Although I've spent a lot of time over the years critiquing freewill theists like Roger Olson and Jerry Walls, because they have a popular following, they're hardly the most able exponents of that position. Far higher on the food chain is Peter van Inwagen. Let's consider his argument.
When I myself look at contemplated future courses of action in the way I have described above, I discover an irresistible tendency to believe that each of them is "open" to me…I find myself with the belief that sometimes more than one course of action is open to me, and I cannot give it up…I don't find the least plausibility in the hypothesis that this belief is illusory. Peter van Inwagen, Metaphysics (Westview Press, 4th ed., 2015), 282-83.
In a sense I agree with PVI, but not with the conclusion that he draws from his experience:
i) He's describing a psychological impression, but psychology isn't conterminous with ontology. Our psychological impressions don't necessarily map onto the extramental world.
ii) Apropos (i), surely it's possible to imagine future courses of action that seem to be open to us, but are in fact infeasible. For one thing, when we contemplate alternate courses of action, that's very sketchy. We don't entertain of all the intervening links that chart a pathway from the present to the future outcome. We don't know enough. There are too many variables. Rather, we take our current situation as the frame of reference. We contemplate different outcomes. But in many cases, those imaginary trajectories may not be open to us, because they depend on many independent variables lining up in one particular direction, and we don't control most of the variables. Sometimes we can manipulate circumstances to achieve our goals, but in many cases our goals are stymied. In fact, after writing this paragraph, I read the following observation:
We all use our imaginations to foresee the future, whether we're planning the rest of the day or the rest of our lives; and despite our best efforts, nothing guarantees that what we imagine will come to pass. Indeed, rarely does the future heed our plans. Each morning I plot the coming hours, and seldom to my designs unfold without a hitch. On most days, unforeseen circumstance interrupt, and I end up improving, deferring activities, changing course. The same is even more true of my long-term planning: the further into the future my imagination has projected itself, the less prescient it's been. Indeed, only rarely have I seen the far future approximately as it's come to pass. My life illustrates chaos theory: I can't predict or control things because there are always too many variables. The result is that most of my personal goals have turned out to be useful fictions. They've given me something to shoot for, but they've rarely been realized, at least in the forms I'd first imagined them. D. Allison, Night Comes (Eerdmans, 2016), 81.
Suppose C is my objective, yet I can't achieve C unless and until I do B, and I can't do B unless and until I first do A. But that means the effort to achieve the goal may break down somewhere in the process.
So there is, in fact, an illusory quality to many of these future courses of action. In some cases they aren't coherent, because they depend on a chain of events that is in conflict with actual and inexorable causal chains that are already in place.
Perhaps what he means, as he expresses himself elsewhere,
It is at least very plausible to suppose that Jack is not, during the course of his deliberations, able to hit the right-hand side and is not able to hit the left-hand side. But such cases are not decisive, since they involve the concept of success or at least the concept of result: they are cases in which in which an agent is now faced with a choice between doing A and doing B, and in which, if the agent should endeavor to do A or should endeavor to do B, whether the agent would succeed in either endeavor is now undetermined. The Oxford Handbook of Free Will, 2nd ed. (Kane, ed.).
This would also dovetail with his denial that access to alternate possibilities is a necessary condition of libertarian freewill. If, however, freedom of choice reduces to mental acts of choosing between one contemplated future course of action and another, without corresponding, extramental forking pathways, then it's unclear what is meant by the claim that future courses of action are open to the agent. If his position boils down to the psychology of choosing, with no matching ontology of doing, then the "courses of action" are imaginary. Figments of the imagination. If you can't act on your decision, then the whole framework of forking paths and future course of action seems to be a rhetorical flourish.
iii) From a Calvinistic or compatibilist standpoint, these aren't illusory in a deceptive or useless sense. The process of deliberation is how we settle on a course of action. By comparing and contrasting hypothetical alternatives and considering the respective consequences, contemplating of the alternatives is what induces us not to opt for any of the alternatives. Their practical value is to make us see that the course of action we actually settle on is preferable.
iv) In addition, there are many situations in which we don't deliberate, either because we don't have a range of options to consider, or because the alternatives are so unappealing. Only one option is viable or attractive.
The way he defends freewill theism seems counterproductive. On the one hand he raises a familiar moralistic objection to determinism. On the other hand, he seems to concede that on his view, indeterminism takes the decision out of the hands of the agent:
If it goes to the left, that just happens. If it goes to the right, that just happens…There is no way to make it go one way rather than the other…It is a plausible idea that it is up to an agent what the outcome of a process will be only if the agent is able to arrange things in a way that would make the occurrence of this outcome inevitable and able to arrange things in a way that would make the occurrence of that outcome inevitable. If this plausible idea is right, there would seem to be no possibility of its being [up to the agent] what the outcome of an indeterministic process would be (278).
But how is the agent responsible for such choices? He toys with agent-causation, but finds that opaque.
The judgment that you shouldn't have done X implies that you should have done something else instead; that you should have done something else instead implies that there was something else for you to do; that there was something else for you to do implies that you were able to do something else; that you were able to do something else implies that you have free will. To make a negative moral judgment about one of your acts is to evaluate your taking one of the forks in the road of time, to characterize that fork as a worse choice than at least one of the other forks open to you. (Note that if you had made a choice by taking one of the works in what is literally a road, no one would say you should have taken one of the other forks if all the other forks were blocked.) A negative moral evaluation of what someone has done requires two or more alternative possibilities of action for that person, just a surely as a context requires two or more contestants (268).
i) That sounds plausible. Indeed, it's the primary argument for libertarian freewill. However, it fails to draw an elementary and fundamental distinction between psychology and ontology. Even if, metaphysically speaking, alternate courses of action are available, that doesn't mean an agent is able to access those alternatives if he's psychologically ill-disposed to avail himself of the opportunities. Take a psychopathic killer. He's too morally hardened to do the right thing. He's lost the capacity for virtuous action. If we view freedom as a relation between deliberation and opportunity, there are two sides to the relation. Even if (ex hypothesi) freedom of opportunity were a necessary condition of freedom, it's not a sufficient condition unless the agent is open to the pathways that are open to him. That's why we say some people are in a state of diminished responsibility. Take a person with senile dementia. They may have the same objective opportunities, but they've lost the capacity to make rational decisions. Or take someone acting at gunpoint.
ii) Predestination or determinism doesn't imply that if (ad impossibile) you were to attempt do something contrary to what you were predestined to do, a mysterious invisible force would block you or impede you. Hypotheticals and counterfactuals aren't illusory in that sense.
In some cases, there are hypothetical pathways that have no obstacles. There is a coherent alternate plot or alternate timeline. But that's represented in a possible world rather than the actual world. Or, if something like the multiverse is true, in an actual parallel universe.
Ask yourself a question. What would happen if some supernatural agency–God, say–were to "roll history back" to some point in the past and then "let things go forward again"? Suppose the agency were to cause things to be once more just as they were at high noon, Greenwich mean solar time, on 11 Marsh 1893 and were thereafter to let things to on their own accord. Would history literally repeat itself? Would there be two world words, each the same in every detail as the wars that occurred the "first time around"? Would a president of the United states call "John F. Kennedy" be assassinated in Dallas on the date that in the new reckoning is called "22 November 1963"? Would you, or at least someone exactly like you, exist? If the answer to any of these question is No, determinism is false. Equivalently, if determinism is true, the answer to all these questions is Yes. If determinism is true, then, if the universe were "rolled back" to a previous state by a miracle (and there were no further miracles), the history of the world would repeat itself. If the universe were rolled back to a previous state thousands of times exactly the same events would follow each of these thousands of "reversions" (270).
i) I think that's generally a good way to expound the distinction between generic determinism and generic indeterminism. However, I'd point out that it can be misleading. There are varieties of determinism. In his essay on "How to Think about Free Will," PVI says "Determinism is the thesis that the past and the laws of nature together determine, at every moment, a unique future" (Journal of Ethics 12: 330). I'd just point out that this definition not only doesn't coincide with theological determinism (i.e. Reformed predestination), it doesn't even intersect with theological determinism. By that I mean, a Calvinist subscribes to exhaustive predestination and providence, yet that's not how he defines theological determinism.
In Calvinism, for instance, predestination doesn't imply that all future events are caused by past events. No doubt many future events are caused by past events. But what makes predestination deterministic isn't physical determinism or nomic necessity. One thing doesn't follow another because there must be an unbroken causal continuum between antecedent states and subsequent states. Rather, it's more like screen play where every event is scripted. Or, to take a related comparison, it's like Alfred Hitchcock who said he filmed what he visualized.
In fairness, his argument isn't directed at theological determinism (e.g. Calvinism). So it's not a flaw, in that respect, if his definition fails to map onto theological determinism. Even so, that's a huge omission. How would he need to change his argument if he were targeting Calvinism?
ii) Notice an implication of indeterminism. On this view, you don't exist because you were a part of God's plan. If history was reset, you wouldn't exist. You don't have God to thank for your existence. If you're healthy, if you have a happy marriage, you don't have God to thank for your circumstances. Everything that happens to you is just the roll of the dice. If the dice were rolled a second time, you wouldn't even be here. Whatever happens to you is a matter of sheer luck. Good luck or bad luck, as the case may be. If indeterminism is true, there's no basis for pious gratitude.
iii) Apropos (ii), his view is that it puts future agents at the mercy of past agents. The options available to future agents, or whether some future agents will even exist, depends on which course of action past agents take. So the fortunes of future agents are enslaved to the often ignorant, capricious, or malevolent actions of past agents. To evoke an illustration he uses in chap. 9, freewill theism is like a situation in which your life depends on drawing the shortest straw, only the drawing in rigged in favor of past agents, because their choices impact the fortunes of future agents.
Mind you, I think there's a sense in which this is true. The problem is if it's just up to human agents. If, by contrast, they are acting in accordance with a divine plan, then there's an ultimate wisdom and justice to how things turn out.
PVI draws a distinction between touchable and untouchable facts. Paradigm examples of untouchable facts include the necessity of the past and the fact that 317 is a prime number (273-76). He compares that to other situations, like whether he has the freedom to stop writing a book (272). If determinism is true, then "all facts are untouchable facts" (276). This is the "hidden mystery" that "lies behind the facade of bluff common sense compatibilism presents to the world" (276).
That, however, is a deceptive comparison. Determinism in general, and predestination in particular, doesn't mean all facts are necessary in the sense that mathematic truths are necessary truths or the necessity of the past is metaphysically necessary. In predestination, everything must unfold due to conditional necessity, not absolute necessity. In principle, God could predestine alternate outcomes. For all we know, God has predestined alternate outcomes–a multiverse. So these are not the same kinds of facts.
