For Theism: Part 1

In which I, among other things, address everything anyone has ever said about psychophysical harmony

0 Introduction

Long lay the world in sin and error pining
'Til He appears and the soul felt its worth
A thrill of hope the weary world rejoices
For yonder breaks a new and glorious morn
Fall on your knees; O hear the Angel voices!

—Oh holy night.

I’ve been an atheist for pretty much all of my life. But here, I thought it would be worth making as compelling a case as I could for theism. There is quite a lot of evidence for theism, and while I don’t think it’s enough to swamp considerations about evil, it is pretty considerable. I think there is enough evidence for theism that if you don’t find the problem of evil persuasive, you should probably be a theist. Here, I’ll begin a series that involves me making the most persuasive case I can for theism.

I’m writing this for another reason, namely that I’d like to believe theism if it’s true. But for most of my life, when I’ve argued about theism, it’s been to argue against theism. When one argues for a position, they become more confident in that position and only see the things favoring it. So, to offset that bias, here’s my attempt to make a maximally compelling case for theism.

1 Psychophysical harmony

Harmony and me
We're pretty good company
Looking for an island
In our boat upon the sea

—Elton John in the song Harmony

The argument from psychophysical harmony from Crummett and Cutter is the best argument you’ve never heard of. So few people have heard of it that in this section I will address every objection anyone has ever made to it, that I know of. Most people who do hear of it misunderstand it and present demonstrably mistaken objections to it. If you don’t want to hear the super-in-the-weeds review of all the objections, you can skip this section after the basic version is presented.

The phenomenon of psychophysical harmony denotes the harmonious pairing between our physical states and our mental states. This takes place in a few ways:

1. When a person has a conscious desire to move their limbs in some way, they do move in that way. When a person wants to speak or act in some way, they are able to manipulate their limbs to do that. Thus, our desires manifest themselves in bodily movements that even enable us to talk in detail about our consciousness

2. Our brain matches reality roughly as it is. I see a computer in front of me because there actually is a computer in front of me.

3. We respond in rationally appropriate ways to stimuli. When we feel pain, we act rationally in response and try to avoid the pain.

4. We can report accurately the content of our mental states—I can say, for example, that there is a computer in my visual field when there is, in fact, such a computer.

But both of these depend on the mapping between the mental and the physical! If the mapping were different, we could have a psychophysically disharmonious world, where our mental states were radically out of accordance with our physical states. For instance, suppose that the pairings were such that when C-fibers fired—which produces pain in the actual world—instead pleasure was produced, but this pleasure caused the same physical effects as pain does. In this case, we would act like we were in pain when we really experienced pleasure.

The basic argument from psychophysical harmony is relatively simple: the world could be radically disharmonious so that there was no harmony between the mind and the body. There could be, for example, a world physically identical to our own that contains only the experience of eating hummus for all sentient beings. Or another world that’s physically identical but contains opposite mental states—when we feel pleasure, they feel pain. Given that each of these is possible, and many more scenarios than this, the odds that the mental would line up with the physical are vanishingly low, especially because each of these psychophysical laws are strictly simpler than our own.

It’s very plausible that a world that is the same in physical respects but differs in phenomenal respects is at least conceivable—there is no lurking internal contradiction in the description of such a scenario the way there is in the description of a square circle. It seems intuitively like there could be a world like ours physically that lacks consciousness altogether or has very different consciousness. But if that’s true then the psychophysically disharmonious worlds swamp in probability the harmonious worlds—most ways the world could be entail radical disharmony. And many of those worlds are simpler—a world where every experience in the universe is just of eating a sandwich is simpler than our complex psychophysical laws that enable a rich phenomenology.

I hope you’ve gotten a sense of the problem, but if you have not, this illustration should help. Suppose we’re epiphenomenalists and we think that consciousness causes nothing. For the epiphenomenalist intuitively there is a problem of harmony—if consciousness causes nothing then the internal states of a person could be anything and it would make no difference. It’s thus miraculous that our internal states line up with our bodily movements and are harmonious more broadly.

Lots of non-epiphenomenalists seem to appreciate that this is a problem for an epiphenomenalist. But the problem is just as acute for interactionist dualists who say that consciousness is causally efficacious. The interactionist says that there’s some brain state that causes a mental state which causes some third physical state. Call the physical brain state A, the mental state B, and the physical effect C. A causes B which causes C. But B could be replaced with D, E, F, G, or pretty much any other mental state and as long as state C is still caused, this would result in a disharmonious world. Thus, even for the interactionist, the mental state could be scrambled while keeping the physical states the same. So then it becomes miraculous that the mental state produced by A is B—which produces harmony—rather than one of the myriad other mental states that it could be.

Now you might think that you can avoid this just by being a physicalist. A physicalist thinks that there is a necessary connection between the mental states and the physical states. But being a physicalist doesn’t get you out of it unless you think that you can rule out disharmonious laws a priori. Which is only possible if you adopt the widely rejected, and in my view deeply implausible, type-A physicalism, which says that without any experimental inquiry, we could know that brain states like ours would result in mental states like ours. If you think we can’t know this a priori, then the fact that the necessary facts produce harmony is miraculous.

To see this, imagine that we discovered that the universe spelled out “God is great,” at the atomic layer. This would be good evidence for God. It wouldn’t solve it to say “I think the universe is necessary so it couldn’t have been otherwise.” As long as you can’t know that some connection holds a priori, the fact that a deeply surprising connection holds is miraculous. Crummett and Cutter expound on this in more detail as does my friend Amos Wollen in an excellent dialogue written for people who don’t really get the argument—I’m providing just the quick and dirty explanation.

The objection everyone thinks of when they hear this argument is “what about evolution?” But this objection is demonstrably mistaken. To see this, remember that the psychophysical laws are about which physical states produce which mental states. The challenge is explaining why the world is harmonious rather than disharmonious. But for every harmonious world, there are huge numbers of physically identical disharmonious ones—for every world physically like ours that’s possible, there are thousands that are identical physically but produce radically different conscious states. Remember, as long as you keep the physical states fixed, you can literally make the mental states any mental states that are possible, and you’d get a physically identical world. The challenge is explaining why we’re in this world rather than the physically identical world with different mental states. If we were in a physically identical world where every being in the universe had the experience of eating hummus all the time, that would produce creatures that evolve just as we do, and that are radically disharmonious. So the challenge is explaining why the psychophysical laws are such that they produce harmony, but the psychophysical laws can’t be changed by evolution any more than the law of gravity can. If you’re confused by this, I highly recommend Amos’s dialogue.

Many more objections have been raised to the argument from psychophysical harmony. Some are lousy, and others are just as bad. Various objections are compiled nicely here. Let’s address every one of these arguments. The first section claims that each of the following views is incompatible with the psychophysical harmony argument having force:

Type A Physicalism

Liberal Naturalism and Kripkean Conceivability (Davidson, Putnam and Kripke)

Eliminative Materialism

Error Theory and Moral Anti-Realism

Axiarchism and Directed Naturalism

You can read their explanation of these in more detail. But the first 3 require denying the manifest fact that zombies are conceivable, that inverted qualia are conceivable, and that Mary learns something when she sees red. On account of denying this, they are extremely implausible. Whether they’re more implausible than theism will depends on one’s credence in theism—but claiming that a blind neuroscientist learns nothing when she sees for the first time is a view I find incredibly absurd.

Error theory says that there are no moral facts. There are two problems for this: first, error theory is obviously false, second, unless you’re almost totally certain of error theory, accepting it does not help you. Error theory says the statement “it’s wrong to torture babies,” or “pain is worse than pleasure,” is false. But those statements are clearly true.

Second, theism entails that error theory is false. So if theism is true, there’s maybe a 50% chance that pain would pair with aversion. If harmony is valuable, we’d observe the harmony that we do. In contrast, atheism doesn’t predict the observed correlations and thus renders it a coincidence of unfathomable proportions. So if we start with a 1% credence in theism and a 50% credence in normative harmony conditional on theism, it will be that almost all epistemically possible harmonious worlds are theistic, because the odds of harmony conditional on atheism is virtually zero.

Finally, as defenders of the argument admit, this only explains normative harmony, not the other three kinds.

The next view that supposedly avoids the problem is called Axiarchism and says that the fact that states of affairs have value somehow explains why they exist. But axiarchism is not a good view, as I’ll argue in later sections. In addition, if the atheist is forced to axiarchism, which identified a minuscule slice of the starting probability under atheism, this argument has still accomplished a lot.

The next objection given comes from Sebastian Montesinos. Notably, this is just an argument against normative harmony—the claim that there’s a rationally appropriate pairing between our mental states and our behavioral states, such that we avoid pain as it’s rational to do so. This only addresses the third of the psychophysical harmony points I raise. He argues that whether we take introspection seriously as a way of learning things or not, the argument from psychophysical harmony is diffused. I think we should take it relatively seriously—we can know that pain is bad, for instance, from its phenomenal character. He argues that if this is so we should conclude that one can’t have the experience of pain without the causal things that go on in the brain when one experiences pain going on.

Problem: this is obviously false. Remember, if you think that pain differing from its functional profile is at least conceivable then the argument has full force. So it has to be inconceivable that, for example, C-fibers firing produces pleasure rather than pain. But that obviously is conceivable! There seems to be no contradiction in the scenario. A scenario where a person acts like they’re in pain but really is experiencing intense pleasure seems to be coherent. If you can conceive of inverted qualia or zombies you can also conceive of scrambled qualia.

Why should anyone accept this view? As already mentioned, there are quite a number of physicalist accounts of consciousness that entail this view, including non-illusionist and non-reductive accounts.

But these views all entail that zombies are inconceivable, a view affirmed by only about 16% of philosophers. And the people who think that zombies are possible, myself included, are often quite strongly convinced of that, such that giving up on the view would require a radical shift to our worldview. The fact that an extreme minority view, whose opponents often find it utterly crazy, explains psychophysical harmony should be of little comfort to most people.

Next, Montesinos argues that through introspection we can see this—and apparently a few undergraduates say we can’t divorce pain from its functional behavior. But this seems clearly false, and undergraduates are often deeply conceptually confused. I place little stock in what undergraduates say about sort of technical philosophical issues—like this one. As for the claim that we can see this through introspection—no, we see the exact opposite through introspection. Disembodied minds could feel pain—it doesn’t need to have the same physical profile as exists in the physical world. Montesinos’s introspection apparently delivers him very different intuitions from me—it seems utterly obvious that you could have a world that is physically identical but that differs with regard to the consciousness facts.

Montesinos discusses the second horn of the dilemma in which we think that phenomenal introspection is unreliable. He ends up here biting the bullet and saying that if a disembodied mind experienced something subjectively identical to excruciating torture, that wouldn’t be bad. He disputes that we can have reliable judgments about this because our judgments are just based on pain in the real world. But even though our judgments are based on pain in the real world, we can see the obvious fact that our behavioral disposition isn’t the reason that it’s bad. That disembodied minds being tortured is bad is the most obvious thing ever! If you found out that you were a brain in a vat for your entire life, your pain would still have been bad.

Montesinos mentions pain asymbolia, a case in which patients experience pain without it causing aversive reactions. They sort of notice the experience but it doesn’t prompt aversive behavior. But crucially, to them it doesn’t feel bad. The thing that’s bad about pain is that it feels bad—if, as pain asymbolia does, you remove its bad feeling, then you’ve removed what makes it suffering, as Hewitt-Rawlette argues at length. But if you have an experience that feels just as bad as being tortured but causes the same physical behavior as experiencing pleasure, it would still obviously be bad!

The next objection comes from Joseph Lawal. This is just an objection to semantic harmony, which claims that there’s an appropriate pairing between mental states and our reports about those mental states. Lawal notes that we might be semantic externalists—thinking that what a thought is about depends in some way on external reality. For example, if I think about water, that thought is about water, while if my twin thinks about what he believes to be water, that is made of XYZ, he is thinking about something other than water, even if we have the same thought. He then says:

It is not hard to see how this bears on the question of psychophysical harmony. Cutter & Crummett suppose that in a world in which I never have any reddish experiences, I can still make the judgment “I am having a reddish experience,” but that judgment will be false. But does reflection on Twin Earth cases not suggest that in the scenarios Cutter & Crummett describe, I cannot form the relevant judgment at all?

Two problems:

1. Semantic harmony is not about judgments about consciousness it’s about reports about consciousness. Even if you couldn’t think you were having a reddish experience without having a reddish experience, you could still obviously say you were having a reddish experience. Why that doesn’t occur cries out for explanation.

2. Most worlds that don’t have semantic harmony don’t have people having beliefs about their mental states. Most of them just have no ordered consciousness—maybe they’ll have people only with the phenomenology of eating Cheetos, for instance. So even if you can’t have mistaken judgments about having reddish experience, this doesn’t explain the vast improbability of having such judgments in the first place.

Next, Montesinos and Lucas Collier have a reply that invokes Nigel Thomas’ 1998 paper "Zombie Killer." This argument has been subject to various criticisms—I’ve criticized it here, so I won’t repeat my criticisms. Notably, it if true would rule out the conceivability of zombies—but zombies are clearly conceivable.

My only comment on the argument is the following: the argument assumes erroneously that if we have non-inferentially justified beliefs then our inverted twin would have the same non-inferentially justified process because they form their beliefs as part of the same cognitive mechanism. But all non-physicalists should deny this assumption—only conscious beings have beliefs and those constitutively depend on their mental states, rather than their physical brain states.

The next objection comes from Lawal again, claiming that interactionist dualists can avoid the argument. Lawal claims that in an inverted world, where pleasure generates aversion, it’s not genuinely pleasure causing the inversion. Lawal claims that as a consequence, believers in interactionism can deny that in such a case the pleasure would be genuinely causing the aversion. He additionally claims that on account of this, interactionists should think that it’s not possible for there to be inversion. A few problems:

1. If interactionists are committed to this, then they should give up on interactionism! Because inverted qualia, even radically inverted qualia is obviously possible. It’s not hard to imagine a world where you act as you do but have radically different experiences.

2. Perhaps Lawal would say that these worlds are possible but are just not interactionist worlds. But then this just means that the vast majority of worlds which are disharmonious are not genuinely interactionists. But then the fact that the actual world is genuinely interactionist becomes miraculous! Whether the disharmonious worlds count as interactionism is irrelevant to whether they’re possible worlds, and if they are possible, their probability utterly swamps that of the harmonious worlds.

3. Lawal’s view commits him to an implausible view of causality. In the case where whenever one experiences pleasure from an activity they act to avoid it in the future, the pleasure still causes their aversion—if they didn’t experience the pleasure, they wouldn’t have done it. This is certainly true on counterfactual accounts, and on pretty much all other causal views.

The next objection comes from Montesinos and Truth Teller and claims that psychophysical harmony understates the evidence. They claim that the real world is much more disharmonious than we think, listing the limits of human memory, human addiction, and errors in perception. Here, two responses are in order:

1. Any explanation of evil will explain these limitations. I’ll argue later that there is a theodicy that is uniquely qualified to explain these limits. But once you think that there’s an explanation for genocide, malaria, flesh eating parasites, cancer, earthquakes, and all the other evil in the world, an explanation of why we can’t see optimally falls out naturally.

2. Even if there isn’t perfect harmony, there’s more harmony than almost all psychophysical laws. Virtually all sets of psychophysical laws will produce radical disharmony—the fact that the world is almost entirely harmonious is quite miraculous. If we’re in the top .00000000000000000000000000000001% most harmonious laws, then this is strong evidence for theism, even if we’re not sure why we don’t have the most harmonious laws.

The last objection comes from Truth Teller and it involves the revenge problem. This objection claims that God’s mind is psychophysically harmonious, so the view assumes harmony rather than explaining it. TT first presents a response given by Crummett and Cutter, namely that psychophysical harmony follows from divine perfection. A being that is maximally perfect will necessarily have the ability to influence the world and thus be psychophysically harmonious.

TT argues that this gets the order of explanation wrong. The fact that a being is perfect must depend on the properties of the being. TT says:

As an example, suppose I claim that my dog is good, you'd take my claim to be about my dog's behavior or other descriptive features, He doesn't bite, He doesn't bark too often, He is obedient, friendly, loyal, cute etc. But suppose I reply, "No, my dog is not good in virtue of any of these good-dog making properties, my dog's goodness is logically prior to these properties but entails those properties". Then it seems like my claim that my dog is good is devoid of any content that would make it intelligible. The same point appears to be true of God's perfection, for it to have content, it has to be true that God instantiates perfect-making properties logically prior to His having the thin axiological attribute of "perfection", for it is those properties which gives His perfection content.

But the proponent of this objection doesn’t have to say that God’s perfection doesn’t obtain in virtue of these properties. They might say that perfection constitutively depends on these properties obtaining but still explains why they happen. An apple pie may constitutively depend on the atoms composing it, but the reason why they are arranged as they are may depend on features about the whole.

In addition, even if there are some questions about whether the revenge problem arises, because psychophysical harmony is so shocking on naturalism, the fact that it obtains is still very strong evidence for theism—if you think it might or might not be a problem, this still gives you major reason to update.

Next, TT addresses the claim that the theists’ psychophysically harmonious laws are very simple and thus more intrinsically probable. The theist can say that the full set of psychophysical laws is “what is willed obtains.” But he said this might be very complicated—it can be rephrased as “if x is willed it obtains, if y is willed it obtains, if z is willed it obtains, etc,” for infinite things, which is infinitely complicated.

I do not have a very strong view on how to count simplicity. But there are many views according to which the psychophysical laws are simple. The approach where you just list all the things that can be done given by TT is clearly unworkable—it implies full omnipotence is no more complicated than omnipotence minus being able to perform one random fact. But if we’re not sure about how to count simplicity, then we’ll still have sizeable credence in views according to which these are simple laws—but those views dominate the probability space for theism, so we’ll update hugely in favor of theism. Remember, this is like a quadrillion-to-one coincidence, so it still massively favors theism.

Richard Chappell has an objection to the argument from psychophysical harmony. Richard argues that just as we should assign a lower intrinsic likelihood to skeptical scenarios in which we’re brain in vats, we should also do that to worlds that are disharmonious. Several worries (for more see this Facebook thread, particularly the responses of Cutter and Crummett):

1. I’m dubious that we can have a brute epistemic principle that skeptical worlds are less likely. That should flow from more fundamental commitments. Why think this? Well, it seems like the plausibility of fundamental laws doesn’t depend on what they’ll result in, but this requires abandoning that principle by holding that the probability of laws depends on whether they’ll result in agents in skeptical scenarios. I think of skeptical scenarios as like the ontological argument—you can just know that they’re nonsense but there will be some deeper explanation of that.

2. It’s plausible that we can have a basic justification for thinking that harmonious worlds are more likely in the absence of a defeater. But here there is a defeater—the fact that nearly all ways the world could be would produce radical disharmony IS a defeater. If we discovered that a non-skeptical scenario requires vast improbabilities—a million consecutive 6s on a 6-sided die or insanely improbable cosmological fine-tuning—then positing that as brute would be unsatisfactory. So too is the explanation that posits that psychophysical harmony is just a brute coincidence not a good explanation.

3. Most of the cases where psychophysical harmony fails flatly don’t involve any bizarre skeptical scenarios. Most of them just involve nothing of value obtaining—for instance because people only experiencing eating fish. Surely those aren’t ruled out brutely!

The notorious alt-right debate blogger Bentham’s Bulldog (I kid you not, that is how I was described once) has his own criticisms of the argument from psychophysical harmony. Now, this blogger’s wit, virtue, and brilliance are surpassed only by his humility. That said, however, his (my) previous objections fail.

In this piece, I start by giving a few reasons to not place a ton of weight in the argument even if one doesn’t have a specific reply. I first note that it’s a new argument. This is true but it has now been subject to lots of criticism from many people. Not one of the criticisms succeeds. It’s unlikely future ones will, especially because the argument is pretty simple—there aren’t many responses left unexplored.

Second I say that theism is very improbable. This will, of course, depend on other evidence.

Finally, I say that consciousness is mysterious. But Chalmers’ dichotomy captures all possible views of consciousness and all except radical physicalism entail radical disharmony. But radical physicalism is clearly false.

I give various objections that have already been covered like the revenge problem, the claim theism is complex, and appeals to natural teleology. My final objection is as follows:

Suppose we accept the following few things.

1. There are huge numbers of universes.

2. Universes have their own psychophysical laws.

3. Beings in the universes can create new universes if they’re sufficiently motivated.

Plausibly this would create an anthropic selection effect for the sets of psychophysical laws that create beings that are more willing to create universes. But beings that are motivated to do good things would be more willing to create universes — after all, new universes are a good thing. Thus, most universes that would come to exist in the long run would have psychophysical harmony, because most of them would be created by good beings that want to create good psychophysical laws.

Additionally, if we expect that beings would evolve morality, they’d be motivated to do good. If they’re motivated to do good, they’d want to create sets of psychophysically harmonious laws.

For one, this view requires lots of assumptions. It requires assuming, based on no evidence, that beings can make new universes, make new psychophysical laws (which is not true of most psychophysical laws), and change the universes. But if a being was disharmonious it plausibly wouldn’t understand consciousness and wouldn’t realize the value of harmony enough to make harmonious psychophysical laws. In fact, these disharmonious beings probably wouldn’t even think about psychophysical laws, much less know to create them.

The next objection I gave to psychophysical harmony claims that once we stipulate that desires have a significant effect, anything other than acting in accordance with desires would result in desires making huge shifts in the brain, which would cause the organism to die. Thus, evolution would result in psychophysical harmony. Problems:

1. Stipulating that desires play a significant role in the brain just is stipulating a solution to psychophysical harmony. Almost all psychophysical laws don’t have that occur.

2. If their role was as significant as this objection claims, such that organisms die unless there is correlation between the mental and the physical, then we’d observe that experimentally—we have not.

3. If this were true the earlier disharmonious organisms would die off. Similarly, this would entail the possibility that many organisms with simple consciousness are disharmonious, which is hard to believe.

The final objection I give to the argument is in this article. I claim that more harmonious laws are more natural in that it posits that things go together in expected ways. But this idea has problems:

1. There’s an obvious debunking account of our belief that this explanation is natural. We think things like our world are natural, even if they are strange and ad hoc. The defender of this explanation owes us an account of why this is more natural which they cannot give.

2. Theism is a much more natural and harmonious hypothesis. Positing that the most fundamental level of reality contains only perfection is quite a natural hypothesis, much more so than the gerrymandered list of physical and psychophysical laws posited by naturalists.

3. Even if naturalness is a theoretical virtue, it’s hard to explain why it would trump the vast implausibility of psychophysical harmony. It would be bizarre to posit a 1 in 100 quadrillion coincidence just because it’s more natural. Being natural might make a hypothesis a billion times more probable, but even this is not enough to swamp the unfathomable improbability.

4. The definition of naturalness given in the article claims that a suggestion is natural if it’s not puzzling or surprising. But psychophysical harmony is not natural—if nearly all sets of psychophysical laws are disharmonious, including the simplest ones, then harmonious ones are puzzling and surprising.

Two more solutions to psychophysical harmony both appeal to panpsychism. The first comes from Goff and is, I think, a pretty fun solution. Goff argues that there is a fundamental law according to which all things in the universe are predisposed to respond rationally to their experiences. Particles have experiences that make it rational for them to act as they do. This explains why evolution would equip us with a complex internal conscious map of reality—when it does, we act rationally according to the map of reality, resulting in evolution selecting for harmony. I think Goff explains the solution more in his most recent book. This has several problems though:

1. It commits you to panpsychism. Problem: panpsychism is probably false; it’s very unintuitive and not the best explanation of consciousness, as I argue here.

2. Even once panpsychism is established, the vast majority of possible worlds won’t have creatures respond rationally to the contents of their experience. Goff has to brutely stipulate that there is a psychophysical law according to which creatures respond rationally to their experience, but that’s no better than stipulating just that there is harmony.

3. Even once most creatures do respond rationally to the content of their experience, it’s not settled that the psychophysical laws would enable the rich range of experiences that they do. Most sets of psychophysical laws would leave creatures unable to have rich internal lives.

The next panpsychism-based solution to psychophysical harmony comes from Bradford Saad. Saad claims that, just as we can explain that a monkey typed Shakespeare by saying that there are a lot of monkeys, we can explain that we’re psychophysically harmonious by positing that we’re a small array of harmony in a vast sea of disharmony. If nearly all conscious entities are disharmonious—which the panpsychist can say because she thinks that almost all conscious agents are atoms—then the fact that there’s one that’s harmonious is not a mystery. It’s a fun solution, but has pretty fatal problems:

1. There’s a big difference between the Shakespeare case and the case of the real world which is that we can only observe one instance of consciousness directly. Finding out that there are lots of other monkeys wouldn’t explain why the monkey in your house typed Shakespeare—similarly, if we only observe one example of something, positing most things are different from it is not a good explanation.

2. THERE AREN’T ENOUGH MONKEYS (I always wanted to say that in an article). Given that virtually zero percent of possible psychophysical laws are harmonious, even if we have a million or two rounds at the roulette wheel, it would still be miraculous that we have harmony at all.

3. The typical panpsychist thinks that there are two types of conscious beings in the universe: us and particles. But regardless of which percentage of the beings in the universe are particles, it’s miraculous that one of the two types of entities happened to be harmonious. It would be like finding that the monkey in your house happens to be one of two types of monkeys in your house—but this kind has the genetic disposition to type out Shakespeare. This could not be attributed to chance even if most monkeys are of the other kind.

The final response that has been given to the argument from psychophysical harmony tries to appeal to anthropic reasoning to resolve the puzzle. This solution is sort of complicated but claims that, because if we were psychophysically disharmonious we wouldn’t know about harmony, that cognitively sophisticated beings are all in worlds like ours, so that it’s no mystery why cognitively sophisticated beings are in worlds that are harmonious. This is similar to the thought that why I in particular exist is no mystery because if I hadn’t existed I wouldn’t be around to wonder about it.

But this has a few big problems. First, there is the same puzzle as there is in the firing squad case. Imagine that there is a firing squad of people to shoot me, but those people have special guns that if they hit me cause me to be psychophysically harmonious. If one thinks that despite nearly all beings being psychophysically disharmonious, it’s no mystery why they aren’t, then if 100 people in the firing squad all miss you, that doesn’t cry out for explanation, because if they had you wouldn’t be harmonious. But it clearly does cry out for explanation.

Second, the paper assumes that rational agents like us are in psychophysically harmonious worlds. But this is flatly false—there could be an agent that is able to think in an advanced way but whose thoughts aren’t manifested in actions.

Third, I think anthropic considerations only work if there’s a multiverse, in both fine-tuning and psychophysical harmony. But a multiverse that varies the physical and psychophysical laws would have to be amazingly complicated, able to vary experience dramatically.

There are more worries about the anthropic reasoning that I don’t think works, but I do not understand anthropics well enough to say much about it.

I have now (I think) responded to everything ever written about psychophysical harmony. I conclude that pretty much every objection to the argument is lousy; a few of them require massive stretches that make them slightly better than the purely chancy explanations of harmony, but still they are poor explanations.

2 Moral knowledge (but also mathematical and metaphysical knowledge and more)

I'm very well acquainted, too, with matters mathematical
I understand equations, both the simple and quadratical
About binomial theorem I am teeming with a lot o' news
With many cheerful facts about the square of the hypotenuse

—Gilbert and Sullivan in modern major general.

Elsewhere, I’ve argued that for us to be justified in our beliefs about morality, modality, metaphysics, mathematics, induction, and more we have to have a general capacity for rationality. Given that we are justified in our beliefs about these domains, our minds enable us to be acquainted with these facts even when they don’t make a difference to the position of atoms.

I won’t repeat my arguments for this conclusion but they are, to my mind, quite overwhelming. Suppose one accepts this. Theism explains this fact far, far better than naturalism—the naturalist is left scratching their head about this puzzling phenomenon. In contrast, the theist has an explanation—God equips us with the faculties for grasping non-rational truths. These allow us to grasp important things, that God would want us to grasp.

Our ability to have a basic capacity for rationality enables us to know highly general truths that point us in the direction of God. Our knowledge of the mathematical facts allows us to see the mathematical beauty of the universe and has value. Our knowledge of the moral facts allows us to have a better grasp of the perfection of God and is enables us to avoid wrongdoing. Our knowledge of the modal, metaphysical, and epistemic facts is indispensable to our being able to reason in the world, which is a great good.

In contrast, the naturalist can’t explain the reason this arises in terms of the goodness caused by it. So then why does it arise? The fact that this is built into the psychophysical laws is, on its face, very puzzling on naturalism. Why would the psychophysical laws have this surprising feature?

3 The conditions for conscious agents that can do things

Everything is just an accident
A happy accident

Limbo and we're going lower still
And we plan love, put it on a window sill
And another time and place
Where I never even had the chance to see your face

—Saint Motel in the song Happy Accident

Theism predicts that there would be conscious agents. Given that conscious agents are required for value, God’s grand plan would necessarily have conscious agents. In contrast, naturalism has a terrible time explaining conscious agents. Specifically, the naturalist account requires accepting the following series of unlikely events:

1. There are basic physical things. Suppose that the fundamental things are bosons, fermions, etc. Well those have to exist on the naturalist model. Such things very well could have not existed.

2. There have to be laws governing their behavior. You could imagine those things existing but having no causal properties. The fact that there is both fundamental stuff and laws governing their behavior is a vast coincidence—given that one could have existed without the other, it’s odd that they both exist at the same time.

3. Even after there are laws and constants, those laws have to generate complexity. Problem: the overwhelming majority of laws do not generate complexity. Conway to make the game of life, which made interesting things happen from simple rules, had to search through and reject tons of rules, because nearly all fundamental rules don’t result in anything interesting happening. Note it doesn’t help to posit a multiverse here, because a multiverse is also a set of fundamental laws—we have to get extremely lucky to get an orderly set of laws.

4. Even once the laws are still roughly like our laws, we have to get extremely lucky. If you modify many of the constants of physics even slightly nothing interesting could arise—no atoms and certainly no complex life.

5. I’ve argued for dualism elsewhere, so I won’t rehash the arguments. But if you grant this broad picture, you next have to suppose that there are psychophysical laws—something that nearly all possible ways the universe could have been wouldn’t have.

6. Then after that, there have to be things that arise that are the types of things that are conscious in combination with the psychophysical laws. This is far from guaranteed. The psychophysical laws could have resulted in any physical states being conscious—that the things that it caused to be conscious happened to exist is another coincidence.

7. Then on top of that, the psychophysical laws have to enable the creation of things of value. If, for instance, we never had valanced experiences or the ability to feel love or have knowledge, all would be for nothing.

Each step along the way is very improbable, yet the naturalist has to accept all of them occur. This is a sequence of unbelievably improbably coincidences that makes the naturalist hypothesis unfathomably improbable.

The fourth concern I raise here is basically the fine-tuning argument, and each of these are somewhat related to fine-tuning. So let me briefly address five objections to the fine-tuning argument.

A first objection appeals to the anthropic principle to explain fine-tuning. If we hadn’t been finely tuned we wouldn’t know about it, so there’s no puzzle that we are fine tuned. But this solution doesn’t work:

1. If your parents had never met, you wouldn’t be around to wonder about it. Nevertheless, it still might be rational to infer design (say their friends set them up) rather than chance if your parents bumped into each other in lots of extremely unlikely circumstances and were put in a series of extremely improbable situation circumstances that lead to them forming a relationship. For instance, suppose that there were 50 dinner parties each with supposedly random seating and 100 people—if your parents always ended up next to each other, it would be rational to infer the explanation of that was design rather than chance. So too in this case.

2. Suppose that an entire firing squad shoots at you but all their guns jam. It’s true that if they hadn’t jammed you wouldn’t be around to wonder about it. Nonetheless, this bizarre event seems to cry out for explanation.

3. It doesn’t follow from “if X hadn’t happened then I wouldn’t know,” that “the fact that X happened doesn’t cry out for explanation.”

A second objection claims that there could be more fundamental physics that entails fine-tuning that we don’t know about. But it’s equally epistemically possible that new physics could result in any physical arrangement—thus, this doesn’t affect the overall probabilities. Fundamental physics could have resulted in fine-tuning, or it could have resulted in a non-finely-tuned universe—either way the probabilities wash out.

A third objection says that the reason the universe is fine-tuned is out of necessity. The laws of physics can’t be different from what they are, on this account. But this fails for the same reason that the last solution did—it wouldn’t be satisfactory to explain why you got 10 royal flushes in a row to say “it’s a necessary fact that I do.” Even if some fact is necessary, it can still be surprising—and it’s extremely surprising that it’s necessary that the fundamental laws are exactly as they are, because epistemically they could have easily been necessarily something else.

A fourth objection claims that fine-tuning is the result of a multiverse—if there are oodles of different universes, then it’s guaranteed that some universes would have life. And in the universes that don’t, we’re not around to wonder about that. Now, various people have argued that this commits the inverse gambler’s fallacy—I think it doesn’t but would have to think more about the topic—which is one worry for this account.

But there’s a much bigger worry which is that the multiverse itself would have to be finely tuned. The multiverse involves various constants that result in many universes—but those are manipulable constants, such that if they were changed, life wouldn’t exist. Given that, as noted before, almost all laws fail to generate complexity, a multiverse is itself a finely-tuned miracle that is wildly implausible. In addition, many of these views struggle to avoid the Boltzmann brain problem—many of the simpler models would entail that most beings that exist happen to exist randomly as a result of a chemical mishap, such that we should expect that we’re short-lived observers who only briefly blip into existence.

The fifth objection comes from the brilliant Neil Sinhababu. Sinhababu argues that because God could design the psychophysical laws to make any physical arrangement produce conscious beings, theism doesn’t explain why the physical arrangement of the world is what it is. But the version of the fine-tuning argument doesn’t claim it does. I claim theism explains why there are embodied agents like us that can have valuable relationships and argue that atheism must jump through many hoops to explain that. Theism may not predict embodied life, but it can still be a much better explanation of conscious agents arising than atheism.

Well, that’s a wrap for part 1. Thanks for reading this roughly 7,300-word article. Hope you enjoyed it! In later sections, I’ll provide more arguments for theism and address various objections to theism.