Pascal’s Wager

Heaven and Hell, or how to hack a hedonic mind

PNRJ cognitive science, core principles, current events, ethics, heuristics and biases, tribal paradigm , , , , , , , , ,

Sep 13 JDN 2461297

I am writing this on the 25th anniversary of 9/11, perhaps the most vivid moment in recent history in which religious beliefs directly caused horrific violence. Yes, there were political motivations for the attacks as well; but the way that the conspirators spoke of Islam and the afterlife and the fact that they were willing—eager, even—to sacrifice their own lives in the service of their cause shows that religion was also a very important motivating factor.

I’m not going to spend this post re-litigating the American and international response to the attacks, except to say what I already said over a decade ago: we sacrificed vital civil liberties for more or less nothing. (I would also like to note how prescient this line reads today, written in 2015: “If we’re worried about millions of people dying, it is far more likely that there would be a repeat of the 1918 influenza pandemic”)

Instead, I’d like to talk about how religion made 9/11 a rational behavior.

If you allow the premise that there is a Heaven in which we will receive eternal bliss if we do the right thing, and a Hell in which we will receive eternal suffering if we do the wrong thing, and then posit that the right thing God wants you to do is to crash an airplane into a skyscraper, it is rational for you to crash an airplane into a skyscraper.

If Heaven and Hell are at stake, any action can be made rational.

In order to avoid this conclusion, you either need to deny the rationality of religion itself, or you need to somehow ensure that whatever actions are prescribed by religion are good ones. But people don’t always agree on what actions are good—especially if they come from different religions—and if you already know which actions are good… what did you need religion for?

Is the idea perhaps that most people are too stupid to understand what is really good, and so instead we have to motivate them with lies? The people who are smart enough to understand ethics come up with the rules, and then they pass down these rules to the priests so that the priests can convince everyone else that following those rules leads to Heaven and breaking them leads to Hell? Is that why religious people seem so afraid of atheists undermining the authority of religion? You think most people can’t handle the truth? That if you stopped telling people that they are ruled by an almighty totalitarian who is willing and able to inflict eternal suffering on the disobedient, they wouldn’t understand why murder is wrong?

The better answer is to deny the rationality of religion itself.

This is the problem with Pascal’s Wager: If you are motivated by expected utility (which, most philosophers agree, is the rational way to behave), and you allow that there are outcomes which result in infinite utility, then only those outcomes matter, and you can be convinced to do literally anything in order to achieve those outcomes, as long as their probability of occurring is not exactly zero.

And if there are two different possible outcomes with infinite positive and negative utility attached (e.g. maybe Christianity is true, or maybe Islam is true)? Then your decision becomes undefined. Every possible action results in infinity minus infinity expected utility.

But it actually makes perfect sense that people would try to motivate you this way.

Given that you are a hedonic mind, that is, one which is motivated to seek pleasure and avoid pain, and I want to give you some belief that will make you do whatever I want, how can I have the best possible chance to convince you to do whatever I want you to do?

The answer, of course, is to convince you that doing what I want will result in infinite positive utility—eternal bliss—while doing whatever I don’t want will result in infinite negative utility—eternal suffering. I should tell you that doing what I want sends you to Heaven and not doing what I want sends you to Hell.

If I can convince you of that, I can make you do whatever I want. Once you hold that belief, obeying my every command is the only rational action for you to take, for whatever harms it may cause, they could never be as bad as eternal suffering, and whatever benefits you must forgo to do them, they could never be as good as eternal bliss.

Heaven and Hell are an optimal hack for a hedonic mind.

They are like the sudo command in Linux: Promising Heaven and threatening Hell grants full root access and allows you to make any possible command with absolute authority. In Linux operating systems, you need to already have administrative privileges to use sudo; in religion, I suppose what you need is the spiritual authority of a religious leader (or enough charisma and skill at deception to found your own cult).

Or to make it a little more concrete: They are like a gun held to your head, a gun that is so powerful it can make you suffer for all eternity.

And hey, maybe you’ll get lucky, and what whoever is in control of your beliefs (your priest? your holy book? your church community as a whole?) will actually motivate you to do good things! Maybe it won’t be so bad to be a mindlessly obedient automaton as long as you are programmed for good!

But what if you don’t get lucky? What if the person holding an infinite gun to your head orders you to do something bad? You’ll do that instead. The moment you accepted the belief that certain actions result in infinite happiness or infinite suffering, you gave up all your rational autonomy. As long as that belief remains in your head, the only rational thing to do will always be whatever sends you to Heaven and keeps you from going to Hell.

This is what Voltaire was talking about when he said:
“Those who can make you believe absurdities can make you commit atrocities.”

The answer, of course, is to never accept that belief.

There are various tricks you could use to do that: One would be to temporally discount the future so that bliss millennia from now means almost nothing to you. Another would be to assign some sort of arbitrary cap on how high or low the utility from a given outcome can be.

Alternatively, you could insist upon assigning a probability of zero to any belief that would result in infinite utility—just categorically say that you will never believe such a claim. This could actually be accomplished with a nice, smooth prior that is a normal distribution over outcomes, for while any particular positive utility has a positive probability, there is no probability of an infinite outcome. And the larger the utility that is promised, the lower the probability you’ll accept it—as I previously suggested as a reply to Pascal’s Wager (and Pascal’s Mugging).

In practice, I think people sort of do this.

I think that, deep in their minds, most people have some sort of “reasonableness checker” heuristic that tries to ensure that nobody is able to convince them to behave in horrible ways by making them adopt crazy beliefs. I believe that the reason most people do not commit atrocities is that they cannot really be convinced to believe absurdities—not completely. They can believe in believing them, but they can’t really believe them.

But then, there are the people who don’t have that heuristic, or for whom it has malfunctioned in some way, and those people actually believe the crazy beliefs they were told to believe. And those people can be motivated to commit atrocities—and so several of them did, 25 years ago.


Want to support this blog? Join my Patreon, and get early access to all draft posts before they go live! https://www.patreon.com/c/pnrjulius

The afterlife

PNRJ cognitive science, core principles, ethics, Logic of Kindness , , , , , , , , , , , , , ,

Dec 1 JDN 2460646

Super-human beings aren’t that strange a thing to posit, but they are the sort of thing we’d expect to see clear evidence of if they existed. Without them, prayer is a muddled concept that is difficult to distinguish from simply “things that don’t work”. That leaves the afterlife. Could there be an existence for human consciousness after death?

No. There isn’t. Once you’re dead, you’re dead. It’s really that unequivocal. It is customary in most discussions of this matter to hedge and fret and be “agnostic” about what might lie beyond the grave—but in fact the evidence is absolutely overwhelming.

Everything we know about neuroscience—literally everything—would have to be abandoned in order for an afterlife to make sense. The core of neuroscience, the foundation from which the entire field is built, is what I call the Basic Fact of Cognitive Science: you are your brain. It is your brain that feels, your brain that thinks, your brain that dreams, your brain that remembers. We do not yet understand most of these processes in detail—though some we actually do, such as the processing of visual images. But it doesn’t take an expert mechanic to know that removing the engine makes the car stop running. It doesn’t take a brilliant electrical engineer to know that smashing the CPU makes the computer stop working. Saying that your mind continues to work without your brain is like saying that you can continue to digest without having a stomach or intestines.

This fundamental truth underlies everything we know about the science of consciousness. It can even be directly verified in a piecemeal form: There are specific areas of your brain that, when damaged, will cause you to become blind, or unable to understand language, or unable to speak grammatically (those are two distinct areas), or destroy your ability to form new memories or recall old ones, or even eliminate your ability to recognize faces. Most terrifying of all—yet by no means surprising to anyone who really appreciates the Basic Fact—is the fact that damage to certain parts of your brain will even change your personality, often making you impulsive, paranoid or cruel, literally making you a worse person. More surprising and baffling is the fact that cutting your brain down the middle into left and right halves can split you into two people, each of whom operates half of your body (the opposite half, oddly enough), who mostly agree on things and work together but occasionally don’t. All of these are people we can actually interact with in laboratories, and (except for language deficits of course) talk to them about their experiences. It’s true that we can’t ask people what it’s like when their whole brain is dead, but of course not; there’s nobody left to ask.

This means that if you take away all the functions that experiments have shown require certain brain parts to function, whatever “soul” is left that survives brain death cannot do any of the following: See, hear, speak, understand, remember, recognize faces, or make moral decisions. In what sense is that worth calling a “soul”? In what sense is that you? Those are just the ones we know for sure; as our repertoire expands, more and more cognitive functions will be mapped to specific brain regions. And of course there’s no evidence that anything survives whatsoever.

Nor are near-death experiences any kind of evidence of an afterlife. Yes, some people who were close to dying or briefly technically dead (“He’s only mostly dead!”) have had very strange experiences during that time. Of course they did! Of course you’d have weird experiences as your brain is shutting down or struggling to keep itself online. Think about a computer that has had a magnet run over its hard drive; all sorts of weird glitches and errors are going to occur. (In fact, powerful magnets can have an effect on humans not all that dissimilar from what weaker magnets can do to computers! Certain sections of the brain can be disrupted or triggered in this way; it’s called transcranial magnetic stimulation and it’s actually a promising therapy for some neurological and psychological disorders.) People also have a tendency to over-interpret these experiences as supporting their particular religion, when in fact it’s usually something no more complicated than “a bright light” or “a long tunnel” (another popular item is “positive feelings”). If you stop and think about all the different ways you might come to see “a bright light” and have “positive feelings”, it should be pretty obvious that this isn’t evidence of St. Paul and the Pearly Gates.

The evidence against an afterlife is totally overwhelming. The fact that when we die, we are gone, is among the most certain facts in science. So why do people cling to this belief? Probably because it’s comforting—or rather because the truth that death is permanent and irrevocable is terrifying. You’re damn right it is; it’s basically the source of all other terror, in fact. But guess what? “Terrifying” does not mean “false”. The idea of an afterlife may be comforting, but it’s still obviously not true.

While I was in the process of writing this book, my father died of a ruptured intracranial aneurysm. The event was sudden and unexpected, and by the time I was able to fly from California to Michigan to see him, he had already lost consciousness—for what would turn out to be forever. This event caused me enormous grief, grief from which I may never fully recover. Nothing would make me happier than knowing that he was not truly gone, that he lives on somewhere watching over me. But alas, I know it is not true. He is gone. Forever.

However, I do have a couple of things to say that might offer some degree of consolation:

First, because human minds are software, pieces of our loved ones do go on—in us. Our memories of those we have lost are tiny shards of their souls. When we tell stories about them to others, we make copies of those shards; or to use a more modern metaphor, we back up their data in the cloud. Were we to somehow reassemble all these shards together, we could not rebuild the whole person—there are always missing pieces. But it is also not true that nothing remains. What we have left is how they touched our lives. And when we die, we will remain in how we touch the lives of others. And so on, and so on, as the ramifications of our deeds in life and the generations after us ripple out through the universe at the speed of light, until the end of time.

Moreover, if there’s no afterlife there can be no Hell, and Hell is literally the worst thing imaginable. To subject even a single person—even the most horrible person who ever lived, Hitler, Stalin, Mao, whomever—to the experience of maximum possible suffering forever is an atrocity of incomparable magnitude. Hitler may have deserved a million years of suffering for what he did—but I’m not so sure about maximum suffering, and forever is an awful lot longer than a million years. Indeed, forever is so much longer than a million years that if your sentence is forever, then after serving a million years you still have as much left to go as when you began. But the Bible doesn’t even just say that the most horrible mass murderers will go to Hell; no, it says everyone will go to Hell by default, and deserve it, and can only be forgiven if we believe. No amount of good works will save us from this fate, only God’s grace.

If you believe this—or even suspect it—religion has caused you deep psychological damage. This is the theology of an abusive father—“You must do exactly as I say, or you are worthless and undeserving of love and I will hurt you and it will be all your fault.” No human being, no matter what they have done or failed to do, could ever possibly deserve a punishment as terrible as maximum possible suffering forever. Even if you’re a serial rapist and murderer—and odds are, you’re not—you still don’t deserve to suffer forever. You have lived upon this planet for only a finite time; you can therefore only have committed finitely many crimes and you can only deserve at most finite suffering. In fact, the vast majority of the world’s population is comprised of good, decent people who deserve joy, not suffering.

Indeed, many ethicists would say that nobody deserves suffering, it is simply a necessary evil that we use as a deterrent from greater harms. I’m actually not sure I buy this—if you say that punishment is all about deterrence and not about desert, then you end up with the result that anything which deters someone could count as a fair punishment, even if it’s inflicted upon someone else who did nothing wrong. But no ethicist worthy of the name believes that anybody deserves eternal punishment—yet this is what Jesus says we all deserve in the Bible. And Muhammad says similar things in the Qur’an, about lakes of eternal burning (4:56) and eternal boiling water to drink (47:15) and so on. It’s entirely understandable that such things would motivate you—indeed, they should motivate you completely to do just about anything—if you believed they were true. What I don’t get is why anybody would believe they are true. And I certainly don’t get why anyone would be willing to traumatize their children with these horrific lies.

Then there is Pascal’s Wager: An infinite punishment can motivate you if it has any finite probability, right? Theoretically, yes… but here’s the problem with that line of reasoning: Anybody can just threaten you with infinite punishment to make you do anything. Clearly something is wrong with your decision theory if any psychopath can just make you do whatever he wants because you’re afraid of what might happen just in case what he says might possibly be true. Beware of plausible-seeming theories that lead to such absurd conclusions; it may not be obvious what’s wrong with the argument, but it should be obvious that something is.

Pascal’s Mugging

PNRJ cognitive science, heuristics and biases , , , , , , , , , ,

Nov 10 JDN 2458798

In the Singularitarian community there is a paradox known as “Pascal’s Mugging”. The name is an intentional reference to Pascal’s Wager (and the link is quite apt, for reasons I’ll discuss in a later post.)

There are a few different versions of the argument; Yudkowsky’s original argument in which he came up with the name “Pascal’s Mugging” relies upon the concept of the universe as a simulation and an understanding of esoteric mathematical notation. So here is a more intuitive version:

A strange man in a dark hood comes up to you on the street. “Give me five dollars,” he says, “or I will destroy an entire planet filled with ten billion innocent people. I cannot prove to you that I have this power, but how much is an innocent life worth to you? Even if it is as little as $5,000, are you really willing to bet on ten trillion to one odds that I am lying?”

Do you give him the five dollars? I suspect that you do not. Indeed, I suspect that you’d be less likely to give him the five dollars than if he had merely said he was homeless and asked for five dollars to help pay for food. (Also, you may have objected that you value innocent lives, even faraway strangers you’ll never meet, at more than $5,000 each—but if that’s the case, you should probably be donating more, because the world’s best charities can save a live for about $3,000.)

But therein lies the paradox: Are you really willing to bet on ten trillion to one odds?

This argument gives me much the same feeling as the Ontological Argument; as Russell said of the latter, “it is much easier to be persuaded that ontological arguments are no good than it is to say exactly what is wrong with them.” It wasn’t until I read this post on GiveWell that I could really formulate the answer clearly enough to explain it.

The apparent force of Pascal’s Mugging comes from the idea of expected utility: Even if the probability of an event is very small, if it has a sufficiently great impact, the expected utility can still be large.

The problem with this argument is that extraordinary claims require extraordinary evidence. If a man held a gun to your head and said he’d shoot you if you didn’t give him five dollars, you’d give him five dollars. This is a plausible claim and he has provided ample evidence. If he were instead wearing a bomb vest (or even just really puffy clothing that could conceal a bomb vest), and he threatened to blow up a building unless you gave him five dollars, you’d probably do the same. This is less plausible (what kind of terrorist only demands five dollars?), but it’s not worth taking the chance.

But when he claims to have a Death Star parked in orbit of some distant planet, primed to make another Alderaan, you are right to be extremely skeptical. And if he claims to be a being from beyond our universe, primed to destroy so many lives that we couldn’t even write the number down with all the atoms in our universe (which was actually Yudkowsky’s original argument), to say that you are extremely skeptical seems a grievous understatement.

That GiveWell post provides a way to make this intuition mathematically precise in terms of Bayesian logic. If you have a normal prior with mean 0 and standard deviation 1, and you are presented with a likelihood with mean X and standard deviation X, what should you make your posterior distribution?

Normal priors are quite convenient; they conjugate nicely. The precision (inverse variance) of the posterior distribution is the sum of the two precisions, and the mean is a weighted average of the two means, weighted by their precision.

So the posterior variance is 1/(1 + 1/X^2).

The posterior mean is 1/(1+1/X^2)*(0) + (1/X^2)/(1+1/X^2)*(X) = X/(X^2+1).

That is, the mean of the posterior distribution is just barely higher than zero—and in fact, it is decreasing in X, if X > 1.

For those who don’t speak Bayesian: If someone says he’s going to have an effect of magnitude X, you should be less likely to believe him the larger that X is. And indeed this is precisely what our intuition said before: If he says he’s going to kill one person, believe him. If he says he’s going to destroy a planet, don’t believe him, unless he provides some really extraordinary evidence.

What sort of extraordinary evidence? To his credit, Yudkowsky imagined the sort of evidence that might actually be convincing:

If a poorly-dressed street person offers to save 10(10^100) lives (googolplex lives) for $5 using their Matrix Lord powers, and you claim to assign this scenario less than 10-(10^100) probability, then apparently you should continue to believe absolutely that their offer is bogus even after they snap their fingers and cause a giant silhouette of themselves to appear in the sky.

This post he called “Pascal’s Muggle”, after the term from the Harry Potter series, since some of the solutions that had been proposed for dealing with Pascal’s Mugging had resulted in a situation almost as absurd, in which the mugger could exhibit powers beyond our imagining and yet nevertheless we’d never have sufficient evidence to believe him.

So, let me go on record as saying this: Yes, if someone snaps his fingers and causes the sky to rip open and reveal a silhouette of himself, I’ll do whatever that person says. The odds are still higher that I’m dreaming or hallucinating than that this is really a being from beyond our universe, but if I’m dreaming, it makes no difference, and if someone can make me hallucinate that vividly he can probably cajole the money out of me in other ways. And there might be just enough chance that this could be real that I’m willing to give up that five bucks.

These seem like really strange thought experiments, because they are. But like many good thought experiments, they can provide us with some important insights. In this case, I think they are telling us something about the way human reasoning can fail when faced with impacts beyond our normal experience: We are in danger of both over-estimating and under-estimating their effects, because our brains aren’t equipped to deal with magnitudes and probabilities on that scale. This has made me realize something rather important about both Singularitarianism and religion, but I’ll save that for next week’s post.