Me (original writeup): "John can be divided into his properties, capacities, and parts. His properties include John's having blue eyes, John's being 5 feet eleven inches tall, John's modest attractiveness, and so on. His capacities include things like John's ability to someday pick up the guitar or learn French. His parts are his arms, legs, torso, and so on. These are John's properties, capacities, and parts. Something has all of John's properties, capacities, and parts. Let's call that a substance. The idea of a substance is coherent, since we rely on it whenever we predicate some property of a thing.
Now think about a chain of causes that has a beginning. A causes B, B causes C, and so on until Z. There is a cause in this chain, A, which has no prior cause. Call A in this chain an ultimate cause. The idea of an ultimate cause seems coherent. We have evidence that it is coherent, since people employ the idea in philosophy and physics.
We can now produce an account of free will: John has free will just in case John's substance is an ultimate cause. This is coherent, so it should dispel the notion that free will is undefinable. If you don't think it's coherent, we have evidence that you are just failing to understand it. After all, we created the definition by combining two prima facie coherent concepts."
Him: "You say that " John's substance is an ultimate cause"; but you've listed a set of components of "John's substance", and I think most people would readily agree that all of these components have causes in turn. Therefore, "John's substance" is not really an ultimate cause."
Me: "That response is impressive to me. What do you think of this: Let's stipulate, pending another really plausible account, that substances are either classical substances or property bundles. I prefer classical substances, but whatever. Now a classical substance is obviously distinct from the properties, capacities, and parts it has. A property bundle is made a bundle by a bundling relationship R. R is distinct from the properties, capacities, and parts it bundles. So either it's possible for the classical substance to be an ultimate cause, or it's possible for the bundling relationship to be an ultimate cause."
Him: "But it is also a possibility that neither the classical substance nor the bundling relationship to be the ultimate cause. One would expect ultimate causes to be rare, and to have happened long, long ago. Quantum mechanics does suggest that ultimate causes are very common on the subatomic level, but it also suggests that they are random, and we generally don't want to say that free will is the result of pure randomness."
Me: "It's possible that a substance fails to be an ultimate cause, and it seems that many do, including the substances of rocks and trees. But it is the case that human substances are ultimate causes regardless of that possibility. I think our expectation that ultimate causes be ancient can be overridden pretty easily, as evidenced by the willingness of so many people to affirm that there are random quantum mechanical events. I don't trust most philosophical arguments from quantum mechanics, and many scientists take a nonrandom interpretation of the quantum mechanical facts."
Him: "But I still don't understand why you say that "human substances are ultimate causes". I can think of precursor causes to most aspects of humans. Things like consciousness are hard to find causes for, but they are also hard to define exactly. I think you'll find that any reasonable definition of consciousness, faith, morality, etc., will suggest a cause by the very act of being defined. Unfortunately, it is nearly impossible to distinguish between 'I don't know the cause' and 'there is no cause', but you can see why the statement "consciousness is uncaused but only occurs in humans" (for example) is problematic."
Me: "(1) A good definition sets the thing before you outside time. (2) It may be that the higher animals also have free will."
Him: "I don't think that I agree that a good definition takes the thing defined out of time. Time is an important part of the definition of many things, and a complete description of something will always include something about how it works in or changes through time. :change of subject: If I were to say anything about anything having free will, it would raise the same questions. "rocks have free will, but sand doesn't" raises causal questions. "higher animals have free will but lower animals do not" also raises questions. Specifically, you have to be curious about the defining features that divide the two groups. If the complexity of the frontal lobe is what divides the higher animals from the lower, then you have to wonder what it is about the frontal lobe that causes this difference. And when you have found the cause, does it undermine the 'ultimate cause' theory?"
Me: "(1) The thing is removed from any particular sequence of causes when it is defined. Reason as such deals with things in an unmoving way, even chains of causes. (2) Many, many things about the brain and mind are hard to explain causally, so the difficulty of explaining free will causally is not a problem. After all, neuroscience and psychology are still relatively young. We can say that the degree of a thing's free will seems to increase with the complexity of its nervous system and the range of its consciousness, which suggests some sort of causal relationship; so if we cannot explain free will, at least it doesn't come "out of the blue.""
Him: "Doesn't it need to come out of the blue in order for it to be considered an ultimate cause?"
Me: "If that were a requirement, then we would have no more reason to expect humans to be free than rocks or trees. Obviously, there are some preconditions to free will, one of which is that there must be a specific sort of substance. That substance then has the capacity to initiate a causal chain A, B, C, ..., Z at t. So long as A isn't necessary given the state of affairs at t, A is an ultimate cause."
Him: "I don't understand that last line: "So long as A isn't necessary given the state of affairs at t, A is an ultimate cause." Does that mean that anything that does not need to be present at all times in order to maintain a state qualifies as an ultimate cause?"
Me: "I'm not entirely sure what you're asking. Maybe I just need to spell this out more. Take a specific, arbitrary time. We're going to call that time, t. Now at t there is a way that all of reality is. Call the way that all of reality is at t, the state of affairs at t. The state of affairs at t necessitates certain things about the state of affairs at t+1 (we'll say that t+1 is some time after t). But for some t+1, part of the state of affairs at t+1 (and we'll call this part A) is not entirely necessitated by the state of affairs at t. Now if A is the beginning of a causal chain, A is an ultimate cause."
Him: "Yes, now I understand. This then falls back on the question, 'what makes ultimate causes different from random events?' If free will is caused by an immediately preceding random event, it isn't very useful; that's just saying that we behave randomly. If the random event was in the past that means that the rules *currently* governing our choices are deterministic, and therefor predictable on some level."
Me: "If a random event is just "an event without a necessitating cause," then everyone already thinks free will is random. I suspect that our concept of randomness is more robust than that, in which case we think that free events are a kind of non-random ultimate cause."
Him: "Non-random and non-deterministic. I'm having a hard time seeing how something can be both."
Me: "An event may be caused by nothing (random), by a prior event (determined), or by a substance (free). I'm not sure about this, but we might have to make the further qualification that if a substance S causes an event E, S must not have been determined to cause E for E to qualify as free. The point is that we can, in some way, make a sharp distinction between determined, random, and free events once we recognize the possibility of substantial causation."
Him: "I'm not convinced that 'being caused by substance' doesn't necessarily fall back into being random or determined if you examine it closely enough, but I'll keep thinking about it. Maybe I'm missing something."