Last week I published an eristic analysis of an exchange of videos between Rationality Rules and Capturing Christianity, on Which Is ‘Rational’: Theism or Atheism? This time I will analyze a previous Capturing Christianity video, “Why Theism Best Explains Reality (philosophy nerds, get in here!)”, featuring the content creators Tim Howard and Kyle Alander defending “the explanatory power of Theism.” These two articles of mine were patron requested and funded, and today’s goal will be the same as last time: to illustrate what is going wrong with the Christian thought process, resulting in their building a bulwark of unwittingly irrational defenses of their delusion that theism is rational. It then becomes evident that atheism does not depend on the same architecture, and why.

Questions of what it means to be delusional or irrational (e.g. irrational beliefs are those arrived at only by fallacious reasoning, and delusional beliefs are those immune to all rational arguments), or what the scope or consequences of those conditions are (e.g. being irrational in defense of a delusion does not make you an irrational person as if it were a global personality trait, and “rationality,” “intelligence,” and “knowledge” are not the same things), are addressed throughout my previous article, and so you may want to read that article first, for a more accurate perspective on this one. Here I will simply jump right to the core of it. But keep your eye out in coming months for a thoughtful debate on this same topic between me and Miles Donahue, who has an approach similar to but different than Howard and Alander.

Summarizing the Video

Although Howard and Alander are only YouTubers (running the channels Doxastic Mastery and Christian Idealism), not PhDs, this is still high-level apologetics, involving sincere thinkers who are competent, intelligent, well-studied, and informed. So it is definitely a steel man of a Christian worldview defense. This is not shallow or disingenuous. It’s not rhetoric or propaganda. This is an example of the best they have.[1] At least for some kind of “theism.” There is a broader problem that Christianity is a lot whackier and harder to justify than just “theism” (as I mentioned last time, you can get a sense of this in my article Christianity Is a Conspiracy Theory, John Loftus’s chapter “Christianity is Wildly Improbable” in The End of Christianity, and Richard Schoenig’s book Where Christianity Errs). But this video isn’t tackling that; it’s just building a groundwork case for a theistic supernaturalism generally.

Their case still fails. It functions well on internal coherence, but it suffers the common flaw of all apologetics: it leaves evidence out that, when put back in, completely reverses its case to the opposite conclusion. All we see on their show is the blinders-on build-out of their edifice, so I cannot assess anything else about why they believe what they do or whether being shown their mistakes will ever impact them (unlike last time, which was a show full of the latter information). Indeed, some of their error derives from their being false-framed—that is, rather than doing the false framing, they fall victim to someone else’s, like Ted Poston’s paper “The Intrinsic Probability of Grand Explanatory Theories,” which never correctly describes any pertinent version of atheistic naturalism, thus creating straw men to compare with theism; conversely, Poston also mis-describes theism, so as to argue it is epistemically simple when in fact (by the correct metrics of specified complexity and explanatory gerrymandering) it can’t be (there is, after all, A Hidden Fallacy in the Fine Tuning Argument, and all explanatorily simple God hypotheses have been falsified).

However, I won’t be teasing out what’s them and what’s Poston (or anyone else they cite or rely on). Here I will only address Howard and Alander’s presentation, as they present it, and thus on its own terms. But the overall gist is that, supposedly (for reasons) theism is a simpler explanation than naturalism because naturalism requires more ad hoc suppositions (though, as usual, when we put back in everything they leave out, this conclusion reverses), and therefore, theism enjoys a higher epistemic probability (which conclusion, of course, flips the other way around when we put everything back in that they left out). I also won’t address their discussion of probability; it’s overly muddled and convoluted, and uses some words weirdly,[2] but the overall point is correct: the prior probability of a theory is inversely proportional to the number of logically independent propositions required to explain observations; it is not inversely proportional to the complexity of the resulting world.

A common example of this is when theists claim multiverse theory is just the brute positing of countless other universes (an arbitrary assumption that would entail massive specified complexity and thus be very intrinsically improbable). But multiverses aren’t brute posits. They are the inevitable outcome of simple cosmological theories (like eternal inflation) that have been verified by specific and unusual observations. It is therefore the simplicity of those theories that defines the prior probability of them being the cause of those observations, not the number of universes they generate. That they will generate countless universes is ~100% certain and therefore has no effect on the probability of those theories as an explanation. This means atheists do not have to “just posit” a multiverse; we have evidence for it, as an inevitable outcome of a simple process that has been empirically confirmed to be operating (see Six Arguments That a Multiverse Is More Probable Than a God). Theists have had no such luck. And this really gets to the heart of the problem that Howard and Alander never escape.

They are also right (later in minute 36) about retrodiction being respectable, but they miss why predictivism is more reliable: it’s easy to gerrymander a theory to existing data; a lot harder to accidentally do that to yet-unobserved data. And that difference skews the likelihoods very strongly in favor of a predictive theory (this is why Superstring Theory, despite its extraordinary retrodictive power, remains merely a contending hypothesis and not a settled conclusion of science). And so on. I won’t trouble myself further with these meta-questions. I’ll simply analyze the arguments Howard and Alander make for God.

Finding What’s Missing

One of the first rules I impart to students in my Critical Thinking online correspondence course is that, when faced with any argument to a claim, always look for what’s being left out. If you don’t find anything, that bodes well for the claim (see The Scary Truth about Critical Thinking). But if you find some stuff and it tanks the claim, you’ve just caught an irrational argument, and thus found a conclusion that’s likely to be false. It will be a strong trend in the deluded to avoid confronting or accepting this missing evidence. It requires from them ever-more elaborate apologetics (often off-the cuff) to get it to “go away” somehow, so they can remain committed to the dubious proposition they are emotionally invested in—all as I explained, and gave examples of happening in real time on this same channel, in my previous article. So let’s look for what’s missing here.

The Weird Argument from Trope Theory

Their first slip is when they try to get around the problem that “God” is a weird explanation (unprecedented and thus unsupported in actual background knowledge). In thousands of years of human experience, that’s just not ever what actually turns out to explain anything (see Naturalism Is Not an Axiom of the Sciences but a Conclusion of Them). They specifically give the problem of minds (intentional agents) only being observed to exist as complex embodied machines (see The God Impossible and the Argument from Consciousness, wherein this is what we expect to see if naturalism is true but not what we expect to see if theism is true). We just never see other kinds of minds, so we have no reason to believe there are any. Whereas we often see blind natural processes producing all the kinds of specified complexity that natural organisms aren’t already responsible for (from crystallization to evolution—even cosmic evolution: see How the New Wong-Hazen Proposal Refutes Theism). Statistically, that has beat the god hypothesis millions of times now. God has never won. So it’s millions to one it’s going to ever beat naturalism. And we should be betting on the winning horse.

How do they get around this? By a bizarre application of trope theory. Now, trope theory actually grounds physicalist reductionism (see the ontology sections of my book Sense and Goodness without God: A Defense of Metaphysical Naturalism and my article Why A Neo-Aristotelian Naturalism Is Probably True; although I don’t use this terminology there, I’m talking about the same things). That is to say, empirically, only physicalist tropes have been confirmed to exist (insofar as you grant “tropes” exist at all in the sense meant, which is not the sense used in literary theory or other fields). So, really, we’re just right back to the same problem: the “trope” of a disembodied or irreducible intentional mind simply doesn’t exist in our observation, so on background knowledge, it’s a poor choice of explanatory framework (without some really superb empirical evidence, but our not having that is the problem).

What is their argument? Because “God” can be modeled in terms of trope theory, and trope theory is in our background knowledge, therefore “a robust view of theism” can “fit and satisfy very well trope theoretic constraints, [and] so it does fit very well with background knowledge” (minute 23). This is a non sequitur. Indeed this is one of those cases where if they actually tried formulating their argument as a syllogism, rather than a stream of words, they might have caught themselves out as making an illogical argument. For example, I can also model the world of Harry Potter in trope theory. That does not mean it will “fit very well with background knowledge.” They have confused the fact that “anything can be modeled in trope theory” (which is true) with the conclusion that “therefore anything that can be modeled in trope theory fits background knowledge” (which is not true). All false worldviews can be modeled in trope theory.

What we are seeing here is a very awkward intellectualism, whereby it is their very intelligence that is trapping them in delusion, shielding from their cognition the irrationality of their argument by making that argument so convoluted and sophisticated that they mistake its sophistication and complexity for its being correct. Indeed I’d say most Christian philosophy is doing this, especially under peer review: by obscuring logical invalidity and unsoundness beneath fancy verbiage and intuitive moves of thought, a bogus argument can be made to appear valid and sound; but as there is no vettable syllogism (at all, or at the key linchpin move in the argument—because valid syllogisms can also be used to argue bogus things), the mistake is too many steps removed to be obvious to all but those committed to actually looking (by actually trying to prove it false before believing it). In result, Howard and Alander’s “Argument from Trope Theory” is illogical nonsense. Yet they remain seduced by it. Again they may be the victims of someone else’s bad reasoning (like Joshua Sijuwade’s paper “The Theoretical Virtues of Theism”). But again, I am evaluating their arguments here, not those of whomever may have misled them.

Here, what’s missing is a premise, “that which can be modeled in trope theory fits background knowledge,” which cannot be defended (that premise is obviously false—lots of things will fit trope theory but not background knowledge). This reflects the dangers of reasoning by enthymeme (a syllogism with a hidden premise), which I teach my students to identify and analyze because a lot of delusion and deception relies on this. By leaving a premise out, you can play on the faulty intuitions, emotions, and assumptions of yourself or others—they fill in the blank themselves, without thinking through whether that infill is actually defensible. It “sounds” right. But, it often won’t be. This is why syllogistic reasoning needs to be spelled out completely. You can’t ride on unstated premises. Because that’s precisely where your delusional mind will hide the truth from you.

What’s also missing here is a cross-test of the entire theory. What happens when we re-frame mind-brain physicalism in trope theory? Does God then gain or lose explanatory virtues relative to that framework? Well, obviously, God loses. Physicalism relies on empirically well-established tropes (one reason why it dominates all the cognitive sciences now). Disembodied minds don’t (much less superminds). Likewise, physicalism reduces consciousness to components far simpler than it (ultimately, subatomic particles so rudimentary they have only a small number of very basic properties). Theism can’t do that with God. Even as tropes, an electron is obviously vastly simpler than God’s Mind (or indeed any mind, even the mind of a worm, which is obviously vastly simpler than God’s). So when we ask, “Wait a minute, what is being skipped over here?” we then discover defects of their case like this.

This is why it is important not to simply address what is being presented, but to look for what is not being presented. There is more information in their omissions than in their assertions. A rational person needs to extract that information to evaluate their case (like missing premises and symmetrical hypothesis comparisons). That they are not doing this tells us they are irrational. And let’s be clear about this. They are very intelligent. They are very knowledgeable. But they are still not being rational. Yet they have deceived themselves into believing they are. Again, as I explained last time, this doesn’t mean they will be irrational in every aspect of their lives; but that is why delusions are so pernicious: they conceal from your own mind the incongruity between what you are doing to defend your delusional belief, and what you would do or even allow to defend any other.

The Other Weird Argument from Trope Theory

They then re-frame the usual “God is simple” arguments in trope theory, and claim that this solves their plausibility problem. But it doesn’t. Because tropes aren’t magic. Indeed, tropes have no causal powers at all. You can define Darth Vader in trope theory. But that doesn’t mean he exists and can choke you out. You can even define God as depicted in Time Bandits with trope theory, but that doesn’t mean God is therefore a cold and confused old man bumblingly chasing alien construction workers through time tunnels and who isn’t sure whether he created Evil because of, “um, I think it’s something to do with free will.” Tropes are just another framework for discussing abstractions (properties, generalizations, universals). But to exist, an abstraction still has to be realized. And all background knowledge (literally all of it) indicates that they have to be instantiated in a material. The trope of a heart cannot pump blood; you need the actual thing to do that. That’s the whole point about observed minds: they are always the outputs of complex physical machines. The trope of “mind” does not cause minds to exist. The assembly of mind-generating machines does. Which observation is also a trope.

So calling God a “single metaphysically simple omnipotence module trope” is just stringing together fancy words for an entity contrary to all existing physics that is possessed of infinite specified complexity and has no evidence of its existence. It’s like calling The Great Pumpkin a “single metaphysically simple gourd trope” and then saying, “See, it’s totally plausible that that exists!” This is not a rational argument. What it takes to have the trope of omnipotence realized is an infinitely organized entity, which entails massive improbability even as a machine, and has no plausibility at all without one (this is what I demonstrated years ago in The Argument from Specified Complexity against Supernaturalism and The God Impossible, all leading to A Hidden Fallacy in the Fine Tuning Argument). Here I think theists are confusing their intuitions with how those intuitions depend on a machine. “Oh, I can imagine a disembodied supermind, therefore it’s possible” is a fallacy. Because you could only imagine it on a physical mainframe—which means that physical mainframe may be required even to imagine a bodiless mainframe, much less have one. And all background knowledge points to exactly that conclusion. Gods are simply massively improbable. And trope theory cannot change that.

Even Howard and Alander admit that tropes “are abstract particular natures that actually instantiate their properties in objects” (minute 25). Emphasis on in objects. But there is no massive machine instantiating a god-mind. So this is a dead end for theists. It is simply lipsticking the pig of Aquinas (see Thomism: The Bogus Science). We see this in their obsessive repetition of the claim that God is “noncomposite,” but without thinking through how that can be (it is logically impossible for a database of infinite correct knowledge not to be composite—this is my point about its specified complexity; much less all the other property-complexes a God requires, like consciousness, rationality, and specified sets of desires and morals); or thinking through how that even helps. Spacetime is noncomposite, too (in the only required sense: spacetime isn’t made of anything but itself, so no matter how much you break it down, you still end up with just spacetime, nor does spacetime require anything else to exist).

Yet spacetime is vastly simpler than a God. There’s no consciousness, rationality, superpowers, or specific and complex intentional and moral furniture to instantiate there—and when anything like these things is ever instantiated, it’s just us, or other mundanely manufactured or naturally evolved machines. So if we’re to start at the simplest noncomposite, we ought to start with spacetime, not God (as I’ve explained in The Argument to the Ontological Whatsit and Superstring Theory as Metaphysical Atheism). That, indeed, gets us to inflationary Big Bang theory and the Standard Model (by one TOE or another) and from that, everything else. This is the same failure I documented in Edward Feser’s identical argument to Howard and Alander’s (just without all this “trope” jargon), over and over again, documenting Feser’s delusional inability to even recognize this fact (in Feser’s Five Proofs of the Existence of God: Debunked! which led to Feser Can’t Read (and Other Astonishing Facts) which led to Feser Still Can’t Read which finally left us in no doubt of this point).

The Logical Entailment of God’s Properties

This is as far as we need go to explain how “their erudition has driven them mad,” so to speak. This is why the intelligent and well-educated are just as susceptible to delusions, because they are smart and informed enough to build the elaborate architecture required to defend them against reality. And that’s what’s happened here. Howard and Alander have constructed such a sophisticated and convoluted analytical edifice that they don’t have to look at it too closely, and thus don’t realize that, at three or four steps of reasoning in, it doesn’t work. And their delusion will discourage them from ever doing that examination, or even seeing what I just pointed out if they do. That’s why delusion is a psychological trap, a hall of mirrors that can be very hard to escape, as I explained last time.

We thus don’t even need to point out other flaws in their reasoning, because those don’t strike at the core of their edifice like this, but are merely peripheral. For example, Howard and Alander think omnipotence logically entails omnibenevolence (as has been claimed by Swinburne and others they cite). But it couldn’t. Merely “knowing everything” is compatible with any moral architecture, or even none at all (complete indifference). The all-knowing will, of course, know what is moral (by whatever criteria), but they won’t, by simple virtue of knowing that, care. You need some extra sauce, some motive or reason to care about what they know, such that they not only know how to reason rationally, but want to reason rationally, and not only know moral truths, but want to embody them. You also need consciousness—since mere knowledge is just a database; to actually draw conclusions from it, and do things, requires a whole other architecture. And so on (see The Objective Value Cascade). None of this just automatically falls out of any property of God.

For example, mere omnipotence is mindless. Being non-selective, it can’t “select” consciousness or moral character or rationality or any set of desires over against any others (or even at all), so as to instantiate them. If all matter and energy are ripples in spacetime, spacetime is omnipotent: it can do every possible thing. But that does not mean it will; because it has no knowledge, no consciousness, and no desires, and thus won’t of itself ever choose anything to instantiate. It will simply collapse into whatever random chaos inevitably results, and proceed mindlessly, evolving on inevitable physical principles alone (this is The Problem with Nothing: uniformities are inevitable).

So omnipotence does not entail a mind (consciousness, rationality, specific desires). It could get you omniscience (if we define that as having all logically possible knowledge), but only when you add in all that other gear. But how do you get that without it being instantiated in a material? And then, won’t the limitations of that material stymie any acquisition of omnipotence (much less omniscience)? Think of what it would take to build a physical computer that could achieve omnipotence first so as to achieve omniscience. All the practical barriers making that impossible will impose on your mind. Then try working out how you can have any power without a body to implement it, much less all powers, and you’ll be in the weeds. This is Why I Think Theology Is Ridiculous. “But wouldn’t a disembodied hammer drive nails better?” Um. No.

But all that is moot. Because we can’t get omnipotence in their sense anyway. As just explained, omnipotence by itself is impotent to choose and thus do anything. Likewise omniscience. Spacetime “knows” everything in the sense that everything that exists or happens does to its direct eyewitness, and everything that could be, can only be through some other arrangement of it; but as it has no mental properties of itself, it doesn’t know anything in the sense required by Howard and Alander. It has no mind. It can only instantiate minds (and thus only finitely and accidentally). So when Howard and Alander talk about “omniscience” they are dragging in other properties, such as consciousness (an ability to process thought): they mean an entity that does not merely possess all knowledge (a database) but that can think about it (a mind). Merely having the power to create a mind (like spacetime can on string theory) does not entail that will happen. Nor are minds, even when instantiated, simple; they are extraordinarily complex. They are highly specified and complex information processors. Nor can they be realized without being instantiated in an object—just as imaginary cars cannot drive you to work, imaginary minds can’t actually think. (For more on all these points, see Evan Fales’ study Divine Intervention: Metaphysical and Epistemological Puzzles).

So without a pre-existing mind to choose that outcome (the realization of a complex thinking mind-structure) from among all other possible outcomes, it can only arise accidentally, which would not likely produce a god (see The God Impossible). Likewise, though spacetime could have the power to create a brain (by realizing all the matter that assembles one), it would have to first instantiate the objects needed to physically do that (it can’t just “do” it; which is the problem with all Boltzmann entities).

So when Howard and Alander talk about “omnipotence” they really mean all possible powers along with a complex thinking machine that just exists, pre-assembled, for no reason, and just “has access” to all those powers, for some reason, and with a strangely select set of desires and moral concerns, and all without any objects to maintain any of this structure and abilities, or even cause them to form. There is no way to get that in any simple way (trope theory or not). And all background knowledge argues against it (where we see that such specified complexity only gets realized through the assembly of a complex physical machine).

The things being left out here are the missing premises that, when brought back in, get the opposite conclusion to theirs: God is not simple, and not a plausible entity by any appropriate metric. Trope theory can’t fix this. But because it is elaborate and complicated, it can be used to trick you into thinking it can. And that’s what we are watching unfold here. Indeed the host had to apologize for its hard-to-follow complexity in minute 30. That’s a feature not a bug: if you struggle to “get it” at the surface presentational level, you may lack the cognitive resources in time and effort to get beneath it so as to discover that all the hidden premises it is depending on aren’t credible, and so neither is the result.

Remembering That Theology Is Ridiculous

I won’t fisk every argument Howard and Alander make. Now that you have seen what goes wrong when they try to reframe classical theistic arguments in trope theory (and even by reformulating theism as a competing explanatory hypothesis analogous to science, which I’ve always championed doing), you can apply the same principles to see why this doesn’t help them in any way. In every single case, they are just saying exactly the same things theists have always said, just in a fancy new vocabulary. So all the refutations that took it down classically, will take it down in any new form like this. You just have to effect the right translations to convert a classical refutation into its corresponding form in trope theory (or scientific methodology, or any other new way they try framing the question). Notably, they never do this. And that’s why this approach keeps them trapped.

For example, their defense of plenitude theory (in minute 40) remains as self-refuting as ever. Quoting Sijuwade, they claim:

God brought brings things into existence in order to communicate his goodness to creatures, and to represent his goodness through them. And since his goodness cannot be adequately represented by any one creature, he produced many diverse creatures, so that what was lacking in one’s representation of the Divine goodness may be supplied by another. For the goodness that exists in a simple and unified way in God exists in multiple and divided ways among creatures. Hence the universe as a whole participates in and it represents God’s goodness in a more perfect way than any one creature does.

This straightforwardly fails as an explanatory hypothesis, because it entails God would realize all possible goods. Yet most logically possible goods aren’t realized. Unicorns that heal the sick by a touch don’t exist. So God failed to instantiate one aspect of his goodness there. No justice system on Earth is wholly reliable. So God failed to instantiate one aspect of his goodness there. And so on. So by this theory’s own construct, God’s creation remains massively imperfect. Therefore this theory of God is a poor explanation of it. And that’s before even getting to all the instantiated evils, which directly contradict this theory (but I covered that already last time). This theory also lacks any useful explanatory specificity, a necessary requirement of any viable theory. For example, you cannot explain by this why Daeodons and Giant Otters no longer exist, but once did, or why tigers specifically exist and not Soot Sprites or Ghibli Fox Squirrels or, indeed, magical unicorns or intelligent jellyfish. But modern scientific naturalism does explain this. I already documented these problems for Thomas Ward’s version of this theory.

The missing step of reasoning here is to try to disprove it by asking what shouldn’t we observe if it were true (that’s The Scary Truth about Critical Thinking), and then looking to see if we see any of that. Which then brings in all the missing data that, once brought back in, reverses our conclusion: their theory has been massively falsified; and lacked any explanatory virtue to begin with. It’s done for. It cannot compete with scientific explanations; and scientific explanations involve no deities.

And that’s all before we even get to the failure of its entailment to begin with. They argue that God being good entails he would do this, but there is a missing premise there. They are irrationally constructing another enthymeme that diverts attention from that missing premise, which hasn’t been justified: that a good person would even need or want to do this. God’s goodness could be entirely realized in himself, requiring no creation at all—in fact, by creating, God is diminishing his goodness, by realizing imperfect copies of it. To get around this, you have to invent some other arbitrary, ad hoc epicycle of an excuse for why God would ignore this outcome measure and adopt this other convoluted outcome measure instead—that he needs to instantiate his various aspects in numerous imperfect ways, yet (for some reason) not all ways, nor (for some reason) all aspects, and (for some reason) littered with evils contrary to his nature as well.

There is simply no logical entailment here. This is quite literally what philosophers usually call bullshit. Yet these guys are seduced by it, believing it profound. It isn’t. It’s dumb. But their clever and erudite construction of it hides this fact from their cognition, preserving their delusion. (Meanwhile, for what a properly-thought-out theory of plenitude as the goal of a good person would predict, see How Not to Live in Zardoz and Ten Ways the World Would Be Different If God Existed.)

They do this again in their defense of what they call the Ultimate Goodness Principle (around minute 43), whereby the hypothesis of God’s goodness remains sound as long as no matter how horribly he fucks everything up, it ends up well. But this is obviously false. Again, look for contrary evidence, which is what they are leaving out: if Donald Trump gassed all of America’s prisons, the crime rate nationwide would measurably decline (as recidivist crime is eliminated by eliminating all captured criminals). Would that make Donald Trump a good person? Not by any metric. It doesn’t even survive utilitarian analysis (the evils produced by such a scale of innocent deaths, as well as the continued threat of it, are measurably enormous—there is a reason our hard-fought Constitution banned such machinery). But it certainly doesn’t survive deontological analysis. The very act of murdering masses of people itself is an evil that cannot be outweighed by any subsequent good precisely because the same effect could be brought about in less evil ways. And that’s true even within human limitations (there is a reason other countries have massively lower violent crime rates than ours, and it’s not because they are gassing their prisons); but certainly without them.

Because, after all…

A god would know who remains a danger, and thus could single out specific threats or monitor them and warn people when a threat approaches (just like a computer does in Person of Interest and, more convolutedly, in Travelers, only God would do it massively better by every metric). God could also teach and thus reform them more successfully. And so on. So God has no need of messily killing millions of innocents who did not, or would not in future, have committed any serious crime, just to lower the crime rate. The very act of choosing that more evil means to the end is itself evil. So in no way would it be at all likely on the hypothesis that God was good.

This is why you cannot defend God’s goodness with “the end justifies the means.” It does not, because good people by definition do not choose evil means to any ends. That’s what makes them good people. And God would be subject to this same principle, since he is a person, just like everyone else (qua person). Yet Howard and Alander have seduced themselves by bullshit, because they aren’t falsification-testing anything. They are just believing anything that sounds good and assuages their concerns. They are not employing any means of testing whether it is true—or even logical. Which is irrational. And this selective irrationality certainly appears to be in service of a delusion.

There is a lot of other sophisticated nonsense they push besides these arguments, but these are enough to illustrate the failure mode plaguing all of them. So, for example, their convoluted Argument from Axiological Relevance (in minute 47) will fall to the same analysis applied above, as soon as you actually test it. They aren’t testing anything. So they are seduced by everything. But you can test every argument they make, by following the same approach I just illustrated. And you will then discover the implausible missing premises and falsifying evidence they are leaving out. And then you’ll see that, when you put all that back in, the conclusion reverses, and it’s atheism that becomes far more probable than their theism (just like I showed for ten standard arguments for God in Bayesian Counter-Apologetics and Ten Ways the World Would Be Different If God Existed, and formalized under peer review as an entire methodology).

Test Case: Morals, Truth, and Beauty

For example, in the model Howard and Alander present in minute 53, they fail to test the likelihoods of there being “morals, truth, and beauty” on naturalism, and they fail to include the data of them that is unexpected on theism. Which is what Christian apologetics always does. Framing it in a Bayesian format does not make this error go away. But it does make clear where the error is. You just have to know what you are supposed to be looking for.

So, for example, what is the probability that there would be morals and beauty on just evolutionary theory alone? Well, near 100%. So their existence can never argue for theism. By contrast, what is the probability of the actual observed data of morals and beauty on their theism? Well, near 0%. The evidence they leave out, put back in, reverses their conclusion. To see what I mean, look at my surveys of the actual data we have on morality (fits atheism; does not fit theism) and beauty (same result). Indeed, naturalism has the very explanatory specificity here that theism lacks (it explains why the specific observed kinds of beauty exist and not others, likewise the specific observed kinds of moralities and their historical development).

So their framework (to test theism by a Bayesian analysis) is correct; but their application is not, because of GIGO: their delusional brains have tricked them into leaving out all the data that would refute them, and so their inputs to the Bayesian formula are bogus, producing bogus results. But because it looks fancy and sophisticated, they seduce themselves into believing they haven’t done this. This is why Bayesianism is useless by itself. You also have to be a sound and reliable critical thinker. Which means you need to be consistently motivated to look for evidence unexpected on your theory, and for evidence more expected on alternative theories to yours. You have to actually try to prove yourself wrong first. Bayesian reasoning will help you with that. But it can never simply replace that. If you skip that step, then Bayesian reasoning, like all logics, will never help you. All logics will simply become another tool of irrationality and delusion.

All the same can be said of their attempt to also squeeze in “truth” here, in the form of mathematical truths (oops, already entailed by atheism), logical truths (oops, already entailed by atheism), and scientific truths (oops, already entailed by atheism). And what’s missing is not only this (their checking to see if these are even unexpected without theism), but the specificity atheism provides as an explanation that theism lacks. For example, if theism were true and entailed what Howard and Alander claim here, then human reason shouldn’t work the way it does (whereas that is exactly what atheism predicts, while theism predicts something else). Likewise, if theism were true, we could have things atheism would not produce, like functioning oracles, truth spells, instructive ghosts and angels, or trials by ordeal that actually worked. Why, heck, God could actually talk to us, and tell us things, teach us things, answer questions.

This doesn’t mean we would have these things; the point is, theism is less specific as to what it predicts, whereas atheism is much more specific, and lo, its specific predictions are the ones that bear out. Whereas if theism were true, we should certainly have “a successful and moral engineer to produce a successful and moral world design,” like “a universe directly governed by justice-laws, or a kind and just stewardship.” For example we should sooner expect laws of physics like “the good become healthier and more invulnerable than the bad, in proportion to their genuine (and not merely feigned) goodness.” Even if not exactly that, then at least something. Instead we have a physics that is completely values-indifferent, exactly as we expect on atheism. Justice should be inherent in the design of any just engineer. It is certainly quite improbable that it wouldn’t be; whereas that weird observation is guaranteed if there is no such engineer.

So their predictive model doesn’t work. And we learn this as soon as we put data in that they are leaving out—such as from the actual science of “aesthetic impulses” and the actual ontology of math and logic and the actual history of human truth-seeking and all the things a good engineer could actually do. When we ask “What would a better engineer do differently?” we do not get their results. When we ask “Why do these things exist, and in the particular ways they actually do exist?” we do not get their results. Everything falls apart when you check. So their delusion motivates them to never check, or even refuse to acknowledge anyone else’s results when they check. For example, they argue from the existence of saints; yet seem strangely unaware that those are myths. Hence sometimes even their data is delusional. That’s another example of leaving evidence out that, when put back in, disproves their thesis rather than confirms it: the thing they said their theory entailed we should observe here, doesn’t exist.

Have there been exceptional people in history? Certainly. This is an inevitable outcome of bell curve statistics. It therefore is more indicative of the absence of intelligent design than its presence. There simply aren’t as many saints as a divine world should have, and none are actually cosmically favored in any way. The only intelligent design we can find here is human: our societies encourage and (sometimes) reward exceptional moral achievement, thus intentionally generating positive pressure that will steer some people into that niche (and never to perfection). But God has no evident role to play in it. Those countless fake stories about the saints, which exaggerate or invent their deeds and downplay or erase their flaws, and which have God doing all sorts of things in aid and reward of them (like making them immune to fire when alive or decay when dead), have to be faked for a reason: there is no evidence God ever does anything for these people, nor is anyone really as amazing as their myths make out. Their only documented incentives or rewards are entirely human; their achievements, likewise. This does not bode well for Howard and Alander’s thesis.

They go on to make many similarly uninformed arguments that flip against them when the evidence they leave out goes back in. And they are aware this is a risk. In minute 1:04 they admit this might happen, but they never explore the possibility (not even in Q&A). For example, contrary to what they had said a few minutes before, the messy fine tuning and vastness of the universe is not aesthetically predicted but actually contradicted by theism (Aristotle’s cosmology is far more beautiful and would actually confirm theism; our actual cosmology is a chaotic nightmare and more what we expect on atheism). And their claim that it “just makes sense” that God would make the universe evolve through stages is void of any logical argument, and trounced by far better-evinced reasons that atheism already entails it. And so on. In result, their math at the turn of their first hour doesn’t work: their equation is correct; their inputs are not. And that’s the end of that.

Their Five Objections

Howard and Alander then survey what they call “the five objections that we thought, we found, to be the most powerful objections” to everything they just said. But what we get here are only straw men, which their minds built for them so that they never have to confront any effective objections.

For example, the omnipotence objection (starting at minute 1:06) is that because God has infinite options, the probability for any one option he chooses is infinitesimal, therefore you can never get high likelihoods out of this. This is inapt. As they explain, what matters are relative probabilities and fields of probability. Even atheism has infinite options, e.g. we could have evolved from jellyfish or fungus or silicone slimes etc.; the number of possible universes, from physical constants to specific developmental histories, that could produce life is also infinite; etc. So what we want to know is the probability of the sum of these options that are each more likely on one theory or another.

A more cogent objection would be to note that, within a wide field of alternatives (not specific ones), God does not need to do it in any of those ways whereas that is the only way atheism could. For example, God does not need brains for us to have minds, and he has built-in motives to specifically not want to use brains to produce minds; whereas on atheism, that’s the only way we could exist. This spans whole regions of options (it holds for all possible species we could have been, on either theory) and it maintains a relative push in favor of one theory over another (God always has incentives to avoid brains while atheism never can). This is because God has options atheism doesn’t. But this isn’t some pedantry about infinities. But they never address the real objection here, only the ridiculous one.

Conversely, the axiology objection (starting at minute 1:12)—that their theory requires arbitrarily specifying God’s values, and this reduces theism’s prior probability (since atheism requires no such tuning of its fundamentals)—is correct, but they never address the actual objection. Instead, they argue, as with the omnipotence objection, that it is not necessary to specify precisely what God’s values and desires would be, since a global range of possibilities is constrained enough to be predictive. In short, their theory need only predict that there will be “morals, truth, and beauty” in the senses they mean, not specifically what those morals or beauties or truths will be.

That’s also correct, but it’s beside the point, because atheism makes exactly the same predictions, and is in fact even more successfully precise in its predictions. Theism is thus explanatorily deficient here, and still at the cost of arbitrary tuning of the hypothesis (God need not care about morals or beauty at all; and he is unlikely to want morals and beauties that look exactly like there is no God producing them, so these problems have to be “tuned out” of the hypothesis). For example, we can predict to near 100% certainty why sexual lust exists on atheism (as opposed to just love, for example); but this is quite unexpected (indeed weird) on any kind of theism.

But more importantly, as I mentioned before, theism needs a ton of ad hoc tuning, because part of this intended objection’s point is that amoral gods are simpler (and mindless powers even simpler still). But to get their predictive framework, their God has to be possessed of a precise and complex consciousness somehow (and prior to exercising any omnipotence to choose it); he has to care about morals and beauty for some reason (and not just know what they are); he has to care about being rational for some reason (and not just know what that is); and he has to have the ability to do and know things without any mechanism for producing effects or distinguishing and storing knowledge, despite this being entirely contrary to our entire vast array of background knowledge. And, on top of all that, God has to have some weirdly complex array of strange desires so as to explain why he does everything to look exactly like it would have to look if there is no God, rather than like an actual good engineer would. And they don’t ever address these actual axiological objections to their theory. They dismiss objections only to a straw man of their theory (that they need to specify exact desires for God, rather than ranges or fields of possible desires within a superset of “personal goodness”), but not the actual objection an astute critic would level at them (as I’ve done here).

The third objection (beginning at the end of minute 1:16) they address is even weaker tea: the idea that their axiology (the values system they are ascribing to the universe as true) is arbitrary. Just as they explain, this makes no sense as an objection, since even the atheist grants there is (or certainly ought to be!) an observable difference between an engineer acting for motives (regardless of what they are) and how things end up without any intentional causes being involved. This is the problem I just noted: it requires bizarre fine tuning of their hypothesis to get God to “want” to make everything look exactly like it would have to look if there is no God. The reason this must cost them, either at priors (from gerrymandering their hypothesis) or likelihoods (as it is a very unlikely outcome of their hypothesis without that gerrymandering), is precisely that their theory entails some observable axiological effects on the world.

So they are not wrong about that. What they are wrong about is that observations actually match their expectation—by leaving out the actual data regarding human morality and aesthetics and natural reasoning, for example, they fooled themselves into thinking God makes these things more probable; but in fact God’s axiological intentionality makes those actual facts very improbable. And that follows regardless of God’s axiology—unless you are finely tuning it to conveniently get exactly that outcome, which is a gerrymander that tanks your probabilities either way. Even an evil or indifferent God would leave more evident effects in the world than we find. But, of course, they are arbitrarily rejecting evil and indifferent gods, and that does lower their priors (since these options must split the prior probability space, leaving them only a fraction of it). It also lowers their likelihoods (because observations don’t fit their selected sub-hypothesis, the subject I already covered last time). But they never address this, the actually potent axiological objection.

Their fourth objection (beginning in minute 1:20) is what they call the Naturalistic Parody objection, although I don’t understand that—the real objections they mean, as from Malpass and Oppy, are not really parodies (they are logical possibilities occupying some of the prior probability space, and formally are arguments a fortiori), while in minute 1:22 Howard and Alander instead say they “parrot” the theistic model, but that’s parroting, not a parody (while in a slide they show in minute 1:26, there appears the word “parrit,” which … I don’t know). In any event, they never correctly describe the actual objections here.

For example, they mention the “Stalking Horse” objection of Alex Malpass, but that is more like the previous objections, whereby he says naturalists can start with any brute facts they want, as long as their theory remains simpler than theism, and thus gain the same predictive power with no relevant cost to prior probability—because those naturalisms may have lower priors than other naturalisms, but they will still have higher priors than theism. For example, just adding a mindless “goodness force” to naturalism would explain all the things Howard and Alander mistakenly think we’ve observed. That is simpler than theism, because it is the same thing as theism, minus a bunch of arbitrary add-ons, like consciousness, omnipotence, disembodied omniscience, and a whole complex set of desires and motives—as compared to only a single mindless motivating tendency (such as we find in philosophical Taoism). I’m not very impressed with this argument (none of these naturalisms are probable), but Malpass isn’t wrong (those naturalisms are more probable than theism).

Howard and Alander never respond to this, Malpass’s actual point. Instead, Alander conflates Malpass’s point with a claim to gerrymandering naturalism, using (in minute 1:27) an example of an extreme gerrymander that could be used to explain any concrete evidence of theism, on which Alander’s position is entirely correct; it just doesn’t have anything to do with Malpass’s argument. Malpass is taking an existing component of theism and porting it into naturalism without all the other baggage of theism, which means it simply borrows the same gerrymander from one side of the equation into the other (producing no net difference in priors for that one gerrymander), without moving the remaining gerrymanders (which still plague theism).

Howard and Alander want to say that “their” goodness force is entailed by their theism, and thus not a gerrymander, because goodness is entailed by omnipotence, via a resulting omniscience, but this confuses knowledge with concern (Satan can well know what is good and still not care) and confuses root with derived complexity (even if goodness is entailed by omnipotence, merely having the gerrymander of goodness remains less complex than the far more extraordinary specified complexity of having the whole shebang of a convenient omnipotence that is then required to “entail” goodness—much less also of the consciousness required to use that omnipotence to know things and act on that knowledge, and without any body to instantiate either, which are all fantastical contraptions not needed by Malpass’s stalking-horse naturalism).

For my own take on similar approaches, see my original article on Defining the Supernatural, which is rife with fun examples, and will better center you on the actual metaphysical difference between naturalism and supernaturalism—to the point in fact that Malpass’s “stalking horse” naturalisms should really be described as atheistic supernaturalisms. Which reminds us that Howard and Alander’s entire presentation functionally breaks the Law of Excluded Middle, by falsely dichotomizing the options as between naturalism and theism, when in fact a lot of atheistic supernaturalisms outperform theism explanatorily and on priors (a possibility, which they sort of umbrella under the term Axiarchalism, that they just dismiss out of hand early on, somewhere around minute 5). I cover this point in Christianity Is a Conspiracy Theory, where I show Christianity is more implausible than many already very implausible things, including various nontheistic supernaturalisms. Which is another example of omitted evidence that, when you put it back in, hoses Howard and Alander’s conclusion.

The Objection of Doom

Which all lands us at their fifth objection (in minute 1:32), which has pretty much been mine this whole time: the problem of Understated Evidence (i.e. leaving things out so as to juice the likelihoods on theism and conceal the actual likelihoods on naturalism). This is the only one that really matters to my mind. The other four, when steel-manned, end up really just additional versions of this argument, or else they make some point about the ad hoc tuning of their hypothesis, which tanks its prior probability—like just sneaking in the brute facts that God “is” good or cares what is good, that God “is” rational or cares about being rational, that God enjoys the entire apparatus of consciousness, that God can do and remember and think things without any thing to do and remember and think them with, that God has all these strange desires that motivate him to make the world just like it would look if he didn’t exist, etc. But even that is the same argument: reintroducing premises that were hidden in the original hypothesis, and then playing out their mathematical effects, which ends up badly for theism (in contrast to naturalism, which doesn’t need anything comparable).

This time, at least, Howard and Alander finally do correctly describe the objection and why it is effective. But their response (beginning in minute 1:33) is illogical. First, they claim that if they can come up with any story that would explain why God “did it that way,” that this will cancel the objection. But this is the same fallacy I called out last time: possibiliter ergo probabiliter, “possibly, therefore probably.” It is not the case that a logically possible explanation is always a probable one. So you cannot remove an argument to a probability with an argument to a mere possibility. This is fundamental to modal logic. Probable defeats possible every time. So it won’t work to come up with some “just so” story that gerrymanders your God back into being probable again whenever disconfirming evidence is introduced. You need to prove that that weird explanation is probable. Which they never do.

But it’s even worse than that. Since the facts are always near 100% certain on atheistic naturalism, it won’t even do to get the theistic gerrymander to be probable—you would have to show it is near certain. Otherwise atheism is still more probable. For example, if “minds require brains” is 100% expected on naturalism, valiantly getting it to be 80% likely on theism gains little. The fact (“minds require brains”) remains more likely on naturalism and thus remains evidence against theism. Worse, you really can’t ever get it to be evidence for theism. Because there is no probability above 100%, and even a fractional difference (say, 99.999% vs. 99.9999%) will cut very little cloth. In short, the best Howard and Alander can hope for here is to get the evidence to be “equally likely” (“near 100%”) on both theism and naturalism, which simply eliminates it altogether as evidence for either theory. Not a very promising horizon for them (especially if their gerrymandering to get that result tanks their prior).

Why are they so doomed? Because the world simply is not as their theory predicts, but is as atheism predicts. The facts just are that way. They could have been different. But they aren’t. So the only thing they can even logically do here is gerrymander their God one last time into it being “likely” that he’d make the entire universe and its history look just like it would with no God, so that then you could at least say it’s “50/50 or so” whether theism or naturalism is true. Not a great outcome for them, but it would still be a far sight better than where they are. And yet even that landing is not in their cards: for what they’d then be defending is a Cartesian Demon, which is always the least likely explanation of anything like this, as I demonstrate in We Are Probably Not in a Simulation—which exemplifies again that, if we did live in an intelligently designed world, it would be far more obvious to us.

Their second attempt at a response is to make the analytically correct point that they can turn things around if they can nevertheless show there is enough stuff that atheism doesn’t make probable—but then they fail to ever give a successful example of that.

I’ll just illustrate this with Howard and Alander’s first example: moral evil (all the others fall to exactly the same analysis). There is a lot more wrong with morality as evidence for God than the mere existence of moral evil; so already Howard has falsely framed the problem so as to exclude all the evidence he is supposed to be explaining—yet again. But let’s play that game here for a bit. Howard argues that God could have made us morally perfect (and then spins a fable as to why he wouldn’t). That is a straw man; no such hyperbolic measure as “making us perfect” is required here, but you can steel man his point by inserting anything less ambitious, like brains that always reason reliably, which would produce a lot less moral evil in the world, without moral perfection. Either way, moral evil is less expected on theism than naturalism, so “morality” is then evidence for atheism and not theism—when you put all the facts back in (we haven’t, but you get the point). Howard proposes that if we can show there is still a lot more moral good than atheism can make likely, then the total field of facts would flip back for theism—the observed facts would not be 100% expected on theism, but would still be at least slightly more expected on theism than atheism, and thus become evidence for theism again, even with moral evils acknowledged.

Howard is entirely correct. It’s just…that never happens. There is no “weird excess” of moral goodness that naturalism cannot explain. Given The Real Basis of a Moral World, morality is both natural and rational. So to find a lot of it in a society of rational animals is as near to 100% expected as makes all odds. Likewise every other argument. The evidence Howard says theists need to claw their theory back does not exist. Anywhere. So really, the “strategies” for rescuing theism that Alander surveys (in minute 1:41) are just apologetics, tactics for reestablishing your delusion against contact with reality.

To wit…

You need to make up evidence that doesn’t exist or exaggerate the evidence that does (his #1; this is more correctly described as doctoring the inputs). Or you need to make up a convenient story that gets God likely again to act that way (his #2; this is more correctly described as gerrymandering, because it gains likelihood by hiding what should be a corresponding loss in prior probability, owing to the ad hoc fine tuning involved). Or you need to make up a story that gets the facts to actually be what God as defined already wants (his #3; which is simply combining #1 and #2 into a combo move). Or you need to make up a story that gets the facts to be unexpected again on naturalism (his #4, which is a converse #3, and amounts to exaggerating or misstating that alternative hypothesis or what it predicts).

You’ll find every one of these strategies in last week’s article. This is a standard apologetics toolkit. And every one of these tools is designed to immunize a delusion from contact with the facts. Alander even slips and actually says this in minute 1:42: that their framework “is immune from certain objections” (not “could be” or “might be,” but “is”). This is of course false. Though his four strategies can be legitimate approaches, they are only legitimate when based in accurate accounts of facts and competing hypotheses. If there is no accurate way to analyze those facts and hypotheses that actually gets to any of his solutions to those objections, his theory isn’t immune to them. And they never get there. So their theory is not immune to their listed objections—as any rational observer can tell.

Conclusion

At no point in these two hours do Howard or Alander present a rational argument for theism. They construct a really sophisticated appearance of one. But all they have done is dress up old arguments that have already been refuted, and try to rehabilitate them with the cloak of this new framework, and then pretend they have thereby answered those refutations. But they haven’t. There is no evidence that is more likely on theism than naturalism. There is a lot of evidence that is more likely on naturalism than on theism. And naturalism requires far fewer starting assumptions than theism. Naturalist cosmology and ontology also do not require any bizarre entities with massive specified complexity; while their theism does.

Their explication of Bayesian reasoning is okay. And their deployments of it are logically valid. They’re just never sound, because all they do is rig the inputs, leaving data out, or making data up, or misstating what each hypothesis makes likely.

This is illustrated in Q&A when they are confronted with a free will paradox at minute 1:45. Howard goes on about neurological reasons people do stupid or bad things even when they are smart or good, and he really seems to think he has relevantly addressed the question. But he never does. Why are we so badly designed as he describes? This is what his theory cannot explain. It is not even remotely likely on their own account of theism. But it is near 100% guaranteed to have been the case on atheism. To avoid confronting this fact, he makes up a “just so” story about how we need to be badly designed so as to be more praised for overcoming that. He gives no logical argument for this even making sense. (What does a quantity of praise have to do with the quality of one’s person? Are we literally saying we need murderers so their victims can feel better about themselves?). Nor does he give any evidence that this is at all the case. (Are rational people worse than irrational people? Is it actually better to struggle with folly than to live with wisdom? Would a world where we all reasoned reliably be worse than the one we are in?)

So Howard essentially ignored the fact challenging his theism, convinced himself something else much easier to answer was said instead, and responded to that with a made-up story he had no actual knowledge was true (or even good reason to believe), and thereby convinced himself his theism is rational—when everything he just did to defend it was entirely irrational.

Cameron Bertuzzi, the show’s host, reveals this again when he essentially tries to argue (in minute 1:51) that it is unexpected on naturalism that he’d see the evidence almost balanced between theism and atheism. Which amounts to saying it is unexpected on naturalism that people would find a lot of evidence for false beliefs convincing. It is, obviously, the other way around: that’s unexpected on theism. God would build us better, and there wouldn’t be climate deniers, holocaust deniers, anti-vaxxers, flat earthers, QAnons, alien lizard conspiracists, young earth creationists, Islamic fundamentalists, Christian nationalists, Biblical inerrantists, ideological racists and misogynists, Raelians, Heavens Gaters, Mormons, Scientologists, Hindus, literal witch hunters and exorcists, and, wow, the list is long. They all think the evidence for their false beliefs is balanced or convincing, too. So in order to defend himself from realizing he’s deluded, Bertuzzi has literally convinced himself delusions don’t exist or that he cannot possibly be trapped in one—and is now using this ancillary delusion as an argument for his primary delusion!

Of course, this isn’t internally coherent, either. The evidence for theism would not be “almost balanced” with atheism if it were true. It would be far more obvious theism is true. Anything else is a gerrymander. But it is remarkable to see a delusional person use the impossibility of their being deluded as evidence supporting their delusion.

Alas, that is what delusion does to you. But I explained all that last time.

-:-

[1] Refreshingly Howard and Alander are Bayesians (as everyone should be, theists and atheists alike), and they are not fraudulent Bayesians (contrast their presentation with Crank Bayesians: Swinburne & Unwin and Crank Bayesianism: William Lane Craig Edition). They are not gaming the premises, but succumbing to convoluted error.

[2] For my own construction of Bayesian epistemology, which is less convoluted and muddled than theirs but ultimately similar, see my book Proving History and my article “Bayesian Reasoning’s Power to Challenge Religion and Empirically Justify Atheism.” See also the category “Bayes’ Theorem” in my right-hand drop-down menu here.