Assume our world is indeterministic and consider the evaluation of (C),
C. If Nixon had pressed the nuclear button, God would have brought it about that the future was no different from the actual future.
The asymmetry of counterfactual dependence in indeterministic worlds mirrors, as a matter of contingent fact, the asymmetry of quasi-miracles. In the most similar worlds where Nixon pressed the nuclear button he performed some free action. No divergence miracle was necessary. But what was the future of that world like? Do the closest worlds in which Nixon pressed the nuclear button reconverge to the actual future? Re-convergence to the actual future requires a quasi-miracle that, as a matter of contingent fact, constitutes a significant dissimilarity between worlds. A quasi-miracle could bring about re-convergence should an extremely improbable pattern of chance events cover every trace of Nixon's having pressed the nuclear button. But it would be remarkable that just the chance pattern needed to cover every effect of that counterfactual hypothesis is the pattern that occurs.
But, now, are quasi-miracles remarkable under the assumption that we inhabit not some ordinary indeterministic world, but a Molinist indeterministic world? There seems no reason to think so. For all we know about God's purposes and goals for individuals, nations, and humankind, and for all we know about his ways of achieving these goals, God is using quasi-miracles constantly to ensure the realization of his providential aims. After all, quasi-miracles require no violation of law, not even probablistic law. So, it seems to be a contingent fact about Molinist indeterministic worlds that quasi-miracles do not constitute a significant dissimilarity between worlds. If the world we inhabit is a Molinist indeterministic world then, at the very least, we are not sure whether (C) is true.
But now to the problem. If we are not sure if (C) is true, then we are also not sure whether (C') is true.
C'. If Nixon had pressed the button, then there would have been a nuclear holocaust.
But, on the contrary, we do know that (C') is true. So we know too that (C) is false. That is, either we are in radical skepticism about which counterfactuals are true or this is not a Molinist world. Therefore this is not a Molinist world.
Molina and modern Molinists like flint deny that there are true counterfactuals of freedom for GOd's case. So we know (say they) that your first one isn't true.
Molina and modern Molinists like flint deny that there are true counterfactuals of freedom for GOd's case. So we know (say they) that your first one isn't true.
Right, but there are counterfactuals of divine freedom indicating what God would do in various circumstances. These are a part of God's free knowledge. (C) is a counterfactual fo divine freedom. (Tom Flint discusses them in DP:TMA, p. 46 ff.)
All you give evidence for is that on Molinism, at best we do not happen to be certain that C is true. It looks to me as if you have confused that actual uncertainty ("at the very least we are not sure C is true") with the necessity that we are unsure. Only if our uncertainty necessarily follows Molinism does your conclusion necessarily follow. That is, certainty about C does not imply the falsehood of Molinism unless uncertainty about C necessarily follows Molinism. But, for instance, there could be Molinism and certainty about C, say through special revelation.
I understand that you have asserted a dilemma to overcome this problem. But I guess it is ultimately that dilemma I am disputing. It does not require radical skepticism to admit the truth of counterfactuals whose only reason for being dismissed is that they have not been and would not normally be expected to be actualized.
I'm no advocate of Molinism, but I'm not sure your argument rebuts it.
Only if our uncertainty necessarily follows Molinism does your conclusion necessarily follow.
I agree that I have not assumed that Molnists necessarily fail to know that C is false. Perhaps there are some worlds in which God does reveals to us that C is false. But it has nothing to do with my conclusion. The fact is, under the assumption that Molinism is true, we do not actually know that C is false. And since, assuming Molinism, we do not actually know that C is false, it follows that we do not actually know that C1 is true. Unfortunately for Molinism, we do know that C1 is true. Therefore either (Molinism is true and) we are in radical skepticism concerning which counterfactuals are true or this is not a Molinist world. Therefore this is not a Molinist world.