Does free will actually exist? Opinions please.

[Dr Roberts]

What do you guys think?

[Stark]

Depends what you mean. No, because the universe is deterministic. Yes, because you get a 'choice', even if it's not really a choice.

I mean, the only thing someone else can make you do is die... but not really. You see?

[madd0ct0r]

I think it exists, even if it dosen't, I have no choice but to decide it does

and it means I don't have to worry about determinism.

[Sinewmire]

Yes, I think free will is real. The universe does not care if I drink milk or water before bed.

[Surlethe]

Hey motherfuckers, stop spamming up this thread with stupid bullshit.

Here's my perspective. Our bodies are essentially chemical machines; in principle, everything we observe about ourselves reduces to some aggregate of many chemical reactions taking place. This precludes "free will" in the sense that all that occurs is physically necessary. However, this reductionism is computationally intractable. It is practically impossible to meaningfully predict human behavior from first principles in all but a very limited number of cases. So treating humans as individual agents with free will (and at the same time admitting that this agency is emergent from deterministic phenomena) works as a reasonable approximation for day-to-day behavior.

(Note that this is all a rough empirical, not moral, argument.)

[Lord of the Abyss]

First, define free will. I don't believe in the vision of "free will" as something that is neither deterministic nor random; that doesn't even make sense. That version of free will is an incoherent, meaningless idea. Something like the version that Surlethe is claiming above however does make sense.

[Stark]

Hey motherfuckers, stop spamming up this thread with stupid bullshit.

Here's my perspective. Our bodies are essentially chemical machines; in principle, everything we observe about ourselves reduces to some aggregate of many chemical reactions taking place. This precludes "free will" in the sense that all that occurs is physically necessary. However, this reductionism is computationally intractable. It is practically impossible to meaningfully predict human behavior from first principles in all but a very limited number of cases. So treating humans as individual agents with free will (and at the same time admitting that this agency is emergent from deterministic phenomena) works as a reasonable approximation for day-to-day behavior.

(Note that this is all a rough empirical, not moral, argument.)

Thanks for repeating the first response in this thread?

If the universe is determininsitc, it doesn't matter if you can work out what is determined or not. However, society pretty much requires the concept of responsibility (or attributed responsibility), so we need to act like people have choices, even if we know that fundamentally this isn't really true. And hey, most people are probably intellectually incapable of even internalising the concept, so you need practical concepts like 'choice' anyway.

[SCRawl]

Spammy posts split out.

[madd0ct0r]

Oh, I thought he was referring to me

to expand: I think it's an irrelevant question.

If I have free will, then i can never behave other then how I choose to (except when half asleep, or intoxicated or other circumstances that most people wouldn't class as free will)

If I don't have free will, then the same above applies.

It's not like it has been determined I'll drink this glass of water, so I will always drink it, even if i actually thought "Orange juice - I choose you!"

I think current research is indicating the brain 'chooses' to do things before the conscious mind rationalises them.

[Alyrium Denryle]

"Free Will" and "choice" are illusions created by the emergent phenomenon commonly known as consciousness. The mind operates on a completely deterministic, but very complex set of chemical reactions, however, physiological processes are better described by chaos theory than standard reductionist models, are highly variable on even small and very precise scales, and highly dependent on the state of prior conditions. Thus, while there is no ability to predict human behavior unless one is willing to accept large confidence intervals, the system itself is no less deterministic than weather patterns.

One can think of consciousness as being a sort of feedback loop in the system. An emergent phenomenon resulting from extremely high processing power in the higher cognitive functions, combined with an ability to extrapolate and react to potential futures. It is self-aware, and can create a model of future environmental states and thus feed back into and "train" what reactions it might take by affecting the system's current state (and thus the inputs into the deterministic model which will affect a future state). However, there is no choice in current decisions, only the chaotic but entirely determined brain.

[K. A. Pital]

It is practically impossible to meaningfully predict human behavior from first principles in all but a very limited number of cases

What if we deciphered the electrochemical reactions and their consequences down to the last bit and bolt for a certain human and placed him in an environment with entirely known properties where we know the position of every atom? Would we be able to predice that human's behaviour (and if the situation is not as ideal, at least predict it with a high probability?).

I think yes.

[Surlethe]

Stark: Um, no. "Yes because you get a 'choice', even if it's not really a choice. The only thing someone else can make you do is die ... but not really" doesn't make any sense. You didn't bother to try to explain yourself. My answer is superior, especially if you were trying to say the same thing.

Stas: I agree. Hence "practical."

[Alyrium Denryle]

It is practically impossible to meaningfully predict human behavior from first principles in all but a very limited number of cases

What if we deciphered the electrochemical reactions and their consequences down to the last bit and bolt for a certain human and placed him in an environment with entirely known properties where we know the position of every atom? Would we be able to predice that human's behaviour (and if the situation is not as ideal, at least predict it with a high probability?).

I think yes.

With confidence intervals. There is variability intrinsic in the system due to things like quantum interactions between electrons that gets amplified up through the entire system, and is present no matter how exacting our measurements. See Chaos theory.

[ray245]

It is practically impossible to meaningfully predict human behavior from first principles in all but a very limited number of cases

What if we deciphered the electrochemical reactions and their consequences down to the last bit and bolt for a certain human and placed him in an environment with entirely known properties where we know the position of every atom? Would we be able to predice that human's behaviour (and if the situation is not as ideal, at least predict it with a high probability?).

I think yes.

With confidence intervals. There is variability intrinsic in the system due to things like quantum interactions between electrons that gets amplified up through the entire system, and is present no matter how exacting our measurements. See Chaos theory.

Surely it is possibility that we can better understand things like quantum interactions in the future and develop a better ways to measure such things?

[K. A. Pital]

It is practically impossible to meaningfully predict human behavior from first principles in all but a very limited number of cases

What if we deciphered the electrochemical reactions and their consequences down to the last bit and bolt for a certain human and placed him in an environment with entirely known properties where we know the position of every atom? Would we be able to predice that human's behaviour (and if the situation is not as ideal, at least predict it with a high probability?).

I think yes.

With confidence intervals. There is variability intrinsic in the system due to things like quantum interactions between electrons that gets amplified up through the entire system, and is present no matter how exacting our measurements. See Chaos theory.

Is quantum chaos a principial obstacle?

[Alyrium Denryle]

Is quantum chaos a principial obstacle?

Well not quantum specifically, but just chaos. No matter what scale you look at, and no matter how exacting your measurements, the coefficient of variation is the same. Irrespective of what your actual measurement error rates are due to equipment sensitivity etc. Quantum variations are one component, other components would be things like how quickly a neurotransmitter is received on one receptor, because its ligand is not in the correct orientation on first contact. Variation like that makes reductionist prediction by taking a system apart and understanding the bits to predict the whole impossible. The best you can do is a confidence interval describing how likely each set of possible behaviors might be. Kind of like predicting the weather.

[K. A. Pital]

Oh, I see. But that's what I asked- I didn't mean complete determinism but rather determinism with a probability of outcome qualifier.

[Molyneux]

If there is no free will, then it doesn't matter whether I believe that there is free will or not - either way, my actions (or possibly, my volition) are pre-determined.
If there is free will, then the only accurate course of action is to take advantage of that free will, rather than surrendering to fatalism.

As far as I can see, regardless of whether or not we have free will, it pays to assume that we do even if evidence points otherwise.

[Kanastrous]

Perhaps a good question is not so much 'do we have free will, or not' as 'are we better served by operating on the assumption that we do, or not.' Like others in the thread, I think the answer to -that- question is 'yes.'

Reminds me a bit of the artificial-brain conversation from Restaurant at the End of the Universe:

"No one would even notice the difference."

"I would!"

"No, you wouldn't. You'd be programmed not to."

[Starglider]

Quantum mechanics introduces real determinism that is not reducible (as far as we know) even with perfectly knowledge of the initial state and an infinite amount of computing power. At least from a single-universe perspective; if the many-worlds interpretation is physically accurate, then the multiverse as a whole is (probably) deterministic. However this is not really relevant for questions of 'free will', which are generally arguments about the hopelessly fuzzy semantics of the term. No worthwhile constructive theory of mind includes such a thing.

[Alerik the Fortunate]

To echo others, free will as something neither determined nor random, that "just wills" is an incoherent concept. But consider, our selves are complex mechanisms for making decisions, made up of huge complexes of competing desires, unconscious biases and snap decision making shortcuts, and conscious, deliberate, rational thought. The amount that each contributes varies per person and over time and in different circumstances, but it makes each of us a machine that returns a decision when faced with particular environmental stimuli, which includes the aggregate effect of the past history of stimuli we have experienced. Our nature is what is responsible for that decision, even if we are not responsible for the origin of that nature itself. Perhaps, one gains enough self knowledge and moral theory to begin reconstructing oneself into someone that one would consciously choose to be, that makes decisions consistent with professed values by means of rational thought. One may take some responsibility for creating the self that follows, yet one was not responsible for the chain of events that brought one to that point. We aren't self-created, self-existent beings without causal antecedent. Yet, even if all our decisions are predetermined by the initial conditions of the universe, to quote Daniel Dennet, "the universe is subjectively open." Even a mechanistic universe is too vast and complex to be predicted by us, so we make decisions as though they were free, because it appears to us that we have choices since we cannot actually observe all the chemical pathways that predetermined our choice. Even if it is predetermined by the nature of the agent, that choice is made in the form of weighing alternatives, and we (rightly) attribute that choice to the agent as the best available model. I really recommend Dennet's Freedom Evolves.

[Stark]

Stark: Um, no. "Yes because you get a 'choice', even if it's not really a choice. The only thing someone else can make you do is die ... but not really" doesn't make any sense. You didn't bother to try to explain yourself. My answer is superior, especially if you were trying to say the same thing.

Stas: I agree. Hence "practical."

You think introducing the irrelevant distinction between 'can work out ahead of time' or 'can't work out ahead of time' is worthwhile? That's like saying a die isn't deterministic because nobody is currently modelling it's movement, or it was less deterministic before we had geometry.

Alarmingly I have to agree with Kanastrod, the only important this to even consider is what extent we judge people as responsible for their choices. After all, even if fairly floss free will exists (because of computation limits, lol) people never have the full range of options and there are always situational limitations. However, we choose to hold people responsible for their decisions anyway, and this is probably socially necessary as discussed.

[Sriad]

It is practically impossible to meaningfully predict human behavior from first principles in all but a very limited number of cases

What if we deciphered the electrochemical reactions and their consequences down to the last bit and bolt for a certain human and placed him in an environment with entirely known properties where we know the position of every atom? Would we be able to predice that human's behaviour (and if the situation is not as ideal, at least predict it with a high probability?).

I think yes.

Very probably.

"Experiments" (actually just standard treatment) done with people suffering from transient global amnesia show that the victim's behavior in a static environment (or with the same stimulus, ie one-side scripted conversation) VERY CLEARLY loops. As the length of their recall abilities lengthen so do the length of the loops until they've recovered and the cycle breaks.

It's covered (and an example recorded) in a moderately disturbing segment on Radiolab in the Loops episode starting at 7:45 here.
http://www.radiolab.org/2011/oct/04/

[Alyrium Denryle]

However, we choose to hold people responsible for their decisions anyway, and this is probably socially necessary as discussed.

And one can hold someone responsible for their actions without the needlessly punitive social sanctions we inflict on people, and with more recognition of cognitive and neural defects. Right now for example, the gold standard for an insanity defense is whether a person knew the difference between right and wrong. Under this standard, we have imprisoned schizophrenics, and people who have had massive tumors in their frontal lobe.

[Surlethe]

You think introducing the irrelevant distinction between 'can work out ahead of time' or 'can't work out ahead of time' is worthwhile? That's like saying a die isn't deterministic because nobody is currently modelling it's movement, or it was less deterministic before we had geometry.

Yeah, practicality is such a useless criterion to apply to models

Alarmingly I have to agree with Kanastrod, the only important this to even consider is what extent we judge people as responsible for their choices. After all, even if fairly floss free will exists (because of computation limits, lol) people never have the full range of options and there are always situational limitations. However, we choose to hold people responsible for their decisions anyway, and this is probably socially necessary as discussed.

And yet (a) that was not what the OP asked and (b) I see you putting no effort into making sure your writing is comprehensible, let alone proposing a coherent moral theory to address your new version of the question.