I found the following in David Wiggins’s paper “Towards a reasonable libertarianism” (from the 1973 anthology Essays on Freedom of Action, edited by Ted Honderlich) and found it relevant to my recent posts on free will.
[C]an the libertarian even specify a possible world, however different from the actual one, in which there are particular responsible actions which people can (in the libertarian’s sense) do but do not do? Hume has been followed by a large number of philosophers in holding that not even a possible world of the required sort could be specified. If it were false that every event and every action were causally determined then the causally undetermined events and actions would surely, to that extent, be simply random. So the argument goes. That a man could have done x would mean no more than that it might have turned out that way — at random. It will be asked if it makes any better sense to hold a man responsible for actions which happen at random than for ones which arise from his character. Surely then, if it doesn’t, we ought to prefer that our actions be caused?
This is essentially the same argument I’ve been making. I might have known that Hume came up with it first. (That I didn’t know Hume came up with it first is a humbling reminder of just how spotty and haphazard my philosophical education has been. And Hume is one of the thinkers I’m relatively more familiar with, compared to some of the other big names in philosophy!)
Considered simply as an argument this objection is circular, and flagrantly so. One cannot prove that determinism is a precondition of free will by an argument which employs as a premiss everything is either causally determined or random. This is nothing other than a form of the conclusion, that whatever is undetermined is random. This is what had to be shown. But in the form of a challenge something in the objection can stand. If an event is undetermined, if an event of different specification might have taken place, then what does it mean to deny that the event is simply random? What is it justifiably to ascribe the action identical with the event or comprised of the event to an agent whom one holds responsible for that action? In the unclaimed ground between the properly or determinatically [sic] caused and the random, what is there in fact to be found?
This is well put. (Okay, I take that back. I mean, determinatically?) The argument is indeed circular, because it isn’t really a proper argument at all, so much as a bare assertion of something which seems self-evident: that everything is either causally determined or random (or some combination of the two), that those two options exhaust the logical possibilities. The challenge for the libertarian is to show this to be a false dichotomy by coherently describing a third possibility. Later in the paper, Wiggins attempts to do so.
For indeterminism maybe all we really need to imagine or conceive is a world in which (a) there is some macroscopic indeterminacy founded in microscopic indeterminacy, and (b) an appreciable number of the free actions or policies or deliberations of individual agents, although they are not even in principle hypothetico-deductively derivable from antecedent conditions, can be such as to persuade us to fit them into meaningful sequences. We need not trace free actions back to volitions construed as little pushes aimed from outside the physical world. What we must find instead are patterns which are coherent and intelligible in the low level terms of practical deliberation, even though they are not amenable to the kind of generalisation or necessity which is the stuff of rigorous theory. On this conception the agent is conceived as an essentially and straightforwardly enmattered or embodied thing. His possible peculiarity as a natural thing among things in nature is that his biography unfolds not only non-deterministically but also intelligibly; non-deterministically in that personality and character are never something complete, and need not be the deterministic origin of action; intelligibly in that each new action or episode constitutes a comprehensible phase in the unfolding of the character, a further specification of what the man has by now become.
This is an intriguing line of thought. Thinking of intelligible (rather than determined) as the opposite of random offers a different angle from which to view the problem. I’m still not sure it really works, though.
Certainly it is possible in principle for something to be non-deterministic and yet intelligible. The proof of this is that extremely complex systems (the weather, the behavior of other people, etc.), even if it happens to be true that they are “really” completely deterministic, cannot be perceived that way by us. We can see some regularities, enough to make them intelligible to us, but we can’t possibly understand all the causes involved. Even if they are deterministic (which they may be), they are psychologically indistinguishable from truly non-deterministic systems; therefore, if they are intelligible to us, it follows that a truly non-deterministic system could also be intelligible.
But surely (and this is again an assertion of what seems self-evident, not an argument) what makes such systems intelligible is that the randomness is constrained, not that it is not random. Even if truly random processes played a role in, say, the weather, the weather would still be intelligible because it follows certain broadly predictable patterns, because the randomness is not unconstrained. A typhoon may happen to grow stronger or weaker, to last a certain number of days rather than a certain other number, to blow back out to the sea rather than making landfall. I don’t understand why it does the one thing rather than the other, and therefore those aspects of its behavior may as well be truly random as far as I am concerned — but it remains intelligible. If, on the other hand, the behavior of typhoons were pure unconstrained randomness, exhibiting no regularities at all, it would be unintelligible.
The same goes for human behavior. If it is truly non-deterministic and yet intelligible, all it seems to me that that could mean is that it contains randomness, but randomness which is constrained enough that it can still be broadly intelligible. But it’s still just a combination of determinism and randomness.