Why would omniscience and free will be mutually exclusive?

How is "the future is already determined" this phrase "ambiguous"? If there are "contengent events" and God has "infallible foreknowledge" then he obviously knows how these contingencies will be resolved.

You seem to hold a logically very inconsistent POV.

 

Log in or Sign up to hide all adverts.

This betrays no indecision. I just thought infallible foreknowledge entailed successful prediction, predictions just being beliefs regarding the future regardless of the belief acquisition method.

What will happen is what will happen. That's tautological and has no bearing on the modal status of the foreknown events, which may still be either necessary or contingent. As I have argued, foreknowledge of an event only seems to entail its necessity when one invalidly infers the claim (K(P) -> Nec(P)) from the claim Nec(K(P) -> P).

I.e. "If (I know that P) then (P necessarily holds)" can't be inferred from "Necessarily (if I know that P then P is true).

I may know that P is true in the actual world, and there are still possible worlds in which P is false (and in which I may merely falsely believe that P is true, or, if the knower is God, in which he knows P to be false). So, my (or God's) knowing P entails the truth of P but not its necessity (i.e. that it is true in all possible worlds).

The natural intuition of people seem to underwrite this invalid inference from necessary conditionality to conditional necessity only because people aren't used to think carefully about the scope of modal operators.

 

Log in or Sign up to hide all adverts.

Contingencies qua alethic modalities need not be "resolved". You are thinking of subjective epistemic modalities: "possible for all I know", and "necessary for all I know". These may be resolved when I acquire more complete knowledge. Alethic modalities regarding future events are objective (just like the sort of necessity that attaches to laws of nature) and they aren't 'resolved' when the the time comes.

The claim that an event is contingent just is the claim that the world could have been such that this event never happened. (This is the same as saying that there is a 'possible world' in which the event never happened). Since claims of (alethic) necessity are predicated not only on what happens in the actual world but also on what happens in all possible worlds, they aren't resolved by what is seen to happen in the actual world. Once the glass of water fell and broke that doesn't settle that necessarily the glass of water must have fallen and broken. That only settles that it actually fell and broke. It is rather such things as natural laws and the truth or falsity of determinism that have bearing on objective alethic modalities like 'historical inevitability' may be one.

 

Log in or Sign up to hide all adverts.

I do have to say however that even if it was somehow possible that this universe could have been different, and that God's infallible foreknowledge could have been of something else, although it might technically satisfy one of the prerequisites for one definition of free-will (which is that we have it if our choices could have been different), I don't think it satisfies the definition of free-will that is consistent with what we intuitively believe free-will to be.

In other words, the existence of omniscience in any particular universe seems to demand that events can only play out one way in that universe, yet our intuitive understanding of free-will (or belief in how it operates) would only be consistent with a universe in which every single choice is an individual variable in that same universe.

I think we're suffering from a lot of different maladies in this thread (a couple of which have been my fault). Probably the most significant of which however is a failure to have properly defined the key elements in play at the very beginning. What definition of free-will are we working with? What definition of omniscience are we working with? What definition of God are we working with, if any? These things have been hashed out on the fly with varying degrees of success, and from my perspective at least, they are all things that form a fundamental part of any conclusion.

 

That's the heart of the matter. The complexity of this discussion/debate is generated by trying to solve, or skirt, the fundamental, apparently irresolvable paradox that

a) the future is already determined

and

b) all future events are (i.e., the future is) contingent.

 

One way to solve a paradox is to show that, despite appearances to the contrary, two statements aren't contradictory. This is the way Einstein solved the paradox of the measured speed of light being both unaffected by the speed of the observer (as couldn't seem to be expected from waves) or by the speed of the source (as couldn't seem to be expected from particles). Einstein showed how a proper understanding of the way space and time measurements relate to various inertial reference frames could lead one to expect both results to hold without contradiction.

Another way is to show that one of the two incompatible claims is actually false.

Now, it seems to me that both claims (a) and (b) above are false. This would solve the paradox.

Please Register or Log in to view the hidden image!

 

The problem with that is that the explanation for how both are false becomes so intricately complex as to be virtually as incomprehensible as the paradox of both being true. Indeed, that to me is what modern physics and cosmology do whenever they venture beyond their bounds into trying to construct answers to the unanswerable philosophical questions.

 

I think you are perfectly right to mention that the kind of modal possibility (of different courses of action) that I have argued isn't precluded by God's having infallible foreknowledge of our actions isn't sufficient for guaranteeing our will to be free. I have only argued that God's foreknowledge didn't entail the modal necessity (of the course of action we actually take) that would directly impugn our free-will (assuming the validity of the requirement expressed by the principle or alternate possibility).

What is needed in addition, I think, for a defence of the possibility of human free-will, once we've skirted the threat from God's infallible foreknowledge, is a more restricted notion of modal possibility consisting in the negation that only one among alternate courses of actions that we intuitively hold to be open to a practically deliberating human agent is historically necessitated. But this is a threat that stems from determinism. Determinism is a thesis that seems to follows both from the (basically correct) idea of causal determination of physical events and some further (questionable) claims of supervenience mental events (including intentions and conscious decision processes) would have over physical events. I think this threat of determinism (and hence of historical necessitation of our actions) can also be skirted by defenders of free-will (not, though, merely through appeal to the indeterminacy of quantum phenomena -- God's proclivity in throwing dices can't make us free!) but further discussion of this could be lengthy and off-topic.

This is a bit muddled. "Universe" isn't synonymous with "possible world". The latter is a technical term that does not denote concrete universes. Talk of possible worlds should not be understood to presuppose the existence of many universes. A possible world is, by definition, the (unique) universe as it could have evolved had its history been different that it actually has been until some definite point in time. In other words, a possible world just is a possible (non-actual) alternate history of the actual world.

So, all that free-will demands is the absence of historical necessitation of our actual actions. It demands that we could possibly have done otherwise, not in another universe, but in this very universe had its history been different in some relevant respect -- which possibility mustn't be precluded by historical necessitation.

I think this is just as things should be with philosophical discussions. Definitions must often remain implicit and terms are explained as the need arises. Philosophy can't emulate sciences in that regard, for there isn't something like "the state of current philosophical knowledge" as there is in science. When there are some broadly accepted theoretical frameworks, there are, correlative to them, precise definitions of the associated theoretical terms. Acceptance of the definitions entails broad acceptance of the framework. Accepting fixed definitions of technical terms in philosophy would prejudge essentially contentious philosophical theses. So, one must leave room for semantic negotiation and attempt to understand how, and why, one's interlocutor insists for using his words in a was that may seem peculiar.

I fully agree, though, that there is a recurrent need to make explicit one's tacit assumptions. This just doesn't often yield fixed and uncontentious definitions.

Incidentally, I am also an advocate of David Wiggins's suggestion that whenever one feels the urge to express a philosophical claim or argument with the help of quasi-technical philosophical vocabulary, one must always show some willingness and ability to rephrase one's claims in plain English (or French, Spanish, etc. -- whatever one's native language is!)

Though doing this, one mostly does away with the need of explicit definitions and this verbal exercise all but guarantees that one wasn't (unintentionally) making use of jargon for covertly speaking nonsense.

 

There is an alternate explanation for your not having understood an explanation than its being incomprehensible. You may not have tried hard enough to understand it or failed to make the proper request for explanation of more obscure points.

I agree that specialised scientists sometimes do a poor job at exploring philosophical problems, and are liable to speak nonsense when they venture outside of their narrow fields of expertise. It is of course the prerogative of philosophers to be able to speak nonsense within their own fields

Please Register or Log in to view the hidden image!

 

You seem to be assuming that all disagreement with your views can be dismissed in the same easy way.

Unfortunately, that oversimplifies things and threatens to reduce opponents' arguments to caricatures.

For one thing, it simply ignores the infallibility condition. That condition introduces peculiar necessities that don't arise in your common-sensical 'Joe' examples.

 

They are? They are the source of what you've been so stoutly denying -- Nec(P). Given God's omniscient and infallible knowledge of P, not only does P have to actually take place, there's not even any hypothetical possibility left that it might not have happened or that anything else might have happened instead.

You apparently believe that you can escape from those implications by insisting that God might have known something else, something other than P.

It surely contradicts the second part of that. God's predicting not-P isn't consistent with God's predicting P.

I'm not arguing that God couldn't have hypothetically done all manner of inconsistent things. I'm just pointing out that the inconsistent things are inconsistent.

Sure, God might have predicted Q instead of P, or even Z maybe. There are all kinds of alternative possibilities, if that's where we want to go.

The thing is, if we were discussing the implications of a hypothetical (if transcendent) state of affairs, namely God's possessing infallible knowledge of P, then that's what we should concentrate on.

And in that case, the necessity does seem to be there. You've acknowledged it yourself.

Why do you assume that God exists in all the possible worlds? Why do you assume that God knows anything? Why do you assume that God's knowledge must be infallible and omniscient?

You do seem to be holding some of the antecendent conditions constant. So why the freedom to change God's knowledge of P into knowledge of something else instead? How did the content of God's knowledge become a variable instead of a constant?

It seems to me that in all possible worlds in which God exists, knows things, and omnisciently, eternally and infallibly knows that P is going to happen, then necessarily P is going to happen. There's no possiblity that anything else could have happened.

If we switch instead to a different subset of possible worlds in which God exists, knows things, and omnisciently, eternally and infallibly knows that Q is going to happen, then Q is going to necessarily happen in all of those worlds.

Arguing that we humans are the free agents in all of this and that God's knowledge is a logical variable that's being assigned a determinate value by whatever it is that we freely choose to do at some point in time, is going to be problematic if we stipulate and agree that God's foreknowledge that P would occur is eternal and infallible.

That's why all the talk about 'Joe' and sugar cubes isn't a very good analogy. Our fallible and temporal human knowledge can't simply be equated with infallible eternal divine knowledge and it doesn't possess the same logical properties.

So the logical formula that you keep repeating isn't necessarily the correct formulization of our problem either. Your formula doesn't even address the infallibility condition or any of the peculiar necessities that appear to be associated with it.

 

I must sometimes restate my points when I feel they have been either misunderstood or ignored. I mean no offence. If you likewise think I have overlooked your rejoinders to some of my points, feel free to remind me of them.

I can clearly see that the infallibility condition introduces another form of necessity. Indeed I have already stated several times what I take this form of necessity to be. (So, I am hardly ignoring the infallibility condition, as you now complain.) This is the necessity that an infallible epistemic subject never holds a belief about some future event that will turn out not to have constituted knowledge. I have noted this Nec(B(P) -> K(P)). Necessarily, if the agent believes P, then he foreknows that P. The most obvious corollary of this necessitation claim, since knowledge entails truth, is that Nec(B(P) -> P). Are there other peculiar necessities entailed by infallible foreknowledge that you think I may have overlooked?

 

Yes I am insisting on the possibility that God may have foreknown that Q if Q had been the case rather than P. That's because that possibility hasn't yet been ruled out by arguments presented so far.

To summarise my position, it just seems false to me that Nec(P) follows logically from the premises

(1) Nec(K(P) -> P), (In each possible world where God knows P, P);
(2) Nec(P -> B(P)&K(P)), (In each possible world where P, God believes it and knows it);
(3) Nec(B(P) -> K(P) (In each possible world where God believes P, he knows it);
(4) P (P is true in the actual world). ('P' now used as a constant)

You are right that for purpose of the present discussion of the implications from God's omniscience -- which is taken to entail infallible foreknowledge -- our attention is restricted to possible worlds in which God exists and he is omniscient. Premise (1) states that knowledge (either fallible or infallible) entails truth. Premise (2) states that the knower is omniscient. Premise (3) states that his knowledge (including his foreknowledge) is infallible. Premise (4) states that P is true in the actual world.

Try as you may, you can't derive Nec(P) from the above premises. This purported conclusion uses the Nec operator with narrow scope. You just can't seem to be able to move from those premises to your intended conclusion without committing, as you've already done several times now (see further example below), the so called modal scope fallacy. You're not alone. Other posters here have seemed oblivious the different scopes of the modal operator in the purported conclusion from its scope in the premises listed above. There is a clear explanation of this modal-logical fallacy here:

http://www.fallacyfiles.org/modlscop.html

This must of course be agreed upon for purpose of exploring what logical consequences the necessary existence of a necessarily omniscient God who infallibly foreknows our actions may have regarding the possibility of our free-will.

That's no 'freedom' of mine. It has been assumed for purpose of discussion that God is omniscient. It hasn't been assumed for purpose of discussion that there aren't actual actions of mine that are merely contingent. This claim rather is the conclusion you seek to derive. But it is assumed that if there are any such contingent actions, then, whatever else I might have done instead, God would have known it. Contemplating the logical possibility (indeed, just failing to ruling it out at the onset) that I may have done something else than what I did doesn't mean that I am "changing" God's actual knowledge into something else. It is just a case of failing to rule out a priori, and with no argument, that he would still have known what I would have done if I had done something different than what I actually did. Not only can't I rule it out -- it even follows from premise (2) above whenever some action of mine is contingent.

You are again trading on equivocation regarding the scope of the modal operator 'necessarily'. Please, have a look at the short note linked above explaining the modal scope fallacy. Pay attention to the comment about the scope of modalities being ambiguous in English grammar.

This sentence makes little sense. You can't speak of things being necessary or possible (while using the modal operators with narrow scope) in individual possible worlds or in a proper subset of the set of all possible worlds. That's because to say that something is necessary (or possible) just is to say that it occurs in all (or at least one) possible world(s). This is how you must speak if 'necessary'/'possible' are meant with narrow scope.

You may instead mean to speak of necessary conditionality (where the modal operator has wide scope). In that case the relevant subset of possible worlds comprises those worlds in which the antecedent of the conditional holds true. Then, necessarily, the consequent must hold true in those worlds. Notice the ambiguous scope of 'necessarily' in the previous sentence? It must be interpreted thus: Nec(God exists & ... & K(Q) -> Q). If you now interpret it to mean Nec(Q), you've just committed the modal scope fallacy again.

 

Where's K(P)?

This whole thread is about the logical implications that God's eternal and infallible foreknowledge of P has for the possibility of something other than P occurring instead. So clearly, the perfect foreknowledge needs to be among our premises.

Then you apparently do accept the idea of conditional necessities, necessities that only hold in that subset of possible worlds in which antecedent conditions hold. You seem to battle that idea elsewhere.

In other words Nec(A => B) implies that B is necessarily the case in all possible worlds in which A holds true. A might indeed be contingent, but the relationship between A and B isn't.

I don't think that anyone in this thread has tried to argue for Nec(P) in all possible worlds, including those with no God, no knowledge and so on. Nec(P) and your strong construal of it are your ideas, not ours.

My own goals are more modest. The thread's about what implications can be drawn from God's supposedly imfallible and eternal knowledge of P. Presumably God knows what you are going to do before you do it. So what freedom is there to do something different than God already knows you are going to do?

Nec(K(P) => P)

If K(P) then necessarily P

Please note that blowing modal-scope smoke won't work here. This is just your (1) from up above and does NOT assert or attempt to prove your stronger version of Nec(P). It's not asserting that P is true in all possible worlds without exception. It's only asserting that it's the case that IF the antecedent K(P) condition holds - If God has infallible foreknowledge that somebody is going to do something - then P necessarily follows and the person is inevitably going to do it.

Nobody is suggesting that P must necessarily be the case even absent K(P)

 

This is too plateresquely Plantingian for me.

 

If we take this into the theistic context, then freedom might mean quite different things than in a secular context.

I would argue that in a theistic context, our freedom has nothing to do with God's omniscience and the thereby implied determinism; but with the kind of actions that we do. In short, in a theistic context, freedom could simply mean having the choice and the ability to behave in a godly manner or in an ungodly manner, to behave righteously or to sin.

Note that angels are said to have no free will and that they cannot but behave righteously; they are unable to desire to sin and they are unable to sin.
Such reasoning suggests that freedom pertains to the kind of acts, not to God's omniscience or determinism.

In a theistic context, God is omnipotent, and so God's will gets done no matter what, trumping everything else (in that sense, it doesn't matter if God has foreknowledge or not). This can be understood as a kind of determinism.
But like I said: perhaps in a theistic context, freedom simply has to do with the kind of acts that we are able to do.

 

K(P) is the trivial consequence of premises (2) and (4) so there is no need to add it as a separate premise.

I've insisted several times on distinguishing (what I dubbed) conditional necessitation from necessary conditionals. The first has the form Nec(P) -> Nec(Q), where 'Nec' has narrow scope. If P is necessary then Q is true in all possible worlds. Necessary conditionals have the form Nec(P -> Q). 'Nec' now has wide scope. Q is true in all possible worlds in which P is true. I have never been "battling" the idea of necessary conditional. I have warned people from deriving invalidly conditional necessitation from necessary conditionals. That is precisely committing the modal scope fallacy.

Early on, in this thread, you wrote:

There, you explicitly contemplated the possibility that Nec(P) might not follow from the premise of God's infallible knowledge. You can't possibly have been thinking of a possibility for Nec(K(P) -> P) not to follow from this premise. It follows trivially from it, doesn't it? Now you are claiming that you always meant Nec(K(P) -> P) whenever you asserted Nec(P) as a consequence of the premise. I'll accept this, from now on, for purpose of discussion.

I would have thought that the scholastic, modern and contemporary literature about the subject of God's foreknowledge and free-will is concerned with the task of assessing whether or not some stronger form of modality than mere necessary conditionality can be derived from the premise of God's infallible foreknowledge. The mere necessary conditionality Nec(K(P) -> P) simply the premise accepted on both sides of this old debate.

For instance, consider the Basic Argument for Theological Fatalism sketched in this Encyclopaedia article, which someone already quoted here a few days ago.

http://plato.stanford.edu/entries/free-will-foreknowledge/

Notice that premise

"(4) Necessarily, if yesterday God believed T, then T. [Definition of “infallibility”]"

Just is assumed with no argument? What the arguments aims to prove is narrow scope necessity.

You may not realize how trivial and inconsequential wide scope necessary conditionality -- that you now say it was all along your "modest goal" to defend -- really is.

To see this consider first that necessary conditionality holds quite independently of the infallibility condition. Suppose God has fallible foreknowledge of some of my actions. Suppose he contingently foreknows one of them: P. What consequence does this have for the necessary conditional Nec(K(P) -> P)? Well, none at all. The necessary conditional still holds intact since it is only concerned with possible worlds in which God's foreknowledge does not fail. That would mean that some of my actions might be free: those that God failed to predict, but whenever he will have managed (contingently) to know what I will have done, those actions will not have been free!

Second, consider the case where we are concerned not with God's foreknowledge but with his infallible 'retro-knowledge', or infallible memory, if you will. Then the premise Nec(K(P) -> P) is replaces with Nec(R(P) -> P). In all possible worlds in which God remembers my having done P, I did P. Would you hold that God's merely having the power to watch me and not ever forget what I did would constitute a relevant restriction on the possibility that I would have done otherwise?

The main trouble with your insistence that it is necessary conditionality that is troublesome for free-will is that necessary conditionality boils down to the trivial claim: Whenever I do something, God knew it and hence I will do it. But unless God's knowledge constitutes some sort of a causal antecedent, and not merely a necessary correlate of my actions, it has no bearing on their causal necessitation or genuine (narrow scope) modal status.

 

K(P) is absolutely crucial to the whole thread because it's what we are talking about. The thread is about the implications of God having eternal and infallible knowledge of something determinate, something that we're calling 'P'. If hypothetically, God has perfect foreknowledge of P, then our problem is determining whether any possibility remains that some not-P, 'Q' perhaps, might happen instead.

That's why K(P) needs to be included among the premises.

This might be where our biggest difference lies.

If we adopt K(P) among our premises, then we can't just switch it for K(Q) without redefining the original problem. And that in turn creates problems for the idea that Q always remained possible and had Q in fact been chosen, then God would have always had foreknowledge of Q instead of P.

You may think that it's "trivial and inconsequential", but I don't.

The whole problem here is determining the implications God's infallible foreknowldge of P. Setting opponents the task of demonstrating that P must be true in all possible worlds regardless of what God supposedly knows is too strong a demand and kind of irrelevant in my opinion.

Well, the word 'knowledge' is often defined as something like 'justified true belief'. Leaving aside the problem of how God justifies his omniscient knowledge, let's agree that the existence of knowledge necessarily implies the truth of what's known.

In that case, the phrase "fallible foreknowledge" is kind of ambiguous. It seems to suggest that what God really had was a belief which might or might not have really been knowledge, depending on whether the thing believed was true or not. If somebody discovers that they were mistaken about something, then they've discovered that they didn't really know it after all.

Of course, none of us can remember something until it's already happened. Memory is only after the fact. The possibility of knowing the past does suggest that the past might be immutible though. Even if freedom exists in the present, once the present recedes into the past, events may very well be fixed and we lose any ability to alter them. That's certainly consistent with human experience. But that's another difficult philosophical issue that's probably best left for another thread.

If God knew beforehand that you are going to do P, then how can you do anything else without contradicting what God knew?

It's the same with the past. If I know that P occurred in the past, then nothing else could have possibly happened instead without contradicting my (now supposed) "knowledge" claim about the past. That's not a big problem for beings like me, since we make mistakes all the time. Not everything that we imagine as being 'knowledge' really is. But supposedly God never makes mistakes.

 

TIME-OUT:


You do realize that the God most of you are talking about here is the "God of philosophy", not of actual religion, and that therefore any statements made (supposedly) about God cannot possibly be authoritative?

In order for one's concept of God truly be about God, it has to come from God, or it's not about God, or remains mere theory.

"God" happens to be a concept or entity which, by its very nature, cannot be sufficiently explained with mere philosophical reasoning (given that we also posit that God is the one who grants people the ability to reason philosophically).

Trying to reason about God on one's own, without reference to a theistic tradition that one apriori accepts as authoritative, only leads to the old problem that Protestants face: their claims about God have no authority, they are merely claims of people.

 

I can't speak for others but I think I clarified my own goals in post #191. What I am mostly concerned with is to examine what implications hypothetical attributes anyone wishes to attribute to God may have for the possibility of the various brands of free-will we may think we possess. I take no stance, myself, regarding the putative truth of such hypothetical claims, neither could I possibly take any as an atheist.