Not really. If we are rewinding time that means that the event has already happened. Just because we cannot change the past, does not mean we do not have free will. If you rewound time, the same thing would happen because it has already happened and the grandfather parodox says we cannot change it.
I mean rewinding time in a "re-enactment" kind of way. Imagine that you save the position and momentum of every particle in the universe, and "rewinding" is just the act of erasing the universe and replacing it with the one we saved. It's like a backup/restore. If the universe is not deterministic, then we expect that it will resume functioning slightly differently.
Some of you are talking about the universe like scientists understand it. We do not yet understand the universe we live in. There are theories, but theories are not fact. Ever. There was a theory the world was flat and that was considered by alot of people to be fact. But it was a theory because there was no proof. Theories on the universe are a dime a dozen, but most likely they will be proved false. So don't instantly believe a theory just because it makes sense.
So theories cannot correspond to reality? That is an odd claim to make. One would imagine that they could, if only by chance.
If everything is predetermined, there is only ever one thing that can happen in a given situation, only one future for every present. There is no choice.
Ehhh... it's not clear. The problem is that when we say "X has a choice", X might not at all be the
kind of thing we think it is. That is, if we were to take a snapshot of the universe at a given moment and define some aggregate of molecules as "you", I believe we would be making a categorical mistake.
Consider for a second that "X" is a
set of things. For instance, the set of all actual and conceivable dogs. We can define "X can do Y" as meaning the following: "
There exists an element x of X such that
if x obtains
then x does Y". For instance, "dogs can bark" means that we can imagine a dog that would bark if it existed. Now let's define "X does Y" as meaning "
There exists an element x of X such that x obtains
and x does Y". Thus, "dogs do bark" if there exists a dog that barks. Now, we can define "X can choose Y" as "X can do Y
and X can do ~Y" ("~Y" meaning "not Y"), and "X chooses Y" as "X can choose Y and X does Y".
Now, what if "you" were a set, very much like the set of all dogs? I mean, think about it - most people would consider that they are the same person as they were one second ago. Yet, their brain changed slightly during that time. Furthermore, if my brain had one more or one less neuron, I would say that it would still be my brain. I would not consider that this would make me a different person. Therefore, it makes sense for me to think that "I" am the set of all Brains that are sufficiently similar to the Brain that currently obtains.
You can see where this is going: when I say "cantab can choose to reply to my post", that means "we can conceive of some cantab which, if he existed, would reply to my post, and we can conceive of some other cantab which, if he existed, would not reply to my post". Of course, what happens next is determined by which cantab actually obtains, but the point is that choice is actually integrating over all
conceivable cantabs. By that definition, you are lacking a choice if and only if no conceivable agent performing that choice is one you can identify with. None of the argument remotely relies on determinism or the lack thereof, yet I believe that this is eerily close to how we understand free will.
The concept can be generalized to programs and machines relatively easily. If a machine can conceive of other machines which it can very strongly identify with, to the point that it conflates them with itself, then in any situation it would conceive of acting in many different ways and thus it would feel as if it had free will. It might even see itself as being all these machines at once (I think this is exactly how most humans unconsciously see themselves), like a sort of (illusory) quantum superposition of machines. In that case, it might very well perceive its behavior as nondeterministic.
Indeed, I might even define 'apparent free will' as the ability to take different actions when placed more than once in an identical environment.
The problem with this is that in so far that you don't control how many times you are placed in an identical environment, a devious djinn could simply make you repeat your choices until you choose what
he wants you to do. That is, anyone who has free will by that definition can be manipulated to "choose" to do exactly what a sufficiently powerful puppeteer wants. In fact, the definition of free will you state pretty much entails a loss of uniqueness - since nothing about you constrains the choices you can make, there's no way to tell two people apart by their behavior. Everyone is reduced to a random agent. If this experiment was performed, I don't think it would appear like free will.
I know this strays from the actual question a bit, but ... it matters not whether we have free will or not. If we assume that we do not, then why should we punish crime? The criminals didn't have a choice if free will in nonexistant. As such, we must assume that free will exists, at least to some degree.
I don't know about you, but if somebody goes on a murdering rampage, free will is the least of my concerns. I want that fucker
out so that I can enjoy life. Without free will, punishment is still a good regulation system for the machines we are. To put it in another way, machines with a justice system will tend to prevail over unruly ones, and free will is a red herring.
