe in him and insist on a better one.

PairTheBoard