Aporia Is Wrong About Douthat

"The God that failed" that failed

1 Introduction

There’s a common phenomenon in philosophy. Some decently smart person without philosophical training will make confident proclamations about a philosophical topic. Often they raise some interesting points, but because they don’t have much background in philosophy, they end up making quite a few significant errors.

Such is the case with the new review out at Aporia titled “The God That Failed,” in which Bo Winegard reviews Ross Douthat’s book defending religion. This is Aporia’s second article titled “The God That Failed,” but this time the failing God in question is God himself! Winegard isn’t convinced by Douthat’s arguments for the existence of God in his new book Believe. Unfortunately, most of his objections are quite standard and it’s been clear what’s wrong with them for decades.

Winegard begins by describing three different approaches to religion: a poetic one which claims that religion is a poem or symbol, a literalist one that claims religion is literally true, and a pragmatic argument that claims we should be religious for practical reasons even if it isn’t really true. Winegard seems sympathetic to the pragmatic and poetic approaches, though he rejects the literal approach as “a serious mistake,” on the grounds that “the doctrines of traditional religion are almost certainly false when treated as empirical assertions.”

I found these paragraphs confusing. I understand what it means to adopt the religious life, without believing it, for pragmatic reasons. But what the hell does it mean to treat religion as a poem or symbol? When people profess their genuine belief that Jesus rose from the dead and has literally become the wine and wafer, in what sense are they engaged in poetry? Certainly that’s not what they believe themself to be doing, and typically when people are engaged in poetry, they are aware of that fact.

But fine, nitpicks aside, to figure out if the literalist approach is right—the one that agrees that, every now and then, distinctly religious claims are true—we’ll have to see if there are convincing arguments for God’s existence. So, um, let’s do that!

2 Fine-tuning

The second section of Winegard’s essay is dedicated to criticizing the fine-tuning argument. The fine-tuning argument, for those unaware, notes that certain parameters in physics fall in an extremely narrow range needed for the formation of life. For example, if the cosmological constant differed by about one part in ten to the power of a hundred and twenty, no life could ever arise. If it were weaker, the universe would have collapsed on itself immediately after the big bang—if it were stronger, the universe would have flown apart.

Winegard raises four main objections to the fine-tuning argument, some of which combine together multiple different objections. Each of the objections have a common defect: they are all mistaken.

2.1 Texas sharpshooter fallacy

Imagine a person throws a dart and hits a random spot on a dartboard. Then, they draw a red circle around the spot they hit and declare they hit a bullseye. This, in essence, is the Texas sharpshooter fallacy. It involves retroactively declaring some non-special value special, and using this as evidence of great skill on the part of an agent.

Winegard thinks that fine-tuning commits this fallacy. He writes:

This, in essence, is what the fine-tuning argument does with the parameters of the universe’s physical laws. It examines these parameters only after they have been set (or manifested) and declares, “What are the odds that the universe would have precisely these conditions—conditions that allow for life? Surely, it must have been designed.”

But this is plainly not what’s wrong with the Texas sharpshooter’s reasoning. It’s fine to examine something after its occurred and declare that its improbability is evidence for a theory. Historians think Julius Caesar existed—if asked why, they’d probably say something like “what are the odds that all these people would write about him if he’d never existed?” Thus, it’s not committing the Texas sharpshooter fallacy to cite an improbable thing you’ve already observed as evidence for a theory.

To give another analogous case, it’s not the Texas sharpshooter fallacy to think your parents had sex, on the grounds that it’s unlikely that you’d exist if they hadn’t. Therefore, it can’t automatically be the Texas sharpshooter fallacy whenever one cites their existence as evidence for a theory—even though they reason about their existence long after they came to exist.

The problem with the Texas sharpshooter’s reasoning is that there’s nothing genuinely special about the region they hit. If they were an expert marksman, there’s no reason to think they’d hit the particular spot that they hit, rather than some other spot. Therefore, the probabilistic inference is unjustified—them hitting the spot they hit isn’t evidence they’re an expert marksman because it’s no likelier if they were an expert marksman than if they were not. Now, Winegard tries to argue the same thing is true of fine-tuning, writing:

But just as any arrangement of pellet holes is as likely (or unlikely) as any other, so too is any particular set of physical laws—assuming, for the sake of argument, that such laws can vary. Declaring our universe’s specific configuration of forces and constants to be uniquely significant simply because it ultimately (after billions of years) allowed for life is question begging.

This is very confused!

Analogy: Winegard and I are in Vegas. He witnesses me get a royal flush, then another, then another, then another, then another. This continues for dozens of hands. He accuses me of cheating. “Hold on,” I declare, “you’re committing the Texas sharpshooter fallacy. Every arrangement of cards is equally likely—the hands I got are no less likely than any other hand. Declaring the specific configuration of cards I got to be uniquely (sic) simply because it allowed me to take your money over and over again is question begging.”

Winegard is right that if we assume naturalism then every value of the constants is equally likely. But if there’s a God, values that produce life are likelier than other values, because it’s not astronomically unlikely that God would want to create life. Therefore, finding out that the values of the constants fall in a tiny range needed for life is very strong evidence for God’s existence. It’s a lot likelier they’d fall in this tiny range if God exists than if God doesn’t.

If Winegard was right, it would be impossible to ever get evidence for God—or for almost anything else. Imagine that the initial conditions of the universe spelled out “made by God.” Would that be evidence for God’s existence? Well, no initial condition is likelier than any other—so by Winegard’s logic, declaring this evidence for God would be question begging.

Winegard is misunderstanding how the fine-tuning argument works. It doesn’t start by declaring the constants “significant” and then say that God has to be behind them. It simply notes that the constants we observe, that happen to produce complex structures, are likelier if God exists than if he doesn’t. This entails that they’re evidence for God.

It’s not true that the fine-tuning argument assumes the constants could be different. Think back to the case where the initial conditions spelled out “made by God.” Would citing this as evidence for God’s existence assume the initial conditions couldn’t be different? No! It would just proceed from the notion that even if the constants can’t be otherwise, it’s highly unlikely, absent God, that they’d spell out “made by God.” Even if one is a determinist and thinks that the physical world can’t be different, there’s still some sense in which getting extreme luck in poker is unlikely and is evidence that the lucky person is cheating. Simply declaring something necessary doesn’t eliminate its improbability.

I discuss this in more detail here, if you’re interested. In fact, I discuss it even if you’re not interested!

2.2 The rarity of life

Winegard’s next section claims that the rarity of life shows that the universe can’t have been made for life. But this isn’t right; a scientists might set up a whole biosphere to study some very small bacteria. Curtis Yarvin might write a very long essay that’s really centered around one point that he describes in just a few sentences, where the remainder of the essay is filled with smug jabs about how he’s telling you deep forbidden knowledge and exhortations to read various old books! Whether some being is small tells you little about their significance.

I won’t go into too much detail about what’s wrong with this argument as I’ve already argued against it at considerable length. So let me just briefly repeat a few of my points:

1. The universe is likely infinite. Thus, there’s infinite life, which is exactly what one would expect given theism. One’s complaint can, therefore, only be that the life is far apart. But this is likely good policy to prevent the different species from colonizing each other and interfering with each other’s life plans.

2. God has no reason not to waste space because he doesn’t face any resource constraints. He’s not like other agents.

3. A vast and explorable universe gives rise to great aesthetic goods. Exploring a huge cosmos is quite a valuable endeavor. Given that there’s no downside to it, it’s likely enough God would create it.

2.3 Does the designer need a designer?

Winegard thinks that the fine-tuning argument is self-undermining. If a big universe needs a designer, wouldn’t God? He writes:

The fine-tuner needs a fine-tuner. The fine-tuning argument undermines itself with its own logic. It begins by claiming that anything as precisely ordered as the universe must have an intentional designer. Then it posits a designer—and stops there, without further analysis.

The fine-tuning argument doesn’t claim that anything that’s ordered needs a designer. It claims that the constants falling in an infinitesimally narrow range needed for life is very unlikely if there’s no God but likely if there’s a God. This is therefore evidence for God. It need not be that everything that’s ordered needs a designer.

Something likely has a designer if it has components that are set to extremely improbable values, where those values are much likelier to be set that way by an agent than by chance. This is why, to use Paley’s famous example, we infer a watch has a designer—it has a series of parts that fit together in improbable ways, but serve some purpose. It’s likelier an agent would put the parts together in that way than that they’d arise that way by chance.

But this plainly doesn’t apply to God! God doesn’t have any parts. He’s generally been conceived to have only one essential property that’s completely without limits. My preferred conception of God is as a limitless agent. An agent is a mind that’s in some way connected to reality—God is a mind that’s unlimited in its connection to reality. This perfect connection means God will is manifested in reality (omnipotence) and what goes on in reality is manifested in his mind (omniscience). He’s motivated to follow what is good because the contents of his attitudes match reality—he is motivated to bring about that which is worth bringing about. Other candidates for God’s essential unlimited property include: value, perfection, consciousness, mind, being, actuality, and so on.

If any of these are right, God has no parts that need to be combined. Nor does he have any values that need to be fine-tuned. Because his only property has no limits, it doesn’t need fine-tuning, anymore than a universe has to be finely-tuned to be infinitely large. God has no parts that need to be combined or parameters that need to be set to improbable values. He is without part or limit.

2.4 Might there be an evil designer?

Winegard’s last argument is that the designer might be evil or incompetent. In my view, this is ruled out by the considerations of the last section! A designer that’s incompetent is in some way limited. They will thus have to be some complex agent with parts that need to be tuned. They don’t have a single unlimited property. It’s only non-anthropocentric agents without limits that avoid the need for a designer, for only they are simple.

An evil designer is similarly unlikely. Why would the designer love evil rather than paperclips? There are an infinite number of motivations that an agent might have, so the odds of them just desiring evil are very low. Such an agent would have to be complicated because a desire for evil is built into their fundamental attributes.

This doesn’t apply to a good God because the good is self-motivating. Recognizing something is good (at least plausibly) gives one a motivation to bring it about. When you see something is desirable, you’ll generally have a desire for it. For this reason, a limitless agent whose mind totally matches reality will be motivated to pursue the good. Given that only agents of this kind are genuinely simple, only they can be fundamental—not in need of a deeper designer.

An evil God also fits much less well with the evidence.

2.5 Parting thoughts on fine-tuning

The fine-tuning is quite a straightforward Bayesian argument. It notes the existence of something very improbable given naturalism but likely given theism. It then claims that’s evidence for theism. It can’t be fallacious—it’s just a simple application of Bayes’s theorem. Winegard hasn’t given any naturalistic explanation of this coincidence on the order of randomly picking out some particular atom out of all the atoms in the known universe! Thus, based on the considerations presented, at the very least naturalism has been falsified.

3 Consciousness

In his book, Douthat argues that consciousness gives evidence for the existence of God. It’s very valuable, but there’s no plausible naturalistic explanation of how neurons firing in the brain gives rise to subjective experience. Winegard writes:

Yet even if the impulse to posit something supernatural about consciousness is understandable, it remains fallacious. Substituting the riddle of the supernatural for the riddle of the mind, it merely compounds the enigma.

…

The most parsimonious reason for the persistence of the mystery of consciousness is that we are constrained creatures—evolved animals—incapable of fully understanding the world. Just as the quantum leaps of electrons defy ordinary intuition, so too does consciousness elude our understanding. Subjectivity may very well present our conceptual systems with an intractable challenge. But that does not mean something supernatural is involved. Rather, it means that the ordinary natural world is strange enough to confound our limited minds.

I don’t quite know what it means for consciousness to be “supernatural.” Certainly I don’t think that arguments for theism from consciousness must make that claim. The only thing an argument for consciousness needs to claim is that the existence of consciousness is likelier given theism than given naturalism. Which it plainly is! If you knew God existed, you’d expect consciousness to exist! In contrast, if you knew there was no God, you wouldn’t expect throwing together a bunch of neurons together to give rise to consciousness.

The reasons to think consciousness aren’t non-physical isn’t just that consciousness is confusing or we don’t know exactly how it works. The reason is that there are arguments that show that reducing consciousness to the physical is impossible even in principle!

4 Conclusion

Winegard raises quite a few other arguments in his essay that I don’t plan to respond to. People seem not to like response articles that go over about 3,000 words. I’m also less sold on some of the other arguments than Douthat is. But hopefully I’ve shown that the fine-tuning argument and argument from consciousness aren’t in serious danger from the criticisms Winegard raises.