I’ve seen the argument that based on the Kolmogorov complexity notion of the razor, the multi-worlds interpretation ends up being the simplest interpretation that fits the evidence, and therefore the one to keep. It’s fully deterministic, apparently without needing the extra axiom of randomness.
Does an electron have ‘free will’? No – it just follows a set of rules that physicists have discovered. What would it even mean for it to have ‘free will’? As far as I can imagine, it would mean that the electron’s behavior did not follow any rules – no one could predict anything about the electron’s behavior if it had ‘free will’. Now, it is true that an electron has ‘free will’ in an absolutely trivial and un-interesting sense – the collapse of its wave function involves a degree of randomness. But I think everyone can agree that purely random behavior doesn’t constitute ‘free will’ in any traditionally agreed-upon sense, and that if it does it is non a particularly interesting form of ‘free will.’ So we have established if an electron has ‘free will’ in a non-trivial sense, then its behavior must be fundamentally unpredictable, yet non-random. To me these seem to be logically incompatible requirements. So the idea of ‘free will’ seems like nonsense to me.
I don’t know that you’d have to drag Kolmogorov complexity into it (though you certainly could), but this is the traditional argument to single out the many-worlds interpretation, that it satisfies parsimony. This appears counter-intuitive to many people, since those ‘many worlds’ appear to be additional ‘entities’ in the Occam’s razor sense at first brush; but one must take care distinguishing between what could be called explanatory and phenomenal entities, of which the razor only limits the former. Thus, many-worlds, containing as an explanatory entity only the wave function, should from this point of view be preferred to Copenhagen, which also needs to posit some reduction mechanism. Since in a sense every possibility occurs in many-worlds, there is no need for selection, and hence, the theory is fully deterministic (actually, quantum mechanics itself, governed by unitary evolution of the wave function, is deterministic; the many worlds interpretation just takes that seriously, while Copenhagen and related approaches posit a non-unitary mechanism to account for experimental results while keeping only one ‘real’ world).
On another note, a more interesting idea of freedom that does not obviously fall into the dichotomy of ‘either determinism or randomness’ is a formulation based on the concept of supervenience of the future on the past. Supervenience of one set upon another (say, set B supervenes on set A) basically means that the properties of the elements of set A determine the properties of the elements of set B. An example is pixels and pictures: The collection of pixels absolutely fixed what picture one sees; it could not be the case that the same collection of pixels gives rise to a different picture. However, an indistinguishable (or nearly so) picture could be generated by a different set of pixels. The picture supervenes on the pixels.
If there is no freedom, then the future supervenes on the past. Consequently, to allow freedom, the future must not supervene on the past; i.e. the same past could give rise to a different future. This does not necessarily introduce randomness: there could, in principle, be elements separate from what we would consider ‘the past’ that determine ‘the future’. Whether or not one could give a meaningful definition of ‘free will’ in this case, though, I’m not sure.
Another interesting version of free will has been hinted at by Voyager: for any sufficiently complex system, it is in general impossible to predict its evolution by any means other than observing the system’s evolution, or its equivalent. So, the only way to find out how one chooses in a given situation is to make the choice. This still isn’t the conventional free will: You couldn’t have done otherwise in the same situation; in fact, that concept is meaningless. But it allows for some measure of novelty: Sure, that you had your eggs poached rather than fried on Monday morning was, in a manner of speaking, fixed from the Big Bang; but the only way to determine which it would be was to actually have the universe evolve to that point.
You still haven’t explained the Free Will bit here.
Is it the fact that you can choose the $1 ?
But that would just mean ‘Having a Choice’’.
What choice you actually make is still predetirmined by your past experiences/programming.
Past experiences might lead someone to suspect that it is some kind of test. That by choosing the $1, he would pass a test others would fail.
Past experiences with you, might lead someone to punch you in the face instead of shaking your hand.
Free will is not 'making choices." Again, everyone agrees that choices are made and there’s nothing to debate or even think about by that definition.
Free will is the ability to make a choice. It’s the choice-making itself. If I sneak up on you, and whack you over the head with a baseball bat, and stuff you into the trunk of my car while you are unconscious, and drive you to Vegas… well, you had no free will in this scenario since it involved no choice on your part.
If I ask you to kindly step into the trunk of my car, and you think about it, and respond with a “fuck no!” then that was your free will in action. Just because everyone could predict your response in advance doesn’t mean a choice wasn’t made. I think this is an important distinction.
Is my view of free will and choice incorrect? I think the problem is that most people feel that because your choice is predictable it means it’s not a choice at all – and this is where I believe these questions of free will arise. A choice doesn’t have to be unpredictable, however. A choice is the process of your mental machinery making it’s internal calculations and spitting out an output. This can occur whether or not it’s predictable.
Does a computer or robot have free will then?
No, that’s a person in action. Of course a choice was made. The point of the debate is whether or not that choice is a free one.
No one would argue that a choice wasn’t made.
Yes. See the Wikipedia article I linked to.
The question of whether or not free will exists is not determined by predictability. We know choices are made by mental machinery. The question is, what’s free about it?
Choice is the process of taking inputs, processing them with your mental machinery, and then producing an output. The key part is the internal processing. Does that happen or not? If you wish to extend the definition of “mental machinery” to what is contained in robots and computers, then yes it has “free will”.
A computer that scans bar codes in a store, and decides upon a cost, is making a choice. If the clerk ignore the computer, and manually punches in a cost, then the computer did not make a choice.
Whether you wish to arbitrarily restrict “free will” to humans is… well… arbitrary. It’s dependant on your definition of “mental machinery”.
Choices are free if they occur without external manipulation, eg mind-control.
I did, and it tells me that I’m correct. “Free will is the putative ability of agents to make choices free from certain kinds of constraints”. No mind control -> free will.
Was the choice made internally in the mental machinery without intervention, or not? If it was made without intervention, it is free. If it was compelled, through drugs, or x-ray mind control, then it wasn’t free.
Der Trihs didn’t ask you if computers or robots make choices; he asked you if they have free will. If you’re going to stick with defining free will as the ability to make choices, the debate will not be a coherent one since Der Trihs isn’t defining it that way and neither is anyone else (except maybe for typoink).
You’re defining “certain kinds of constraints” the way that suits you. Read further. No one, not here or in that article, is defining free will as being free from “mind control”.
But the choice is precisely determined through the way the internal machinery is wired up. The register’s computer’s output is exactly determined by its input: upon being given one barcode, it will display one price, and always the same price; its choice is constrained to one option. It could not display any other price under identical circumstances; hence, the circumstances uniquely determine its choice, and thus, it is not free.
If you deny free will, then you must either (a) blame people for things over which they have no control or (b) refuse to cast blame under any circumstances. This strikes me as decidedly problematic.
or past experiences, or instinct, or hunger
argumentum ad consequentiam
If you’re saying that the unpleasant consequences do not make something untrue, then you are correct. That is not what I was claiming, though. Rather, I was pointing out that the problem of accountability does NOT vanish if we deny the existence of free will. Quite the contrary; it is exacerbated.
Yeah, sorry, I realized that after I posted it, but I think it applies generally to a few places in the discussion, so I left it.
The idea of determinism doesn’t require that everything was set in stone since the Big Bang. Quantum uncertainty keeps that from happening. But determinism means that what your brain outputs is a result of the internal state of your brain at that time, which is a result of all the things that have happened to it before, some of which would be unpredictable from too far in the past because of quantum uncertainty.
Determinism does not mean we’re trains on a set of rails that we can’t break free from. It means we can make decisions within the confines of circumstance and mind-states with predictable effects. With determinism, I can predict the effects of a murder, and as such I would never do such a thing. With free will, cause and effect flies out the window, and consequences become random and arbitrary.
People tend to think free will is the ability to decide something, and by that definition it would exist, but that’s not what the concept really is, strictly speaking.