If materialism is true, God exists

Causal finitism is the doctrine that backwards infinite causal histories are impossible.

1. If the xs compose y, then y cannot have caused all of the xs.

2. If materialism is true and causal finitism is false, then it is possible to have a human being that (a) is composed of cells and (b) caused each of its cells via a backwards infinite regress.

3. So, if materialism is true, causal finitism is true. (1, 2)

4. If causal finitism is true, then God exists.

5. So, if materialism is true, God exists. (3, 4)

6. If God exists, the materialism is false.

7. So, materialism is false. (5, 6)

Premise (1) is a strengthening of a plausible principle banning self-causation.

Premise (2) follows from the fact that we are causes of all our present cells. If presentism is true, that completes the argument against materialism as in my previous post. But if eternalism or growing block are true, then we may also be composed of our past cells. And we didn’t cause our first cells. However if causal finitism is false, then it’s very plausible that backwards infinite causal regresses are possible, and so we could have existed from eternity, continually producing new cells, with the old ones dying.

Premise (4) is backed by a version of the kalaam argument.

Premise (6) is definitional if we understand materialism strongly enough to apply not just to us but to all reality. If we understand materialism more weakly, then the argument “only” yields the conclusion (5) that if materialism is true, God exists.

If presentism is true, materialism is false

1. If the xs compose y, then y cannot have caused all of the xs.

2. I caused all my present cells.

3. If presentism is true, then all my cells are present cells.

4. So, if presentism is true, then I caused all my cells. (2, 3)

5. If materialism is true, then I am composed of my cells.

6. If materialism is true, then I did not cause all of my cells. (1, 5)

7. So, if presentism is true, materialism is not true. (4, 6)

At least sometimes parts aren't prior to wholes

1. Efficient causes are explanatorily prior to their effects.

2. Circularity of explanatory priority is impossible.

3. I am the efficient cause of my teeth—I grew them!

4. Therefore, my teeth are not explanatorily prior to me. (1–3)

5. My teeth are parts of me.

6. Therefore, at least some parts are not explanatorily prior to the wholes. (4, 5)

Why I can't believe in a God other than of classical theism

I can’t get myself to believe in a God who is an old bearded guy in the sky. That would be just a fairy tale.

What’s wrong with such a concept of God? It’s the beard! Seriously, the problem is that a guy who has a beard has parts and changing. Whether the parts are material or immaterial does not seem of very deep metaphysical significance. But having parts or changing, either one of these is an absurd anthropomorphism.

And hence I can’t get myself to believe in a God who changes or has parts. That leaves classical theism and atheism as the options. And atheism leads to scepticism, I think.

More on doing and allowing

Let’s suppose disease X if medically unchecked will kill 4.00% of the population, and there is one and only one intervention available: a costless vaccine that is 100% effective at preventing X but that kills 3.99% of those who take it. (This is, of course, a very different situation than the one we are in regarding COVID-19, where we have extremely safe vaccines.) Moreover, there is no correlation between those who would be killed by X and those who would be killed by the vaccine.

Assuming there are no other relevant consequences (e.g., people’s loss of faith in vaccines leading to lower vaccine uptake in other cases), a utilitarian calculation says that the vaccine should be used: instead of 316.0 million people dying, 315.2 million people would die, so 800,000 fewer people would die. That’s an enormous benefit.

But it’s not completely clear that this costless vaccine should be promoted. For the 315.2 million who would die from the vaccine would be killed by us (i.e., us humans). There is at least a case to be made that allowing 316.0 million deaths is preferable to causing 315.2 million. The Principle of Double Effect may justify the vaccination because the deaths are not intentional—they are neither ends nor means—but still one might think that there is a doing/allowing distinction that favors allowing the deaths.

I am not confident what to say in the above case. But suppose the numbers are even closer. Suppose that we have extremely precise predictions and they show that the hypothetical costless vaccine would kill exactly one less person than would be killed by X. In that case, I do feel a strong pull to thinking this vaccine should not be marketed. On the other hand, if the numbers are further apart, it becomes clearer to me that the vaccine is worth it. If the vaccine kills 2% of the population while X kills 4%, the vaccine seems worthwhile (assuming no other relevant consequences). In that case, wanting to keep our hands clean by refusing to vaccinate would result in 158 million more people dying. (That said, I doubt our medical establishment would allow a vaccine that kills 2% of the population even if the vaccine would result in 158 million fewer people dying. I think our medical establishment is excessively risk averse and disvalues medically-caused deaths above deaths from disease to a degree that is morally unjustified.)

From a first-person view, though, I lose my intuition that if the vaccine only kills one fewer person than the disease, then the vaccine should not be administered. Suppose I am biking and my bike is coasting down a smooth hill. I can let the bike continue to coast to the bottom of the hill, or I can turn off into a side path that has just appeared. Suddenly I acquire the following information: by the main path there will be a tiger that has a 4% chance of eating any cyclist passing by, while by the side path there will be a different tiger that has “only” a 3.99999999% chance of eating a cyclist. Clearly, I should turn to the side path, notwithstanding the fact that if the tiger on the side path eats me, it will have eaten me because of my free choice to turn, while if the tiger on the main path eats me, that’s just due to my bike’s inertia. Similarly, then, if the vaccine is truly costless (i.e., no inconvenience, no pain, etc.), and it decreases my chance of death from 4% to 3.99999999% (that’s roughly what a one-person difference worldwide translates to), I should go for it.

So, in the case where the vaccine kills only one fewer person than the disease would have killed, from a first-person view, I get the intuition that I should get the vaccine. From a third-person view, I get the intuition that the vaccine shouldn’t be promoted. Perhaps the two intuitions can be made to fit together: perhaps the costless vaccine that kills only one fewer person should not be promoted, but the facts should be made public and the vaccine should be made freely available (since it is costless) to anyone who asks for it.

This suggests an interesting distinction between first-person and third-person decision-making. The doing/allowing distinction, which favors evils not of our causing over evils of our causing even when the latter are non-intentional, seems more compelling in third-person cases. And one can transform third-person cases to be more like first-person through unencouraged informed consent perhaps.

(Of course, in practice, nothing is costless. And in a case where there is such a slight difference in danger as 4% vs. 3.99999999%, the costs are going to be the decisive factor. Even in my tiger case, if we construe it realistically, the effort and risk of making a turn on a hill will override the probabilistic benefits of facing the slightly less hungry tiger.)

Is it prudent to start drinking alcohol?

Here is an interesing exercise in decision theory.

Suppose (as is basically the case for me, if one doesn’t count chocolates with alcohol or the one or two spoonfuls of wine that my parents gave me as a kid to allay my curiosity) I am a man who never drunk alcohol. Should I?

Well, 6.8% of American males 12 and up suffer from alcoholism. And 83% of American males 12 and up report having drunk alcohol. It follows that the probability of developing alcoholism after drinking is around 8%, whle the probability of developing alcoholism without drinking is around 0%.

Family history might provide one with some reason to think that in one’s own case the statistics on developing would be more pessimistic or more optimistic, but let’s suppose that family history does not provide significant data one way or another.

So, the question is: are the benefits of drinks containing alcohol worth an 8% chance of alcoholism?

Alcoholism is a very serious side-effect. It damages people’s moral lives in significant ways, besides having serious physical health repercussions. The moral damage is actually more worrying to me than the physical health repercussions, both because of the direct harms to self and the indirect harms to others, but the physical health considerations are easier to quantify. Alcoholism reduces life expectancy by about 30%. Thus, developing alcoholism is like getting a 30% chance of instant death at a very young age. Hence, the physical badness of facing an 8% chance of developing alcoholism is like a young child’s facing a 2.4% chance of instant death.

At this point, I think, we can get some intuitions going. Imagine that a parent is trying to decide whether their child should have an operation that has about a 2.4% chance of death. There definitely are cases where it would be reasonable to go for such an operation. But here is one that is not. Suppose that the child has a condition that makes them unable to enjoy any food that has chocolate in it—the chocolate is harmless to them, but it renders the food containing it pleasureless. There is a childhood operation that can treat this condition, but about one in forty children who have the operation die on the operating table. It seems to me clear that parents should refuse this operation and doctors should not offer it, despite the fact that chocolate has great gustatory pleasures associated it. Indeed, I think it is unlikely that the medical profession would approve of the operation.

But I doubt that alcohol’s morally legitimate pleasures exceed those of chocolate.

I said that the moral ills of alcoholism are larger than the physical, but harder to quantify. Still, we can say something. If the physical badness of an 8% chance of alcoholism is like a young child’s facing a 2.4% chance of instant death, and there is a worse moral effect, it follows that the overall badness of an 8% chance of alcoholism is at least as bad as a young child’s facing a 5% chance of instant death. And with a one in twenty chance of death, there are very few operations we would be willing to have performed on a child or ourselves. The operation would have to correct a very dangerous or a seriously debilitating condition. And alcohol does neither.

This suggests to me that if the only information one has is that one is male, the risk of alcoholism is sufficient that the virtue of prudence favors not starting to drink. If one is female, the risk is smaller, but it still seems to me to be sufficiently large for prudence to favor not starting to drink.

There are limitations of the above argument. If one has already started drinking, one may have additional data that goes beyond the base rates of alcoholism—for instance, one may know that in twenty years of drinking, one has not had any serious problems with moderation, in which case the argument does not apply (but of course one might also have data that makes alcoholism a more likely outcome than at the base rate). Similarly, one might have data from family history showing that the danger of alcoholism is smaller than average or from one’s own personal history showing that one lacks the “addictive personality” (but in the latter case, one must beware of self-deceit).

I am a little suspicious of the above arguments because of the Church’s consistent message, clearly tied to Jesus’s own practice, that the drinking of alcohol is intrinsically permissible.

It may be that I am overly cautious in thinking which degree of risk prudence bids us to avoid. Perhaps one thing to say is that while there are serious reasons of prudence not to start drinking, I may be underestimating the weight of the benefits of drinking.

I also think the utilities were different in the past. If spices and chocolate are unavailable or prohibitively expensive, wine might be the main gustatory pleasure available to one, and so the loss of gustatory pleasure would be a more serious loss. Likewise, alcoholic drinks may have health benefits over unsafe drinking water. Finally, even now, one might live in a cultural setting where there are few venues for socialization other than over moderate alcoholic consumption.

Of course, in my own case there are also hedonic reasons not to drink alcoholic drinks: the stuff smells like a disinfectant.

Is it permissible to fix cognitive mistakes?

Suppose I observe some piece of evidence, attempt a Bayesian update of my credences, but make a mistake in my calculations and update incorrectly. Suppose that by luck, the resulting credences are consistent and satisfy the constraint that the only violations of regularity are entailed or contradicted by my evidence. Then I realize my mistake. What should I do?

The obvious answer is: go back and correct my mistake.

But notice that going back and correcting my mistake is itself a transition between probabilities that does not follow the Bayesian update rule, and hence a violation of the standard Bayesian update rule.

To think a bit more about this, let’s consider how this plays out on subjective and objective Bayesianisms. On subjective Bayesianism, consistency, the Bayesian update rule and perhaps the constraint that the only violations of regularity are entailed or contradicted by my evidence. My new “mistaken” credences would have been right had I started with other consistent and regular priors. So there is nothing about my new credences that makes them in themselves rationally worse than the ones that would have resulted had I done the calculation right. The only thing that went wrong was the non-Bayesian transition. And if I now correct the mistake, I will be committing the rational sin of non-Bayesian transition once again. I have no justification for that.

Moreover, the standard arguments for Bayesian update apply just as much now in my new “mistaken” state: if I go back and correct my mistake, I will be subject to a diachronic Dutch Book, etc.

So, I should just stick to my guns, wherever they now point.

This seems wrongheaded. It sure seems like I should go back and fix my mistake. This, I think, shows that there is something wrong with subjective Bayesianism.

What about objective Bayesianism? Objective Bayesianism adds to the consistency, update and (perhaps) regularity restrictions in subjective Bayesianism some constraints on the original priors. These constraints may be so strict that only one set of original priors counts as permissible or they may permissive enough to allow a range of original priors. Now note that the standard arguments for Bayesian update still apply. It looks, thus, like correcting my mistake will be adding a new rational sin to the books. And so it seems that the objective Bayesian also has to say that the mistake should not be fixed.

But this was too quick. For it might be that my new “mistaken” posteriors are such that given my evidential history they could not have arisen from any permissible set of original priors. If so, then it’s like my being in possession of stolen property—I have posteriors that I simply should not have—and a reasonable case can be made that I should go back and fix them. This fix will violate Bayesian update. And so we need to add an exception to the Bayesian update rules: it is permissible to engage in a non-Bayesian update in order to get to a permissible credential state, i.e., a credential state that could have arisen from a permissible set of priors given one’s evidential history. This exception seems clearly right. For imagine that you are the mythical Bayesian agent prior to having received any evidence—all you have are your original priors, and no evidence has yet shown up. Suddenly you realize that your credences violate the objective rules on what the priors should be. Clearly you should fix that.

Thus, the objective Bayesian does have some room for justifying a “fix mistakes” exception to the Bayesian update rule. That exception will still violate the standard arguments for Bayesian update, and so we will have to say something about what’s wrong with those arguments—perhaps the considerations they give, while having some force, do not override the need for one’s credences to be such that they could be backtracked to permissible original priors.

Considerations of mistakes gives us reasons to prefer objective Bayesianism to subjective Bayesianism. But the objective Bayesian is not quite home free. Consider first the strict variety where there is only one permissible set of original priors. We have good empirical reason to think that there are about as many sets of original priors as there are people on earth. And on the strict version of objective Bayesianism, at most one of these sets of original priors is permissible. Thus it’s overwhelmingly unlikely that my original priors are permissible. Simply fixing my last mistake is very unlikely to move me to a set of posteriors that are correct given the unique set of permissible original priors and my evidential history. So it’s a matter of compounding one rational sin—my mistake—with another, without fixing the underlying problem. Maybe I can have some hope that fixing the mistake gets me closer to having posteriors that backtrack to the unique permissible original priors. But this is not all that clear.

What about permissible objective Bayesianism? Well, now things depend on our confidence that our original priors were in fact permissible and that no priors that generate our new “mistaken” posteriors given our evidential history would have been permissible. If we have a high enough confidence in that, then we have some reason to fix the mistake. But given the obvious fact that human beings so often reason badly, it seems unlikely that my original priors were in fact permissible—if Bayesianism is objective, we should believe in the “original cognitive sin” of bad original priors. Perhaps, just as I speculated on strict objective Bayesianism, we have some reason to hope that our actual original priors were closer to permissible than any priors that would generate our new “mistaken” posteriors. Perhaps.

So every kind of Bayesian has some difficulties with what to do given a miscalculation. Objective Bayesians have some hope of having an answer, but only if they have some optimism in our actual original priors being not too far from permissibility.

It is interesting that the intuition that we should fix our “mistaken” posteriors leads to a rather “Catholic” view of things: although doubtless there is original cognitive sin in our original priors, these priors are sufficiently close to permissibility that cognitive repairs make rational sense. We have depravity of priors, but not total depravity.

How I learned to be a bit less judgmental about social distancing

Earlier in the pandemic, I was very judgmental of students hanging around outside in groups and not respecting six-foot spacing. Fairly quickly I realized that it is inadvisable for the university to rebuke students for doing this, since such rebukes are likely to lead to their taking such interactions to private indoor venues, which would be much worse from a public health standpoint. But that practical consideration did not alleviate my strong judgmental feelings.

Eventually, however, these observations have made me realize that in our species, there is a natural desire to spend time in relatively close physical proximity to each other. And indeed, this is quite unsurprising in warmblooded social animals. Realizing that social distancing—however rationally necessary—requires people to go against their natural instincts has made me quite a bit less judgmental about noncompliance.

It took observation of others to realize this, because apart from practicalities, I find myself to prefer something like two meter spacing for social interaction with people outside my family. Greater physical distance from people outside my family has been quite pleasant for me. A conversation at two meters feels a little bit less stressful than at one meter. But looking at other people, it is evident that my preference here is literally unnatural, and that for other people such distancing is quite a burden.

Of course, sometimes it is morally necessary to go against one’s natural desires. It is natural to flee fires, but fire fighters need to go against that desire. And in circumstances where it is morally necessary to go against natural desires, people like me who lack the relevant natural desires are particularly fortunate and should not be judgmental of those for whom the actions are burden.

Desires for another's action

Suppose that Alice is a morally upright officer fighting against an unjust aggressor in a bloody war. The aggressor’s murderous acts include continual slaughter of children. Alice has sent Bob for a mission behind enemy lines. Bob’s last message said that Bob has found a way to end the war. The enemy has been led to war by a regent representing a three-year-old king. If the three-year-old were to die, the crown would pass to a peaceloving older cousin who would immediately end the war. And Bob has just found a way to kill the toddler king. Moreover, he can do it in such a way that it looks like it is a death of natural causes and will not lead to vengeful enemy action.

Alice responds to the message by saying that the child-king is an innocent noncombatant and that she forbids killing him as that would be murder. It seems that Alice now has two incompatible desires:

• that Bob will do the right thing by refraining from murdering the child, and

• that Bob will assassinate the child king, thereby preventing much slaughter, including of children.

And there is a sense in which Alice wants the assassination more than she wants Bob to do the right thing. For what makes the assassination undesirable—the murder of a child—occurs in greater numbers in the no-assassination scenario.

But in another sense, it was the desire to have Bob do the right thing that was greater. For that was the desire that guided Alice’s action of forbidding the assassination.

What should we say?

Here is a suggestion: Alice desires that Bob do the right thing, but Alice wishes that Bob would assassinate the king. What Alice desires and what Alice wishes for are in this case in conflict.

And here is a related question. Suppose someone you care about wants you to do one thing but wishes you to do another. Which should you do?

In the above case, the answer is given by morality: assassinating the three-year-old king is wrong, no matter the consequences. And considerations of authority concur. But what if we bracket morality and authority, and simply ask what Bob should do insofar as he cares about Alice who is his friend. Should he follow Alice’s desires or her wishes? I think this is not so clear. On the one hand, it seems more respectful to follow someone’s desires. On the other hand, it seems more beneficent to follow someone’s wishes.

Regular Hyperreal and Qualitative Probabilities Invariant Under Symmetries

I just noticed that my talk "Regular Hyperreal and Qualitative Probabilities Invariant Under Symmetries" is up on YouTube. And the paper that this is based on (preprint here) has  just been accepted by Synthese.

Obedience to God out of gratitude

Some philosophers want to ground our duty to obey God’s commands in the need to show gratitude to God for all the goods God has done for us. I think there is something to that, but I want to point out a complication.

When someone has done something good for us, that generates moral reasons to do something good for them. But that is different from generating moral reasons to obey their commands. After all, being obeyed need not be good for the person issuing the commands. Imagine that you are on a ship along with someone who has already done many good things for you and your family. The ship is sinking. There is one last space left in a lifeboat. You start to push your benefactor into that space. Your benefactor interrupts: “Don’t! Get into it yourself!” In this case, it would be bad for your benefactor for you to obey their commmand, and obedience to this command would not, I think, be a right expression of gratitude. You might have other moral reasons to obey the command, such as that if someone is offering to make the ultimate sacrifice, you should not deprive them of that choice. But gratitude for past benefits is not a reason to obey your benefactor when your benefactor would not in fact benefit from your obedience.

So the mere fact that you are commanded something by your benefactor does not generate a gratitude-based reason for obedience. It is only when your benefactor would benefit from your obedience that such a reason is generated.

Now things get a little complicated in the case of God. It seems we cannot benefit God, as God has perfect beatitude. But as we learn from Aquinas’ discussion of love for God, it’s more complicated than that. There are internal and external benefits and harms a person might receive. Internal benefits and harms affect the person’s intrinsic properties in a positive way—think here of pleasure and pain, virtue and vice, etc. But there are external benefits and harms: when people speak badly about you behind your back, the loss of reputation is an external harm, even if you never find out about it and it never affects any intrinsic property of yours. Similarly, because friends are other selves, if x loves y, then benefits and harms to y are benefits and harms to x, albeit perhaps only external ones. Thus, we can benefit God by benefiting those that God loves, namely everyone.

So, when God commands us—as in fact he does—to love our neighbor, then our obedience to that command does benefit God, albeit externally. So it seems we do have a gratitude-based moral reason to do what God says. But notice that as far as the argument goes right now, that is not a reason to love our neighbor out of obedience to God. The reason to love our neighbor out of gratitude to God would remain even if God did not command us to love our neighbor.

But this isn’t the whole story. For, first, when we show gratitude to a benefactor by bestowing some internal or external benefits on them, gratitude seems to call on us to have a preference for bestowing those benefits that the benefactor asks us to bestow. Thus, the fact that some benefit to the benefactor is requested by the benefactor adds to the gratitude-based reasons for bestowing that benefit. And, second, it seems plausible that having one’s commands be obeyed is itself an external benefit—it is a way of being honored.

Thus, that God has commanded us to love our neighbor intensifies our reasons based on gratitude to God to love our neighbor. And even if God were to command something seemingly arbitrary and not in itself beneficial, like abstinence from pork, we externally benefit God insofar as we obey him. But in the latter case something stronger is needed than that: for the obedience to be a form of gratitude, it needs to be the case that on balance we benefit God through the obedience. That being obeyed is good as far as it goes does not show that being obeyed is on balance good. It is good as far as it goes for our benefactor to be obeyed when they say to go into the lifeboat, but it might actually on balance be better for them to be pushed into the lifeboat.

Thus there is a limit to how far this justification of divine authority goes. If we obey God out of gratitude, our reason for obedience cannot simply consist in the facts that God has commanded us and that God has bestowed great benefits on us. Our reason would also need to include the fact that on balance it bestows a benefit—an external one, to be sure—on God if we obey.

Is it the case that whenever God commands something, it always bestows a benefit on balance on God that we obey him? That initially sounds like a reasonable thesis, but we can easily imagine cases where it is not so. Consider cases where my disobedience to God would prevent massive disobedience by others. For instance, a malefactor offers me a strong temptation to disobey God, and tells me that if I refuse the temptation, then a thousand other people will be offered the same temptation, but if I give in, I will be the only one. Looking at how strong the temptation is, I conclude that if it’s offered to a thousand people, about 500 of them will succumb to it. Thus, if I obey God, there will be much less obedience of God in the world. Hence, my obedience to God actually leads to God being less honored and receiving less external benefit. But nonetheless I need to obey. Hence, the duty of obedience is not grounded in gratitude.

In summary: Normally, gratitude does give us moral reason to obey God. But if I am right, then the moral reason to obey God that comes from gratitude needs to include the assumption that it is good for God to be obeyed. And we can imagine cases where that assumption is false and yet obedience is still required.

An exercise in vacuity

I’ve been thinking what an Aristotelian would say about the thesis that the normative supervenes on the non-normative. That thesis holds that:

1. any two possible worlds that have the same non-normative facts have the same normative facts.

So, the first question I asked myself was: What sorts of non-normative facts are there on the Aristotelian view of the world? First, every fact involving a natural kind is normative, since natural kind concepts are normative concepts—sheep are the sorts of things that should have four legs (among other things) and electrons are the sorts of things that should repel other electrons (among other things). Second, because natural kind membership is essential, it seems that any facts about particulars will also be normative, since all particulars other than God (and by divine simplicity there are no non-normative facts about God) are essentially members of natural kinds. Could there at least be some non-normative facts about how many objects exist? I don’t think so. For to exist is to be a substance, or to be related appropriately to a substance (say, by being an accident of a substance). But a part of what it is to be a substance is to have a form which governs how one ought to behave, and that’s normative. So existential facts are normative, too.

In fact, it’s looking to me like there are no non-normative facts on an Aristotelian view. Hence, if (1) were to hold, we would have to have:

2. any two possible worlds have the same normative facts.

But since all facts are normative, it would follow that:

3. any two possible worlds have the same facts.

But since possible worlds are distinguished by their facts, we see that on an Aristotelian view, the supervenience thesis basically says:

4. there is only one possible world.

And that’s false.

Aquinas and Descartes on substance dualism

Roughly, Aquinas thinks of a substance as something that:

1. is existentially independent of other things, and

2. is complete in its nature.

There is a fair amount of work needed to spell out the details of 1 and 2, and that goes beyond my exegetical capacities. But my interest is in structural points. Things that satisfy (1), Aquinas calls “subsistent beings”. Thus, all substances are subsistent beings, but the converse is not true, because Aquinas thinks the rational soul is a subsistent being and not a substance.

Descartes, on the other hand, understands substance solely in terms of (1).

Now, historically, it seems to be Descartes and not Thomas who set the agenda for discussions of the view called “substance dualism”. Thus, it seems more accurate to think of substance dualists as holding to a duality of substance in Descartes’ sense of substance than in Aquinas’.

But if we translate this to Thomistic vocabulary, then it seems we get:

3. A “substance dualist” in the modern sense of the term is someone who thinks there are two subsistent beings in the human being.

And now it looks like Aquinas himself is a substance dualist in this sense. For Aquinas thinks that there are two subsistent beings in Socrates: one of them is Socrates (who is a substance in the Thomistic sense of the word) and the other is Socrates’ soul (which is a merely subsistent being). To make it sound even more like substance dualism, note that Thomas thinks that Socrates is an animal and animals are bodies (as I have learned from Christopher Tomaszewski, there are two senses of body: one is for the material substance as a whole and the other is for the matter; it is body in the sense of the material substance that Socrates is, not body in the sense of matter). Thus, one of these subsistent beings or substances-in-the-Cartesian-sense is a body and the other is a soul, just as on standard Cartesian substance dualism.

But of course there are glaring difference between Aquinas’ dualism and typical modern substance dualisms. First, and most glaringly, one of the two subsistent beings or Cartesian substances on Aquinas’s view is a part of the other: the soul is a part of the human substance. On all the modern substance dualisms I know of, neither substance is a part of the other. Second, of the two subsistent beings or Cartesian substances, it is the body (i.e., the material substance) that Aquinas identifies Socrates with. Aquinas is explicit that we are not souls. Third, for Aquinas the body depends for its existence on the soul—when the soul departs from the body, the body (as body, though perhaps not as matter) perishes (while on the other hand, the soul depends on the matter for its identity).

Now, let’s move to Descartes. Descartes’ substance dualism is widely criticized by Thomists. But when Thomists criticize Descartes for holding to a duality of substances, there is a danger that they are understanding substance in the Thomistic sense. For, as we saw, if we understand substance in the Cartesian sense, then Aquinas himself believes in a duality of substances (but with important structural differences). Does Descartes think there is a duality of substances in the Thomistic sense? That is not clear to me, and may depend on fine details of exactly how the completeness in nature (condition (2) above) is understood. It seems at least in principle open to Descartes to think that the soul is incomplete in its nature without the body or that the body is incomplete in its nature without the soul (the pineal gland absent the soul sure sounds incomplete) or that each is incomplete without the other.

So, here is where we are at this point: When discussing Aquinas, Descartes and substance dualism we need to be very careful whether we understand substance in the Thomistic or the Cartesian sense. If we take the Cartesian sense, both thinkers are substance dualists. If we take the Thomistic sense, Aquinas clearly is not, but it is also not clear that Descartes is. There are really important and obvious structural differences between Thomas and Descartes here, but they should not be seen as differences in the number of substances.

And here is a final exegetical remark about Aquinas. Aquinas’ account of the human soul seems carefully engineered to make the soul be the sort of thing—namely, a subsistent being—that can non-miraculously survive in the absence of the substance—the human being—that it is normally a part of. This makes it exegetically probable that Aquinas believed that the soul does in fact survive in the absence of the human being after death. And thus we have some indirect evidence that, in contemporary terminology, Aquinas is a corruptionist: that he thinks we do not survive death though our souls do (but we come back into existence at the resurrection). For if he weren’t a corruptionist, his ontology of the soul would be needlessly complex, since the soul would not need to survive without a human being if the human being survived death.

And indeed, I think Aquinas’s ontology is needlessly complex. It is simpler to have the soul not be a subsistent being. This makes the soul incapable of surviving death in the absence of the human being. And that makes for a better view of the afterlife—the human being survives the loss of the matter, and the soul survives but only as part of the human being.

A metaphysical argument for survivalism

Corruptionist Thomists think that after death and before the resurrection, our souls exist in a disembodied state and have mental states, but we do not exist. For we are not our souls. Survivalist Thomists think we continue to exist between death and the resurrection. They agree that we are not our souls, but tend to think that in the disembodied we have our souls as proper parts.

Here is a metaphysical argument against corruptionism and for survivalism.

1. An accident that has a subject is a part of that subject.

2. There are mental state accidents in the disembodied state.

3. All mental state accidents in the disembodied state have a subject.

4. The soul does not have accidents as parts.

5. Therefore, the mental state accidents in the disembodied state have something other than the soul as their subject.

6. The only two candidates for a subject of mental state accidents are the soul and the person.

7. Therefore, the mental state accidents in the disembodied state have the person as their subject.

8. Therefore, the person exists in the disembodied state.

(This argument is a way of turning Jeremy Skrzypek’s accident-based defense of survivalism into a positive argument for survivalism. Maybe Skrzypek has already done this, too.)

The argument is slightly complicated by the fact that Thomists accept the possibility of subjectless accidents existing miraculously (in the Eucharist). Nonetheless, I do not know of any Thomists who think the disembodied state is such a miracle. Given that Thomists generally think that the survival of the soul after death is not itself miraculous, they are unlikely to require the miracle of subjectless accidents in that case, and hence will accept premise 3.

Premise 2 is common ground between survivalists and corruptionists, as both agree that there is suffering in hell and purgatory and joy in heaven even in the disembodied state.

I think the controversial premises are 1 and 4. I myself am inclined to deny the conjunction of the two premises (even though I think survivalism is true for other reasons).

Premise 1 is a core assumption of compositional metaphysics, and compositional metaphysics is one of the main attractions of Thomism.

One reason to accept premise 4 is that the soul is the form of the human being, and one of the main tasks for forms in Aristotelian metaphysics is to unify complex objects. But if forms are themselves complex, then they are also in need of unification, and we are off on a regress. So forms should be simple, and in particular should not have accidents as parts.

Another reason to accept 4 is that if the soul or form has mental state accidents as parts, it becomes very mysterious what else the form is made of besides these accidents. Perhaps there is the esse or act of being. But it seems wrong to think of the form as made of accidents and esse. (I myself reject the idea that objects are “made of” their parts. But the intuition is a common one.)

Pascal's wager and decision theory

From time to time I find myself musing whether Pascal’s Wager doesn’t simply completely destroy ordinary probabilistic decision theory. Consider an ordinary decision, such as whether to walk or bike to work. There are various perfectly ordinary considerations in favor of one or the other. Biking is faster and more fun, but walking is safer and provides more opportunity for thought.

But in addition to all these, there are considerations having to do with one’s eternal destiny. It is hard to deny that there is a positive probability that we will have an eternal afterlife and that our daily choices will affect whether this afterlife is happy or miserable. But even tiny differences in the probability of eternal happiness infinitely swamp all the ordinary considerations in the decision whether to walk or bike. If the opportunity for more leisurely reflection afforded by walking even slightly increases one’s chance at eternal happiness, that infinite contribution to expected utility completely overcomes all the ordinary considerations. But on the other hand, biking would allow one to arrive at work earlier, and thereby take on a larger share of work burdens, which would lead to growth in virtue, and increase chances of eternal happiness. So in the end, it seems, many of our ordinary everyday decisions end up turning into exercises in balancing tiny differences in the probability of eternal joy, as these swamp all the other ordinary consdierations. And that seems wrong.

One move here is to say that the question of how ordinary approximately morally neutral decisions affect the afterlife is one that we have so little information on that we should bracket the infinities, and just focus on the finite stuff we know about. But on the other hand, does that make sense? After all, perhaps we should put all our mental energies into figuring out this stuff that we have so little information on, as the infinities in the utilities swamp everything else?