Sunday, May 5, 2019

Radical skepticism, revived


Radical skepticism gets a bad rap. Sure, it's possible to question everything about your reality, but what's the point? Countless freshman philosophy majors and enterprising potheads have done just that, and look where it got them. Better to face up to the cold, hard facts!

But maybe they simply didn't know how to wield the tool properly. Let's take a trip down the rabbit hole and see if we can't do better....

Ompha, Lompha

Young Earth creationists believe that the earth was created in the past ten thousand years. According to their so-called Omphalos Hypothesis, dinosaur bones and other seemingly-old artifacts were planted there by God as a test of our faith.

As ludicrous as this may strike you, it contains no logical contradiction. In fact, there is no way to disprove that the universe sprang into existence, fully formed, last Thursday. Mathematician-philosopher Bertrand Russell takes it one step further (as mathematicians and philosophers are wont to do):
There is no logical impossibility in the hypothesis that the world sprang into being five minutes ago, exactly as it then was, with a population that 'remembered' a wholly unreal past.
But why stop there? For all we know, this may all have appeared freshly in this very moment. But while such a claim is not logically impossible, it is certainly very improbable. Right?

To help answer that, let's look at a technique called Bayesian inference.

Bayesian inference

Suppose I am hiding a coin from you. I tell you it is either a (US) penny or nickel, and you have to guess which. What kind of evidence would help you decide?

If I tell you that there is a US president on one side of the coin, that is no help at all, because it applies equally well to both coins. Whatever you thought the relative odds were before, they should remain unchanged after.

On the other hand, suppose I tell you that I flipped it and it landed on its edge. Both coins can theoretically do this, but because the nickel is thicker, it is probably more likely to land that way. Depending on how much more likely, this evidence should nudge your bias toward the nickel to a corresponding degree.

So we see that evidence is one crucial component in determining your beliefs. The second, equally important piece, is something that we hinted at earlier: your pre-existing bias, technically known as your prior.

If you're like most people, you probably started with the assumption that both the penny and nickel were equally likely. And why not? In lieu of any information, this seems like a reasonable starting point. But you don't have to start at 50-50. You might look up some stats on the number of pennies and nickels in circulation, and use that ratio as your prior. Or perhaps you know that I secretly love pennies, and use that to inform your initial bias. It's totally up to you.

Whatever your prior, you must combine it with evidence, to get an updated bias (called the posterior). You can either stop there or use this posterior as a new prior, to be combined with further evidence (generating yet another posterior), and so on. Over time, as the evidence accumulates, any mistakes in your initial prior will get ironed out.

In this way, Bayesian inference is a formal framework for doing something that's already very natural to us. We may start off uncertain about something, but we allow the weight of evidence to bring our beliefs into closer alignment with reality.

Let's apply this to our question about the past.

Evaluating the past

First let's try to pick a good prior.

Basic physics tells us that fully-formed worlds are very unlikely to just pop into existence. Therefore we should a priori be very biased against this possibility, right? Unfortunately not: we could only have learned about physics in the past, which we cannot trust without resorting to circular reasoning.

Okay, what about Occam's Razor? It tells us that we should prefer simpler theories to needlessly complicated ones. But this is a statement about preference, not likelihood. Moreover, it runs into the same problem as before: when, exactly, did we collect evidence that justifies the Razor?

Try as we may, we cannot logically justify any particular prior. So let's throw up our hands and start at 50-50. The evidence should settle it, right?

Well, this picture I took yesterday seems like pretty good evidence at first. But remember that our hypothetical sudden-world is designed to provide such fabricated evidence. So this supposed "picture from yesterday" fits both models equally well, and it is therefore of no use. By design, neither is anything else. Drat!

A final attempt might be to say fine, let's accept that maybe the world sprang into being five minutes ago (or whenever you began this exercise). But since then, you've been collecting evidence that justifies your trust in physics, Occam's Razor, and the rest. This ought to restore your faith, right?

But notice that you can repeat the thought experiment right now. How do you know that this is not the first moment? When you try to work out the answer, you will find that your supposed "evidence from the past five minutes" goes out the window just like our supposed "picture from yesterday." So this approach fails, too. There's nowhere to get a grip!

Strange as it may seem, we are not rationally justified in saying that a real past is "more likely" than a fake one. It certainly feels like the real past is more rational, but this is an illusion. For now, simply notice how powerful this illusion is. We will return to it later.

Choosing our beliefs

We do not necessarily need to rely on reasoning to choose our beliefs. Discussing a different skeptical hypothesis, physicist Sean Carroll points out:
There is no way to distinguish between the scenarios by collecting new data.

What we’re left with is our choice of prior credences. We’re allowed to pick priors however we want—and every possibility should get some nonzero number. But it’s okay to set our prior credence in radically skeptical scenarios at very low values, and attach higher prior credence to the straightforwardly realistic possibilities.

Radical skepticism is less useful to us; it gives us no way to go through life. All of our purported knowledge [...] might very well be tricks being played on us. But what then? We cannot actually act on such a belief [...]. Whereas, if we take the world roughly at face value, we have a way of moving forward. There are things we want to do, questions we want to answer, and strategies for making them happen. We have every right to give high credence to views of the world that are productive and fruitful, in preference to those that would leave us paralyzed with ennui.
Russell also recognized this:
Skepticism, while logically impeccable, is psychologically impossible, and there is an element of frivolous insincerity in any philosophy which pretends to accept it.
In other words, given that evidence cannot help us, we might as well pick whichever belief is most helpful. And our commonsense notion of time is the most helpful of all.

But is this really true? Have you ever been genuinely free of the belief in a real past, in order to make a fair comparison? Is such a state even possible or desirable? Wouldn't you be like a proverbial goldfish, totally unable to function?

As it turns out, that is not what it is like to be free of conviction about the past. It is entirely possible -- though not easy -- to drop your belief in a literal past while retaining the ability to function as though it were real. You can (and should) try to enter that state right now, but unless you've cultivated exceptional facility with your most subtle mental processes, you will attain at best a vague facsimile of it. Something deep inside you is unwilling to genuinely and completely let go. It protects itself by telling you "well, our belief is the best one anyway!"

The authentic state of not-knowing belongs to the purview of mystics and contemplatives. It is unfair to expect mathematicians, physicists, or even philosophers to find the time to explore it properly.

The future

Let's visit another oddity. 18th-century philosopher David Hume popularized something called the Problem of Induction. Roughly, it goes as follows.

It seems painfully obvious that the past provides good evidence for the future, right? But why do we believe this? Well, it has certainly been true in the past. Okay, but so what? We'd like to say "... and therefore it will be true in the future," but then we are assuming what we set out to prove. There's that nasty circularity again! So we cannot justify our belief that the past will continue to provide evidence for the future.

Thus, the fact that the laws of physics have faithfully operated for billions of years gives us no reason at all to believe that they will continue to operate even one second from now [1]. As before, notice how ridiculous and illogical this feels, despite being impeccably rational. This is a powerful clue, if used properly!

Unfortunately Hume, too, loses his nerve before fully taking on board the implications:
Should it be asked me whether I sincerely assent to this argument which I have been to such pains to inculcate, whether I be really one of those skeptics who hold that everything is uncertain, I should reply that neither I nor any other person was ever sincerely and constantly of that opinion. I dine, I play backgammon, I converse and am merry with my friends and when after three or four hours of amusement I would return to these speculations, they appear so cold and strange and ridiculous that I cannot find in my heart to enter into them any further. Thus the skeptic still continues to reason and believe, though he asserts he cannot defend his reason by reason.
The nature of reality

It does not stop with time. Consider the hypothesis that this world is a simulation -- and further, that its creators are clever enough to hide all evidence of this fact from us. Then, by assumption, there is no way to test the hypothesis, and we are again free to discard it on the basis of practicality, even if not on pure rationality. In fact, there are an infinite number of strange and seemingly-pointless possibilities that we can cut away with one fell swoop in this way. This can make us feel supremely confident in our default worldview. But there is at least one strange possibility that comes away unscathed.

Consider the hypothesis that the world is a dream of sorts. This possibility may seem inconsequential, as with the previous examples: if the dream behaves exactly as a physical world would, then there's no point in pursuing the belief further. On the other hand, if it actually is a dream -- and moreover, your dream, in a sense -- then your participation might be pivotal in gathering evidence of the fact. How might this work?

In a nighttime dream, naively questioning your surroundings is often insufficient to expose the dream's unreality. In most cases, the dream will fabricate an explanation that -- despite being nonsensical from a more awakened perspective -- will nonetheless suffice to quell your suspicion. Nothing to see here, move right along! To expose the sham, you must question the dream in just the right way, at which point you might become lucid.

So how does the analogy extend to this reality? How would one question it in "just the right way?"

The mind doth protest too much

You might begin by noticing how illogical it is to feel so dead certain about the past despite having no rational basis for the belief. Next, watch as your mind tries to wriggle out of this accusation: "well, such a belief is evolutionarily adaptive, so it's probably embedded deep in our ancient limbic system, beyond the reach of higher cognition...." Bam! In a finger snap, circular reasoning again magically restores your faith: I believe in the past because the past made me do it, duh. Whew! Nothing to see here! 

This should trigger a great deal more suspicion in you than it probably does.

If you were to sit and grapple with this conundrum very sincerely -- not just thinking harder about it, but experientially penetrating the very heart of the discrepancy -- you might have a mind- and reality-shattering "aha!" moment, not unlike what Zen Buddhists call kensho or sudden awakening. It might reveal that you've been taking life utterly for granted [2], subtly (but erroneously) assuming that you have the slightest inkling of what it is and how it works. Side effects may include an overwhelming flood of gratitude, awe, wonder, love, joy, and humility, beyond what you believed possible. There may even be insights about the nature, purpose, and evolution of this dream, though they may be hard to prove or even communicate via the standard channels.

Why is it impossible to prove that the past really happened, or that it's even likely? Sure, maybe this is all just sophistry and there's a perfectly reasonable explanation. Or maybe something much more curious is going on, right under your nose but hidden by a cleverly self-protective veil.

If this all sounds like too much for you, well, then, luckily you have the option to continue to blindly trust your default beliefs. And why not? After all, some Very Smart Scientists have assured you that being too skeptical would be "ridiculous," "frivolous," and "paralyzing." And who are you to question that? Nothing to see here, move right along....

"It’s a good test of whether someone has actually understood Hume’s argument that they acknowledge its conclusion is fantastic (many students new to philosophy misinterpret Hume: they think his conclusion is merely that we cannot be certain what will happen tomorrow.) ... [But] if Hume is right, the belief that the sun will rise tomorrow is as unjustified as the belief that a million mile wide bowl of tulips will appear over the horizon instead. We suppose the second belief is insane. But if Hume is correct, the first belief is actually no more rational. ... 
[T]he onus is on these defenders of “common sense” to show precisely what is wrong with Hume's argument. No one has yet succeeded in doing this (or at least no one has succeeded in convincing a majority of philosophers that they have done so)."
[2] Notice that to take something "for granted" can mean either to be unappreciative of it, or to logically presuppose it. Here it takes on both meanings: we are quite confident that we've been experiencing life for a long time, which is why this moment feels so mundane. And now notice that "mundane" means both "pertaining to physical reality" and "tedious, repetitive, dull." Again, this is not a coincidence.