Wednesday, September 1, 2021

Models of libertarian agency, and some more on divine simplicity

Here is a standard libertarian picture of free and responsible choice. I am choosing between two non-mental actions, A and B. I deliberate on the basis of the reasons for A and the reasons for B. This deliberation indeterministically causes an inner mental state W(A), which is the will or resolve or intention to produce A. And then W(A) causes, either deterministically or with high probability, the extra-mental action A.

Now notice two things. First, notice that my production of the state W(A) is itself something I am morally responsible for. Imagine that I have resolved myself to gratuitously insult you. If it turns out that my vocal chords are paralyzed, my resolve W(insult) is itself enough to make me guilty.

Second, note that that my production of W(A) could involve the production of a prior second-order state of will or resolve, a willing W(W(A)) to will to produce A. For there are times when it’s hard to resolve ourselves to do something, and in those cases we might resolve ourselves to resolve ourselves first. But at the same time, to avoid an infinite regress, we should not adopt a view on which every time we responsibly produce something, we do so by forming a prior state of willing or resolve or intention. In light of this, although my production of W(A) could involve the production of a prior second-order state W(W(A)), it need not do so. In fact, phenomenologically, it seems more plausible to think that in typical cases of free choice, we do not go to the meta level of producing W(W(A)). We only go to the meta level in special cases, such as when we have to “steel” ourselves to gain the resolve to do the action.

Thus we have seen that, assuming libertarianism, it is possible for me to be responsible for indeterministically producing a state of affairs W(A) without producing a prior state of willing or resolving or intending in favor of W(A). The state W(A) is admittedly an inner mental state. But the responsibility for W(A) does not seem to have anything to do with the innerness of W(A). We are responsible for W(A) because our deliberation indeterministically but non-aberrantly results in W(A).

Here is a question: Could there be cases where we have libertarian-free actions where instead of our deliberation indeterministically non-aberrantly resulting in W(A), and thereby making us responsible for W(A) as well as A, our deliberation directly indeterministically and non-aberrantly results in the extra-mental action A, without an intervening inner mental state W(A) that deterministically or with high probability causes A, but with us nonetheless being responsible for A?

Once we have admitted—as a libertarian has to, on pain of a regress of willings—that we can be responsible for producing a state of affairs without a prior willing of that state of affairs, then it seems hard to categorically deny the possibility of us producing an extra-mental state of affairs responsibly without an intervening prior willing. And in fact phenomenology fits quite well with the hypothesis that we do that. We do many things intentionally and responsibly without being aware of a willing, resolve or intention to do them. If we stick to the initial libertarian model on which there must be an intervening mental state W(A), we have to say that either the state W(A) is hidden from us—unconscious—or that these actions are only free in a derivative way. Neither is a particularly attractive hypothesis. Why not, simply, admit that sometimes deliberation results in an extra-mental action that we are responsible for without an intervening willing, resolve or intention?

Well, I can think of one reason:

  1. It seems that you can only be responsible for what we do intentionally, and we cannot do something intentionally without intending something.

But note that if this reason undercuts the possibility of our responsibly directly doing A without an intervening act W(A) of intention, it likewise undercuts the possibility of our responsibly directly producing W(A) without an intervening W(W(A)) act, and sets us on a vicious regress.

I actually think (1) can be accepted. In that case, when we directly responsibly produce W(A), the intentionality in the production of W(A) is constituted by the non-aberrant causal connection between deliberation and W(A), rather than by some regress-engendering intention-for-W(A) prior to W(A). And the occurrence of W(A) means that we are intending something, namely A.

But what would it be like if we were to directly responsibly produce A, without an intervening act of intention W(A)? How would that be reconciled with (1)? Again, the intentionality of the production of A would be constituted by the non-aberrant causal connection between deliberation and A. And the content of the intention would supervene on the actual occurrence of A as well as on the reasons favoring A that were instrumental in the deliberation. (There are some complications about excluded reasons. Maybe in those cases deliberation can have an earlier stage where one freely decides whether to exclude some reasons.)

Call the cases where we thusly directly responsibly produce an extra-mental action A cases of direct agency.

A libertarian need not believe we exhibit direct agency. Perhaps we always have one level of resolve, willing or intention as an inner mental state. But the libertarian should not be dogmatic here, given the above arguments.

Our phenomenology suggests that we do exhibit direct agency, and indeed do so quite commonly. And if God is simple, and hence does not have contingent inner states, all of God’s indeterministic free actions are cases of direct agency.

In fact, independently of divine simplicity, we may have some reason to prefer the direct agency model in the case of God. Consider why it is that sometimes we go to the meta level of W(W(A)): we do so because of the weakness of our wills, we have to will ourselves to will ourselves to produce A. It seems that a perfect being would never have reason to go to the meta level of W(W(A)). So, the remaining question is whether a perfect being would ever have reason to go to the W(A) level. I think there is some plausibility in the idea that just as going to the W(W(A)) level is a sign of weakness, a sign of a need for self-control, going to the W(A) level is also a sign of imperfection—a sign that one needs a tool, even if an intra-mental tool, for the production of A. It seems plausible, thus, that if this is possible and compatible with freedom and responsibility, a perfect being would simply directly produce A (where A is, say, the action of the being’s causing horses to exist). And I have argued that it is possible, and it is compatible with freedom and responsibility.


Wesley C. said...

First, why view the deliberation as causing the will to A though? Why not the will itself? Or the whole person-as-substance if one accepts agent causation? Scotus, IIRC, argued that the will is self-moving and that the principle behind self-motion is an active principiative potency that actualises a passive principiative potency - so something that is in act reaches out to a potency, which is how the will moves itself.

Second, let's say we view indeterministic free will as self-movement of the will, and say we accept the idea that the metaphysical explanation is that there is something in act within the will that actualises another aspect of the will that is in potency. How would this apply to God specifically, since He is pure act and so doesn't have a potency within Himself to be a self-mover or have a self-moving will?

Third, have you seen this interesting paper by Mark Spencer on reconciling Scotus, Aquinas and Palamas?:

One of the arguments he makes (citing Scotus I think) about divine freedom and simplicity is that the cognition of the intellect and will don't require any intrinsic gain in perfection or actuality - so when our intellect knows something, or the will wills something, it's not a perfection added to the intellect or will intrinsically as such, nor is it an intrinsic change. Now there really is a difference in conscious experience depending on what you know or will, but this difference isn't intrinsic to the intellect or will as such, so it's not an intrinsic actualisation. In a similar way, God's conscious "experience" would be different in different worlds, but neither His knowledge nor His will would be intrinsically perfected or actualised or changed, so a difference in what God knows or wills doesn't imply anything intrinsic to them.

Another thing he points out is Norris Clarke's coining of the intentional relation - a sort of middle between real relation and relation of reason, and is based on what I laid out above. What do you think?

Unknown said...

Dr. Pruss, what's your height

Alexander R Pruss said...


It doesn't matter for the purposes of the point I am making whether one views the deliberation or the agent-qua-deliberating or the agent's will as the cause of W(A).

I doubt that the "intentional relation" story is much more than just giving a name to the mysterious phenomenon of God being intrinsically the same while having different experiences and acts of will.

I try to never talk about potency. :-)

Wesley C. said...

Here's a quote from the paper in case it helps to clarify what is meant:

"We human persons bring about free acts in ourselves in virtue of our substantial actuality. As agent causes, we can, by immanent causation, without change in being to our substance as such, bring immediately into existence new free acts of the will,which are actualities that participate in our substance and terminate that substance in a new way but are really distinct from that substance. These acts in turn bring about effects entirely outside of us. Though our substance is terminated in a new way by these acts, it is not thereby changed inasmuch as it is this substance. But then it is not implausible, contrary to objection three, to think that a thing can act without being changed in its being, for we agent causes act without being so changed.Since all beings are participations in God and God always has all the actuality possible to have in himself, God can, without change in being to himself, freely cause there to be beings distinct from himself participating in him and terminating his substance in new ways. God’s formalities as terminated by creatures just are God’s contingent acts or energiai, the ways that he is present to, knows, and wills creatures."