Saturday, December 5, 2009

Two Exchanges on Foreknowledge and the Necessity of the Past

Below are two recent exchanges I had with Paul Manata and Ron Di Gacomo on foreknowledge and freedom; specifically the argument from the necessity of the past. In both I argued that the classic argument based on the necessity of the past conflates truths and the basis of truth; there's a difference between me and propositions about me. In Paul's case I asked him to try to reformulate the argument, but he declined. In Ron's case I tried to reformulate the argument for him and he rejected my reformulation. Either way, these exchanges lead me to be more convinced of my hunch that the classic formulation is a train wreck in light of the distinction between truth and the basis of truth.

I will be in blue; Paul and Ron in red.

Exchange with Paul:

The ability to do ~A being consistent with God's forebelief that you will A strikes me as obviously false given the foreknowledge argument. In the least, that's what is up for debate.

Apparently, what you mean is that you have the "ability to do otherwise" in the sense that you have the ability to do something, such that if you were to do it, the past would have been different. Classical compatibilists say the same.

The argument mirrors Van Inwagen's transfer of powerlessness argument against compatibilists. TOP: If P is true, and P entails Q, then if no one has the power to bring it about that not-P, and no one has the power to bring it about that it is not the case that P entails Q, then no one has the power to bring it about that not-Q.

P stands for either propositions about the history of the world conjoined with the laws of nature, and Q stands for some action that obtains. P can also stand for God's forebelief.Thus, we can argue:1. P is true.2. No one has the power to bring it about that not-P (no one can change the past).3. P entails Q.4. No one has the power to bring it about that P does not entail Q.5. Therefore, no one has the power to bring it about that not-Q.(1 states God has always had a particular belief; 2 states Plantinga and Flint &co. claim that no one has the power to bring about God's past belief; 3 states a necessary connection between omniscience and truth; if 3 is true, then 7 is necessarily true, and 8 goes through)

This argument has been championed by incompatibilists. It has also been champtions by incompatibilists of another stripe, incompatibilists about foreknowledge and LFW.

Clearly you don't like TOP, so you, like Flint, will offer TOP*TOP* If P is true, and P entails Q, then if no one has the pwer to bring about not-P, and no one has the power to bring it about that it is not the case that P entails Q, one may nevertheless have the power to bring it about that not-Q.Butr TOP* can be used by compatibilists of another variety. Just switch out P = God's forebelief with P = God's decree/ the history of thr world and laws of nature/ etc.Both compatibilists will claim that one has the power to do soemthing such that if he were to do it the past would have been different.Both sides reduce to claiming to have a power that is, given the current circumstances, impossible to be exercised.

I made two distinctions that your arguments don't take into account. The first was the difference between doing otherwise and the ability to do otherwise. The second was between truth and the basis of truth. Compatiblist define the ability to do otherwise in terms of a counterfacual past I could have chosen otherwise if I had wanted to or I could have done otherwise if I had chosen to. Libertarians (Occhamists) do not. They say that given the actual past doing otherwise is within my power. So comatiblists posit an ability requiring a counterfactual past, but Occhamists do not. So comptiblists face the consequence argument but Occhamists don't.

Q is an event. P is a proposition. How could a proposition entail an event? It could entail a proposition about an event but not an event. If you try to rephrase keeping the distinction in mind I think you will see the issue with the argument. Or if you try to construct a new argument based on the relationship between propositions and events, I think you will see the force of Daniel N’s point about the relationship between God’s foreknowledge and the event. (Daniel N had said "I think we need to make a distinction between logical and chronological order. God knows you will do A at t3 because you will choose A, (rather than you choosing A because God knew it).


"Q is an event. P is a proposition. How could a proposition entail an event?"

In the anti-determinist argument, P is propositions about the history of the world, conjoined with laws of nature. In the foreknowledge argument, P was God's belief.

"If you try to rephrase keeping the distinction in mind I think you will see the issue with the argument."

Sorry, but I took my phrasing from Plantinga, Flint, Freddoso, Hasker, Hunt, Zagzebski, etc. Every one of them agrees that "If God believes that Q will happen, then Q will happen." This argument is then used, as the logical steps made clear, to show that you do not now have the power to do otherwise.P entails q =df ~Poss(p & ~q).

If you disagree then you must hold to Open Theism.

I agree "if God believes that Q will happen, then Q will happen." But what is entailed is not Q; rather it’s the proposition PQ: ‘Q will happen’. Events are not the same as truths about events. I understand that this distinction is not consistently maintained by all parties in the literature, but that doesn’t mean it should not be.

-------------------------
Exchange with Ron:

This argument is from your previous post:http://reformedapologist.blogspot.com/2007/12/molinists-and-calvinists-agree-in.html

Establish the necessity of God’s belief about Tom’s choice:
1. 100 years ago God believed that Tom will do x tomorrow
2. If x is believed in the past, it is now necessary that x was believed then
3. It is now necessary that 100 years ago God believed that Tom will do x tomorrow

Establish the necessity of Tom’s choice, given the necessity of God’s belief:
4. Necessarily, if 100 years ago God believed Tom will do x tomorrow, then Tom will do x tomorrow
5. If p {i.e. God's historical belief about Tom's choice} is now necessary (3), and necessarily if p, then q; then q {i.e. Tom's choice of x tomorrow: (consequent from 4)} is now necessary [transfer of necessity principle]
6. Therefore, it is now necessary that Tom will do x tomorrow [3, 4 and 5]
Establish that Tom does not act freely, given the necessity of Tom’s choice:
7. If it is now necessary that Tom will do x tomorrow, then Tom cannot do otherwise
8. Therefore, Tom cannot do otherwise than x tomorrow
9. If one cannot do otherwise, then one does not act freely
10. Therefore, when Tom does x tomorrow, he will not do it freely

The problem I see with it is in 4 & 5; specifically, it conflates truth with the basis of truth. The proposition 'Tom does X' isn't the same thing as Tom doing X; one is a proposition, the other is a person performing an action.

4 should state:

4~. Necessarily, if 100 years ago God believed Tom will do x tomorrow, then the proposition 'Tom will do x tomorrow' is true.

4~ is true, but 4 is not. A truth cannot entail an event, only another truth. 4 conflates truths with the basis of truths. 4 also slides from God’s belief (which is a historic event or state) to the truth of God’s belief. It moves from ‘the necessity of the past’ as argued in 1-3 to necessity based on God’s essential omniscience. The necessity of the past is causal; since it assumes causation works forward in time. God’s essential omniscience means it’s not logically possible for God’s beliefs to be false (i.e. ‘God’s false beliefs’ implies a contradiction). So the argument has changed its basis from causal necessity to logical necessity.

With 4~ in mind, 5 should be revised to:
5~. If p {i.e. God's historical belief about Tom's choice} is now necessary (given God’s essential omniscience), and necessarily if p, then q; then q {i.e. that the proposition 'Tom will choose x tomorrow' is true: (consequent from 4~)} is now necessary [transfer of necessity principle]

5 subtly and invalidly moved from the logical necessity of the combination of two truths to the causal necessity of a past and future event.

From this, it should be reasonably clear that the type of necessity transferred is not causal necessity or the necessity of the past. Rather it's the logical necessity of the combination. P and not-Q are logically incompossible. But this does not threaten libertarian freedom. LFW assumes the causal possibility of choosing otherwise (i.e. it’s with Tom’s power to choose X or nonX and the logical possibility (i.e. ‘Tom chooses X’ is not a contradiction.)

The rest of the argument does not follow after 5 is corrected.

"The problem I see with it is in 4 & 5; specifically, it conflates truth with the basis of truth. The proposition 'Tom does X' isn't the same thing as Tom doing X; one is a proposition, the other is a person performing an action. 4 should state: 4~. Necessarily, if 100 years ago God believed Tom will do x tomorrow, then the proposition 'Tom will do x tomorrow' is true.4~ is true, but 4 is not. A truth cannot entail an event, only another truth. 4 conflates truths with the basis of truths.”

First off, given the antecedent of 4, that which the consequent contemplates is true, regardless of any metaphysical cause of the consequent. Secondly, your strictures regarding the proposition “Tom will do x tomorrow” and the proposition: “the proposition Tom will do x tomorrow is true” affords you nothing. Neither addresses the metaphysical question that is apparently driving your arbitrariness.

4 also slides from God’s belief (which is a historic event or state) to the truth of God’s belief.”

I’m happy to state the argument in terms of God’s knowledge because what he believes he knows. Moreover, God can be taken out of the argument because the issue pertains to truth, which so happens to be believed and known.

It moves from ‘the necessity of the past’ as argued in 1-3 to necessity based on God’s essential omniscience.”

My argument has nothing to do with God’s essential omniscience. That’s something you’ve imposed upon it. All that is in view is one bit of knowledge, not omniscience. Moreover, God’s knowledge of a past proposition that contemplates an act still future could just as easily been worded as your knowledge of a past proposition that contemplates an act still future; knowledge presupposes the truth of the proposition, no matter who knows it, which is why we can leave knowledge out of it; the issue is that the proposition is true and not whether it is known.

“The necessity of the past is causal; since it assumes causation works forward in time.”

That’s a bit ambiguous.

“God’s essential omniscience means it’s not logically possible for God’s beliefs to be false (i.e. ‘God’s false beliefs’ implies a contradiction). So the argument has changed its basis from causal necessity to logical necessity.”

Even allowing for your premise that my argument somehow pivots on omniscience, you have just taken a quantum leap in reason. Allow me to jump down to the bottom of the funnel. You closed with: “LFW assumes the causal possibility of choosing otherwise (i.e. it’s with Tom’s power to choose X or nonX and the logical possibility (i.e. ‘Tom chooses X’ is not a contradiction.)

Why should we believe that if it is “logically impossible” for x to be caused that it is indeed true that x might be caused? You are left with something that is logically impossible to be caused, yet metaphysically possible to be caused! Need not your metaphysic comport with your logic, or may the two be isolated and independent? In sum, you have an illogical metaphysic because it need not comport with what you would like to call logical truth.

You seem to be arguing that the truth of the proposition is part of the past and therefore necessary. I disagree. Even if we grant that the truth of propositions is part of the past (which seems true in one sense and not in another) I don't think we can conclude they are necessary. Past events and states are necessary because they have been caused to be and also because there is no such thing as retro causation but past truths are not necessary.

Perhaps we also disagree on the definition of logical necessity. A truth is logically necessary if its negation is self contradictory. This applies to a truth by itself or multiple truths taken in combination. But if two truths in combination are logically necessary that does not mean each is separately. Assuming that they are is a division fallacy. So I maintain 'Tom does A at T 'is logically possible but the combo of 'Tom will do A' and 'God's past belief that Tom will do B at T was true' is logically impossible. Similarly, the combo of 'Tom will do A' and 'Tom will do non-A' is logically impossible, even if 'Tom will do A' is logically possible.

So looking at 4 & 5 once more and with your kind permission changing God's past belief with a true proposition and also understanding the consequent as a propositional truth as well we get:

4*. Necessarily, if 100 years ago 'Tom will do x at T' was true, then 'Tom will do x T' is true

Thus 4 is reduced to a totology and we can get rid of it.

Now 5, in light of 3...

3. It is now necessary that 100 years ago God believed that Tom will do x a T5. If p {i.e. God's historical belief about Tom's choice} is now necessary (3), and necessarily if p, then q; then q {i.e. Tom's choice of x tomorrow: (consequent from 4)} is now necessary [transfer of necessity principle]

First, 5 should be changed to:

5.* If p {i.e.'Tom will do x at T' was true} is now necessary (3), and necessarily if p, then q; then q {i.e. 'Tom will do x T' is true} is now necessary [transfer of necessity principle]

Second, to assume 5 follows from 3 is an invalid conflation of a truth ('Tom will do x at T'was true) and a past event (God's past belief). So 5 as it stands should be rejected and it doesn't seem in can be repaired.


You are not interacting with the refutation that is before you.

“You seem to be arguing that the truth of the proposition is part of the past and therefore necessary.”

Yes, that is the argument but it is not assumed; it is deduced. You have yet to interact with the proof. You are simply presupposing that it cannot be true and then simply asserting the same, that it cannot be true.

“I disagree.”

Of course you disagree - you’re an Arminian. I’m waiting upon your interaction with the refutation of your assertions which, by the way, never addressed the original proof.

“Even if we grant that the truth of propositions is part of the past (which seems true in one sense and not in another) I don't think we can conclude they are necessary.”

Ah, but a proof with a valid form concludes they are necessary. The only question is whether the premises within the proof are true, which you have yet to deal with.

“Past events and states are necessary because they have been caused to be.””

In this discussion the only thing that is past is the future tense truth proposition about a future creaturely choice. That the proposition is caused is irrelevant to the question of whether the action the proposition contemplates will be caused. That this escapes you indicates that you are a novice in this matter. That, of course, is not shameful (apart from the fact that you are posturing yourself as one who has something significant to offer).

“Perhaps we also disagree on the definition of logical necessity. A truth is logically necessary if its negation is self contradictory.”

Dan, I’m well aware of what a logical necessity entails. What you fail to grasp is that a logical necessity about a metaphysical occurrence implies the necessity of the occurrence. To assume the contrary is to affirm that something that must logically occur need not metaphysically occur; which is to assert that causality is not logically relevant – which is to suggest that causality is not logical. No Molinist has dealt with the weight of the quandry. They assume contradictory truths, that something will occur yet might not occur. As for the rest of your post, you’re simply repeating yourself and in doing so showing yourself unwilling (or unable) to deal with your internal inconsistencies and the clear argument that remains before you.

Dan, that is strike 2. One more time and you're out and if you do nothing more than repeat yourself, don't expect to for your post to be published.

No comments: