A nuanced compatibilism and the problem of heavenly freedom

The problem of heavenly freedom is the apparent tension between these two claims:

1. The blessed in heaven are free

2. The blessed in heaven cannot sin.

One solution is compatibilism, but as Pawl and Timpe note, this undercuts the Free Will Defense.

But there is another move. One can be a compatibilist and say that while one can have freedom without the ability to do otherwise, nonetheless freedom with the ability to do otherwise is better. If one accepts this version of compatibilism, one can affirm (1) and (2) while yet offering a Free Will Defense.

This, however, leads to an obvious riposte: If freedom with the ability to do otherwise is better, why don’t we have that kind of freedom in heaven? Isn’t heaven supposed to be the best state for us?

One can, however, add another nuance. There are some activities that it is good to have done at some point, but repetition significantly diminishes the value. It is of some value to have read The Murder of Roger Ackroyd. To re-read it, not so much. Or for a religious example, think of the Hajj. Suppose freedom with the ability to do otherwise is like that. Perhaps, then, it is valuable to have made the choice for God with the ability to do otherwise. But a repeat of that choice is of rather lesser value. So much lesser, that if on earth one has made the choice for God with the ability to do otherwise, in heaven the value of doing so again is outweighed by the value of making guaranteed righteous choices.

This is not too different from Pawl and Timpe’s preferred solution of allowing for derivative freedom in heaven. But there may be an advantage to the above solution. Pawl and Timpe’s solution doesn’t solve the problem of infants who go to heaven without being able to make a free choice in this life—they don’t seem to have derivative freedom. (One of my undergraduate students has ably pressed this problem.) The nuanced compatibilism I have suggested can help with that: the infants in heaven genuinely have freedom. Granted, their death has denied them one of the goods proper to earthly life—the good of choosing righteousness with the ability to do otherwise. But that they have lost something by their untimely death is indeed rather intuitive.

We might ask: But why wouldn’t God then give them a chance to make a decision with ability to choose otherwise after death? Wouldn’t that be better? In one respect, it would indeed be better: in the respect of choosing with the freedom to choose otherwise. But in another respect, it would be less good: in the respect of having the risk of choosing wrongly. These are incommensurable considerations, and God can reasonably follow either one.

Granted, this move weakens the force of the Free Will Defense. We can no longer say that it’s better all things considered for God to give us the kind of freedom that allows us to reject him. For while that’s a better kind of freedom, it comes with an incommensurable cost—the risk that we will reject him. However, we can still say that God can rightly choose to follow either of the incommensurable considerations. In our case, he has opted to give us the better freedom despite the risk; in the case where he has taken some infants to himself, he opted for the guarantee of freedom being rightly used.

I don’t endorse the above solution. But I think it’s possible.

Predictability and epistemic utility

You’re thinking whether to become an assembly-line worker or an artist. Then you reflect on the value of knowledge. And you become a factory worker, on the grounds that if you become an assembly-line worker, you will know what you’ll be doing every working day of your future, but if you’re an artist, your activities will be unpredictable.

Some remarks. First, there is something perverse about using the value of knowledge in this way. The normal way to pursue the value of knowledge is to find out things that are independent of your pursuit. But here you are pursuing knowledge by making there be less to know about the world (or your world). Yet, paradoxically, it sure seems like the line of thought above makes sense.

Second, the my initial story depends on Molinism being false. For if there are comprehensive subjective conditionals of free will, then by becoming an artist you get to know the conditionals about what you would do in the various artistic situations you’re in. But on the assembly line story, you don’t get to know these. So the Molinist doesn’t have the paradox. I suppose that’s a bit of evidence for Molinism.

An argument against the Thomistic primary/secondary causation account of strong providence

Most people agree that one cannot have circularity in the order of explanation when one keeps the type of explanation fixed. Some like me think one cannot have circularity in the order of explanation at all. I argued for this thesis in my previous post today. Now I want to draw an interesting application.

On one influential (and I think exegetically correct, pace Eleonore Stump) reading of Aquinas, God decides what our free choices will be. Our free choices cannot be determined by created causes, but they are determined by God. This is because God’s causation is primary causation which is of a different sort from the secondary causation which is creaturely causation. God can primarily cause you to freely secondarily cause something, and this is how providence and free will interact. Often the analogy between an author and a character is given: the author decides what the character will freely do and this does not infringe on the character’s freedom.

But now observe this (which was brought home to me by a paper of one of our grad students). On this picture, God will presumably sometimes providentially make earlier actions happen because of later ones. Thus, God may want you to perform some heroic self-sacrifice in ten years. So, right now God prepares you for this by having you freely engage in small self-sacrifices now. In the “because” corresponding to the explanatory order of providence and primary causation, we thus have:

1. You engage in small self-sacrifices because you will engage in a great self-sacrifice.

However, divine primary causation does not undercut secondary causation, and we have the standard Aristotelian story of habituation at the level of secondary causation in light of which we have:

2. You will engage in a great self-sacrifice because you are engaging in small self-sacrifices.

These explanations form a heterotypic explanatory loop (i.e., we have explanations of two different sorts in opposite directions). But if I am right that no explanatory loop is possible, the above story is not possible. However, there is nothing to rule out the above story if the above Thomistic account of primary and secondary causation’s role in providence is correct. Hence, I think we should reject that account.

Three fixity principles

In debates about free will and foreknowledge as well as about compatibility and incompatibilism, fixity-of-history theses come up. Here is such a thesis:

1. If a decision is causally or logically necessitated by the history behind the decision, then one could not have decided otherwise.

But now we have a crucial question as to what is meant by “the history behind the decision”. There are at least two takes on this. On the temporal version, the history behind the decision is the sum total of what happened temporally prior to the decision. On the causal version, the history is the sum total of what happend causally prior to the decision.

This is not just a nitpicking question. Linda Zagzebski for instance nicely shows that if we go for the causal-history version of (1), then the main argument for the incompatibility of free will and foreknowledge does not get off the ground assuming God’s forebelief is not causally prior to the action. On the other hand, if we go the temporal-history version, then we have a prima facie argument for such incompatibility (though I think it’s blockable).

I am pretty confident that we should go for the causal-history version, and this has to do with the fact that the temporal-history version is not strong enough to capture our fixity intuitions. Suppose that we live in a world with simultaneous causation—say, a Newtonian world with rigid objects such that if you push object A and A pushes B, then B begins to move at exactly the same time as you start pushing (rather than with a delay caused by then need for a compression wave traveling through nonrigid materials at less than the speed of light). Then we could imagine cases where someone’s decision is causally necessitated by something outside the agent that is simultaneous with the decision. Such causal necessitation would just much make it true that one could not have decided otherwise as would causal necessitation by something in the past.

Furthermore, if backwards causation is possible, then a neurosurgeon in the future who used a backwards-causing machine to determine your decision would clearly prevent you from deciding otherwise, even though the neurosurgeon’s action was not in the temporal history. We may not believe backwards causation is possible, but it is clear that if it were possible, then deterministic backwards causation would be just as threatening for free decisions as deterministic forwards causation. This shows that causal determination is indeed a threat.

Of course, my above argument only shows that if we need to choose between the causal and temporal history versions of (1), we should definitely go for the causal one. But perhaps we don’t need to choose. We could accept both versions. But if we think we accept both versions, I think what we really should accept is an even stronger principle, where “history” is causal-cum-temporal (cct). On that stronger principle, event A counts as in the cct history of event E provided that it is either temporally or causally prior to E. The resulting fixity principle is pretty strong principle, but also a bit gerrymandered. And I think accepting this principle not that plausible, because the much simpler causal version captures our intuitions about all the ordinary cases (not involving God, or backwards or simultaneous causation), since in all ordinary cases causal and temporal history coincide, and we should not go for a more complex principle without pretty good reason.

Causal histories and freedom

Linda Zagzebski proposes the plausible principle that one is able to ϕ only if ϕing is compatible with one’s causal history relevant to ϕing.

Suppose Alice is considering whether to rob a bank. While she is doing so, God loudly announces to her nearby friend Bob that Alice will not rob the bank. God’s announcement is in Swahili, which Bob knows and Alice used to know in childhood but completely forgot. But the sound of the language her loving parents spoke to her as a child leads to Alice putting more emphasis on virtue in her deliberation, and she freely decides not to rob the bank.

Since God cannot lie, and since God’s announcement is a part of Alice’s causal history in her deliberation, Alice’s robbing the bank is incompatible with causal history and by Zagzebski’s principle, Alice cannot rob the bank. Yet it is unclear how God’s announcement removes her freedom to rob. After all, had God announced in Swahili that Alice will have breakfast, that would have influenced her deliberation in the same way, and yet obviously she would still have been free to rob the bank. But since Alice doesn’t know Swahili, the content seems causally irrelevant.

I think there are two ways out of this. First, we might cut events very finely. There is (a) God’s saying something or other in Swahili and there is (b) God’s saying in Swahili that Alice will not rob the bank. To determine the causal history, we pare away from the events all that’s causally irrelevant, and so we include (a) but not (b) in the causal history.

Alternately, we might say this. Whether or not Alice knows Swahili, her decision is affected by the detailed facts about the sounds in God’s announcement. Indeed, by essentiality of origins, her deliberation is a numerically different process because of the difference of sounds. And now we can say that God cannot make the announcement, because doing so would result in a circularity in the explanatory order: God would be making the announcement because Alice is not going to rob and Alice is not going to rob because God is making the announcement because she is not going to rob. So it is not so much that the announcement takes away God’s freedom, but that God cannot produce explanatory circularities.

It’s worth noting that Molinism does not seem to help. Sure, the subjunctive conditional of free will

1. Were God to announce in Swahili that Alice won’t rob, Alice wouldn’t rob

is true. But it is necessarily true independently of Molinism!

From three or four problems of omniscience down to one

The three most influential problems of omniscience are:

1. Boethius’ problem of foreknowledge: What is known is necessarily so, and thus if God knows what you will do, you will necessarily do it.

2. Pike’s problem of foreknowledge: If you can act otherwise, you can thereby make it be that either God didn’t exist or that God wasn’t omniscient or that God had believed otherwise than God actually did, and you just can’t do that.

3. The simplicity and knowledge of contingents problem: If the world had been different, God’s beliefs would have been different, which implies that God’s beliefs are accidents of God, contrary to divine simplicity.

Of these, (1) is fully solved by Boethius/Aquinas by distinguishing between necessity of consequence and necessity of consequent. The problem in (1) is just a simple matter of a fallacy of modal scope ambiguity. It’s a non-problem.

I now want to argue that the most widely accepted solution to (3) also solves (2).

This solution, likely already known to Aquinas, is that God’s belief in contingent facts is partly extrinsically constituted by creatures, and all the contingency is on the created side. For instance, God’s belief that there are zebras is grounded in essential facts about God that do not vary between possible worlds and the actual existence of zebras, which only obtains in some possible worlds.

Suppose we apply this solution to (3). Then God’s belief that you will ϕ at t is partly grounded in your ϕing at t and partly in essential facts about God. At this point it is obvious that:

4. If you were not to ϕ at t, God wouldn’t have believed you would ϕ at t.

For the contingent part of the grounds of God’s believing that you would ϕ at t is your actually ϕing at t, so when you take that away, God’s belief goes away. And if instead you ψ at t, your action thereby constitutes the contingent part of the grounds of God’s believing that you would ψ at t, and so:

5. If you were to ψ at t, God would have believe you would ψ at t.

If God’s past belief is partly constituted by our actions, it is no surprise that there is counterfactual dependence between our actions and God’s past belief. In other words, the classical theist who accepts divine simplicity has a way out of Pike’s argument that is motivated completely independently of considerations of time and freedom, namely by embracing counterfactuals like (4) and (5) that Pike considers absurd.

Of course the extrinsic constitution of divine beliefs is somewhat hard to swallow, notwithstanding excellent work by people like W. Matthews Grant to make it more plausible (I myself have swallowed it). But once we do that, problem (2) is gone, and problem (1) was never there as it was based on a fallacy.

There is a fourth problem, a more recondite one, which is about the incompatibility between God’s knowledge of what time is objectively present (assuming the A-theory of time) and divine immutability. Probably the most extensive pressing of this problem is in Richard Gale’s On the Nature and Existence of God. But Aquinas (according to the very plausible interpretation by Miriam Pritschet in an excellent paper I heard yesterday at the ACPA) responds to the fourth problem precisely by using the extrinsic constitution of God’s knowledge of continent facts (indeed this is why I said that the solution to the simplicity problem was likely known to Aquinas). So even that fourth problem reduces to the third—or just doesn’t get off the ground if the B-theory of time is true.

Freedom: a problem for presentism and growing block

A number of people have told me that they have the intuition that a four-dimensional picture of reality like that in the B-theory undercuts free will.

I want to suggest that there is one way in which it is a presentist picture of temporal reality that undercuts free will. (A similar argument applies to growing block, but curiously enough not to shrinking block.)

Assume that open future views are false: there are always determinate facts about contingent future events. (If your reason for thinking that four-dimensional theories undercut free will is because you are an open futurist, then you won’t be impressed by what I say.) Suppose it is a fact that tomorrow morning I will have oatmeal for breakfast. On presentism, this fact can only be grounded in what is present, since on presentism, what is present is all there is. Maybe it’s grounded in the present existence of a future-tensed fact or maybe it’s grounded in my having a future-tensed property of being such that I will eat oatmeal in nine hours. But in any case, things right now are already such as to ground and guarantee that I will have oatmeal for breakfast. Moreover, this was already true five minutes ago—five minutes ago, things were also already such as to ground and guarantee that I will have oatmeal for breakfast tomorrow. This sure feels like it should undercut free will! It seems pretty intuitive that freedom isn’t compatible with there existing grounds that guarantee the action prior to the choice.

On the other hand, on a four-dimensional view while it is a fact that I will eat oatmeal for breakfast tomorrow, the grounds of this fact are not located in the present—and were never located in the past. Rather, the grounds of this fact are where they should be—at tomorrow morning. How things are on the present slice of reality, or on past slices, does not determine (assuming indeterminism) what I will have for breakfast tomorrow. That’s left for tomorrow.

The neatest way out for the presentist is to deny with Merricks that contingent truths about the future and past have any grounds. But that’s also costly.

After writing the above, I came across this related paper by Hunt. No time to revise right now to see what similarities or differences there are.

A Stoic thesis

Combine these two rather Stoic theses:

1. No one can make the fully virtuous person worse off

2. Doing what is morally wrong always makes you worse off

and you get:

3. No one can make the fully virtuous person do what is morally wrong.

It is going to be crucial to this post that (3) includes cases of inculpably doing what is morally wrong. I myself think (1) is false, say for the that Aristotle cites, namely that severe pain makes even the virtous worse off. But nonetheless I want to defend (3).

It may be possible to first destroy a virtuous person’s virtue, and then get them to do what is morally wrong. Hitting someone on the head or brainwashing them can severely damage the psyche in a way that can remove the rational habits that constitute virtue. I do not count this a counterexample to (3), because in this case when the victim performs the wrong action, they have previously lost their virtue.

One might try to rule out the case of head injury and brainwashing by restricting (3) to culpable wrongs, but I don’t want to do that. I want to defend (3) in the case of inculpable wrongs, too.

A consequence of (3) is a fairly strong source incompatibilism about our action. Not only is it that neural manipulation cannot make you perform a free action, but it cannot make you perform an action. This fits well with dualism, but does not require it, because it might be that brain states that constitute acts of will have to have functional characteristics incompatible with outside control.

As a final clarification, I understand “making” as reliable, but not necessarily 100% reliable. Someone with significant free will cannot be 100% reliably made to do wrong, even if they are not virtuous. But at the same time, while a fully virtuous person cannot be reliably tempted away from right action, they might still have significant free will and be able to do wrong, so a temptation might unreliably get them to do wrong. Furthermore, I am thinking of “making” on something like a specific occasion. Thus, perhaps, if you tempt a virtuous person a million times, while restoring their brain to the pre-temptation state between temptations, by the law of large numbers you can expect them to fail at least once.

Let’s think about (3) some more.

Threats aren’t going to reliably get the fully virtuous person do the wrong thing. Sometimes, it is reasonable to bow to a threat. If someone holds a gun to your head and tells you to cover the side of your neighbor’s house with a giant smiley face graffiti, it’s reasonable to go along with it. But morality is reasonable, and where it is reasonable to bow to a threat, doing so is not only not culpable but simply not wrong. Indeed, it would typically be a failure of respect for human life to refuse to paint the graffiti when one’s life is threatened. In cases, however, where it is unreasonable to knuckle under, the fully virtuous person will reliably withstand the threat.

Physical control of another’s body or brain isn’t going to produce morally wrong action, because it doesn’t produce action at all. It is not wrong to kill someone by being pushed off a cliff on top of them, because it’s not an action to fall off a cliff. Similarly, if someone implants a remote control for one’s muscles, even if in the brain, then the resulting muscle spasms are not actions, and hence are not morally wrong actions.

Cases of omissions are interesting. It is easy to make someone fail to do what they promised, say by imprisoning them. If one thinks that such a failure counts as an inculpable wrongdoing, and if (3) is supposed to apply to omissions as well, then we have a counterexample. I do want (3) to apply to omissions. But I think that all that’s morally required by a promise is a reasonable amount of effort—where what counts as reasonable depends on the case. If you promise to come to a party but are in the hospital after a serious accident, it’s not morally required—indeed, it’s morally forbidden—that you rip the IV out of your arm and drag yourself on hands knees to the party. Indeed, (3) is a part of my reason for thinking that promises only require reasonable effort, so this is the first example where I have (3) giving us evidence for a substantive moral thesis. I think something similar is true in the case of commands, legislation and the like. You can’t ask for more than reasonable effort! Asking for more adds insult to injury. The parents whose children starve because the parents were unjustly imprisoned have not done wrong in failing to feed them.

Cases of ignorance are also interesting. If Alice is serving wine to her guests and Bob pours poison in the wine whe she isn’t looking, some might say that Alice has done wrong in poisoning her guests. Certainly, actual-result act utilitarianism implies this. But so much the worse for actual-result act utilitarianism. It is much better to say that Alice has done no wrong, as long as it was reasonable for her to have no suspicion of Bob. Cases of ignorance of through-and-through moral facts, on the other hand, are arguably incompatible with full virtue.

Where I think (3) becomes most interesting is in cases where we have a normative power over what is right or wrong for another to do. Using our normative powers, we can make someone who would otherwise have done wrong not be doing wrong. There is a story of a hasid whose house is being robbed, and when the thief is carrying his property away, the hasid yells: “I renounce my property rights.” In doing so, the hasid releases the thief from the duties of restitution, and makes it be the case that the thief is not sinning by continuing to carry the goods away.

Are there cases where we can use our normative powers to reliably make someone do wrong? Definitely. You can know that someone under your authority will very likely refuse to follow a certain command, and you can nonetheless issue the command. But this is obviously a case of someone who is lacking full virtue.

I think the best bet for using our normative powers to reliably make someone perfectly virtuous do wrong is when our exercise of normative powers creates a duty but does so in a way that the perfectly virtuous person does not know about. For instance, one might command the fully virtuous person in circumstances where they will likely not hear the command. Or one might pass legislation that they won’t know about. There are two ways to defend (3) in these cases. The first is to have a communication condition on commands and legislation—they are only morally binding when person subject to them either is informed about them or ought to be informed about them. The second is to say that all that’s morally required is that one make a reasonable effort to obey commands and laws in general, not that one make a reasonable effort to obey each specific command or law (since if one doesn’t know about a command or law, one doesn’t need to make a reasonable effort to obey it). I somewhat prefer the first option of a communication condition—the vicious lawbreaker does wrong in disobeying a specific law, and not just law in general (though according to James 2:10, they are also doing wrong in disobeying law in general).

In any case, I think (3) puts some significant constraints on the shape of moral obligation and the nature of action, but these constraints seem defensible. Though maybe I am failing to notice some better counterexample.

Divine willing

A correspondent asked me how a simple God can choose. I've thought much about this, never quite happy with what I have to say. I am still not happy (nor is it surprising if "how God functions" is beyond us!) but the following helps me a little.

Suppose I am choosing between making a brownie or a smoothie, and end up making a smoothie. Then there are four stages with each stage causing the next:

1. Deliberation between brownie and smoothie (and any other options).

2. An internal intention for a smoothie.

3. Physical movements.

4. Output: smoothie!

At Stage 1, I am still open between brownie and smoothie. Starting with Stage 2, I am internally set on the smoothie, and at that point I become morally responsible for setting myself on the smoothie.

Now one great thing about God’s power is that God doesn’t need means: he can produce effects directly.

In particular, God will omit stage 3: he doesn’t need moving limbs (nor anything else beyond himself) to produce the smoothie in the way that I need them.

Now suppose we apply perfect being theology to God. It’s a perfection of power not to need means. But stage 3 is not the only means in the above story: stage 2 is also a means. If God really doesn’t need means, then stage 2 will also be omitted in God, and we will have the two (non-temporal) stage production:

1. Deliberation between brownie and smoothie (and infinitely many other options).

2. Output: smoothie!

In particular, nowhere in this account is there an internal intention. It’s not needed: God acts directly on the external world.

We might ask: How does God know that he intends to create a smoothie? I think it’s by direct observation of the output, stage B in the divine case. (And God’s knowledge of contingencies is extrinsically grounded in the contingencies.)

If that sounds wrong, let’s ask how we know what we intend. Sometimes we know what we intend by introspection: by observing the internal intention of stage 2. But not always. Sometimes stage 2 is not conscious—we deliberate, we presumably form an internal intention but we are not directly aware of it, and then we act. When we deliberate whether to do something minor like whether to lift the left or the right hand, sometimes the first thing we are aware of is not an internal intention or specific act of will, but the movement of the hand itself. Thus, even in us, knowledge of our intentions is sometimes read off from stage 3.

Moreover, in some cases, for us, stage 3 is not distinct from stage 4. For we have bodies that we move, and sometimes—as in the hand-lift case—the output is the same as the physical movements. In some such cases, then direct knowledge of the merges stages 3 and 4 is how we know our own intention. There may even be rare cases where stage 3 and stage 4 are distinct, but in our consciousness, the knowledge of stage 4 comes first. Suppose I am deliberating whether to press the space-bar or the enter key in response to the computer saying “Press any key”. I choose to press the space-bar. It may well be that in the order of consciousness, I first feel the impact of my finger on the keyboard, then I discern the subtler kinesthetic sensation of my finger having moved through the air, and then only then do I realize what key I intended to press. I don’t know if this is how it happens—it’s too fast to be confident of phenomenology—but it seems like an intelligible possibility.

In any case, there is nothing absurd about knowing an intention by direct observation of the output.

While writing the above, it occurred to me that perhaps we shouldn’t be all that confident that we always have a stage 2. In the cases where our knowledge of what we intend comes from knowledge of stage 3 or 4, and we do not have direct conscious access to an internal act of intention, the internal act of intention is a mere theoretical posit. Perhaps that theoretical posit is correct, but perhaps it is not. If it is not, then one can intend a specific output without having an internal specified act of intention.

Open Theism and divine promises

Open Theist Christians tend to think that there are some things God knows about the future, and these include the content of God’s promises to us. God’s promises are always fulfilled.

But it seems that the content of many of God’s promises depends on free choices. For imagine that all the recipients of God’s promise freely choose to release God from the promise; then God would be free not to follow the promise, it appears, and so he could freely choose not to act in according to the promise. Thus there seems to be a sequence of creaturely and divine free choices on which the content of the promise does not come about.

This argument may not work for all of God’s promises. Some of God’s promises are covenants, and it may be that covenants are a type of agreement in which neither party can release the other. There may be other unreleasable promises: perhaps when x promises to punish y, that’s a promise y cannot release x from. But do we have reason to think that God makes no “simple promises”, promises other than covenants and promises of punishment?

I do not think this is a definitive argument against open theism. The open theist can bite the bullet and say that God doesn’t always know he will fulfill his promises. But it is interesting to see that on open theism, God’s knowledge of the future is even more limited than we might have initially thought.

A really bad moral dilemma

Here would be a really bad kind of moral dilemma:

• It is certain that unless you murder one innocent person now, you will freely become a mass murderer, but if you do murder that innocent person, you will freely repent of it later and live an exemplary life.

If compatibilism is true, such dilemmas are possible—the world could be so set up that these unfortunate free choices are inevitable. If compatibilism is false, such dilemmas are impossible, absent Molinism.

We might have a strong intuition that such dilemmas are impossible. If so, maybe that gives us another reason to reject compatibilism and Molinism.

Moral conversion and Hume on freedom

According to Hume, for one to be responsible for an action, the action must flow from one’s character. But the actions that we praise people for the most include cases where someone breaks free from a corrupt character and changes for the good. These cases are not merely cases of slight responsibility, but are central cases of responsibility.

A Humean can, of course, say that there was some hidden determining cause in the convert’s character that triggered the action—perhaps some inconsistency in the corruption. But given determinism, why should we think that this hidden determining cause was indeed in the agent’s character, rather than being some cause outside of the character—some glitch in the brain, say? That the hidden determining cause was in the character is an empirical thesis for which we have very little evidence. So on the Humean view, we ought to be quite skeptical that the person who radically changes from bad to good is praiseworthy. We definitely should not take such cases to be among paradigm cases of praiseworthiness.

Culpability incompatibilism

Here are three plausible theses:

1. You’re only culpable for a morally wrong choice determined by a relevantly abnormal mental state if you are culpable for that mental state.

2. A mental state that determines a morally wrong choice is relevantly abnormal.

3. You are not culpable for anything that is prior to the first choice you are culpable for.

Given these theses and some technical assumptions, it follows that:

4. If determinism holds, you are not culpable for any morally wrong choice.

For suppose that you are blameworthy for some choice and determinism holds. Let t₁ be the time of the first choice you are culpable for. Choices flow from mental states, and if determinism holds, these mental states determine the choice. So there is a time t₀ at which you have a mental state that determines your culpable choice at t₁. That mental state is abnormal by (2). Hence by (1) you must be culpable for it given that it determines a wrong choice. But this contradicts (3).

The intuition behind (1) is that abnormal mental states remove responsibility, unless either the abnormality is not relevant to the choice, or one has responsibility for the mental state. This is something even a compatibilist should find plausible.

Moreover, the responsibility for the mental state has to have the same valence as the responsibility for the choice: to be culpable for the choice, you must be culpable for the abnormal state; to be praiseworthy for the choice, you must be praiseworthy for the abnormal state. (Imagine this case. To save your friends from a horrific fate, you had to swallow a potion which had a side-effect of making you a kleptomaniac. You are then responsible for your kleptomania, but in a praiseworthy way: you sacrificed your sanity to save your friends. But now the thefts that come from the kleptomania you are not blameworthy for.)

Premise (2) is compatible with there being normal mental states that determine morally good choices, as well as with there being normal mental states that non-deterministically cause morally wrong choices (e.g., a desire for self-preservation can non-deterministically cause an act of cowardice).

What I find interesting about this argument is that it doesn’t have any obvious analogue for praiseworthiness. The conclusion of the argument is a thesis we might call culpability incompatibilism.

The combination of culpability incompatibilism with praiseworthiness compatibilism (the doctrine that praiseworthiness is compatible with determinism) has some attractiveness. Leibniz cites with approval St Augustine’s idea that the best kind of freedom is choosing the best action for the best reasons. Culpability incompatibilist who are praiseworthiness compatibilists can endorse that thesis. Moreover, they can endorse the idea that God is praiseworthy despite being logically incapable of doing wrong. Interestingly, though, praiseworthiness compatibilism makes it difficult to run free will based defenses for the problem of evil.

Do I need to be aware of what I am intending if I am to be responsible?

I am going to argue that one doesn’t need to be conscious of intending to ϕ in order to be responsible for intending to ϕ.

The easiest version of the argument supposes time is discrete. Let t₁ be the very first moment at which I have already intended to ϕ. My consciousness of that intending comes later, at some time t₂: there is a time delay in our mental processing. So, at t₁, I have already intended to ϕ. When I have intended to ϕ, I am responsible for ϕ. But now suppose that God annihilates me before t₂. Then I never come to be aware that I intended to ϕ, but yet I was already responsible for it.

Here are three ways out:

1. I am not yet responsible at t₁, but only come to be responsible once I come to be aware of my intention, namely at t₂.

2. My awareness is simultaneous with the intention, and doesn’t come from the intention, but from the causal process preceeding the intention. During that causal process I become more and more likely to intend to ϕ, and so my awareness is informed by this high probability.

3. My awareness is a direct simultaneous seeing of the intention, partially constituted by the intention itself, so there is no time delay.

Open futurism does not save free will

In yesterday’s post, I showed that if an open-futurist is impressed by a certain plausible-sounding logical fatalism argument based on bivalence, and hence opts for truth gaps, then they should also be impressed by another logical fatalism argument based not on bivalence but on truth gaps.

However, there was a weakness to my logical fatalism argument. It was based on the principle:

1. If something true now is incompatible with it’s being true that p, then p is not within your power.

But perhaps our open futurist will deny (1) on the grounds that a present action can be within our power, even though it is presently true that we will do it. (I think this is a problematic concession for the open futurist to make, but let’s bracket that.) Such an open futurist will instead run arguments based on:

2. If q is a past-tensed truth, and q is incompatible with p, then p is not within your power.

Well, here is perhaps a truth value gap counterexample to (2).

• Alice freely ϕs at 5 pm.

• At 1 pm it is true that no indeterministic events will happen between 2 and 4 pm.

(To get the second part, we can suppose that the laws of nature are such that they only allow indeterministic events after 4:30 pm each day, or maybe God just promises not to allow any indeterministic events between 2 and 4 pm.)

So, consider the following complicated past-tensed statement, which is true at 3 pm:

• q: Two hours ago [i.e., at 1 pm], it was true that no indeterministic events would happen between an hour ago [2 pm] and an hour from now [4 pm], while half an hour ago [2:30 pm], it was neither true nor false that Alice freely ϕs at 5 pm.

Now, on the open futurist’s view, time-indexed propositions can only gain truth value as the result of indeterministic events. It logically follows from the ban on indeterministic events between 2 and 4 pm that any time-indexed proposition that was neither true nor false at 2:30, is also neither true nor false at 3 pm. Or to put it in a tensed way, q entails:

3. It is neither true nor false that Alice ϕs at 5 pm.

But (3) is logically incompatible with Alice ϕing at 5 pm, since, necessarily, if Alice ϕs at 5 pm, then it’s true that Alice ϕs at 5 pm. Since q entails (3), it follows that:

4. q is logically incompatible with Alice ϕing at 5 pm.

Hence it follows from (2) (since q is a past tensed truth) that at 3 pm it is true to say:

5. It is not within Alice’s power that Alice ϕs at 5 pm.

Now, I said that “perhaps” this was a counterexample to (2). Besides objecting to the Tarski T-schema, there is one powerful response an open futurist can make. They can just embrace (5) and say: it’s only at 5 pm, or shortly prior to it, that it comes to be within Alice’s power to ϕ.

But I think the open futurist’s intuitions behind (2) also support:

6. If q is a past-tensed truth and p is time-indexed, and q is incompatible with p, then p will never be within your power.

(The reason for the restriction to time-indexed p is to avoid this counterexample. Let q be the proposition that there was no wine in the world a minute ago. Let p be the proposition that you are drinking well-aged wine. Then p and q are incompatible. But if you make wine, and age it, then it can come to be the case that drinking well-aged wine is in your power.)

And now (4) and (6) imply:

7. It will never be within Alice’s power that Alice ϕs at 5 pm,

which is just false in our story, since she does ϕ at 5 pm! (Alternate phrasing: replace “within Alice’s power” with “up to Alice”.)

What about open futurists who instead of supposing a truth value gap think that statements about contingent future events are all false? Well, such open futurists will not accept q (at 3 pm). But they will accept:

• q′: Two hours ago [i.e., at 1 pm], it was true that no indeterministic events would happen between an hour ago [2 pm] and an hour from now [4 pm], while half an hour ago [2:30 pm], it was false that Alice freely ϕs at 5 pm.

Again, on their view, time-indexed propositions only change truth value when indeterministic events happen. Thus, q′ entails that presently (i.e., at 3 pm) it is still false that Alice freely ϕs at 5 pm. And the rest of my argument goes through.

So it pretty much seems like I’ve shown that the only person who can accept a principle like (6) is someone who doesn’t believe in the possibility of free will.

Maybe what this is really an argument for is that the open futurist needs to deny the T-schema, which I had used to argue that if something is incompatible with it’s being true that Alice will ϕ at 5 pm, then it’s incompatible with Alice ϕing at 5 pm. Some open futurists do do that (Keith DeRose, for instance; I wonder now: do they do it because of an argument like this one?)

I have to confess a nagging suspicion of an error somewhere. I already found one that I just corrected—I had to restrict (6) to time-indexed truths, which forced me to remove an argument that would work even without the T-schema.

Does denying bivalence get us out of the logical argument for fatalism?

Consider this seemingly standard argument for logical fatalism.

1. It is true that you will ϕ or it is true that you will not ϕ.

2. If something true now is incompatible with it’s being true that p, then p is not within your power.

3. If you are free with respect to ϕing, then it is within your power that you will ϕ and it is within your power that you not ϕ.

4. That you will ϕ and that you will not ϕ are incompatible.

5. So, if it is true that you will ϕ, then it is not within your power that you will not ϕ. (2, 4)

6. So, if it is true that you will ϕ, then you are not free with respect to ϕing. (3, 5)

7. Also, if it is true that you will not ϕ, then it is not within your power that you will ϕ. (2, 4)

8. So, if it is true that you will not ϕ, then you are not free with respect to ϕing. (3, 7)

9. So, you are not free with respect to ϕing. (1, 8)

Many open futurists want to refute arguments for logical fatalism by supposing that in cases of freedom, that you will ϕ is indeterminate (and hence neither true nor false), and that you will not ϕ is also indeterminate, which allows them to deny premise 1 of the above argument.

But now consider this argument.

10. It is now indeterminate that you will ϕ.

11. Necessarily, p if and only if it is true that p.

12. So, it is true that it is now indeterminate that you will ϕ. (10, 11)

13. That it is indeterminate that you will ϕ and that it is true you will ϕ are incompatible.

14. That it is indeterminate that you will ϕ and that you will ϕ are incompatible. (11, 13)

15. If something true now is incompatible with it’s being true that p, then p is not within your power.

16. If you are free with respect to ϕing, then it is within your power that you will ϕ.

17. So, that you will ϕ is not within your power. (10, 14, 15)

18. So, you are not free with respect to ϕing.

Premise 15 of this argument is the same as premise 2 of the first argument. Premise 16 is an even less controversial version of premise 3. So anybody who is impressed by the first argument will be impressed by premises 15 and 16. Premise 13 is obviously true, and is an immediate consequence of the fact that a proposition that is indeterminate is neither true nor false.

Premise 11 is the plausible Tarski T-schema (necessitated, because we can think of the T-schema as an axiom). It has been questioned, but it is still very plausible.

Finally, premise 10 is a commitment of our open futurist.

So, unless our open futurist denies the T-schema, the supposition of indeterminacy leads to fatalism just as determinacy did!

Suppose we deny the T-schema. Nonetheless, even without the T-schema to back them up, 12 and 14 are still plausible as they stand, and so we still have a pretty plausible argument for fatalism, at least one that should be plausible by the open futurist’s lights.

I am not an open futurist. I just get out of the arguments by denying 2 and 15. Easy.

Forced choice and deontology

Suppose the only way to save five innocent people is by killing one innocent person. The deontologist says: you must refrain.

But what you have a forced choice between killing one and killing five?

How could that be? I know of two ways.

First, a psychological block. Perhaps you are brainwashed into killing, but it’s left to your free will whether you are to choose one victim or five.

Second, you can tie the outcomes to cases where effort is required to maintain the status quo but where at least mental effort is needed to stop maintaining it.

As a little introspective experiment, I held my breath for 20 seconds. It took a mild to moderate amount of effort to do so, increasing towards the end of the time period. As our language of “holding” indicates, holding one’s breath is an action. But at the same time, it was clear at all the times that breathing would also be an action—a deliberate interruption of the holding that would also take a mental and physical effort.

Similarly, imagine you’re holding an extremely heavy suitcase. The on-going holding is an effort. But at the same time, to let the suitcase go would also be an effort: you would need to bend your knees to lower it to the ground, or at least move your fingers to release your grip.

In both the breath and suitcase cases, there is no such thing as refraining from action. Holding is an action and letting go is an action.

Very well, now imagine that an evildoer has set things up as follows. They informed you that if you don’t kill the one innocent, five innocents will die. And then they set up a machine that will shoot the one innocent if you let go of the suitcase in the next thirty seconds. What should you do?

If you hold on for thirty seconds, then your effort will ensure that overall four people will die. Even if we grant that you are not intending this tragic consequence, it is wrong to act in a way that produces such a consequence. Think about this in terms of Double Effect (I am grateful to one of our grad students for the connection): holding on to the suitcase has an evil consequence that is disproportionate to whatever goods are involved in holding on.

If you let go, however, then only one person will die. This seems better. But if that’s why you let go, then you are letting go in order that that one person’s death should prevent the deaths of the five. And that violates deontology.

Here is a tentative suggestion. Standard deontological principles have an unstated presupposition: refraining from action is possible. If refraining is impossible, the principles apply at best in modified form.

Choices on a spectrum

My usual story about how to reconcile libertarianism with the Principle of Sufficient Reason is that when we choose, we choose on the basis of incommensurable reasons, some of which favor the choice we made and others favor other choices. Moreover, this is a kind of constrastive explanation.

This story, though it has some difficulties, is designed for choices between options that promote significantly different goods—say, whether to read a book or go for a walk or write a paper.

But a different kind of situation comes up for choices of a point on a spectrum. For instance, suppose I am deciding how much homework to assign, how hard a question to ask on an exam, or how long a walk to go for. What is going on there?

Well, here is a model that applies to a number of cases. There are two incommensurable goods one better served as one goes in one direction in the spectrum and the other better served as one goes in the other direction in the spectrum. Let’s say that we can quantify the spectrum as one from less to more with respect to some quantity Q (amount of homework, difficulty of a question or length of a walk), and good A is promoted by less of Q and incommensurable good B is promoted by more of Q. For instance, with homework, A is the student’s having time for other classes and for non-academic pursuits and B is the student’s learning more about the subject at hand. With exam difficulty, A may be avoiding frustration and B is giving a worthy challenge. With a walk, A is reducing fatigue and B is increasing health benefits. (Note that the claim that A is promoted by less Q and B is promoted by more Q may only be correct within a certain range of Q. A walk that is too long leads to injury rather than health.)

So, now, suppose we choose Q = Q₁. Why did one choose that? It is odd to say that one chose Q on account of reasons A and B that are opposed to each other—that sounds inconsistent.

Here is one suggestion. Take the choice to make Q equal to Q₁ to be the conjunction of two (implicit?) choices:

1. Make Q at most Q₁

2. Make Q at least Q₁.

Now, we can explain choice (a) in terms of (a) serving good A better than the alternative, which would be to make Q be bigger than Q₁. And we can explain (b) in terms of (b) serving good B better than the alternative of making Q be smaller.

Here is a variant suggestion. Partition the set of options into two ranges R₁, consisting of options where Q < Q₁ and R₂, where Q > Q₁. Why did I choose Q = Q₁? Well, I chose Q over all the choices in R₁ because Q better promotes B than anything in R₁, and I chose Q over all the choices in R₂ because Q better promotes A than anything in R₁.

On both approaches, the apparent inconsistency of citing opposed goods disappears because they are cited to explain different contrasts.

Note that nothing in the above explanatory stories requires any commitment to there being some sort of third good, a good of balance or compromise between A and B. There is no commitment to Q₁ being the best way to position Q.

Compatibilism and servitude

Suppose determinism and compatibilism are true. Imagine that a clever alien crafted a human embryo and the conditions on earth so as to produce a human, Alice, who would end up living in ways that served the alien’s purposes, but whose decisions to serve the alien had the right kind of connection with higher-order desires, reasons, decision-making faculties, etc. so that a compatibilist would count them as right. Would Alice's decisions be free?

The answer depends on whether we include among the compatibilist conditions on freedom the condition that the agent’s actions are not intentionally determined by another agent. If we include that condition, then Alice is not free. But it is my impression that defenders of compatibilism these days (e.g., Mele) have been inclining towards not requiring such a non-determination-by-another-agent condition. So I will take it that there is no such condition, and Alice is free.

If this is right, then, given determinism and compatibilism, it would be in principle possible to produce a group of people who would economically function just like slaves, but who would be fully free. Their higher-order desires, purposes and values would be chosen through processes that the compatibilist takes to be free, but these desires, purposes and values would leave them freely giving all of their waking hours to producing phones for a mega-corporation in exchange for a bare minimum of sustenance, and with no possibility of choosing otherwise.

That's not freedom. I conclude, of course, that compatibilism is false.

Necessity and the open future

Suppose the future is open. Then it is not true that tomorrow Jones will freely mow the lawn. Moreover, it is necessarily not true that Jones will freely mow the lawn, since on open future views it is impossible for an open claim about future free actions to be true. But what is necessarily not true is impossible. Hence it is impossible that Jones will freely mow the lawn. But that seems precisely the kind of thing the open futurist wishes to avoid saying.