Open Theism and Causal Determinism

Dear Jon, one solution would be to focus on warranted assertability. For any future tensed statement, presumably God knows how likely it is to be true. (I am considering the likelihood of an event to be the degree of warrant that God would attach to a statement to the effect that the event will take place) Even for undetermined parts of the future, God knows that it is, for example, extremely likely that Curley will accept a bribe of $1500, and extremely unlikely he wil accept a bribe of $1.50 Given the laws of nature, and the current position of the Earth and the Sun, the only thing that could prevent the rising of the Sun tomorrow is a miracle. But God, let us suppose, has decided not to perform such a miracle. So the evidence that he has means that he is right to be 100% certain that the Sun will rise tomorrow, and this maximal warrant is equivalent to truth.

This raises many interesting questions of course: does the Open Theist hold, as a general thesis, that truth is maximal warrant, or is this something that applies only to future-tensed statements? In the latter case, to say that some future tensed statements are true seems to be saying that, when it comes to the future tense, maximal warrant serves the same purpose as true, that we can treat statements with maximal warrant with the same confidence that we treat true statements, and so on.

I'm using the term 'maximal warrant' as a place-holder for whatever epistemic value the Open Theist chooses. Candidates would include Putnam's idealized warranted assertability, Wright's super-assertability and so on. The groundwork has already been laid by anti-realist theorists, its a matter of choosing the best available version of anti-realism.

I think it would fit the case for the Open Theist to be an anti-realist about the future but not about the present or the past, since I think one of the key intuitions that underlies Open Theism is that there is an ontological difference between the future on the one hand, and the present and past on the other, and you are correct that, in that case, it seems odd to describe any part of the future as 'true' period.

If you think of miracles as things God worked into the laws to begin with that just go against what we would think the laws are from what we've observed, then this might be fine. But then God couldn't work any miracles in to respond to unforeseen events as a result of indeterministic processes, whether free will or otherwise. So I'm not sure many open theists would want to reduce miracles to that sort of thing the way some compatibilists do.

Ben:
I think it would fit the case for the Open Theist to be an anti-realist about the future but not about the present or the past, since I think one of the key intuitions that underlies Open Theism is that there is an ontological difference between the future on the one hand, and the present and past on the other, and you are correct that, in that case, it seems odd to describe any part of the future as 'true' period.

Isn't it also odd for open theists to be presentists rather than growing block theorists? Unless backward determinism is true while forward determinism isn't, there are similar problems with truth about the past unless you're working with a growing block view of time. These might be overcome some way, but it's an obstacle that I don't think presentist open theists have spent a lot of time on,

Hi Jon,

I'm not sure I understand your points about miracles, but I think I see the point re: that brand of open theism. As I'm that kind of open theist (presentist, open future, bivalence-denier), I'll bite.

I'm not sure I understand your concern. But what this sort of open theist needs is just something to ground the truth-values of some but not all future-tensed statements. e.g. Suppose it is now true that: Rosie O'Donnel will die on 1/1/08) This requires it to be the case that she will, in fact, die then. How could that be a fact? Because of some further fact, perhaps this one: God is setting or has set an inalterable intention to "take Rosie home" on that date. It seems to me that just about any fixed element of the future a Christian open theist could want, can be secured in this way - no need to worry about divine providence and how it relates to laws of nature or causal conditions.

Both Ben and Dale appeal to God's intentions or decisions about the future to provide the needed basis. That's fine unless God can form such an intention or make such a decision and take it back tomorrow. If the decisions or intentions are metaphysically revokeable, then the open theist is in trouble (since then the only ground would be the fact that God will freely not alter his intention or decision, and such claims are purportedly ungrounded). And if the view is that it is metaphysically impossible for God to change his mind, it's hard to see why that would be true. Especially since open theists delight in Biblical passages where God is said to change his mind.

Jeremy, the view I was alluding to about laws of nature and miracles is the view that all laws of nature include ceteris paribus clauses regarding intrusion from outside the system in question. When a miracle occurs with respect to such laws, no violation of the law occurs. Given such laws and the initial conditions, you can't deduce what will happen next. You also need the premise that no force outside the system will intervene. That's why I wondered why "laws-plus-initial-conditions" is thought by open theists to provide a suitable ground for the parts of the future that are not undetermined.

Hi Jon,

There's not as much trouble for us on that score as you think. We neither need nor want the (unattractive) claim that God can't ever change is mind. The fact that he can is part of what makes a genuine, push and pull, give and take relationship with him possible. (compare: his literal arguments with Moses)

Rather, all we need is the claim that he can form an unrevokable intention - one such that he knows that no condition - internal or external to him - is going to move him off his current course. It is plausible that an omniscient and omnipotent being can do that. Of course, you might think that it involves a kind of self-limitation, so there would be a problem with naive accounts of omnipotence. But those should be avoided in any case.

Just to illustrate, suppose that God intends to take Rosie home on 1/1/08. Well, that's one thing. If between then and now she gets right with God, he may drop that intention. (Compare: king Hezekiah) Contrast this with a more extreme case, where God has (to steal some OT language) "set his face against" her - he's just going to whip the rug out from under her life on 1/1/08, no matter what anyone does. It's a more severe kind of judgment and rejection.

In sum, you're right - we'd be in trouble IF for some reason ALL God's decisions were revokable. But why think that?

Dear Jon, I don't think its accurate to lump together Dale's position and mine. I was recommending that the Open Theist treat future-tensed statements as having degrees of warrant, rather than trying to find a ground for their truth-conditions. In other words, I'm advising the Open Theist to make a major concession in response to your initial posting.

A better way of expressing my point would be this. I'm suggesting that the Open Theist say that for all truths, God knows they are true - so what is true has maximal warrant for God. Future-tensed statements have degrees of warrant for God, up to and including maximal warrant.

Now I don't think that this requires God's decisions to be metaphysically irrevocable. It is just that sometimes, God says "I've made up my mind, I'm definitely going to do this. Nothing else will take priority," and because of this firm intention, he knows what will happen. It is true that there are unknowns for the God of Open Theism, but they are known unknowns: he knows that he doesn't know what action a free creature will perform, but he isn't going to be taken by surprise by an asteroid, or by the sudden appearance of a new free being that will confound his intentions. So sometimes he is entitled to be 100% sure that, for example, the Sun will rise tomorrow.

Jon,

Suppose there are miracles. Does it follow that causal determinism does not ensure the truth of some future propostions? I don't think so. Miracles are not, as you note, exceptions to exceptionless regularities. Otherwise they'd be metaphysically impossible. Better to think of miracles as contravening "almost-laws" (to adopt Lewis and Johnson's terminology). That is, miracles are events that are exceptions to extremely highly confirmed regularities. Regularities that have the strong appearence of being laws. But those regularities fall (at least, just) short of being laws.
This is all compatible with there being laws and there being events determined by those laws and antecedently knowable. I mean, this is compatible with there being exceptionless regularities (regularities that do not permit miraculous violation) on the basis of which the future is determined. And that determined future could be known.

Jon,

I have small sympathy for the open theist position, but I don't think they're vulnerable in the way you suggest. The idea is that some limited part of the future can be causally determined, and God can know what will happen then. You raise the point, what about the possibility of a miracle between now and then? That would un-determine it. True. But it seems to me that the reply here should be: (i) only God can perform such a miracle; (ii) God can infallibly know whether he will perform such a miracle between now and then. God can know this, not because it's impossible for him to revoke an intention, but because he would know that he would have no reason to do so--by hypothesis, there are no other free actions in this stretch of time which might make him change his mind.

OTOH, challenging (i) above would be a game-changer.

This question is for Dale. Suppose God did form an intention to "take Rosie home," as you say, on the first of next year (call the date "T"). I don't see how that grounds the truth of "Rosie will die at T" UNLESS God insures a lot of other things. For instance, he would have to insure that Donald Trump doesn't freely decide to kill Rosie before then (or, as is more likely, HIRE someone to kill her). He has to insure that she is not in a wreck with a drunk driver, etc. In short, he has to prevent any number of events that may or may not be very likely to occur and that would thwart his plan to take Rosie home at T, some of which require him to limit free will. But I wonder if, in order to ground the truths of statements like the one we're considering, he might have to tinker a little too much for the open theist's liking. (So, for instance, in order to prevent Trump from hiring an assassin, he might know that George cannot suggest the idea to the Donald because the Donald's mental state is such that if George were to suggest this to him, he would, given his character and current state of mind, very likely agree to the plot. And so on.) What do you think?

Justin, God wouldn't have to prevent anything unless it's about to cause her to die. It might not have to limit free will, either. Suppose Trump hires someone to assassinate her. God could allow that transaction. The assassin then gets set up to fire at hire at a good opportunity and then proceeds to fire. God could allow that. So far no prevention of free will. For example, God could ensure that the wind speed would change just enough for the bullet to hit but not kill her or for it to miss entirely. Or the assassin's gun could explode, or a bird could fly in the path of the bullet. Not one of these things violates the assassin's free will. They just don't allow the choice of the assassin to be successful in its aim.

What's a lot more difficult for the open theist is if free beings have to do something very specific for God's unalterable plan to happen (e.g. that Judas has to betray Jesus) while also insisting that it be done freely. Perhaps something like that would be important enough that God could violate his free will if he didn't do it freely, or maybe he'd violate Satan's free will to make him violate Judas' free will if for some reason that's not as bad morally.

Jeremy,

I pressed this same point (or one very similar to it) a few years ago at a conference, and Alan gave an answer that satisfied my misgivngs somewhat (I think Hasker said the same thing in correspondance) If my memory serves me correctly: The open theist is simply not committed to thinking that there are very many true future tense propositions of the form "X will do A." So they're not committed to the idea that it was Judas who was appointed to betray Christ. Pretty much anyone could have played Judas's role. So, again no limiting of free will.

Perhaps a more interesting question than the one I ask ed above is how many and what sorts of future tense propositions of the form X will do A are open theists committed to affirming as true? This may be more of a theological/biblical interpretation question though.

dale, you hold out hope for OT through the possibility of an irrevokable intention. The irrevokability has to be metaphysically impossible revokability; anything less implies that the intention is revokable in precisely the sense needed to keep the claim that the sun will rise tomorrow from following from the information in the premises.

But nothing metaphysically impossible ever comes to be so. Only other forms of necessity allow for that. And notice that you do not describe the intention as being metaphysically necessary. You say it is one such that God "knows that no condition internal or external is going to move him from it." That's not metaphysical necessity. It's, as the medievals say, necessity of a consequence rather than necessity of a consequent.

Ben, God says, "I've made up my mind that the sun will rise tomorrow." It doesn't follow from this that the sun will rise tomorrow. It only follows if God's decisions are metaphysically irrevocable.

Mike that's really interesting. So you have laws, and miracles, and exceptionless regularities. The laws imply the regularities, and miracles are not violations of laws, but of supposed laws. That's awkward, though, at least if it is a sufficient condition for being a miracle. I believe it is a near-regularity that no one in history has been named "jon kvanvig". But it isn't a miracle that I am so-named.

Heath, it is not clear to me why it is easier for God to infallibly know what he will do in the future than it is for him to infallibly know what humans will do. You give an argument that he has to have a reason and he can know that he won't have one, but I doubt there's a good reason to adopt this application of the principle of sufficient reason. Why can't he change his mind for no reason? If both doing A and doing B have the same support, then he can change from one to the other with no reason at all.

to all,

it might be worth noting that the desire for a "ground" of the truth of a claim needs to the desire for something like a truthmaker. and nothing is a truthmaker for p unless it is impossible for the truthmaker to obtain and p to be false.

Justin, I don't think that helps much, actually. Even if the irrevocable decree is that someone or other freely chooses to betray Jesus, how could an open theist ensure that it would happen. Even if it doesn't have to be Judas, there's still the problem of assuring that someone will do it.

. . . that's really interesting. So you have laws, and miracles, and exceptionless regularities. The laws imply the regularities, and miracles are not violations of laws, but of supposed laws. That's awkward, though, at least if it is a sufficient condition for being a miracle.

I agree. Clearly this cannot be a sufficient condition and neither Lewis nor Johnson think so. Sufficient conditions would include all the features that distinguish laws from accidental generalizations: for instance, that they support counterfactuals and so on. It is the extremely(!) rare exception that distinguishes them from genuine laws.

Jon, first I find your clarification of 'ground for P' is useful - you mean what I take you to mean. The ground for P is the truth-maker(s) for P, and it is impossible for the truth-maker to obtain and for P to be false.

Now I am saying that the Open Theist should say that there are no truth-makers for future-tensed statements, only evidence. But, in some cases, the evidence for a future-tensed statement might be overwhelming.

One reason the evidence would be overwhelming, the best reason, is if an omnipotent and omnibenevolent being has made a commitment.

I was always taught not to make a rash promise: only promise to do something if you are really sure that you will do it. Being true to your word means that no matter how difficult, you have to try to keep your promise.

I take it that God sometimes makes promises, and that having done so, he keeps them: he is a God of his word, and does things for his own name's sake and so on. I don't see the need to say that it is metaphysically impossible for him to change his mind, or metaphysically impossible for him to act on a whim. But I don't think that he would make a promise and then break it on a whim - his whims are not going to lead him into a surprising change of policy. Nor, having made a promise, would he be led to break it because of some unexpected event beyond his control. There are events he does not expect - actions of free beings - but he is sensible enough not to make rash promises that he might be unable to keep because of such events.

I've been invoking God's making a commitment and keeping his promises. I'm not saying that he only knows the future when he's made a promise, just that he has the capability of making promises and commitments because sometimes he forms a definite intention such that he knows that he is not going to change this intention on a whim.

Heath's point about God's knowing he will have no reason to change his mind is important.

As I understand it, the point of Open Theism is that the world changes and God changes and his knowledge adapts to the changing world. Instead of a static God we have a dynamic God. But that doesn't imply that we were wrong to praise God for being 'still the same as ever, slow to chide and swift to bless', or that we can no longer say 'We blossom and flourish as leaves on a tree, and wither and perish but naught changeth thee.' We are confident that God will always be a loving God, a God of justice and mercy. I can say 'I will always pursue truth, justice and the Panamanian way of life', and that is admirable, but perhaps presumptuous as well - I cannot be sure that I will not become corrupt. But we know God will never be corrupted and, presumably, he knows himself even better than we do. He doesn't just judge on the evidence 'Well, so far I've been a God of Love, I don't suppose I'm likely to change.' He is love, after all.

Another analogy. Robin Hood sees a target and fires his arrow, without hesitation: he has no moment of doubt. It is as though the arrow is an extension of his arm - he does not wonder whether his arm will respond to his command, he just moves it. And he does not wonder whether his arrow will hit the target - he just knows it will. Of course, even Robin Hood's best shot may be blown off target by a sudden gust of wind, but not God's.

God's son lies in the tomb, and his mother is weeping. But he will rise again on the third day. God doesn't pin a note to his fridge because he might forget, or put off the resurrection until the 4th day because of bad weather. He knows what he is going to do.

Maybe his intention was irrevocable maybe not. I suppose 'he forms a metaphysically irrevocable intention' could be a translation of what I've said above into philosophical language. But, for what its worth, I don't think that God forms an intention to raise Jesus on the third day, and then adds metaphysical irrevocability to that intention to prevent any future whimsical lapses.

Returning now to metaphysical terminology. My Open Theist says that the objects that make P true may exist before P is true, even though they are not yet making it true. I hope that Manchester United will win the Premiership. If they do, the Manchester United team will make it true that they win (let's say). The team exist now, but they are not yet making it true that they win - so the truth-makers can exist before they do the truth-making.

I am the truth-maker for my own free actions, and until I make truths, God does not know what those truths will be. But for most truths, truths that do not depend on free beings other than God, God himself will be the ultimate truth-maker, since he controls the things on which the truth depends. He knows what truths he is going to make, and he can, after all, make and do all things.

Mike, I'm not sure I quite get this. The supposed laws don't support counterfactuals, do they? They don't even support regularities.

Jeremy,

Well, he might create a situation in which either A B C D or E (and so on)will, say, betray Christ. And if it turns out that none of them do, that is, if God sees that his last option is about to (or is probably going to) do other than what he wants him to do, then God can limit free will in that case to enuure that his ends are accomplished. But he only has to do this as a last resort. This is problematic for open theists only if there are a lot of specific future tense propositions that they are theologically or exegetically required to affirm the truth of.

Ben, you agree to the truthmaker idea, but then you don't honor it. You aren't the truthmaker of your own free actions, if you are capable of doing otherwise. Your obtaining doesn't guarantee that the action in question occurs. You also say that the objects that make P true may exist before P is true, even though they are not making P true. But then these objects aren't truthmakers.

the appeal to evidence and overwhelming evidence won't be relevant here unless it entails truth. if you claim that it does, then we can just talk about truth and bypass the talk about evidence and warrant.

Hi guys,

Justin - Jeremy basically answered for me. Thanks, Jeremy. :-)

Jon - sorry, no sophmoric fallacies here. God's intention would have to have a kind of metaphysical necessity, but not broadly logical necessity - rather, it's now-unchangability - being such that given the history of everything up till and including now, it can't not be. And this is a valid argument (N = broadly logical nec, and U = now-unavoidability)

1. N (p -> q)
2. Up
3. Uq

Dale, you're going to have to fill in what the values are for p and q. I take it q is the claim about the future. presumably, p is something like "God intends q". But then premise 1 is false. So 1 needs to have the U operator on P as well.

In any case, the point I've been making is about premise 2 anyway. Premise 2 is a claim about the future (or at least implies a bunch of such truths). For simplicity, I'll just assume that it is 2 itself that is about the future. It is a logically contingent one. So what is its truthmaker? It doesn't have one if truthmakers are a subclass of entailers.

Dale, I noticed that you said that God's intention would have to have a kind of metaphysical necessity. that's enough to show that it can't come to have such necessity. Presumably the kind of unchangeability in premise 2 is something that can be acquired. Nothing is either metaphysically or logically necessary at any time if it isn't so at every time.

Mike, I'm not sure I quite get this. The supposed laws don't support counterfactuals, do they? They don't even support regularities.

The 'almost-laws' support counterfactuals in the same way that all other laws do. So from (1),
(1) Pr([c](Vx)(Fx -> Gx))=.999

You can infer,

(2) Fa []-> Ga

That is for an extremely highly confirmed causal necessity in (1), we can infer the corresponding counterfactual in (2). This sort of inference better go through, since all we have concerning all of our laws (genuine or not) is extremely highly confirmed regularities.

Dear Jon, my point about my being the truth-maker of my own free actions was not that I am now making those future actions true. I am not currently the truth-maker of my future actions. However, the object that will be the truth-maker of those future actions, I, is something that now exists. Then I was pointing out that for many future events, God is the thing that will be the truth-maker, and he knows what he will do.

Let me explain some of the other points more clearly.

Some actions are such that they intrinsically take time - Robin Hood hitting a target with an arrow, or God smiting the Egyptian harvest with locusts (the latter action requires the locusts to eat the harvest - of course God could just destroy the harvest instantaneously, but he chooses to use locusts). In both cases, I'm supposing, the agent has a single intention, and does it.

If such an action fails, the agent has performed a failed action, demonstrating a lack of omnipotence. The action could fail because of external interuption, or because the agent changed their mind half-way through. The change of mind half-way through is indicative of an imperfection: the good agent is a confident agent, sure of doing the right thing, not bothered by wishy-washy last minute doubts or whims.

So, God acts to smite the harvest with locusts. The smiting takes time, but if it is stopped half-way through, contrary to the original intention, God's action has failed.

So, is there a metaphysically irrevocable intention?

Well here, another analogy occurs to me. I remember when I did a capsize drill in a kayak. The idea is to wobble the kayak until it capsizes. I was very nervous and wobbled it gently, but at last I wobbled it sharply to the left and saw the water rapidly approaching. At the last second, I lost my nerve and pulled back sharply to the right. I pulled back so sharply that I found myself tipping over the other way and, to my horror, my momentum was not too great for me to pull back. I'm happy to say I survived the experience.

Now, could God, when he smites, build in a sort of metaphysical momentum to his smiting of the locusts so that, even if he wants to, he won't be able to withdraw the action? I'm not going to discuss whether he could, because I don't see why if he could, he would. He isn't worried that in the future, he might lose his nerve or change his mind, and so doesn't need metaphysical irrevocability to keep him on track. Much more perfect as an agent is his free refusal to revoke because he is confident he is doing the right thing.

Confidence that one's actions will succeed is a mark of a perfect agent, and God is a supremely perfect agent, entitled on the grounds of confidence in himself to be 100% certain that his action will succeed, even though there is no metaphysical guarantee of irrevocability.

This doesn't ground truth or entail truth of 'this will happen', but it does entitle God to 100% confidence, maximal warrant in the statement 'it will be true that this will happen'.

To put it another way, God decides to smite the harvest with locusts, and orders that this be done. After the harvest is smited, he feels no sense of relief, 'Thank me that went according to plan', he does not become any more certain after the success of the smiting that it would be successful than he was before the smiting. The truth-makers that ground the truth of the event give him no extra degree of certainty beyond the certainty he already had in his own ability to carry out his intention.

Mike, let's distinguish two things. The first possibility is that we have a high degree of confirmation for the categorical component of 1. The second is that we have a high degree of confirmation for 1 itself. that's important, because there's no plausible account of how 1 implies 2 that doesn't treat counterfactuals as probabilities. The probabilistic view of counterfactuals founders on a number of points, however. The one that strikes me as especially telling is that true antecedent and false consequent refutes 2 decisively, but doesn't refute any probabilistic claim.

Jon,

I don't see the worry. No doubt, the closest world in which I flip a fair penny, it comes up heads or tails. The fact that there is some small chance that it lands on its side--and indeed there is such a small chance--does not affect the truth of the counterfactual.
Again, the closest world in which I write my name on the blackboard is a world in which I use my left hand. But no question there is some smallish chance that I use my right hand.
The probabilistic formulation in (1), in normal contexts, does entail (2).

Mike, what you do here is go lax with the supported counterfactuals but precise with the laws. If you go precise with the supported counterfactuals, the only thing the laws imply is that if you flip a fair penny it is really, really likely to come up heads or tails. If you go lax on the laws, then our evidence confirms the categorical claim rather than the one with a probability operator, and then the laws in question imply the counterfactual you want.

But more to the point. I know of no semantics for counterfactuals and probability that will undergird the entailment here. Do you?

Well, yes, I think both Lewis and Stalnaker semantics would work. Clearly, Lewis is not claiming that it is CERTAIN that the tossed coin comes up head or tails. That is evidently not certain. And equally clearly he is not claiming that, given the uncertainty in that proposition, it is false that we're I to toss the coin, it would come up heads or tails. For, the counterfactual is indeed true.
Similar counterfactuals can be based on infinitesimally small chances that the atoms in my chair all pull apart at the same time. There is some chance of that, of course. Byt the counterfactual--that were I to sit in the chair, the atoms would pull apart--is obviously true.

Mike, you are explaining how the counterfactual is true even given the small chance that the consequent won't happen in the presence of the antecedent being true. That's not the question, though. the question is about entailment from a probabilistic law to a counterfactual claim. For that, you have to say something more than this. You have to show that it is impossible for the false consequent world to be as close as a true consequent world. On any proposed measure of closeness I know of, that doesn't follow.

You have to show that it is impossible for the false consequent world to be as close as a true consequent world. On any proposed measure of closeness I know of, that doesn't follow.

There is no law that is anything more than a highly confirmed regularity (unless you're a necessitarian about laws, which I'm guessing you're not). The violation of extremely highly confirmed regularities (Lewis was, after all, a regularity theorist) is one of the major sources of dissimilarity between worlds. In case it matters, he is explicit about that in 'Counterfactual Dependence and Time's Arrow'.
I don't think I've been off topic. I've been insisting all along (I think) that the very laws that everyone takes to uncontroversially support counterfactuals are themselves widely regarded as nothing more highly confirmed regularities. Those who deny this seem to have some a priori access to their necessity. I don't.

Mike, there's a mistake in your last note that is important. Chisholm pointed it out in the 50's. We have evidence E and conclusion p. We draw a line between them to signal that p is the conclusion and E the premise. We assume E doesn't entail p but is strong evidence for it. The question Chisholm asked is: where is the probability that is involved in non-necessitated argumentation of this sort? You've put it on the conclusion. This is precisely the mistake Chisholm addressed. The probability is in the line separating premises and conclusion, not an operator on the conclusion.

That point aside, however, it may be that the laws of nature are pretty generally only probabilistic. But then they don't entail counterfactuals, but only probabilified counterfactuals.

Mike, there's a mistake in your last note that is important. Chisholm pointed it out in the 50's. We have evidence E and conclusion p. We draw a line between them to signal that p is the conclusion and E the premise. We assume E doesn't entail p but is strong evidence for it. The question Chisholm asked is: where is the probability that is involved in non-necessitated argumentation of this sort? You've put it on the conclusion.

Whether there is any mistake or not depends on the kind of argument one is offering. I do not so much as offer an argument in my last note (2/4) or even in the note before that (2/2) that, as far as I can see, might be relevant.

That point aside, however, it may be that the laws of nature are pretty generally only probabilistic. But then they don't entail counterfactuals, but only probabilified counterfactuals.

I don't think I can offer many more counterexamples to this claim. Here is a probabilistic generalization that again entails a counterfactual.

1. For all fair coins P, if I toss P, P will come up heads or tails.
----------------------
2. :. If I were to toss a fair coin P, it would come up heads or tails.

Are you claiming that (2) is not a true non-problified counterfactual?

Mike, you say that there is no law that is any more than a highly confirmed regularity. But you formulate the law as a probabilified regularity rather than letting the confirmation itself encode the probability. That's the mistake Chisholm wrote about. If the evidence only highly confirms the conclusion, it is a mistake to claim that the conclusion is a probability claim.

Your example involving (1) and (2) drops the probability operator on the premise, as you were insisting on earlier. As long as the probability operator isn't there, there's no problem as long as the premise is lawlike.

God does not require determinism in order to foreknow anymore than He needs pre-existing matter in order to create. Open Theism is motivated by the presupposition (stolen from Hellenic metaphysics) that since God is a being his *knowing* that X will happen makes X’s happening *necessary.* Why does free will require the possible falsification of God’s foreknowledge? Why not simply deny that divine foreknowledge is opposed to human freedom in the first place? The assumption that it does has some intuitive appeal but no theological or Christological basis. God cannot sin but that does mean that He is good out of necessity rather than free choice.

A bigger and more obvious problem in Western theology is libertarian freedom in the eschaton; both Catholics and Protestant deny that the saints in the eschaton have libertarian freedom or freedom with alternative possibilities. There is only one object of choice for them: God. This has three flaws: first, it implies that Christ’s human will in the Garden was sinful or evil; secondly, it makes it impossible to explain the fall of Adam or Satan, and thrirdly, it makes it impossible to explain why God would allow the possibility of evil in the first place if we end up right where we started.

Jon, you say,

Your example involving (1) and (2) drops the probability operator on the premise, as you were insisting on earlier.

This can't be the resolution. We all know that (1) is implicitly probablified. To put it another way, no one would take (1) as expessing an exceptionless causal necessity. Nonetheless, they will draw (2).

Yes, Mike, I realize this. (1) is in fact false. But people assent to it. So perhaps what they are thinking is that the probabilistic version of it is true. And (2) doesn't follow from that version, but people draw that conclusion. There's no coherence to be found here.

(Note: I am definitely not an open theist.)

Why can't God today directly cause an effect tomorrow? If so, then God could today cause tomorrow's sunrise, and could then know that tomorrow's sunrise will occur, even if open theism is true.

This is different from God's forming an irrevocable intention to cause a sunrise tomorrow. Rather, it is God's today performing the action of causing tomorrow's sunrise.

Now there is the standard problem for presentists and growing block theorists about non-simultaneous causation. However, presentists and growing-block theorists presumably all take themselves have a solution to that problem (or else they couldn't rationally remain presentists or growing block theorists), say Bigelow's solution or some other. And it does not seem to me that there is a more serious problem with respect to temporally gappy than temporally non-gappy causation.

Now, I suppose there might be a worry about omnipotence. Suppose that an hour ago God has already caused tomorrow's sunrise. It seems to be contrary to divine omnipotence to suppose that God can't counteract the causation.

However, it does not seem at all a restriction on God's power that God cannot prevent things that he has caused. If one thinks this is a restriction, then one can respond that either God is able today to cause an effect tomorrow or not. In either case there is a restriction on omnipotence, then. The restriction involved in allowing God today to cause an effect tomorrow appears to be the preferable one given a perfect being theology, since then God's power is limited by God's earlier choices, and that does not seem a problem.

Hi, Alex, very nice suggestion! I think, though, that the problem of action-at-a-temporal-distance is worse in the case of God than in other sorts of non-simultaneous causation. In the usual cases of diachronic causation, we can appeal to the the ever-present causal structure of the universe to try to explain diachronic causation. But then this explanation won't result in the future being metaphysically guaranteed, as is needed to answer the worry here. So, I think the right reply here is that while God can cause tomorrow's sunrise today, there is no metaphysical structure to exploit to make this causation result in the metaphysical impossibility of the sun's not rising tomorrow. Since that would then be a kind of action-at-a-distance that no possible individual could accomplish, the usual restrictions in the definition of omnipotence will leave God's omnipotence intact.

Jon:

Personally, I find simultaneous causation more mysterious than diachronic causation. (This may, of course, have something to with my being a B-theorist.) I also find diachronic causation mediated by a continuum of intermediate causes more mysterious than direct diachronic causation across a temporal gap. So, oddly enough, the case of causation that you find to be the most problematic is the one I find the least problematic.

I am afraid I don't know what kind of a metaphysical structure makes diachronic causation easier. I can see how on certain stories on which the obtaining of causal relations reduces to nomic facts this would be true, but all such stories seem clearly false to me.

But that is just autobiography (with maybe an implicit invitation for some similar autobiography from you).

On the non-autobiographical side, why can't the open theist say that there is a metaphysical structure in the case of absolute divine promises, namely God's metaphysical inability to act unjustly, and hence his inability to choose contrary to what he had absolutely promised. A restriction of omnipotence to actions that are not unjust seems much less problematic than a restriction of omniscience to things outside the contingent future (in the Christian tradition, it is knowledge of the contingent future that has traditionally been seen as the most distinctive aspect of divine knowledge).

On the other hand, here's a little argument that you're right. Apart from gerrymandered cases, a truthmaker for a disjunction will be a truthmaker for at least one of the disjuncts. There are no disjunctive truthmakers. But if there can be some determined truths about the future, then there can be determined truths like: "Jones will choose A or Jones will choose non-A". (For, surely, it could be determined that Jones will in fact be in a position of a choice between A and non-A, with no possibility of escaping the choice.) This is not a gerrymandered case. Hence, a truthmaker for it will have to be a truthmaker for at least one of the disjuncts. But if the choice is free, this the open theist will not allow.

Yes, talking about causation will be fun. I'm OK with diachronic causation involving events that butt up against each other. And I'm OK, or at least willing to grant, the idea of action at a distance that is undergirded by whatever the truthmakers are for inertial principles that keep in place the causal impetus across the temporal gap. What I find mysterious would be some event in 1920 that leaves no mark in the intervening time period but has an effect in 1980, e.g., with no story at all of any mediation or mark transmission across the intervening time span. I find inertial principles in science very mysterious, because I'm not quite sure what their truthmakers are; but then I'm not sure what the truthmakers are for laws of nature either!

I invite open theists to explain the metaphysical structure that carries anything about God's promises or intentions into the future. We classical theists have an advantage here: immutability. Maybe the moral perfection of God can do the work, but I'm not convinced yet, since breaking promises, even unconditional ones, can be OK (maybe you are more Kantian than I am here, though?), at least for beings without infallible knowledge of the future.

I like your truthmaker argument, though I know the OT reply: give up bivalence!

Leibniz was deeply concerned about Newtonian action at a distance, finding it much more mysterious than sane mechanistic causation by proximity. I think we can learn from Hume that causation between billiard balls is no less mysterious than Newtonian action at a distance. Or, to put it more optimistically, we learn from Hume that Newtonian action at a distance is no more mysterious than causation between billiard balls in contact. (Another nice lesson we learn from this is that the interaction problem for dualism is no more--though no less--a metaphysical difficulty than causation between billiard balls in contact.)

I think the same kind of lesson can be learned about transtemporal causation. I just don't see why two events abutting each other temporally makes it any easier for a causal relation to take place between them than if they are separated.

(Unless maybe one defines time in a Kantian way in terms of causation in such wise that it comes out that wherever there is a basic causal relation there is temporal contiguity? But while I am sympathetic to defining time in terms of causation, I do not see why the consequence should follow.)

As for giving up bivalence, that is indeed the reply, but the weirdness of a truthmaker for a disjunction without a truthmaker for either disjunct remains. In other words, it seems to me that No-Bivalence & Truthmaker is more problematic than No-Bivalence & No-Truthmaker.

I think I'd say this about causation, instead of saying that mechanistic causation is mysterious. If Hume showed us something, it is that reductive attempts to understand causation make it quite mysterious, since none of them work! One reason for thinking that mechanistic causation is less mysterious is that you can see it happen. To the extent that perception is a source of knowledge, any perceptible relation between objects or events has an advantage in terms of our ability to understand it over relations that aren't perceptible. Something like this has to lie at the heart of why we find telekenesis so mysterious, and the claims for such powers beyond belief.

On logic, I expect we agree on this: the path to philosophical destruction is littered with denials of bivalence, excluded middle, abandonments of the T-schema, etc. Maybe the philosophical hell at the end of this path is dialetheism!