Carrier's Extreme Blunders Continue

Contra Carrier concerning cosmology, constants, categorization

1 Introduction

Richard Carrier is a blogger and Jesus mythicist who spends much of his time arguing that Christians are wrong on the internet. I’ve mentioned recently that Carrier has taken a break from his busy schedule of calling people liars, misunderstanding Bayes theorem, bogusly suing people, and has replied to various arguments of mine. I’ve already addressed, at some length, his embarrassingly failed replies to the anthropic argument. In this article, I’ll discuss what’s wrong with the rest of his claims made here and here, in which he criticized the analysis in my tier list of arguments for God, as well as his recent reply to my rebuttal of him on the anthropic argument.

I want to make something very clear (I’ll even put in in bold, so that Carrier, despite his frequent tendency to overlook things his interlocutor said, might see it): I am quite happy to have a public debate with Carrier about the anthropic argument. Given that Carrier’s responses to it involve saying false things, then lying and backpedaling, such a debate would go quite poorly for him. Richard, if you’re reading this, pick any platform you want and any moderator you want—I would be more than happy to have a verbal debate about the anthropic argument.

Carrier comes out of the gate swinging, saying of the arguments I list in the tierlist:

Most of [them] are already addressed in the article you are commenting on (or its mirror article, Bayesian Counter-Apologetics), especially The Fine Tuning and Biological (Design) Arguments

I gave the biological design argument D-tier—I did not defend it. I’ll address what he had to say about fine-tuning later in this article.

The Moral Argument

Which I gave F-tier.

the Argument from Consciousness, the Arguments from Miracles, the Cosmological Argument (what they call the Kalam), and their Common Consent Argument, which is just a fallacious restructuring of an Argument from Religious Experience.

The common consent argument is about widespread theistic belief, not widespread theistic experience. So no, it’s not the same argument. Analogy: there’s widespread belief in a round Earth (for good reason) but few claim to have experienced the roundness of the Earth.

Even what they call the Psychophysical argument is just a fallacious restructuring of the Argument from Consciousness (so they illogically count the same argument twice).

Nope! The argument from consciousness claims that the fact that consciousness exists is likelier given theism than naturalism. The psychophysical harmony argument (read more about it here—it’s a confusing argument) claims that the fact that consciousness pairs harmoniously with the physical—so that, for instance, when you have a desire to move your limbs, your limbs move—is likelier given theism than given naturalism. Psychophysical harmony doesn’t assume dualism; the argument from consciousness does. They’re just different arguments. They are alike in that they are arguments having something to do with consciousness, but they are clearly not the same argument.

(Well, perhaps the psychophysical argument is, indeed, a fallacious restructuring of the argument from consciousness. Fortunately, however, I did not make the psychophysical argument, but the argument from psychophysical harmony).

And what they call the Nomological Argument and Argument from Laws and Argument from Physical Reality are just restructured Fine Tuning Arguments (or badly restructured Arguments from Uniformities), and thus a quadruple-dip.

I gave the nomological argument F-tier! Carrier is right that the nomological argument is the same as the argument from uniformities. The argument from laws and physical reality are different both from each other and the nomological argument—they’re about the fact that there is a physical universe at all and that it behaves in various ways rather than lying dormant (certainly lying dormant would be simpler). A helpful way to think of the differences between the arguments is displayed by the features each one hones in on:

1. There’s a physical universe (the argument from physical reality).

2. It does stuff (the argument from laws).

3. It follows uniform laws all throughout the universe (the argument from uniformities).

4. It has finely-tuned constants that fall in a narrow range needed for the formation of complex structures (fine-tuning).

It would be surprising if I was quadruple counting the same argument, as I rate the first argument mediocre, the second a bit better, the third so bad as to be evidence for atheism, and the fourth S-tier. Normally when one quadruple counts evidence they do not give it wildly different rankings.

Likewise their Contingency Argument is just a redo of the Cosmological Argument (also addressed here) and duplicated under their Argument from Motion (so, a triple-dip).

These are not the same argument. The argument from motion is about changes needing an outside explanation, while the argument from contingency is about stuff that could be otherwise needing an explanation, and the “cosmological argument” is a broad family of arguments that includes the Kalam, the contingency argument, and the argument from motion. This is a bit like saying “the problem of evil, divine hiddenness, and arguments for atheism are just a redo of each other.” What?

(Also, notably, I didn’t rate any of these arguments very highly, and I’ve probably soured on the contingency argument a bit since making the list).

And the “The Neo-Platonic Proof” falls apart exactly as they themselves admit, which fact I covered in my series on Feser (start with Feser’s Five Proofs of the Existence of God: Debunked!).

Which I agree with!

Now that we’ve gotten the misrepresentations of the various arguments out of the way, I’ll address Carrier’s linked articles that allegedly rebut each of the arguments I list. I’ll also, in section 2, address Carrier’s pitifully bad responses to my rebuttal to him on the anthropic argument.

2 Carrier’s anthropic argument flop

Recently, I wrote a reply to Carrier on the anthropic argument. Not to toot my own horn, but the reply was fairly devastating; the problem was Carrier didn’t even faintly grok my arguments, and made a series of wild and crazy claims about probability that can be shown to be false by two seconds of reflection. One might expect Carrier to write a more detailed reply (he has, after all, written a 7,000-word reply to an Amazon review of his book by a random person). Unsurprisingly, however, Carrier only briefly fired off a few comments beneath one of his own blog posts about how his original response refuted me, so he wasn’t going to bother responding to the numerous points where I showed him to be wrong. I obviously won’t rehash every claim I made in the article, but let me just list a few where Carrier erred quite egregiously and explicitly:

1. He interpreted my argument to be about finite people existing when it was about infinite people existing. In fact, in a sentence where I said that infinite people exist (throughout the multiverse, of course) he interpreted that as saying that I was suggesting that finite people exist.

2. He claimed I was making a basic error in the following case. A coin gets flipped. If it comes up heads, one person gets created. If it comes up tails, ten people get created. I suggested that in this case, after being created, you should think at 10:1 odds the coin came up tails. He suggested that this was a basic statistical error because coins have a 50% chance of coming up heads. But, of course, this is wrong: while coins come up heads half the time, sometimes you should think at more than 50% odds that a coin came up heads. If your parents flipped a coin after meeting, and only had sex if a coin came up tails, then you should think at 100% odds it came up tails. It can’t have been a basic statistical error when this judgment is the plurality view among philosophers; it’s unlikely that most philosophers would just not realize something so basic.

3. Carrier suggested I was making a basic mathematical error by suggesting one probability was infinitely greater than another, because probabilities can’t be infinity—they max out at 100%. But for A to be infinitely greater than B, A doesn’t have to be infinity—B can be zero or infinitesimal. To see this, imagine that infinity people are put to sleep. A coin is flipped. If it comes up heads, they’re all woken up. If it comes up tails, a random one is woken up. Upon being woken up, you should think the probability the coin came up heads is infinitely greater than the probability it came up tails.

4. Carrier asserts that there’s no evidence for a big multiverse with lots of people. But he ignores my arguments for such a multiverse.

5. Carrier accuses me of neglecting total evidence, while explicitly neglecting total evidence. More specifically, he says your evidence is “someone exists.” The requirement of total evidence says you take the most specific version of your evidence. But the most specific version of your evidence, regarding your existence, is that you exist, not just that someone does.

Anyway, from Carrier’s new reply, we get a snarky, overconfident soup of “I already refuted you, but I’m not going to bother specifying where.” This is the desperate flailing of a man who has been decisively rebutted but doesn’t want to admit defeat, a rather common pattern with Carrier.

As I noted, I think he is mostly guilty of badly wording things, and then I try to think of what he might have meant to say instead (to steel man his argument). His reply mostly consists of running with those steel mans (fixing his language to be more clear as stated). Which is progress. But then his argument falls to my subsequent analysis of those steel mans. So I need not respond further. He simply ignores my actual refutations and picks on the trivia (of my pointing out his poor wording). So one can already refute his new reply with my original comment.

What subsequent analysis? I went line by line and addressed every single sentence Carrier said about the anthropic argument. There was nothing of substance other than blatantly false statistical claims, and claims I had no evidence for the presence of an infinite multiverse, which I rebutted at length in my reply. Carrier’s response has devolved to the childish “I’ve refuted you but I won’t even bother to tell you where.” Just a complete waste of time.

Carrier constantly accuses people of lying. But here, he is actually lying, in the sense of saying things that he knows to be false. If he has read my post, he knows I addressed everything he said. His original post did not rebut my replies—he’s just ignoring them because he knows there’s absolutely nothing he can say.

There are also a few occasions where he fails even to steel man (he keeps conflating “infinitely likelier” with infinite rather than diminishing gains in probability, which I think is confusing probability with odds, as I explained; or “you should think the number of people that exists is the most that there could be,” which is a non sequitur, as I explained). I suspect this is just more poor wording, and he really means what I propose as the steelman of these points instead, which again I already addressed in my comment, and he makes no reply to in his.

It’s true, of course, that as the odds ratio of some bit of evidence goes up, it has diminishing effects on the percent of the hypothesis in light of the evidence (you can’t get big gains over 99%, even with an infinity to one odds ratio). But what the hell does that have to do with what I said? Carrier suggested that you can’t have a probability be infinitely greater than another. That’s just wrong, and obviously so.

His claim that thinking the number of people that exists is the most it could be “is a non sequitur,” just shows he doesn’t know what he’s talking about. There’s a view of anthropic reasoning called the self-indication assumption which says your existence is likelier by a factor of N if there are N times as many people that exist. If that view is true, then your existence gives you infinitely strong evidence that the number of people that exists is the most it could be. I argued for that view at length—Carrier doesn’t address any of what I said about it. He’s just not engaging with the argument, preferring to take random confused swings at random statements I made, each one involving egregious philosophical and statistical errors.

It would be a non-sequitur if I’d just asserted it without justifying it. But I included quite considerable defense of that view, in the linked post having included over 4,600 words defending it. In my most recent reply to Carrier, I urged at length that he read that post for the full argument, but even went out of my way to include one of the arguments for the self-indication assumption. What does Carrier say in reply? Nothing, just bluster. Next, he says:

And sometimes he just gets wrong what I said (e.g. he mistakes what I meant by my remark about multiverse theory, as already explained in the other thread here).

But I quoted and rebutted what he said about the multiverse theory! What he said in the quoted thread was:

[An infinite multiverse] is not observed. So it cannot be cited as evidence confirming the theory. That’s my point.

That apologetic doesn’t work anyway because the prediction should be observed in this universe, not unobserved ones. So the prediction failed. Whereas atheism exactly predicts what we observe. So this is evidence against theism.

In my reply to Carrier, I quoted that statement of his and said:

But it follows from the self-indication assumption—the principle that you’re likelier to exist if there are more people—that your existence gives you infinitely strong evidence for an infinite multiverse. If SIA is right—which I’ve argued for at length—then your existence does give you infinitely strong evidence for an infinite multiverse. Carrier says that the evidence must be in this universe, but it is! The odds that this universe with you would exist is infinitely higher if an infinite multiverse is created, for the reasons I’ve already given (in the brief argument I gave for SIA a few paragraphs back, just replace people with universes containing exactly one person). See also the other arguments I give in the piece arguing for SIA.

So Carrier asserts that I’ve given no evidence for a multiverse. In response, I note that a view that I’ve spent upwards of 5,000 words arguing for implies that your existence is evidence for a multiverse. In reply Carrier says he’s already rebutted me and so doesn’t need to repeat himself. Could a more blatant display of incompetence and dishonesty even be imagined?

Lastly Carrier writes:

In the end, all his handwaving doesn’t change the fact that there is no reason to believe there should be endless people on atheism; atheism predicts there should be pretty much the number we observe (owing to rareness of biogenesis, evolutionary and terrestrial timelines, and planetary carrying capacity, the observed number of people is pretty much in line with prediction) whereas theism predicts there should be many more (every planet should be inhabited; we observe they are not, which falsifies rather than argues for theism). His giant wordwall never addresses this, my actual point. And generating giant wordwalls to avoid ever addressing a point in an effort to claim to have addressed the point is not the sign of sound reasoner.

Rather amusingly, Carrier is so breathtakingly incompetent that he triumphantly declares he’s refuted my argument by parroting arguments that are integral steps of the argument. He says “there is no reason to believe there should be endless people on atheism,” and “atheism predicts there should be pretty much the number we observe.” Indeed! This is the entire point of the argument.

The argument claims that your existence gives you evidence that there are more people. In the coinflip case discussed a few paragraphs back, where a coin is flipped that creates one person if heads and ten if tails, you should think at ten to one odds it came up tails if you’re created. Your existence is likelier if more people come to exist. But if this is right, and ten people existing makes your existence ten times likelier, infinity people existing makes your existence infinity times likelier. So by the very same logic, your present existence gives you infinitely strong evidence that there are infinite people, spread throughout an infinite multiverse. More specifically, it gives you infinitely strong evidence that there is a big infinity worth of people.

Given that, as Carrier himself says “there is no reason to believe there should be endless people on atheism,” once we conclude from your existence that there are, in fact, endless people, we get evidence for theism. That is the anthropic argument that Carrier has wildly flailed at without even remotely grokking. Put succinctly, the argument is: your existence gives you evidence that there are endless people, and this fact confirms theism.

Thus, Carrier is right that “atheism predicts there should be pretty much the number we observe.” The problem is he ignores my arguments for why your existence gives you justification for inferring that there are endless people. So if Carrier is right that atheism predicts “the number we observe” but we have anthropic evidence for there being many more people than we observe, then we have evidence against atheism. Now, Carrier might disagree with that anthropic evidence, but he has yet to address even a single one of my arguments, and has ignored my detailed takedowns of his laughably bad replies.

Carrier says that given theism every planet should be inhabited and claims that my “giant wordwall never addresses this, [his] actual point.” I can only conclude, once again, that Carrier either is a liar or cannot read. In my reply, I said “Regarding his final claim that theism predicts the entire universe being inhabited, see here for my reply to this.” The linked article is entirely about the question of whether the scale of the universe favors theism or atheism. Carrier’s just ignored it.

Thus, while Carrier’s reply to me carries his characteristic bluster, it’s filled exclusively with falsehoods. Every one of his points has been rebutted at length, and he’s just ignored the replies.

3 Fine-tuning

The first article that Carrier linked in which he allegedly refuted the fine-tuning argument is titled Bayesian counter apologetics. The basic point that he makes in the article is that if one takes into account the total evidence regarding the data cited in favor of theism, it flips to be against theism. He claims this is true of fine-tuning, writing:

This one’s easy. If you made a world, you simply would not muck about with the Standard Model, or physical constants of any kind, much less waste billions of years fiddling idly and trash countless lightyears with lethal useless junk-and-radiation-filled vacuum just to share the prospect of life with the people you meant to make, and then make them so badly designed they are plagued with illnesses and vulnerabilities to a still-hostile environment even in the one microscopic place they are supposed to live. To the contrary, that is what the world would have to look like if there was no God; because then, that is the only universe that could produce observers to notice, as all that waste is necessary for a random chance accident to generate us.

But not if there’s a God. God does not need to roll dice. Literally. He can just make what he wants. Like, poof!

I agree with Carrier here that on theism, one would expect the world to be a lot better than it is. The problem of evil is significant evidence against the existence of God. But it’s not totally decisive evidence; if we conjoin theism with the decently plausible archon abandonment theodicy, according to which archons (angel- or demon-like creatures) are tasked with making the world very good but have failed, it predicts a world filled with pointless evil.

God would be likely to set up such a scheme because it brings about a great potential good. In the scenario where the archons succeed, their sacrifice for our sake forges a valuable connection with us that lasts forever, thus being infinitely valuable. But for that connection to exist, the archons have to make the difference regarding our welfare—they can only do that if God doesn’t intervene in the event that they fail.

When we conjoin this assumption, which is not terribly unlikely, to theism, it perfectly predicts the data. It predicts a world with various improbable features that are needed for any value to arise, but one also filled with evil and indifference. The people tasked with making the world non-indifferent have failed. While God allows us to exist so that we can serve as “backup archons,” he otherwise doesn’t really meddle. This perfectly explains the world we observe. On the one hand, it has many indifferent looking features. On the other, it has many improbable features needed for agents: consciousness, psychophysical harmony, moral knowledge, and so on. Naturalism can’t explain the features needed for agents, while theism conjoined with the theodicy can explain the indifference of the world.

I admit that theism takes a hit from adding an auxiliary hypothesis. But once theism has that auxiliary hypothesis, there’s virtually no evidence favoring naturalism over it, while lots of evidence favoring it over naturalism. Theism thus wins by quite a lot in the overall balance of considerations. And this is just one theodicy; there are many others. This theodicy also explains divine hiddenness; the failed archons were tasked with bringing the world out of ignorance about God, but failed, leading to inconsistent religious beliefs and divine hiddenness.

Carrier asserts that God wouldn’t bother making a universe governed by the standard model or other physical laws. But why think that? Carrier doesn’t give any reason. In fact, I think that a physical universe is evidence for theism over atheism; under atheism, there’s no reason to expect a physical universe or laws at all. In contrast, if God is going to create conscious beings, which he likely will because it’s good to create, he can do this in ~3 ways:

1. An idealistic universe with no physical stuff.

2. A physical universe where all creatures are created directly by miracle.

3. A physical universe where creatures are created by workings of the physical laws.

None of the options seems obviously better than the other. It’s therefore not terribly unlikely that God would choose option 3. It should be given at least 10% odds. That’s way likelier than laws + a physical universe is on atheism, so I think this is a point in favor of theism.

Carrier also suggests that the scale of the universe is evidence for atheism in that it took time for life to arrive and the universe is mostly barren. Regarding taking a long time for life to arise, this isn’t very surprising. I think probably we preexisted our birth on earth (for anthropic reasons). Thus, God had no particularly strong reason not to wait.

Furthermore, there’s no strong reason to think that God wouldn’t wait. It makes sense not to waste time if you’re finite and have limited time that you’re alive for. But if you’re eternal and have infinite resources, time isn’t a cost. I’ll maybe give the fact that we were born billions of years after the universe a slight point in favor of atheism, but not by much.

Regarding the scale of the universe, I think that’s a point in favor of theism. I’ve written about this here. In short:

1. Our best physics points to an infinite universe which is much likelier given theism than given naturalism. God would be almost certain to create an infinite universe given that it’s good to create.

2. It doesn’t matter that the universe is barren because God has no resource constraints. He can create as many people as he wants while leaving however much of the universe he wants barren. Wasting space isn’t a problem for God when he can make infinite space with a thought.

3. There wouldn’t actually be more people if the universe was more densely packed. If the universe is infinite, there would be the same number of people no matter how they were distributed. This can be seen by the fact that you can change around the density of people by moving them around—it can’t be a viable proxy for the number of people that exist.

So let’s sum up the facts we’ve analyzed so far. We’ve talked about:

1. The universe being barren.

2. The universe probably being infinite.

3. The universe taking a long time for life to develop.

4. The existence of physical laws.

5. The existence of a physical universe.

6. Evil.

I think the first 5 are roughly a tie. Evil is a big piece of evidence for atheism, but it’s outweighed by the many bits of evidence for theism. Once we have evil in the background, there isn’t much more evidence for atheism; hiddenness doesn’t add much force to the problem of evil, because whatever it is that explains pediatric cancer, malaria, natural disasters, and animal suffering will probably also explain why some people don’t believe in God.

But it’s a bit disingenuous to say that the fine-tuning argument is defective because it neglects the problem of evil. They’re just different arguments. You don’t rebut an argument by pointing to another argument that’s slightly similar to it. The arguments should be considered one by one.

Anyway, with the above first five facts roughly cancelling, we have the problem of evil left on the atheist side. On the theist side, we have the following facts:

1. The laws, unlike the overwhelming majority of possible laws, including all the simplest laws, produce something interesting and valuable.

2. The constants in the laws are finely-tuned in the sense that if they were a bit different no interesting structures would arise.

3. So are the initial conditions.

These are all very surprising. I’ve written about why there’s no good naturalist explanation of this here—the multiverse just pushes the fine-tuning back a level. Being generous, I think the problem of evil and these considerations roughly cancel out—the atheist’s best argument and the theist’s best argument cancel out. So at the end of all this bloodshed, atheists can mitigate the evidential force of the fine-tuning argument, but only if they’re willing to sacrifice their best argument.

Carrier writes:

All such excuses thus fall to the same fact: if you were God, this is not the world you would make. No moral entity would, who had the option not to.

God is omniscient, knowing many things we don’t. Thus, even if I wouldn’t make this world, it may be that God has a reason to. I wouldn’t make the move Magnus made here, but that doesn’t mean it’s not a good move; it just means he knew stuff I didn’t.

Now, it’s true that the probability that this universe would exist given theism is low. But it’s also low given atheism for the reasons discussed—an atheist universe would be nearly guaranteed to be barren and lifeless. Whichever worldview you adopt, the world is surprising. It has too much of value for naturalists to expect it, too much of disvalue of theists to expect it. As Kenny Boyce says:

My live hypotheses are that you intentionally arranged the marks on the paper or the marks were generated at random. I look at the paper and find a crude limerick not to be repeated in polite company. That's really surprising given the hypothesis that you wrote it (I would have expected another interesting argument for the existence of God). But it's less surprising given that you wrote it than it is given that the marks were generated at random.

Most of Carrier’s article, therefore, just involves pointing to other arguments that are sort of related to fine-tuning but mostly different. It’s little different from the cheap trick employed by apologists of saying, in response to the problem of evil, “evil proves God because you need God for morality.” Even if it were correct, it would be obviously changing the subject. Most of his arguments are not very successful—the only one that is is the problem of evil. While the POE is a good argument, it’s a different argument, and thus fine-tuning still has lots of force.

Carrier has another article explaining why the fine-tuning argument is allegedly a bad argument titled Why the Fine Tuning Argument Proves God Does Not Exist. Well, actually Carrier has a lot of articles about this, but they’re mostly pretty redundant and I can’t address them all.

Carrier begins his article by noting that adding to your theory to explain the data is a cost. It’s not virtuous to posit, after getting some highly specific sequence of cards when you flip over a deck, that there’s a fairy that loves that sequence. While that theory does predict that sequence, what it gains in explanatory power it loses in prior probability.

Carrier is right. Notably, however, a theory with an auxiliary hypothesis can still often be superior to other theories. If you see a video of John committing a murder, but his DNA isn’t on the murder weapon, it makes sense to think that he committed the weapon while wearing gloves. This adds an auxiliary assumption, but it’s likely given the total evidence. Thus, if theism can explain lots about the world with just one auxiliary assumption (which it can) it does pretty well.

Next, Carrier explains:

[T]he only way we could [observe ourselves existing] without a God is by an extremely improbable chemical accident, and the only way an extremely improbable chemical accident is likely to occur is in a universe that’s vastly old and vastly large; so atheism predicts a vastly old and large universe; theism does not (without fabricating excuses—a bankrupt procedure, as I already explained … ).

Similarly, the only way we could [observe ourselves existing] without a God is by an extremely long process of evolution by natural selection, beginning from a single molecule, through hundreds of millions of years of single cells, through hundreds of millions of years of cooperating cells, to hundreds of millions of years of multicellular organisms; so atheism predicts essentially that; theism does not …

Likewise, … we should expect [the universe we observe] to be only barely conducive to life, indeed almost entirely lethal to it (as in fact it is), since there are vastly more ways to get those universes by chance selection, than to get a universe perfectly suited to life throughout (indeed … by countlessly many trillions to one). Design predicts exactly the opposite (again, without a parade of convenient excuses).

I’ve already addressed his point about the scale and hostility of the universe. As I’ve argued, the scale of the universe is actually an own goal—it favors theism. But Carrier’s argument goes wrong in several other ways.

First of all, even if what he says is correct, the fine-tuning argument can still be evidence. Carrier argues that given that we exist, the world we observe is much likelier given naturalism than given theism. I might even agree with that claim! But crucially, the odds that we’d exist are astronomically lower given theism than naturalism.

I mean, first of all, for us to exist, one needed the following ingredients, each improbable:

• a physical universe.

• laws

• nomological harmony

• the laws to be able to produce complex structures.

• fine-tuning.

• certain physical arrangements producing consciousness.

• those arrangements arising.

• psychophysical harmony.

• you happening to exist, out of the infinite possible people.

And quite a bit more. Thus, even if naturalism better explains the world given that you exist than theism, theism can still win by better predicting your existence. If at this point you’re thinking that this argument fails because you wouldn’t be around to observe anything if those things hadn’t happened, I’ve explained why that’s irrelevant here (see the section on the anthropic principle). Carrier is therefore egregiously wrong when he says:

Without any evidence that God is any more likely than not to have made a universe to produce life in exactly the same way as a godless universe, the probability of each decision from God is at best 50/50. Whereas the probability is in every case 100% if there is no God, because the probability that we’d observe ourselves in a Fine Tuned universe of this kind is 100%.

Um, no. The odds of complex chemistry arising given naturalism are near zero given the above steps needed for it.

In other places, Carrier has suggested that the fact that we exist can’t be evidence either way, because if we hadn’t existed we wouldn’t have known this fact. But this is irrelevant. The fact I exist is evidence my parents didn’t use effective contraception, even though if they had, I wouldn’t be here (see also the firing squad case). Your posterior that you exist is 1, but your prior is not.

Second, Carrier is just wrong that naturalism predicts an old universe with billions of years of chemical accidents. On naturalism, the most likely scenario where we’d exist would be one where we’re Boltzmann brains. Even if our total evidence rules out our being Boltzmann brains, they’re generally easier to create than more complex observers.

Naturalism, by itself, gives no reason at all to expect that the way to produce consciousness would be through biological evolution. It’s logically consistent with naturalism that any arrangement of matter produce consciousness. In fact, probably naturalism predicts panpsychism, because consciousness arising at the deepest levels is simpler than it randomly kicking in at higher levels. A young universe could also give rise to life by chance.

Thus, taking into account the total evidence regarding your existence, we get strong evidence for theism! Carrier’s counterapologetics has been countered!

Fine Tuning is therefore evidence against intelligent design. It could never be evidence for it, because gods don’t need fundamental constants at all, much less all the weird ones we have.

This is a complete non-sequitur. Compare: “the fact that Jim there was a building painted blue is evidence against it being designed. It could never be evidence for it, because builders don’t need to paint things blue at all. They could pain it any color, like green.” While this is true, it’s completely irrelevant; the fact an agent doesn’t need to do A to achieve B means that the odds of them doing A if they’re doing B are less than 1, but it doesn’t mean B isn’t evidence that the agent did A. It depends on what the alternative probabilities are.

Carrier ends his article by linking to two articles where he makes the case that “the God hypothesis also requires positing a chance accident—indeed, one even more improbable.” I won’t address the second one because this article is already too long, but I’ll note that the saintly Aron Wall rebuts it in detail.

The first of these pieces involves raising three basic points—I’ll explain the claim and then link to where I’ve addressed it.

1. Fine-tuning assumes you keep all the other constants the same and change one of them, which neglects the possibility of life arising if you change a bunch of constants at once.

2. The constants can take on values over an infinite range, so the probabilities turn out undefined.

3. A finely-tuned God is even more improbable than a finely-tuned universe.

I’ll just link to these, given that nothing Carrier says addresses what I have to say in those articles.

4 Consciousness

Carrier claims to have rebutted the argument from consciousness in his article Bayesian counterapologetics. He begins by saying:

Theists try to focus just on the fact that conscious phenomena are weird and not yet scientifically explained, “therefore” God is the best explanation of it. But that’s a non sequitur. When we don’t know an explanation, the most likely explanation will be the one that has most commonly succeeded before when we thought something couldn’t be explained.

???

I have never heard a theist make the argument that consciousness must not be reductively explained because we haven’t reductively explained it yet. We also have not explained high temperature superconductivity—a mysterious phenomenon in physics. Until a few years ago, we had not explained holes in swiss cheese. Despite this no one argued God must be directly behind it.

Instead people argue that there are good philosophical reasons to think that consciousness can’t be reductively explained. These reasons are recognized by many atheists of a sharply materialist bent, like Chalmers. Thus, consciousness must be built into the fundamental laws. But the odds of the fundamental laws giving rise to consciousness are high given theism while low given naturalism. Conscious laws are near certain given theism, while they fit poorly with the rest of naturalism. Giving them 1% odds conditional on naturalism is generous.

(Regarding his inductive trend argument, see here).

For example, that we need brains to generate conscious phenomena is quite unexpected if God exists. Because if God exists, disembodied minds can exist, and are the best minds to have, therefore we should also have disembodied minds.

Why are disembodied minds the best minds to have? Carrier doesn’t explain. This doesn’t seem obvious. Now, it’s true that our brains are inefficient and kludgy, but I assume the explanation for that will be the same as the explanation for all the other evils of the world. If you just keep pointing out that stuff is poorly made in the world, you’re quintuple counting the problem of evil. Thus, pointing in response to theistic arguments “but the thing that you pointed to is bad in various ways,” isn’t a devastating rebuttal; it’s just double-counting.

Indeed, when you include the other fact about consciousness—that it is psychophysically harmonious—that gives even more evidence for theism. Psychophysical harmony has force like few other arguments, majorly reducing the odds of atheism. So when you take into account the full range of facts regarding consciousness—both that it exists and is harmonious—naturalism starts to look quite poor.

I thus conclude that Carrier has completely failed to rebut the argument from consciousness.

5 Skepticism

The argument from skepticism makes two claims with similar justification. First, atheism implies skepticism—one who thinks atheism is true loses their justification for trusting their senses. Second, the fact that we’re not in a skeptical scenario is evidence for theism. I’m quite confident in the success of the second argument, while less sure about the first.

On atheism, it’s pretty likely we’d be in a skeptical scenario. Default atheistic universes produce huge numbers of Boltzmann brains. If entropy keeps rising, there will be an infinite period of time with only Boltzmann brains and only a finite period without. By default, an infinite atheistic universe collapses all probabilities.

In contrast, if there’s a perfect God, he’d want to keep us out of skeptical scenarios. He wouldn’t want people to be globally deceived. Thus, because atheism predicts skepticism, atheists should be skeptics, and if as is obvious we’re not in a skeptical scenario, we have good evidence atheism is false. Carrier claims he addressed this point here and here. So let’s see what he says.

The basic argument that Carrier makes in the first article is that we should think we’re not in a skeptical scenario because it poorly explains the data. If there’s a cartesian demon or we’re Boltzmann brains, it’s unlikely everything in the universe would seem to obey orderly and predictable laws. The mental life of Boltzmann brains are mostly random chaos.

The problem is this doesn’t touch the second argument. Consider the following dialogue:

Richard (after being kidnapped): hey, I think I was captured by the notorious helicopter capturer gang, who, when they capture people, almost always puts them in helicopters.

Me (also captured): well, that’s pretty unlikely. We’re in a van, not a helicopter.

Richard: Look, I know we’re not in a helicopter. Thinking we are involves committing various fallacies. Thus, the argument from not being in a helicopter against being captured by the helicopter capturer gang is fallacious.

In this case, the helicopter is analogous to the skeptical scenario, the helicopter capturing gang is analogous to naturalism, and the theory someone else captured us is analogous to theism.

In other words: even if we’re are not in skeptical scenarios, the fact that we’re not in skeptical scenarios is surprising on naturalism but not on theism. It’s therefore evidence for theism.

But does this work to rebut the claim that naturalists should think they’re in skeptical scenarios? It’s not at all clear. Richard notes that if we’re in a skeptical scenario it’s not likely that we’d observe orderly physical laws. But it’s easy to cook up a scenario, like one where there are infinite Boltzmann brains, where infinite people have that experience. And that’s what matters probabilistically—the number of people having an experience.

In his next supposed rebuttal, Carrier claims that the Boltzmann brain problem is “logically invalid.” This would be news to the many physicists who work hard specifically to rid their models of Boltzmann brains. Carrier’s argument:

It is a fact that the vast majority of people on Earth will observe themselves to be non-philosophers. Therefore, since I’m not in fact a non-philosopher, I can conclude that the rest of the population of Earth doesn’t exist. Needless to say, something terribly illogical has just been said. And yet this is the same…

…

The fact of the matter is that it would not matter how many “more” Boltzmann brains there are. Since a vast multiverse logically entails there will also be countless non-Boltzmann brains—whole civilizations arising from causal sequences in stable environments—it follows that the probability that there will also be persons observing themselves in one of those conditions is as near to 100% as makes all odds. So a merely immediate observation alone can’t tell you which kind of brain you are; because there will be infinitely many of both kinds of brain.

…

The number of Boltzmann brains in the universe has no relevance to whether “we” would be one. By definition, brains that evolved into civilizations won’t be Boltzmann brains. Therefore there is a near zero probability that a brain observing itself in an evolved civilization will be a Boltzmann brain.

Suppose we’re considering between two theories: the first one says that everyone is a philosopher. The second one says that only one person is. If I learn that I’m a philosopher, I get evidence for the first theory. (Whether you’ll get overall evidence for the nonexistence of non-philosophers will depend on your exact theory of anthropics—SIA and CC hold that you’re likelier to exist if more people do, all else equal, so you won’t get an overall update in favor of solipsism. On SSA, you will. If you don’t know what that means, don’t worry, it’s irrelevant to the larger argument).

Carrier says that “it would not matter how many “more” Boltzmann brains there are.” This is patently false! As an analogy: suppose that 99% of people have birthmarks. Well, then you should think that probably you have a birthmark if you haven’t looked. If 99.9999999999999% of people have birthmarks, then even if you have looked and didn’t see one, you should assume you just missed it.

Similarly, if the overwhelming majority of people are Boltzmann brains, it’s surprising you’re not. At the first moment of birth, you should have expected to be one. The fact that you aren’t one is thus strong evidence that most people aren’t Boltzmann brains.

Or here’s perhaps a better analogy. Suppose that there are two groups: the longlasters and the shortlasters. The longlasters live for a long time. In contrast, the shortlasters live very short, simulated lives. The shortlasters only live about a day, while the longlasters live 100 years. If there were enough more shortlasters than longlasters, then after living for a day, you should think that probably you’re a shortlaster with fake memories. The same basic principle is true of Boltzmann brains.

Carrier seemed to suggest that there couldn’t be more simulated brains than real ones. After all, in an infinite multiverse, it would be infinite in both cases. But this will break down all probabilities. There will also be infinite Boltzmann brains with Richard’s exact set of experiences; thus, Richard has no grounds to think he’s probably not a Boltzmann brain. If you have aleph null people with any of a range of properties, your credence in having each of those properties will turn out undefined—dividing infinity by infinity gets you an undefined answer, by default. This is the infamous measure problem (well, technically the measure problem is a bit broader, but this is certainly a major part of the measure problem).

Now, you can solve this by taking a measure over observers. A measure is a way of calculating probabilities using math over infinites. So, for example, by taking a measure over the natural numbers, you can get intuitive results like that only a small percentage of naturals are primes (though it’s infinite in both cases).

(This math is all way above my paygrade, so I might be getting some stuff wrong. But this is my best understanding from talking with people who do get the math).

But once you start taking measure over observers, it’s no longer obvious that the probability of someone with your experience being a Boltzmann brain isn’t high. Worse, measures over infinite numbers of observers are sensitive to how observers are arranged—thus, it implies that your credence in your having some property would change if people just moved around.

(I explain why this problem won’t apply to theists here).

Thus, Carrier has completely failed to address the Boltzmann brain problem for atheism.

6 The problem with the problem with nothing

The last piece of Carrier’s I’ll respond to is titled The Problem with Nothing: Why The Indefensibility of Ex Nihilo Nihil Goes Wrong for Theists. I was going to respond to his objections to my argument from haecceities, which Carrier says:

fallaciously gets wrong the entirety of identity theory (and even they suspect this, as they grade it poorly; in any event, I explain correct identity theory in comments under Kastrup’s Idealism, but I will pull all that together today in A Quick Brief on Identity Theory).

It would be rather impressive if I managed, in just over 1,000 words, to get wrong “the entirety of identity theory (and fallaciously, no less)!

Reading his articles, however, I do not see anything in them which explains where my argument goes wrong. They’re just not about the argument from haecceities. Thus, I shall refrain from commenting until I know where Carrier thinks my argument went wrong.

In his article titled The problem With Nothing, Carrier has a proposal for how you can get a universe like ours out of nothing. While it doesn’t strictly refute any of my arguments, if it’s right, it would lower the probability of theism relative to nothing-first atheism. As such, it’s worth mentioning.

Carrier summarizes his basic argument as the following:

A quick and dirty way to phrase that argument is: if nothing exists, then by definition no rules exist limiting what will happen to it; if no rules exist limiting what will happen to it, it is equally likely it will become one of infinitely many arrays of things (including remaining nothing, which is just one of infinitely many other things no rule exists to prevent happening); if we select at random from the infinitely many arrays of things it can become (including the array that is an empty set, i.e. continuing to be nothing), the probability is infinitesimally near 100% the array chosen at random will be a vast multiverse whose probability of including a universe like ours is infinitesimally near 100%. Because there are infinitely more ways to get one of those at random, than to get, for example, the one single outcome of remaining nothing. There is no way to avoid this. Unless you insert some law, power, rule, or force that would stop it, or change the outcome to something not decided at random. But once you do that, you are no longer talking about nothing. You have added something. Which you have no reason to add. Other than your human desire that it be there. Which is not a compelling argument for it being there.

He more precisely lays out the steps here:

Proposition 1: That which is logically impossible can never exist or happen.

Proposition 2: The most nothingly state of nothing that can ever obtain, is a state of affairs of zero size lacking all properties and contents, except that which is logically necessary.

Proposition 3: If there was ever Nothing, then nothing governs or dictates what will become of that Nothing, other than what is logically necessary.

Proposition 4: If nothing governs or dictates what will become of Nothing (other than what is logically necessary), then nothing (other than what is logically necessary) prevents anything from happening to that Nothing.

Proposition 5: Every separate thing that can logically possibly happen when there is Nothing (other than Nothing remaining nothing) entails the appearance of a universe.

Proposition 6: If there is Nothing, then there is nothing to limit the number of universes that can logically possibly appear.

Proposition 7: If nothing (except logical necessity) prevents anything from happening to Nothing, then every logically possible thing that can happen to Nothing has an equal probability of occurring.

Proposition 8: If every logically possible thing that can happen to Nothing has an equal probability of occurring, then every logically possible number of universes that can appear has an equal probability of occurring.

Proposition 9: If when there is Nothing every possible number of universes has an equal probability of occurring, the probability of Nothing remaining nothing equals the ratio of one to n, where n is the largest logically possible number of universes that can occur.

Proposition 10: If Nothing produces a random number of universes, nothing exists to prevent the contents of each of those universes from being equally random.

This is stupendously wrong. First of all, if it were right, because the huge majority of possible worlds involve us being in skeptical scenarios, if nothing really just produces a random possible world, we’d be nearly guaranteed to be in a skeptical scenario. After all, for every world where the laws continue working as they have up until this point, there are infinity possible worlds where they horrendously self-destruct one second from now—perhaps where everything in the cosmos is replaced by a goose or thing of wood or a new law appears that causes everything to collapse in on itself. Thus, Carrier’s proposal is completely untenable.

Second, Carrier’s premises are wrong. Specifically, in premise 3 he claims that “If there was ever Nothing, then nothing governs or dictates what will become of that Nothing, other than what is logically necessary.” But this is false; there might still be metaphysical principles that constrain what happens, even if they’re not logically necessary. In fact, there necessarily have to be metaphysical principles that govern what will happen, because logic alone doesn’t tell you what will happen. The principle that “every logically possible thing that can happen to Nothing has an equal probability of occurring,” certainly does not fall out of the laws of logic. They just don’t tell you about the relative likelihoods of different worlds becoming actual.

Indeed, it’s generally thought that simpler worlds are likelier to be actual. We should not expect a unicorn to pop into my room uncaused because unicorns are complicated things. But if this is right, then probably a nothing universe would stay in its nothing state—that would be much simpler and more elegant than the alternative.

Indeed, though I do not like the cosmological argument, the following does seem a perfectly good metaphysical principle:

It’s highly unlikely that things would appear uncaused.

But if this is right, then nothing becoming something is very unlikely, if it’s even possible, which isn’t obvious.

Additionally, as my good friend James has noted, Carrier seems to accept a powers based view of laws (though I admit to finding his commentary on such matters confusing). On a powers based view of laws, stuff happens because objects have the power to make it happen. The reason that everything attracts is that every particle has the power to attract every other particle.

But if behavior is grounded in powers, then nothing would stay dormant. Nothing has no powers. (The same will also be true if one adopts a governing account of laws. Only a humean account of laws leaves open the possibility of something appearing for no reason, but Carrier rejects such an account, and for good reason).

Thus, Carrier’s proposal involving getting a universe from nothing is badly flawed to the point of being completely unworkable.

7 Conclusion

Richard Carrier is quite a prolific writer. He writes at considerable length about many theistic arguments and concludes, without exception, that they are dismal failures. However, when one examines his arguments in detail, they mostly rest on extreme error and wild overconfidence. Specifically, Carrier often makes the following errors:

1. Pointing out some phenomenon is unlikely given theism and treating that as entailing it’s evidence against theism. But one must consider the probability of it given atheism—if that probability is lower then it will be evidence for theism.

2. Neglecting the improbability, given theism, of your existence.

3. Double-counting the problem of evil.

4. Double-counting the problem of evil.

5. Double-counting the problem of evil.

6. Double-counting the problem of evil.

7. Misunderstanding why Boltzmann brains are supposed to be problematic.

8. Thinking you can get something from nothing with high probability.

His objections to the anthropic argument were even more embarrassing. It’s clear that he doesn’t even understand what the argument is saying, and thinks that each step is exactly the opposite of what it actually is. This is a consequence of his creative readings, where he, for instance, interprets my stating that there are infinite people as claiming that there are finite people. It’s easy to get confused if you bizarrely, for no reason, interpret a person’s statement as being the exact opposite of what they actually said.

Well this is Carrier’s shtick. Insane overconfidence, errors and misreadings that verge on the infinite (and no, this is not me saying that they verge on the finite), and constant accusations of lying. When these errors are pointed out, he never corrects them, preferring to call the person who rebutted him a liar or charlatan. One would almost pity such profound philosophical incompetence if it weren’t paired with extreme aggression and overconfidence.