Saints and Simulators 17: #PascalReloaded

Seventeenth in an ongoing series about the interface between religion and technology. The previous episode is here.

Last week we discussed Newcomb's Paradox, a thought experiment about the rational response to an omniscient being, and also Roko's Basilisk, the frightening digital boogeyman the paradox spawned in the minds of those who pursued the train of thought too far.

It was finding myself in a similar bind as a young child that first pushed me in the direction of religious faith. I suffered from chronic insomnia, and was often troubled at night by the sheer terror at the thought of witches, vampires, and other supernatural monsters. I was not sure they existed, but if they did, I seemed entirely and helplessly at mercy of their sinister forces, against which I could not hope to protect myself. Eventually, however, I reasoned that it was ridiculous for me to believe only in supernatural forces that could harm me, yet not also in a supernatural force, God, that could help me. After all, the same things that made God difficult for me to believe in—that I could not see God, and had no direct evidence of God’s existence—were equally true of the evil spiritual forces that I had no trouble envisioning as absolutely, terrifyingly real.

Although I did not realize it at the time, this was in spirit, if not in detail, the same as the move undertaken by the great mathematician, inventor, and all-around genius Blaise Pascal, the father of modern statistics. A gambling man if there ever was one, Pascal reasoned that the gains of believing in God, if God exists, are infinite, and the losses from disbelieving in God, if God exists, are likewise infinite, while the gains and losses either way if God does not exist are nil. Therefore, the expected payout is all on the side of belief.

The setup of Pascal’s Wager, as this argument is generally known, is quite similar in form to Newcomb’s paradox. The glass box with the visible $1000 bill is your ordinary life on earth: you know it exists, and is yours to spend. The opaque box is your eternal reward. It might be empty, or it might be filled with a vast reward far beyond the one in the glass box. You will discover which one is the case only when you die and the box is opened. Do you take the glass box with the known, but finite reward, or the opaque box that could have nothing or everything inside it?

The big difference in Newcomb’s paradox and Pascal’s Wager is that in Newcomb’s paradox, the existence and the nature of the predictor is given as an assumption, and the core question is what prediction the predictor will make. In Pascal’s Wager, the core question is whether the predictor exists at all. Accordingly, criticisms of the Wager have centered around questioning three major assumptions: First, why assume there is only one opaque box, instead of hundreds, only one of which can be chosen, and only one of which has a reward; which is to say, how does Pascal know that he is betting on the right version of God, given all the hundreds of competing religions? Second, why assume that the predictor, God, will stock the box with an eternal reward, even in the case that God exists; which is to say, why assume the nature of the predictor is benevolent? Third, why is the act of claiming the box enough to receive it; which is to say, does it not take a more significant and consequential act of faith than making a wager to gain the rewards of the religious?

Pascal’s Wager is so often reduced to a slogan that it is not generally recognized that Pascal did consider each of these objections and provide a reply to them. For that reason, it is worth considering his actual argument at length:

If there is a God, he would be unfathomable, and infinitely so, since, having neither parts nor boundaries, he would exist in a way that we could not even imagine. Thus, we have no power to know either his nature of his existence. And because this is the case, it is impossible for anyone to solve this problem. Who would dare? Not we mere mortals, who are so disproportionate to him.

Who, then, will complain about Christians when they are unable to rationally account for their beliefs, since in truth no such account can be rendered? In proclaiming the faith to the world, they say it is a folly… and then you complain that they did not prove it! If they had managed to prove it, they wouldn’t be keeping their word, because it’s by lacking a proof that they show their sense.

This is just an introduction. Pascal acknowledges that we have no way of knowing God’s nature for sure. In other words, there is much that is unknown about the predictor, including whether or not the predictor even exists. He defends, however, the Christian belief in God as an act of pure faith, not based in reason.

'Yes,' you say, 'but even though this might excuse those who offer this religion as it is, and might free them from the blame of offering it without a rational argument, it doesn’t excuse those who accept it.'

Let’s examine that point, then: let’s say that God does or does not exist. Which side should we choose? Reason is powerless before such an issue. There is an infinite abyss separating us. At the far end of this infinite universe, a coin is tossed –which will turn up, heads or tails? What will you wager? Relying merely on reason, you can’t decide. You can’t rationally bet either way, for you can’t defend either choice.

Thus, don’t call people who have made a choice fools, for you know nothing about it.

'No, but I’ll blame them,' you say, 'for making any choice at all, because even if one man picks heads and the other man picks tails, both are equally at fault, for the right thing is not to bet at all.

Pascal suggests that we all must gamble either that God does or does not exist, and warns against criticizing anyone who has chosen one way or another, since we are all equally in the dark on the question. His hypothetical interlocutor then replies that since that is the case, no one should make any pretensions to having come to a meaningful decision either way.

But I say it’s necessary to bet. You cannot avoid it, for you are already launched on the waters. This being the case, which one will you take? How will you decide? Come now, since you must choose, let’s consider which one is of less importance to you. You have two things to lose –the true and the good, and two things at stake –your reason and your will, your knowledge and your bliss, and your nature has two things to shun –error and misery. Since you absolutely must choose… your intelligence will not be offended by one choice any more than by the other.

Pascal’s reply that we must wager may seem to deny the possibility of agnosticism. However, the binary choice he is presenting is between either becoming a person of faith, or not. It is either one or the other. Suspending judgment is still making the choice to not become a person of faith. As far the content of this forced choice, Pascal insists that one option is as rational as the other, since the actual situation is unknown.

Let’s play the game and weigh the consequences of playing if you take heads—that is, that God exists. Now let’s evaluate these two cases: if you win, you win everything, and if you lose, you lose nothing; so, consider that you needn’t hesitate. It’s a wonderful situation!

'Yes, I must wager,' you say, 'but I don’t want to bet too much.'

Here Pascal describes the potential outcomes, for all the world like a bookie in Vegas creating a payout sheet. His next move is his famously unorthodox introduction of statistical analysis into the world of religion:

Let’s see now, since the chance of winning and the chance of losing are even, then perhaps let’s say that you would win two lives for one, you could still bet, but if there were three to be won, you have to wager (since life forces you to play) and you’d be a fool, being forced to wager, not to risk your one life for three lives when the chances of losing and winning are even.

But there is is an eternity of life and happiness at stake. You have one chance of winning, and a finite number of chances of losing and what you are risking is itself finite, but what you could win is infinite.

Although it may be a blink-and-miss-it moment, this is actually Pascal’s answer to the scenario of hundreds of boxes (hundreds of religions) to choose from. Even if only one box is the right one, your possible reward is so big, and your risk so small, you should still take a chance on taking one of the boxes. If none of the boxes have anything in them, you are not losing anything by taking one, but if one box has an infinite reward, and you take none of them, you have lost all chance at that reward.

The choice is clear: there can be no excuses for timidity when an infinity of life is to be won in a game where there is a finite number of chances to lose as opposed to a single chance of winning…

'I admit it, I confess, but isn’t there any way of "peeking at the cards"?' Yes, the scriptures and the rest, etc.

The above elides a more repetitious part of Pascal’s argument in favor of another “blink-and-miss-it” moment, where he explains why he thinks God is benevolent. As it turns out, it is not an abstract God he is betting on after all, but the one that he has been told about by friends and mentors in his own Christian religious tradition. This might well be considered a bit of a cheat on Pascal’s behalf, since it contradicts his opening premise that we are going into this bet completely blind –a fact that he tacitly acknowledges by describing it as “peeking at the cards.”

'Yes, but my hands are tied [and] my lips are silenced. I’m being forced to bet and I’m not at liberty, they won’t let me go, and my nature is such that I haven’t the power to believe. What, then, would you have me do?'

It’s true, but at least wake up to the fact that your inability to believe comes from your passions, since reason induces you to believe but you still can’t. So, don’t strive to persuade yourself by counting up proofs of God’s existence; strive to diminish your passions. You want to find faith, but you don’t know the way. You want to cure yourself of your unbelief and you’re asking for remedies; learn from those who were once tied up like you and are now throwing the dice. They are people who know the path you’d like to follow; they are people cured of a disease from which you’d like to be cured –follow the way they started on.

They acted as if they believed… that will make you believe quite naturally, and make you more pliable to the faith.

Here is Pascal’s response to the final objection, that just claiming the box is not enough to actually receive it, that merely making a wager is not much of a profession of faith. His answer is that the wager is just the first step, and that entering a community of faith and taking on the rituals and practices of that faith will naturally and eventually lead to the genuine achievement of faith.

'But that’s what I fear.' Why? What do you have to lose?

How will you be harmed by choosing this path? You will be faithful, honest, humble and grateful; you will be full of good works, and will become a true, good friend to those who know you. What will you lose? Noxious pleasures, vainglory and riotous times, but those losses will be easily supplanted by other, greater joys.

This is the final kicker to Pascal’s argument, the dramatic revelation that you do not have to risk the $1000 in the glass box, your ordinary life on earth, to gain the opaque box after all. Instead, your ordinary life actually gets better. If you claim the opaque box, whether or not it has anything in it, the money in your glass box actually doubles. You lose nothing, and gain something by believing in God, Pascal claims, whether or not God actually exists.

That was the conclusion my younger self came to as well. In my case, what I was gaining was a sense of protection from unseen spiritual dangers, and my reward for my newly embraced beliefs was better sleep. Of course, Roko’s Basilisk, the boogeyman lurking in the closets of the Less Wrong community, is neither precisely a supernatural nor a spiritual threat (although, as an all-powerful entity that can neither be seen nor perceived, and that may or may not exist, it is difficult to characterize as either naturalistic or empirical). Is the Pascalian move I made available to those oppressed by this concept? Can they escape the terror of the Basilisk by belief in an even more powerful good entity? God, as a concept, is considered unfashionably musty and out-of-date in modern intellectual circles; but is there, somewhere out there, an Even-Friendlier AI?

It is at this point that we return to Nick Bostrom, the theorist we met in the all the way back in the first installment, who believes it likely we live in a computer simulation, and who has demonstrated this as a logical and apparently inescapable conclusion of a belief in a naturalistic, physicalist universe, to the satisfaction of billionaires and bank analysts alike. In particular, we highlight the surprising fact that he himself identifies our hypothetical simulator as God—or at least as “a god”:

Although all the elements of such a system can be naturalistic, even physical, it is possible to draw some loose analogies with religious conceptions of the world. In some ways, the posthumans running a simulation are like gods in relation to the people inhabiting the simulation: the posthumans created the world we see; they are of superior intelligence; they are “omnipotent” in the sense that they can interfere in the workings of our world even in ways that violate its physical laws; and they are “omniscient” in the sense that they can monitor everything that happens. [emphasis added]

Behold, then, the great prodigy of our age: a man capable of demonstrating the probable existence of God as a casual aside to his main argument!

But are all those who accept Bostrom’s argument, as given, de facto people of faith (seemingly believing in not just one, but an entire hierarchical pantheon of simulationist gods)? If so, that fact might come as a shock to the many committed atheists among them.

References