Pascal’s wager revisited

I have previously written about how the ‘assume correct god’ objection destroys Pascal’s argument for believing in ‘God’ in principle. I’ll delve a bit deeper into this topic in the following. A particular feature of the model I first had in mind when discussing the subject was that I implicitly assumed an expected post-life utility of 0. You go to hell, that’s infinitely bad, you go to heaven, that’s infinitely wonderful, no gods means no afterlive, so all in all that would point to an expected post-life utility of zero. Believing in a God doesn’t change that, so no point in doing that. But is this argument correct? Let’s look a bit closer at the likelihoods.

The a prior expected post-life utility P(U) is: P(heaven)*U(heaven) – P(hell)*U(hell). [we don't need the 'no god' scenario in this calculation, just let U('afterlife' given no god)=0]

The utilities cancel out in the equation. But do the probabilities necessarily cancel out? I’d like to ponder this question a bit in the following.

The first thing to consider is that by picking monotheist god X, you loose the chance to bet on god Y (because god Y like God X do not like polytheists. We don’t consider the ‘no-hell’ alternatives here, they are not interesting). But if there’s an infinite amount of (potential) monotheist gods, which there is, it would seem that you would be ill adviced to bet on any of them. At first glance, it looks as if you pick a monotheist god, the likelihood of going to hell is infinity minus epsilon. If you assume that the likelihood of going to hell if you don’t believe is = 1, then of course you’re better off with a god than without one, but a difference of epsilon isn’t really anything to cheer about considering the outlook. The message here is: Pascal really should have focused more on hell, he really shouldn’t have wasted his time on heaven, the likelihood of any of us going to heaven given the above considerations is lower than the likelihood of me winning every lottery in Denmark for the next 100 years without ever buying a lottery ticket.

No, picking just one god doesn’t give you many winning chances at the lottery. So how do we improve the odds? Well, polytheism is a place to start. Picking at least a few different Gods would surely be a better idea than to stick with just one, that way you get to hedge your bet a little. But only a little. And you still have the problem that the monotheists might be right. Not only you might pick the wrong 10 gods, you still have the problem with the infinite amount of monotheist gods to deal with. This of course also leads back to an additional consideration when it comes to monotheism: By picking one god, you not only risk pissing off all the other monotheist gods, you might also anger Zeus and Vishnu. Yes, the likelihood of going to hell would still in such a model be 1 minus epsilon, but it sort of puts things in perspective, no? The other side of this coin is that even if you’re a polytheist, you can’t hedge your bet 100% by picking ‘all gods’ – even if you pick all the possible polytheist gods (which of course would be impossible) there would still be that infinite amount of annoying monotheist gods to deal with. Besides, for every ‘openminded/tolerant’ polytheist god you can think of, I can think of just as many that would never accept you worshipping both Odin and Zeus. Actually, the ‘hedging’ you’re doing isn’t really hedging at all, you just exchange one set of risks with another, without changing the overall risk level at all.

Now, it’s probably not true that the game is rigged so that we have such a low chance of picking the right God. God rigged the game himself, and for every god that invents a system rigged like this, I can think of a god that would rig it differently. So the problem of our low chance of success, which leads to an E(U-afterlife) much lower than zero, can thus be solved by stating the obvious fact that for every vindictive and egocentrical god that demands you worship him (/her) in order for you to go on to ‘eternal happiness’, there are just as many that just don’t give a damn what you do ‘down here’, they let you ‘move on’ no matter how much you abuse them during your life. In this way the expected post-life utility of zero can be restored.

One last thing to note. If any of the popular gods are true, those gods are necessarily ‘evil’ in that they all send a majority of humans to hell, keep on reincarnating us, or whatever else it is they do to punish us when we misbehave and don’t follow the divine plan, do as we’re told or whatever. Either your God-model, with a God that punishes unbelievers, is true, or God is good. You can’t have it both ways, for if you disagree, what we disagree about is the meaning of the word ‘good’. And if your model of God is the real deal, then that God to me is a terrible [something]. No matter which one(s): Pretty much all the Gods we humans have invented so far are a bunch of scumbags.

Lidt overvejelser omkring ‘Pascal’s Wager’

Hvorfor tror du på Gud?

Pascal’s Wager er manges svar. Lidt kritik her. Den del jeg vil bruge lidt tid på her er ‘assumes correct god’ argumentet. For mange af dem, der svarer ved at henvise til Pascal, er formentligt slet ikke klar over, at de reelt ikke svarer på det spørgsmål, jeg stillede dem.

Hvorfor lige netop GUD? Hvorfor ikke Zeus? Cæsar?

Det er ret væsentligt man får fat i den rigtige gud at tro på, hvis man accepterer Pascal’s argument. Hvis de accepterede argumentet, burde religiøse mennesker faktisk efter min opfattelse bruge noget nær al den tid, de overhovedet kunne slippe af sted med at sætte af til det, på at finde ud af, hvilken en af guderne, der er den rigtige. I mange af de populære religiøse varianter (kristendom, islam, jødedom) har man kun et gæt, og kan kun vælge en gud. Dertil kommer, at straffen for at vælge den forkerte gud i mange af varianterne er aldeles grusom. Hvis de religiøse troede så meget på helvede som de troede på himlen, ville de færreste af dem spilde al deres tid på en gud. Og de ville have langt mere udviklede argumenter for, hvorfor de troede på lige netop deres gud, end de fleste af dem har. ‘Mine forældre …’ holder ikke i byretten. At næsten alle religiøse overtager deres forældres religion peger på, at de ikke rationelt og aktivt selv søger at optimere ved at udvælge den religion, som de mener har størst chance for at holde dem ude af strafvarianterne (‘helvede’) og give dem en teoretisk mulighed for belønningsvarianten (‘himlen’/evig lykke/evig tvekamp på et gammelt slot med masser af mjød og vildsvin). Illusionen om den rationelle troende er, også på dette punkt, lige netop det: En illusion.

Det er i øvrigt værd at bemærke, noget jeg næsten altid kommer til at tænke på i relation til Pascal, at han udover at give en et lille argument for at vælge en eller flere af de mange guder at tro på, også giver et særdeles solidt argument for at nytænke lidt på det religiøse plan. Måske er den sande gud slet ikke opfundet endnu stadig et ubeskrevet blad? Måske er der mere end en?

Det synes på mig tydeligt, at langt de færreste troende, på trods af indsatserne der efter sigende står på spil, har tænkt særligt langt når det kommer til deres valg af trosretning. Det undrer mig. Og så alligevel ikke. Det er trods alt altid De Andre der kommer i helvede.

De fleste tror på noget, fordi døden er noget lort. De har ret: Døden er noget lort. Men det ændrer religiøse vildfarelser altså ikke på.