Agent causation

I have long identified as having an agent-causal theory of free will. But I have just realized that my Aristotelian take on agent causation is far enough from the most common agent-causal theories—those of Clare and O’Connor—that it may be misleading to talk of myself as accepting an agent-causal theory of free will.

Standard agent-causal theories distinguish between agent causation and event causation as two distinct and real things in the world. For instance, they hold that I agent cause my writing of this post but there is an event causal relation between my being in this armchair and the cushion being squished. Agent causation has the agent as the cause and event causation has an event as the cause. But I think in both cases the cause is the same: it is a substance, namely myself. I cause the writing of this post and I cause the cushion to be squashed. On the standard view on which agent causation is distinguished solely by the fact that the cause is the agent, both are cases of agent causation. But that would be misleading to say.

So, if we take seriously the Aristotelian account of causation as substance causation, we shouldn’t distinguish agent causation from other kinds of causation by whether the cause is an agent or something else. But we can still make the distinction. My writing this post is an actualization of a power of my will (or my practical rationality, if you prefer). My squashing the cushion is an actualization of the power of my weight. Agent causation is distinguished from other kinds of causation not by what does the causing, but how it does the causing. In agent causation, the substance causes by actualizing its will (and any substance with a will is an agent). In other kinds of causation, the substance causes by actualizing a different power.

So, I think the fundamental relation underlying causation is actually at least ternary: agent X causes event E by actualizing power P.

This neatly integrates agent causation with reasons causation. There are more and less proximate powers. I now have a nearly proximate power to speak Polish (some motivation might be needed to make it proximate). When I was an infant, I had a remote power to speak Polish, by having a nearly proximate power to learn Polish. When the power P in the causal relation is specified as the maximally proximate power, in the agential case, it is a maximally proximate power of the will. And a maximally proximate powers of the will are tied to reasons: it is only by having a reason to do something that I am able to will it (one wills under the guise of the good). So, my reasons for action supervene on the maximally proximate power for action. Thus, the ternary description of agent causation neatly includes the reasons.

Randomness and freedom

Consider cases where your decision is counterfactually dependent on some factor X that is not a part of your reasons and is outside of your (present or past) rational control. The kind of dependence that interests me is this:

• In the presence of X, you decided on A, but had X been absent, you would have decided on B on the basis of the same set of reasons.

It's important here that X isn't just an enabler of your making a decision, nor is it one of the reasons—your reasons are the same whether X is present or not—but is an extrarational difference-maker for your action.

As far as rationality is concerned, these are cases of randomness. It doesn't matter whether X's influence is deterministic or not: the cases are random vis-à-vis reason.

In these cases, the best contrastive explanation of your decision is in terms of your reasons and X. And the counterfactual dependence on X, which is outside of your control, puts your freedom into question.

I think many cases of conflicted decisions have the following property:

1. If determinism is true, then the case involves such counterfactual dependence on a factor outside of one's reasons and rational control.

But I also think that:

2. Some of these cases are also cases of responsibility.

It follows that:

3. Responsibility is compatible with such counterfactual dependence

or:

4. Determinism is false.

If (3) is true, then a fortiori the kind of causal undeterdetermination that is posited by event-causal libertarians does not challenge freedom.

I think the right conclusion to draw is (4). I think the counterfactual dependence here does indeed remove freedom. But I do not think the mere absence of a determiner like X is enough for freedom. Something needs to be put in the place of X. What? The agent! The problem with X is that it usurps the place of the agent. Thus I am inclined to think that freedom requires agent causation. I didn't see this until now.

Substance causation, agent causation and time

Aristotelians about causation think all causation is substance causation. Events are causes only derivatively. What does the real causing are substances. This should make Aristotelians very sympathetic to the use of agent causation in the theory of free will. And insofar as the theory of agent causation is just that the agent is the cause of free actions, the Aristotelian who believes that we are substances[note 1] is surely going to agree that we are the causes of our free actions, and we are both agents and substances, so the agent is the cause of her free actions.

So far so good. But there is more to agent causation in regard to free will. Typically, agent causalists invoke agent causation to solve problems such as the randomness problem for libertarians. Agent causation is what makes an action be genuinely one's own action rather than a random blip. But the Aristotelian's embrace of substance causation is too broad. For not only does the Aristotelian think that her free actions are caused by her, she also thinks her non-free actions are caused by her, and even things like the circulation of the blood, which isn't an action at all, are caused by her. Moreover, since she is an agent, she thus thinks all of these things are caused by an agent. But if agent causation metaphysically lumps free actions with non-free ones, and doesn't distinguish them metaphysically from the circulation of the blood, then agent causation can't do the job it's designed for. The Aristotelian believes in agent causation, of course, and may do so with good metaphysical reason, but this agent causation cannot be used to solve the problems that the free will theorists want it to solve.

This line of thought might lead some Aristotelians about causation to accept a version of Cartesian dualism on which we are souls. For then one might hold, contrary to Aristotle, that only our free actions are caused by us and that the circulation of the blood and so on is not caused by us, because we are immaterial beings whose only direct effects are in whatever the equivalent of the pineal gland on this theory will be. This is not in the Aristotelian spirit, though, and it leads to unhappy ethical conclusions (bodies as akin to property).

I think there is something else one should say here. One shouldn't say that agent causation just is causation by an agent. Rather, agent causation is causation by an agent qua agent. You cause your free actions qua agent and you circulate your blood qua mammal, though of course you are both agent and mammal. It is a bit odd to say that you don't perform your non-free actions qua agent, though. After all, how can there be an action without an agent? Aren't all actions, free or not, the actions of an agent qua agent? Maybe. But maybe the distinction is still of some help, for maybe the kinds of mere randomness we want to rule out with the distinction isn't an action at all when looked at more closely.

There is another issue around here. There needs to be more to substance causation than the simple structure substance x causes event E. For paradigmatic substances persist over a long time, but many of their effects happen only at particular times in their existence. And there is an explanation of why the substance causes an effect at one time or another. For instance, I caused oatmeal to be assimilated to me earlier to day because I was hungry. The explanation includes not just the substance, but a state of a substance (and maybe other substances—but that's for another day). I caused the assimilation of my breakfast not only qua agent, but qua hungry agent. Likewise, I circulate the blood not just qua mammal, but qua mammal with a brain stem that sends such-and-such electrical signals to the heart.

Apart from considerations of free will, then, Aristotelians should say that the structure of substance causation is something like: substance x qua in state S causes event E. If we have an Aristotelian constituent ontology, then the state will be a mode (an essence, a necessary accident or a contingent accident) of the substance, and the causation relation will be a ternary relation between the substance, the mode and the event.

But now that we have all of this detail in place, we can go back and ask whether we still get benefits of an agent causal theory in regard to free will. That's not so clear. For instance, when I qua hungry agent caused my breakfast to be assimilated to me, the work distinguishing this from non-actions like circulation is being done by my state of hungry agency. It is because this state is involved in the causation—and not just involved, but involved in the right way (my being a hungry agent could cause me to grow a tail if I was rigged the wrong way)—that I am acting. But event causalists can say something exactly parallel. What distinguishes my eating the breakfast from my circulating the blood is that the former is caused by the event of my being a hungry agent qua my being a hungry agent.

So I do not know that Aristotelian agent causalists can claim to do better than event causalists. In fact, for certain ends they might well want to join cause with the event causalists.

The randomness argument against compatibilism

The randomness argument insists that if our actions are not determined by our character, then they are a matter of chance and hence not free. This is the most powerful argument against libertarianism. I want to think about whether the argument presents a challenge to the compatibilism. (Compatibilists think freedom is compatible with determinism; libertarians think that (a) freedom is incompatible with determinism, and (b) there is freedom. I shall assume that freedom and responsibility go together, though a more careful examination will require dropping the assumption.)

The compatibilist had better have something to say about how on her preferred view of action the actions do not come out as "a matter of chance". As a warmup, observe that for the purposes of evaluating whether something is "a matter of chance" in our ordinary intuitive sense—which is presumably the sense at play in the argument—the question isn't whether the event is determined or not. We think of coin flips as a matter of chance, and we would think this even if it turned out that the outcomes of coin flips are determined by the precise values of the initial conditions, so a slightly different angle or velocity would produce a different outcome, and we do not have precise enough control over the angle or velocity. If our actions came from a deterministic coin flip in the head, they would be no less objectionably random than if they came from an indeterministic coin flip.

Now, add this observation. People sometimes make seriously responsible decisions where their desires are very close in strength. For instance, no one would get off on a murder charge just because her desire to do the right thing was almost as strong as her desire to inherit the money from her uncle. But when desires are sufficiently close in strength, then which desire is the stronger is "a matter of chance" in exactly the same sense in which a deterministic coin flip is "a matter of chance". What sense is that? It's the sense that when something depends on very fine differences in conditions, too fine for us to discern or control, it's "a matter of chance." Whether one is seriously responsible for murder or not responsible for murder at all shouldn't depend on such tiny differences in desires: if it depends on the fact that one's desire to commit the murder has strength 27.4837 while one's desire not to commit the murder has strength 27.4836, that's just as chance as a truly indeterministic choice would be.

I think three families of answers are available to the compatibilist.

Answer 1: Insist that as the strengths of desires get closer and closer, the degree of freedom and responsibility decreases to zero.

Objection: It doesn't! While perhaps we want to especially criticize the character of the person whose desire for murder was by far the stronger desire, the person for whom the desires were closely balanced, and who agonized over the decision is still seriously responsible, and not merely vanishingly responsible.

Answer 2: Go for Markosian's agent causal compatibilism.

Objection: I have some problems with Markosian's story, but once one goes for agent causation to respond here, one should allow the libertarian the courtesy of relying on agent causation to respond to the randomness argument as well. But if one does that, one loses the strongest argument against libertarianism.

Answer 3: The person whose desire for murder has strength 27.4837 and whose desire not to commit the murder has strength 27.4838 is still a pretty bad person, even if she does not commit murder. Thus, our moral evaluation of the character of the person who in fact commits the murder because her desire strengths are 27.4837 and 27.4836 for and against respectively and the person who does not commit the murder because her desire strengths are 27.4837 and 27.4838 should be very similar, all other things being equal. The latter person does not commit the murder, but that's just her luck, and it is not a very different kind of luck from the luck of a person who does not murder her uncle because he happens to die before she gets around to trying to kill him. In all of these cases, ordinarily, we are dealing with persons with a corrupt moral character, persons who can be appropriately criticized for that character. Largely for pragmatic reasons, however, we do not punish the person who does not commit the murder, whether because her uncle dies on his own or because her desire for murder is slightly the weaker.

Objection: I do not think this fits with how we ordinarily think about the situations. We think there is something significantly praiseworthy when someone with very closely balanced desires for good and evil does the right thing. Yet the difference between the significantly praiseworthy and the significantly blameworthy action shouldn't be "a matter of chance".

Here, the compatibilist pressing Answer 3 might bite the bullet and say that while for pragmatic reasons we praise the person with such closely balanced desires, perhaps in the hope that the balance is shifting in a good direction and our praise will push it further there, she is a really bad person. But I still doubt this: Is she morally really about as blameworthy as an actual murderer? It still seems that the blameworthiness in the action is objectionably a matter of chance, unless one takes Answer 2 as the way out.

In any case, let's explore this compatibilist answer some more. Even if the bullet has been bitten, it's only going to have any hope of working in cases where the person is responsible for her character. For if her character is not something she's responsible for, it might as well be a matter of chance (and indeed may be a matter of chance circumstances and genetics). In particular, it won't work for the first responsible choices a person makes. The story has to be a story of bootstrapping: one has a very low level of responsibility in the initial choices, but then as one's choices build up a character in concert with them, the level of responsibility rises.

I doubt this this bootstrapping. If the choice itself does not inject any new responsibility, as I think on a deterministic story it does not since it is simply an outcome of earlier character, then one is no more responsible for the up-built character than for the initial character. But that is a different and independent objection to compatibilism, I think.

In sum, I think the randomness argument isn't so much an argument against libertarianism as an argument against the possibility of freedom: it affects compatibilists and libertarians alike. Oh, and I think it doesn't work, either because of agent causation or because it fails to notice how well indeterministic explanation can be made to work.

Agent and substance causation

Some people think that events are never causes, except in a derivative sense.  It is substances that are causes (and when one or more substances are cause because they stands in some relations, then their standing these relations is an event, and we can derivatively count it as a cause).  It seems very natural for someone who takes a substance-theory of causation to take an agent-causal theory of action.  Doing so does not carry the cost that agent-causal theories of action normally carry, namely the cost of supposing two kinds of causation.  So a substance-theory of causation would seem to be a great match for an agent-causal theory of action.

However, I think that the substance-causal theorist may lose one of the benefits of agent-causal theories of action.  The traditional agent-causal theorist can make a neat distinction between my voluntarily doing something and my "doing" something in the non-agential way in which I depress the grass when I lie on it or circulate the blood throughout my body.  The non-voluntary "doing" is a matter of event-causation, while the the voluntary doing is a matter of agent-causation.  But on the substance-causal view, both the non-voluntary and the voluntary cases are instances of substance-causation, with one and the same cause--namely me.  Granted, the substance-causal theorist can distinguish the voluntary doings from the non-voluntary "doings" by saying that reasons enter in a certain way into the explanation of the former but not into the explanation of the latter, but this is exactly the sort of thing the event-causalist would say--an advantage of agent-causation has been lost.

This isn't really an argument for or against any theory.  The loss in this regard is balanced by a greater overall theoretical simplicity in having only one kind of causation.

Agent-causation and God

In us, it is a different thing for an event to be caused by our internal psychological makeup and for it to be caused by us simpliciter. Causation by internal psychological makeup is event-causation while causation by us simpliciter is agent-causation. But in the case of God, the distinction seems to disappear, given divine simplicity. For God's internal psychological makeup is God.

An adverbial model for agent causation

The big problem for libertarian views of free will, especially agent-causal ones, is how to make the action come from both the agent and the agent's reasons. The compatibilist gives up on the agent part—or, more charitably, we should say that, roughly, she analyzes the action's originating from the agent in terms of the action's originating from the agent's reasons.

Here is a model. In the world, there is nomically explained causation. Maybe, charged particle A causes charged particle B to move away, because of the laws of electromagnetics. Maybe, massive particle A causes massive particle B to approach, because of the law of gravitation. Here is a very natural way to say what is happening here:

1. A electromagnetically causes B to move away.
2. A gravitationally causes B to approach.

The laws that are explaining the causation can be included adverbially in the causal statements. The laws from which the causation comes tag the causation, modify it. (In Aristotelian terms, we might even be tempted to say that electromagnetic causation and gravitational causation are analogically cases of causation—causation takes multiple forms.) The adverbial part here is crucial—the law really is doing much of the explaining here. In some sense, even, I would say that the lawmaker (that in virtue of which the law is a law) causes the movement of B or maybe causes A's causing of that movement. (I somehow like the latter, but in the free will case I think the former works better.) For some relevant background, see an unpublished paper of mine.

Suppose now that Plato writes a book because of love of truth and Euthydemus fools Callias out of a desire to impress. Then, very roughly:

3. Plato's love of truth Platonically causes Plato's writing of the book.
4. Euthedemus' desire to impress Euthydemically causes Euthydemus' fooling Callias.

The nomic case provides us with a way in which causation has three relata[note 1]: the reasons, the agent and the action. But the agent and the reasons enter differently.

Strictly speaking, the analogy shouldn't be between the agent and the law, but between the agent and the lawmaker, or, even better, between the agent's form and the lawmaker.

An equivocating argument?

1. If I am responsible for an event E, then I am responsible for at least one of the causes of E. (Premise)
2. There are no infinite regresses or circles among the events I am responsible for. (Premise)
3. Therefore, any event E for which I am responsible has among its antecedents a cause that I am responsible for and which is not an event. (By 1 and 2)
4. All causes are either events or agents. (Premise)
5. Therefore, any event E for which I am responsible has among its antecedents an agent that I am responsible for. (By 3 and 4)
6. Therefore, if I am responsible for any event, agent causation occurs.

I think the argument is unsound because (4) is false. Causes can be events or substances. (Actually, I think only substances, but the weaker claim is all I need.) With this substitution, I get the conclusion that substance causation has occurs if I am responsible for any event. But then when we examine the substances I am responsible for, I think we will eventually get another regress. For substances other than myself, such as my dog (if I have any), I am responsible for only because I took on that responsibility. To avoid regress, I must arrive at a substance that I innately have responsibility for. And that's myself. Hence, there there is an agent who is a substance cause (namely I myself), and an agent who is a substance cause is an agent cause.

The equivocation is in (1), in that I am differently responsible for myself and for events. Maybe it's better to call it an argument from a disjunctive notion of responsibility rather than an equivocating argument.

Moral evil and naturalism

The only argument against theism that is worth considering at all seriously as an argument against theism[note 1] is the inductive problem of evil.

The inductive argument from evil holds that it is unlikely that God would have created a world containing the degree, kind and amount of evil that we observe.

Now, one might offer the inductive argument from evil as an argument for some supernaturalistic hypothesis, such as that there is an evil god (either alone or along with a good or neutral one), or that there is an indifferent god, or the like. In that case, what I will say below does not apply. But, among contemporary philosophers, the primary hypothesis competing with theism is not one of these unorthodox supernaturalistic ones, but naturalism. And then the philosopher offering the inductive argument from evil needs to show not just that the degree, kind and amount of evil that exists in the world is unlikely on the theistic hypothesis, but that it is more unlikely on the theistic hypothesis than on the naturalistic hypothesis.

But to show this is tricky. Much of this point comes from C. Stephen Layman (I am grateful to Todd Buras for letting me know of Layman's argument). Let us grant for the sake of argument that the conditional probability of the degree, kind and amount of evil that we observe is, say, 10⁻²⁰ given theism. Denote theism by T and let E is the event of there being the degree, kind and amount of evil that we in fact observe. Then it's granted for the sake of the argument that P(E|T)=10⁻²⁰. Do we know that the probability of the degree, kind and amount of evil that we observe is more than 10⁻²⁰ given naturalism, i.e., that P(E|N)>10⁻²⁰, where N is naturalism?

Many of the evils that we observe are of such a kind that they are only evil because of conscious life—pains are like that. Therefore, E entails the existence of conscious life, but says a lot more than that conscious life exists. Hence: P(E|N)<P(C|N), where C is the claim that there is conscious life. If P(C|N) is no greater than 10⁻²⁰ then we do not have P(E|N)>10⁻²⁰. Thus the naturalist to use the argument from evil against theism must be able to estimate P(C|N) as greater than 10⁻²⁰. I do not think we are in a position to make that estimate. And without it, we are not in a position to know that the inductive argument from evil supports naturalism over theism.

Moreover, many of the evils that we observe are moral evils. That there are moral evils entails that there are morally responsible persons, and that there are moral truths. IF R is the claim that there are morally responsible persons and M the claim that there are moral truths, then in order to use the inductive argument from evil as an argument against theism and for naturalism, one would have to show that P(R and M|N) is bigger than 10⁻²⁰. But I don't think we are in a position to know that. First, it may be very unlikely that persons would arise given N. Second, it seems quite plausible that for moral responsibility one requires persons who engage in a form of causation very different from the natural causation around them—if all we have is the kind of causation that ordinarily goes on in nature, we don't have moral responsibility. If so, then the likelihood of R given N is nil, or perhaps very small. Furthermore, these persons would have to have moral beliefs to be morally responsible. But to secure intentionality for moral beliefs is a tough task for the naturalist. It's hard enough to secure intentionality for empirical beliefs on naturalistic theories, but perhaps some kind of causal theory of content can be made to work. But it does not seem likely that in a naturalistic setting (except an Aristotelian one—and that's not what people usually mean by "naturalism") we could get intentionality for moral beliefs.

And while I do not know of an entirely convincing argument that the existence of moral truths is incompatible with naturalism, the existence of moral truths is not as prima facie likely given naturalism as given non-naturalistic hypotheses like theism. (In fact, theism entails the existence of moral truths.)

Moreover, among the evils that E reports, there are some truly horrendous moral evils. While bare-bones theism may have difficulties with explaining these, naturalism likely has a difficulty with explaining them, too. But theism, unlike naturalism, is at least compatible with positing a supernatural tempter, a devil, to help explain these horrendous moral evils. So, once again, it is not clear that naturalism does better.

Of course one can tell stories about the evolution of persons, responsibility and even the capacity for horrendous moral evil. But are these stories any better than stories that the theist can tell—stories about divine design, the value of human and demonic freedom, etc.?

If I am right, then the argument from evil does not support naturalism. The naturalist can still offer the argument as an ad hominem, but if she is serious about thinking that the argument is a good reason to reject theism, she needs to be open to the possibility that it is a good reason to abandon naturalism.