The Strong Free Will Theorem (2009) [pdf] • Hacker News Board

Re: The Strong Free Will Theorem (2009) [pdf]

#81

Post by tasty_freeze » Wed, Nov 07, 2018, 9:12 PM UTC

michael2l wrote: post #79

Quite apart from the math and science of deterministic physics and probabilistic quantum physics, there's a broader evolutionary question. Why would we evolve this hallucination that we have free will, if in fact we fundamentally don't? It seems like a lot of effort and metabolic energy on a moment-by-moment basis to maintain the illusion. I guess you could make an argument we're in some sort of simulation the maker…

Why do you think maintaining the sense of free will is evolutionarily costly? If anything, to me, it seems like a heuristic shortcut that is imperfect but is useful most of the time, and saves resources overall.

As a close analogy, we've all said something like, "My car wants to pull to the right." Nobody claims the car has free will, but "wanting to pull to the right" is a useful way to both communicate and to think about it. What is is in the car's makeup that causes it to want to pull to the right?

There are plenty other examples where people have false beliefs yet still cling to them. If you ask drivers if they are better or worse than average, about 10% will say worse. My point is, having false beliefs doesn't always exert a strong evolutionary pressure.

Re: The Strong Free Will Theorem (2009) [pdf]

#82

Post by whack » Wed, Nov 07, 2018, 9:12 PM UTC

simonh wrote: post #7

I really think this, along with much philosophical musing about free will, is entirely missing the point. Pre-determination is essential to free will. You literally cannot have free will without it. Random choice, or even non-predetermined correlated outcomes of entangled objects don't lead to free will, they lead to uncertain decisions and those are not the same thing. If my decisions are not a product of my prior s…

> If my decisions are not a product of my prior state, then they are not my decisions. The definition of 'me' is my prior state.

There are two problems with this definition of free will.

First, it implies that your present self (and future selves) are all entirely subservient to your prior state, and surrounding environment. Ie, the present you has no way of overcoming your past self or your environment. This means that you're essentially sitting in a roller-coaster ride, with no levers to pull or opportunity for change. Sure, your past self made the decision to get on the ride. But your present and future self are all helpless from that point on.

Second, did even your past have any free will? The chain of causation extends all the way back. Your past self at time T-X is itself entirely subservient to your past self at time T-X-1. Extend this all the way back, and you realize that you're subservient to your "self" at the moment of conception. At no point did your consciousness have any ability to change the course of the life that was already plotted for it. If I predict at the moment of your birth, every single thing that will happen in your life and every decision you will make, can you really claim to possess free will?

Hence why people try to cling on to randomness and uncertainty. "If the future can be perfectly predicted, people definitely do not have free will => ergo, if the future cannot be perfectly predicted, then maybe we have free will?" I agree with you though that this is shoddy reasoning. Just because there exists randomness at the quantum level, just means that we're passengers in a driverless car that's rolling dice. Randomness that we have no control over, does not grant us free will in any way.

Re: The Strong Free Will Theorem (2009) [pdf]

#83

Post by cryptonector » Wed, Nov 07, 2018, 9:17 PM UTC

simonh wrote: post #7

I really think this, along with much philosophical musing about free will, is entirely missing the point. Pre-determination is essential to free will. You literally cannot have free will without it. Random choice, or even non-predetermined correlated outcomes of entangled objects don't lead to free will, they lead to uncertain decisions and those are not the same thing. If my decisions are not a product of my prior s…

I think there's no such thing as free will in any case. At least for the religions notion of free will floating around in my mind:

  if (!is_fully_deterministic(universe) &&
      have(universe, RANDOM_CHOICE))
      assert(!have(universe, FREE_WILL));
  if (is_fully_deterministic(universe))
      assert(!have(universe, FREE_WILL));

The only remaining way to have that sort of free will is to have non-random laws that somehow are not fully deterministic. I'm not sure what that would mean. Non-random -> deterministic, no?

However, even in a fully-deterministic universe, or a mostly-deterministic universe, we will appear to have free will in so far as we cannot read minds and we cannot predict next states for the universe on any scale, with any accuracy, and in any amount of time that would be practical for determining e.g. whether someone is at fault for their actions. And it's pretty clear that this will always be so.

The question of whether we have free will, then, is sort of moot. From a social perspective we have little to choice but to act as if we do have free will.

Re: The Strong Free Will Theorem (2009) [pdf]

#84

Post by tylerjwilk00 » Wed, Nov 07, 2018, 10:02 PM UTC

adrianN wrote: post #70

tylerjwilk00 wrote: post #56

Earlier quoted context omitted.

It's not random though - it is probabilistic. And even if it was an internal 'RNG' it is is still _their's_ not someone else's.

If probabilistic outcomes are the same as having a free will, then elementary particles also have free will. That doesn't fit the intuitive fuzzy definition of free will.

That is exactly the point!

From the paper:

Some readers may object to our use of the term “free will” to describe the indeterminism of particle responses. Our provocative ascription of free will to elementary particles is deliberate, since our theorem asserts that if experimenters have a certain freedom, then particles have exactly the same kind of freedom. Indeed, it is natural to suppose that this latter freedom is the ultimate explanation of our own.

Re: The Strong Free Will Theorem (2009) [pdf]

#85

Post by mreome » Wed, Nov 07, 2018, 10:02 PM UTC

whack wrote: post #82

simonh wrote: post #7

I really think this, along with much philosophical musing about free will, is entirely missing the point. Pre-determination is essential to free will. You literally cannot have free will without it. Random choice, or even non-predetermined correlated outcomes of entangled objects don't lead to free will, they lead to uncertain decisions and those are not the same thing. If my decisions are not a product of my prior s…

> If my decisions are not a product of my prior state, then they are not my decisions. The definition of 'me' is my prior state. There are two problems with this definition of free will. First, it implies that your present self (and future selves) are all entirely subservient to your prior state, and surrounding environment. Ie, the present you has no way of overcoming your past self or your environment. This means t…

I think simonh was not arguing against the chain of causation, but rather agreeing with it, and arguing against "If the future can be perfectly predicted => people definitely do not have free will" and in favor of "If the future can not be perfectly predicted => people definitely do not have free will"

If an entity's decision process is is not not dependent and repeatable based entirely on it's internal state and perceptions of it's environment, then it must be dependent on some external random element, and that element not being part of that entity, means it's choice is not entirely of it's own free will.

This argument rejects duality of course, as one's "will" itself could be the external element.

Re: The Strong Free Will Theorem (2009) [pdf]

#86

Post by prmph » Wed, Nov 07, 2018, 10:10 PM UTC

whack wrote: post #82

simonh wrote: post #7

I really think this, along with much philosophical musing about free will, is entirely missing the point. Pre-determination is essential to free will. You literally cannot have free will without it. Random choice, or even non-predetermined correlated outcomes of entangled objects don't lead to free will, they lead to uncertain decisions and those are not the same thing. If my decisions are not a product of my prior s…

> If my decisions are not a product of my prior state, then they are not my decisions. The definition of 'me' is my prior state. There are two problems with this definition of free will. First, it implies that your present self (and future selves) are all entirely subservient to your prior state, and surrounding environment. Ie, the present you has no way of overcoming your past self or your environment. This means t…

What if current behavior is not completely determined by previous state, and yet is also not random?

Re: The Strong Free Will Theorem (2009) [pdf]

#87

Post by starbeast » Wed, Nov 07, 2018, 10:12 PM UTC

mreome wrote: post #85

whack wrote: post #82

Earlier quoted context omitted.

> If my decisions are not a product of my prior state, then they are not my decisions. The definition of 'me' is my prior state. There are two problems with this definition of free will. First, it implies that your present self (and future selves) are all entirely subservient to your prior state, and surrounding environment. Ie, the present you has no way of overcoming your past self or your environment. This means t…

I think simonh was not arguing against the chain of causation, but rather agreeing with it, and arguing against "If the future can be perfectly predicted => people definitely do not have free will" and in favor of "If the future can not be perfectly predicted => people definitely do not have free will" If an entity's decision process is is not not dependent and repeatable based entirely on it's internal state and per…

Gödel had an interesting argument that I encountered via Rudy Rucker's blog;

“There is no contradiction between free will and knowing in advance precisely what one will do. If one knows oneself completely then this is the situation. One does not deliberately do the opposite of what one wants.”

http://www.rudyrucker.com/blog/2012/08/01/memories-of-kurt-g...

Re: The Strong Free Will Theorem (2009) [pdf]

#88

Post by tylerjwilk00 » Wed, Nov 07, 2018, 10:36 PM UTC

esotericn wrote: post #58

tylerjwilk00 wrote: post #56

Earlier quoted context omitted.

It's not random though - it is probabilistic. And even if it was an internal 'RNG' it is is still _their's_ not someone else's.

If you take an action based on determining the probabilities of an event, you would take that action every single time because the state would be identical every time. You look at the roulette board. You bet on red. We rewind time. You still bet on red. The state is identical. Did _you_ come up with that decision? Of course you did, who else could have? My claim is not that there's no "you". It's that "free will" is…

Let's excuse that we have to invoke time travel, and that once the choice has been made it is now part of the past of our collapsed reality...

The point you seem to be making:

A) If I make the same choice everytime there is no free will.

B) If I make different choices everytime there is no free will? Or perhaps even there is no "me"?

Free will could entail the will to choose the same thing every time AND the will to choose something different.

If we can't accept that then perhaps an argument could be made that there is no such thing as "you".

Re: The Strong Free Will Theorem (2009) [pdf]

#89

Post by whack » Wed, Nov 07, 2018, 10:44 PM UTC

mreome wrote: post #85

whack wrote: post #82

Earlier quoted context omitted.

> If my decisions are not a product of my prior state, then they are not my decisions. The definition of 'me' is my prior state. There are two problems with this definition of free will. First, it implies that your present self (and future selves) are all entirely subservient to your prior state, and surrounding environment. Ie, the present you has no way of overcoming your past self or your environment. This means t…

I think simonh was not arguing against the chain of causation, but rather agreeing with it, and arguing against "If the future can be perfectly predicted => people definitely do not have free will" and in favor of "If the future can not be perfectly predicted => people definitely do not have free will" If an entity's decision process is is not not dependent and repeatable based entirely on it's internal state and per…

> I think simonh was ... arguing against "If the future can be perfectly predicted => people definitely do not have free will"

Just so I'm understanding you correctly: Are you claiming that even if the future can be perfectly predicted, people can still have free will?

At the moment of your conception, if I were to predict with 100% accuracy and confidence, every decision you will ever make in your life, would you still claim to possess free will?

Re: The Strong Free Will Theorem (2009) [pdf]

#90

Post by narag » Wed, Nov 07, 2018, 11:10 PM UTC

whack wrote: post #82

simonh wrote: post #7

I really think this, along with much philosophical musing about free will, is entirely missing the point. Pre-determination is essential to free will. You literally cannot have free will without it. Random choice, or even non-predetermined correlated outcomes of entangled objects don't lead to free will, they lead to uncertain decisions and those are not the same thing. If my decisions are not a product of my prior s…

> If my decisions are not a product of my prior state, then they are not my decisions. The definition of 'me' is my prior state. There are two problems with this definition of free will. First, it implies that your present self (and future selves) are all entirely subservient to your prior state, and surrounding environment. Ie, the present you has no way of overcoming your past self or your environment. This means t…

Ie, the present you has no way of overcoming your past self or your environment.

You are answering to someone that clearly expressed that you are your past self. Why would you want getting rid of yourself?

This means that you're essentially sitting in a roller-coaster ride, with no levers to pull or opportunity for change.

Not at all. Human being can learn, practice and make all kind of efforts to change ourselves. Sports, arts, science are all fields where humans show how to improve.

You are getting to the no-levers problem by drawing a line between "you" and "what you do", as if you have nothing to do with what you do. Why?