Indeed, I explained that the tensed language was unnecessary for my argument by providing an alternative phrasing that avoided it, whilst retaining what I would consider the argument’s actual substance, in the same message in which I made that claim.
And, indeed, I do, or have for most of this discussion, held this view of infallible knowing that you find strange, expressed in my Definition 1[1]. In the message where I first introduced it, I also introduced another, discussed it, and explained why I found it less compelling. In my next message after that I treated three more alternative definitions.
I introduced a modified version of Def. 2 in Definition 6[2], and conceded that it may be an agreeable alternative. My argument as it stood so far would not formally hold under it. If a variant of my argument would, I do not as of yet know, for I have not yet spent the time analyzing its full implications regarding the topic at hand, and nobody else presented an analysis of it here either.
Good. Because i was actually triggered by @Faizal_Ali 's fixation on the phrase “no alternative(s)”.
His wording seemed designed to confuse: equivacating between scenarios which present with no alternatives and those that DO have alternatives, but [of course!] the choosing agent only chooses one.
Of course. Apologies if you misunderstood me as saying any such thing. But your language in turn just gets in the way. Let’s try this another way.
Forget free will. Let’s talk about quantum indeterminacy. Suppose I have an atom of an isotope with a short half-life and I want to know when it will decay. Obviously I can’t predict that moment, since the decay occurs at random; present conditions, no matter how perfectly known, will tell me nothing. No problem: I set up my detector to record it when it happens, jump into my time machine, go several half-lives into the future, and check the recorder. Then back to the present, with the knowledge that decay will happen in 3 days, 5 hours, 47 minutes, and 6 seconds. Which of course happens, right on schedule. Have I just invalidated quantum randomness?
Similarly, prediction of the future might kill free will, because it depends on the past causing the future, but seeing the future wouldn’t, because in this case the future in some sense causes the past. Do you see the difference? God, presumably, sees all time, but this affects neither the randomness of nuclear decay nor the free will of humans.
“Could God have created a universe without these?” – “No” would not contradict omnipotence if such a universe were logically impossible (as I believe some have argued). It has been argued that even an omnipotent being cannot create a logical impossibility.
“Then why doesn’t he?” – “Free-will” raises the question of whether heaven has free-will without evil? And if it is possible for heaven, then why not elsewhere (or alternatively, why not just created everybody in heaven)?
The question is imprecise. What does “God’s knowledge about this discussion” mean, specifically? Does God know “[such-and-such] discussion will happens at [such-and-such] time”? In that case, if it is possible for the discussion to not happen, then it is possible that the statement is false. There exists therefore a possible world, wherein the content of God’s knowledge does not obtain.
If we go with the JTB definition of knowledge, then there exist a possible world where God does not have this item of knowledge. It is not necessarily the case that God knows, but it is the case that He does.
We may out of caution choose to call it contingent knowledge. If it is not fair to say that it is fallible in such circumstance, then when would it be fair to call it so?
Ah, but it is a problem. By the way, this is a great example, and thank you very much for the opportunity to consider it.
See, the mistake here is to think that present conditions are not changed when the experimenter returns back from the future. But they are. Information is physical. Even setting aside that having information about the moment the decay occurs is a different state from not having it, obtaining information – transfering it from one system to another – takes time.
So if the experimenter vanishes into the future and returns back to the moment she departed from, with knowledge of the time at which decay occured, she will have spent some time in the future to obtain this knowledge. During this time period, the wave function describing her body, her time machine, her lab book, etc. will have evolved in accord with Schrödinger’s equation.
When the experimenter returns then, be it to the same time she departed, or any other prior to the decay, in so doing she discontinuously alters the universe’s wave function. For all intents and purposes, she renders the measurement “already performed” from that point onwards, even if the event being measured is yet to transpire.
Let’s grant that this time jumping trick yields perfectly reliable knowledge of an otherwise unpredictable quantum event. The statement that no such prediction is possible without/before taking the measurement still holds. The time traveling experimenter can only render the “prediction” after returning from a trip to the future wherein she took the measurement.
What this violates, to put it into words you chose earlier, is some sort of integrity of time and causation. Information in this hypothetical can jump to and from any point in time.
However, we must be cautious about drawing conclusions about knowledge in general from this particular hypothetical example of physically acquired[1] knowledge. With that in mind, let’s consider what you go on to say:
Indeed, this “some sense” is in great need of specification here. Seeing the future should mean having some kind of access to it, but it is not at all clear that conventional cause-and-effect are what facilitate such access, and I for one would hesitate to even call it causation, without explicitly assuming or proving that there is an actual parallel between what is usually meant by causation[2] and what ever the exact nature of this future-seeing is.
In the case of the time travelling nuclear physicist one could argue that the note she took in her lab book is caused (through some chain of causally linked events) by the detector having recorded the decay time stamp, and that what ever transpires because of that knowledge in the past, after the time of her return, is likewise retro-caused by an event in their timeline’s future, because there is a forward-through-time path from said future, through the jump, and into the past.
In the case of a god who sees all time, what exactly it means to say that its knowledge of an event at some time is “caused” by that event occurring at the time in question is nowhere near as clear.
Regardless, if that god has knowledge of the occurrence of all events at all times (including all time travel jumps, for that matter), then it has knowledge of the one and only way the universe can go. For it cannot have knowledge that the proposition “event E occurs at time t” is true, if that proposition is in fact false. One could call it a block universe, therefore, or a universe with superdeterminism, or what ever other expression we may find for what is at root just fatalism. Quantum indeterminacy and libertarian free will could seem perfectly realistic from a perspective inside that universe, but neither would be actually there.
We did not actually establish the physical plausibility of time travel here, so whether this hypothetical is an example of anything is an open question. ↩︎
As if that was not enough of a can of worms in its own right. ↩︎
Not sure what “facilitate” is supposed to be. In the scenario, my experience in the future causes me, in the present, to know what will happen.
I’d say it has knowledge of the one and only way the universe does go. That doesn’t mean it couldn’t have gone another way. It merely means that free will is embedded in time. Of all the alternatives, only one actually happens. It’s the same with nuclear decay. An atom could decay at any time, but it decays at one particular time. That time is not determined.
No problem. It was a thought experiment, and in fact an analogy to God’s omniscience. TIme travel is the least of the impossibilities under discussion when we’re talking about God and about libertarian free will, neither of which I think exists.
But is the time travel example an analogy to God’s omniscience, though? Does God need to be caused to know what events will happen the same way the time travelling nuclear physicist is, or is it conceivable that God “just knows” without being caused to know? This is what I meant, when ever I questioned whether anything like cause and effect are what is required for a being to know the future.
All analogies fail at some point. The question is whether this one fails in a way that matters to the matter at issue. I don’t think it does. I think it shows that knowledge of the future prevents neither quantum randomness nor free will, as long as the knowledge doesn’t arise from the ability to predict a future state based on the current state.
Well, you do formulate the future-knowing in a way that makes it caused by the future events in question. And assuming there is a way for that to happen, I agree that such causally produced future-knowing might not require a fatalistic universe. In your example and in the argument hence built, the future is known as a consequence of the future being what it is. So the future gets to freely be as it likes, affected by the whims of free agents or the proverbial quantum dice. Once a path is assumed, it then gets to “mold” any knowledge time travellers or gods have of the events in question, even if the knowledge can then be transported back into the respective events’ past.
I believe this nature of future-knowledge acquisition is rather crucial for your conclusion. If the future the universe assumes does not get to produce god’s knowledge of it, and the knowledge is rather fixed and cannot be retroactively re-written by what actually transpires, then how would the universe retain multiple possible futures?
I am willing to accept that you have found a way to construe the original scenario so that it doesnt produce the original conclusion.
But the devil is in the details that you have described in a plausible way. But how do you prove the 2 metaphysical scenarios are equivalent?
Perhaps you have heard the assertion that a non-physical demon cant harm anyone because it is non-physical. But making such an assertion is hardly proving it.
The original description is intentionally worded similarly to this:
"If a deity, or device, can reliably predict, for a being, any future choice from among 2 or more alternatives, then we can conclude that this being does not have freewill. Regardless of the presence (or lack) of known or suspected causal chains leading up to the choice, IF freewill existed there would be a signifcant possibility that the prediction would fail.
The devil, as you so correctly point out, is in the details. This argument requires more assumptions than it makes. A case could be made, for example, that a free agent would make, in one sense or another, “rational” choices, when given the opportunity. Knowing their personality, past history, personal preferences, and current conditions of their mood, it is entirely conceivable that one could “reliably predict” choices without said choices being either causally determined or pre-destined. The possibility that an agent would choose poop-flavoured ice cream over vanilla for dessert, for example, when they have no history of such peculiar dietary preferences, can arguably only be “significant” if the agent is in that moment not equipped with sufficient information, or healthy emotional and intellectual faculties, that their choice might be fairly described as “free”.
The argument, as it stands, is crude. Far too much wiggle-room is left available by wording like “reliably predict”, “significant possibility”, “leading up to”, etc… All of this needs to be made precise, which several days’ worth of recent discussion has been all about.
The argument cannot be recovered nor made more compelling by reverting back to a more clumsy phrasing. If anything, this would render it even more of a non-sequitur than its critics found the more precisely articulated variants.
Sure. I think my objection scales well, though. Given enough information, arbitrarily reliable predictions of all choices should be just as conceivable, and without any more robust dismissals of free will for it.
The way you layed it out, it is not “a diagnostic tool” either. Because you made no restrictions as to how these “predictions” are to be rendered, the rate at which they succeed or fail cannot indicate anything about the presence or absence of free will.
You agreed with this, too. I do not understand what went awry between that message and the one I’m responding to now…
There is no “retroactivelly re-written” involved. Many things could happen, but only one does happen, and that’s the one that makes the future. Free will is in the “could” (if it’s anywhere), not in the “does”. God sees the universe that gets actualized. He probably sees the ones that don’t too, but he can tell the difference. He sees all time. To say otherwise is to embed God in the present, experiencing time in the same way we do. Anyway, there is no interference with free will.
Which two?
If that was intentional, it’s unfortunate. I see the biggest problem with the word “predict”. Only a being trapped within the present can predict. The distinction is between predicting, on the basis of the current state of the universe, and seeing the future. Seeing how things turn out by not being stuck in the present causes no problems for free will.
That’s quite a different question, and I would have quite a different answer.
True. Nobody ever said it did. This is irrelevant to the analogy, though.