Is Free Will a Foundational Assumption in Quantum Theory?

[Stephen Tashi]

1. The Experimenter's choice has to be a free choice or there would be some hidden variable that determines what measurement can or can't occur prior to an experiment taking place.

Post #2 addresses that point.

We don't require actual human beings selecting what to measure do we? Visualizing human beings as selecting the measurements is merely a compelling way of describing situations - e.g. the "Bob and Alice" stories. Is there an example where the human beings must make decisions by conscious deliberation? Or is the role of the human beings in the story merely to choose what to do in a probablistic manner? The common notion of "free will", as applied to myself, doesn't say I make random decisions according to certain probabilities - although a person observing my behavior might model it as stochastic! If I had to make a sequence of decisions in a truly stochastic way, I'd make a decision about what random number generator to use or what apparently random phenomenon to use and then let that process make the individual decisons. Otherwise, I suspect my own habits and biases would violate whatever proabilities I was trying to implement.

The common notion of free will as involving conscious deliberation may have nothing to do with QM, but since the original poster is asking about "free will" in a broad sense, we should clarify what aspects of "free will" in the broad sense are needed in various interpretations of QM - as opposed to what aspects of free will are useful in illustrating QM with stories.

 

[WWGD]

I hope I am not too far off but how could someone argue for determinism in a situation such as the three-pendulum experiment? Are outcomes predetermined to be essentially unpredictable/random?

 

[PeterDonis]

how could someone argue for determinism in a situation such as the three-pendulum experiment?

Determinism is not the same thing as predictability. Nonlinear deterministic systems can exhibit chaos, which means that even though they are deterministic, they have such a sensitive dependence on initial conditions that their behavior is unpredictable in practice.

 

[WWGD]

Determinism is not the same thing as predictability. Nonlinear deterministic systems can exhibit chaos, which means that even though they are deterministic, they have such a sensitive dependence on initial conditions that their behavior is unpredictable in practice.

Edit: So I guess we may just not have a precise-enough calibration of measurements to pinpoint the initial conditions closely-enough.

 

[PeterDonis]

I guess we may just not have a precise-enough calibration of measurements to pinpoint the initial conditions closely-enough.

Yes.

 

[WWGD]

Yes.

So is this view of superdeterminism somehow "orthogonal " to that of MWI? If the measurement is pre-determined by the initial tuning then there is only one possible outcome?

 

[PeterDonis]

is this view of superdeterminism somehow "orthogonal " to that of MWI?

I wasn't describing superdeterminism, I was just describing determinism. Determinism does not necessarily imply superdeterminism.

As far as the MWI is concerned, the unitary dynamics of QM, which is what the MWI claims governs everything, is deterministic, but it is not nonlinear; the Schrodinger Equation is linear. So there is no chaos--no sensitive dependence of the initial conditions--in the unitary dynamics of QM. The issue is that the unitary dynamics of QM is for the wave function, which is not something we can measure or observe directly, and it is dynamics in an abstract configuration space, not the spacetime we actually observe.

If the measurement is pre-determined by the initial tuning

It isn't. That's the issue. What is determined by the initial conditions is the wave function, but the wave function does not include any single unique measurement result.

 

[Lynch101]

To add to my post above. Let's simplify things and have us roll dice instead. You can if you wish check that:

You get 1-6 with equal frequency ( I.e probability) and I get 1-6 also with equal probability.

You can also check that our dice rolls are independent and uncorrelated.

Thanks Perok, I was having trouble with the first example you gave, this one I understand better. My understanding of probability stretches this far :)

It might not be necessary, but just in case, I do want to stress that I am not advocating in favor of superdeterminism. I would question the common sense notion of free will which has led me to try and get a better understanding of statements I've heard in relation to Free Will in the context of the foundations of Quanutm Mechanics.

This is the basis of applying probability theory, as it is applied in Bell's theorem.

If we use this data (our dice rolls) to make decisions in a QM experiment that leads to an "unexpected" correlation in the results of that experiment. Then either we accept the correlation is inherent in the experiment. Or, we reject the lack of correlation in the dice rolling. This latter idea is not the loss of free will, but superdeterminism: somehow everything in the universe conspires to produce data that looks uncorrelated on one level, but at some other level is deeply correlated.

It's nothing to do with free will, per se.

I've heard the argument that the act of choosing to use dice itself involves free will, or the freedom of choice of the experimenter. I'm wondering why we would need to use dice in the first place though? Is the purpose of using dice to eliminate [or close the loophole of] Free Will in the first place? If so, would this imply that the commone sense notion of Free Will is a foundational assumption?

I guess, I'm not necessarily interested in the specifics of superdeterminism (SD), rather I'm just wondering if SD is a necessary consequence in the absence of Free Will.

As you allude to above, I have heard the idea that it is not necessarily Free Will as it is commonly thought of, but then I see suggestions - such as the use of dice - as a means to eliminate the common notion of Free Will.

 

[Lynch101]

Post #2 addresses that point.

We don't require actual human beings selecting what to measure do we? Visualizing human beings as selecting the measurements is merely a compelling way of describing situations - e.g. the "Bob and Alice" stories. Is there an example where the human beings must make decisions by conscious deliberation? Or is the role of the human beings in the story merely to choose what to do in a probablistic manner? The common notion of "free will", as applied to myself, doesn't say I make random decisions according to certain probabilities - although a person observing my behavior might model it as stochastic! If I had to make a sequence of decisions in a truly stochastic way, I'd make a decision about what random number generator to use or what apparently random phenomenon to use and then let that process make the individual decisons. Otherwise, I suspect my own habits and biases would violate whatever proabilities I was trying to implement.

The common notion of free will as involving conscious deliberation may have nothing to do with QM, but since the original poster is asking about "free will" in a broad sense, we should clarify what aspects of "free will" in the broad sense are needed in various interpretations of QM - as opposed to what aspects of free will are useful in illustrating QM with stories.

Thank you Stephen!

I have heard the argument made that the simple act of choosing the random number generator involves Free Will. Does Quantum Alchemy's point apply here, about hidden variables that determine what measurement can or can't occur prior to an experiment taking place?

 

[DarMM]

Take for example what is done in some actual Bell Tests. Here we have a pair of entangled photons. A device can measure one of two observables on one of the photons and another device can measure one of two observables on the second.

In each round we get a pair of distant quasars to select which of the observables each device will measure, this is based on the wavelength of light received from the quasars. Bell proved that local causal single valued hidden variables being responsible for the outcomes puts a bound on the correlations in such an experiment. This bound is broken, verifying QM and rejecting such hidden variables.

However it assumes that the distant quasars are uncorrelated with the photons, i.e. that the wavelength of the light from the quasars has nothing to do with the photons created in our lab.

Superdeterminism rejects this. It says that the photons are described by such hidden variables, but that the initial state (or at least a suitably early one) of the universe was such that the state of photons in our lab happen to be in sync with the quasar emission wavelength in such a way as to make it look like QM is true.

The analogue for General Relativity would be that Newtonian Mechanics is correct even for Mercury's orbit, but our telescopes happen to warp or distort images of Mercury due to flaws that emerge thanks to initial conditions in their construction in just such a way as to make it look like General Relativity is true.

 

[Lynch101]

Take for example what is done in some actual Bell Tests. Here we have a pair of entangled photons. A device can measure one of two observables on one of the photons and another device can measure one of two observables on the second.

In each round we get a pair of distant quasars to select which of the observables each device will measure, this is based on the wavelength of light received from the quasars. Bell proved that local causal single valued hidden variables being responsible for the outcomes puts a bound on the correlations in such an experiment. This bound is broken, verifying QM and rejecting such hidden variables.

However it assumes that the distant quasars are uncorrelated with the photons, i.e. that the wavelength of the light from the quasars has nothing to do with the photons created in our lab.

Superdeterminism rejects this. It says that the photons are described by such hidden variables, but that the initial state (or at least a suitably early one) of the universe was such that the state of photons in our lab happen to be in sync with the quasar emission wavelength in such a way as to make it look like QM is true.

The analogue for General Relativity would be that Newtonian Mechanics is correct even for Mercury's orbit, but our telescopes happen to warp or distort images of Mercury due to flaws that emerge thanks to initial conditions in their construction in just such a way as to make it look like General Relativity is true.

Thanks DarMM. I haven't had a chance to try to wrap my head around SD and it certainly sounds like it will take quite a bit of wrapping. I also gather that the consensus is seriously against SD. I'm not particularly interested in it at the moment, and so I'm not trying to defend it. I might be misinterpreting the purpose of your reply, but when I see a critique of SD in this manner it reads like a defense of Free Will by attempting to demolish the alternative. Is that how it can or should be taken?

 

[DarMM]

I might be misinterpreting the purpose of your reply, but when I see a critique of SD in this manner it reads like a defense of Free Will by attempting to demolish the alternative. Is that how it can or should be taken?

It's just a description of Superdeterminism. "Free Will" in the sense used in quantum foundations theorems basically means telescopes don't distort in just the right way when we go to look at Mercury and so on. That the means of observation used to analyse remote systems are not correlated strongly with those remote systems in a way that prevents correct analysis of them.

 

[Elias1960]

John Conway [of the Conway-Kochen-Specker theorem] talks unequivocally about "Free Will" in the common sense form of the idea.
...
Is there any sense in which the common sense notion is related to the usage in QM foundations?

To make common sense free will compatible with classical determinism is not easy, but one can at least try. I would not see any chance to make it compatible with superdeterminism. So, I think common sense free will is something stronger than the rejection of superdeterminism.

 

[Lynch101]

To make common sense free will compatible with classical determinism is not easy, but one can at least try. I would not see any chance to make it compatible with superdeterminism. So, I think common sense free will is something stronger than the rejection of superdeterminism.

I guess the question is, is it necessary to make Free Will compatible with QM?

 

[Demystifier]

I guess the question is, is it necessary to make Free Will compatible with QM?

As long as free will exists, it's necessary to make it compatible with QM. But whether free will exists or not depends on what exactly one means by "free will".

 

[Lynch101]

As long as free will exists, it's necessary to make it compatible with QM. But whether free will exists or not depends on what exactly one means by "free will".

Is there currently an interpretation of it in use in QM?

 

[Demystifier]

Is there currently an interpretation of it in use in QM?

Interpretation of what?

 

[Lord Jestocost]

As long as free will exists, it's necessary to make it compatible with QM. But whether free will exists or not depends on what exactly one means by "free will".

"Despite their claim that they are better equipped than scientists to make conceptual distinctions and evaluate the cogency of arguments, professional philosophers have mistakenly conflated the concepts of "free" and "will." They (con)fuse them with the muddled term "free will," despite clear warnings from John Locke that this would lead to confusion.

Locke said very clearly, as had some ancients like Lucretius, it is not the will that is free (in the sense of undetermined), it is the mind." [bold by LJ]

From: https://www.informationphilosopher.com/freedom/free_will.html

 

[Lynch101]

Interpretation of what?

You said that if Free Will exists then it needs to be accounted for in QM. I'm wondering if there is an interpretation of Free Will in use in QM?

 

[Lynch101]

"Despite their claim that they are better equipped than scientists to make conceptual distinctions and evaluate the cogency of arguments, professional philosophers have mistakenly conflated the concepts of "free" and "will." They (con)fuse them with the muddled term "free will," despite clear warnings from John Locke that this would lead to confusion.

Locke said very clearly, as had some ancients like Lucretius, it is not the will that is free (in the sense of undetermined), it is the mind." [bold by LJ]

From: https://www.informationphilosopher.com/freedom/free_will.html

Is the freedom of the mind the fundamental assumption in QM then and in what sense would you say that "the mind" is free?

EDIT: At the risk of devolving into unrelated philosophical territory, I'm wondering how these issues relate to QM or the foundational questions of QM.

 

[Lord Jestocost]

Is the freedom of the mind the fundamental assumption in QM...

Is there a fundamental assumption in QM?
As Aage Bohr, Ben R. Mottelson and Ole Ulfbeck put it in "The Principle Underlying Quantum Mechanics":
"In fact, the quantum mechanical formalism was discovered by ingenious guesswork which was given an interpretation in terms of probabilities for the results of measurements."

 

[Demystifier]

You said that if Free Will exists then it needs to be accounted for in QM. I'm wondering if there is an interpretation of Free Will in use in QM?

See e.g. my http://de.arxiv.org/abs/1006.0338

 

[Lynch101]

See e.g. my http://de.arxiv.org/abs/1006.0338

Thanks Demystifier, the abstract sounds very intersting. I'll give a run through it

 

[Lynch101]

Is there a fundamental assumption in QM?

I guess that's partially what I'm trying to find out. I've heard from various sources that Bell's Theorem implies that one of either realism, locality, local realism, or free will must be given up because they are fundamental assumptions of Bell's theorem and the violation of the inequalities implies one or more of the 4 must be jettisoned.

The notion of free will that gets cited by various sources, including Bell and Conway (of the free will theorem) appears to be the common sense notion of free will.

 

[Demystifier]

Neither deterministic nor probabilistic fundamental laws are compatible with a true free will. If the behavior is deterministic that it's not free, if the behavior is probabilistic then it's not controlled by a will.

 

[A. Neumaier]

If the behavior is deterministic then it's not free

No. Free only means that it is determined by nothing else than the person whose free will is under discussion, not that it is not determined at all!

 

[Demystifier]

No. Free only means that it is determined by nothing else than the person whose free will is under discussion, not that it is not determined at all!

With that definition of "free", a human is not more free than a robot. I don't think it is what most people mean by "free".

 

[Lynch101]

Neither deterministic nor probabilistic fundamental laws are compatible with a true free will. If the behavior is deterministic that it's not free, if the behavior is probabilistic then it's not controlled by a will.

This is the position that Sam Harris outlines in his book titled "Free Will". He makes some reference to Heisenberg (I think it was Heisenberg, I must re-read it) and other "compatbilists". It's one of the reasons I'm interested to find out what role - if any - free will plays in QM. It's an interesting topic!

 

[Demystifier]

He makes some reference to Heisenberg (I think it was Heisenberg, I must re-read it) and other "compatbilists".

Compatibilism is an attitude that I never understood.

 

[Lynch101]

No. Free only means that it is determined by nothing else than the person whose free will is under discussion, not that it is not determined at all!

Compatibilism is an attitude that I never understood.

I can't remember the details of it now, I need to go back over it

 

[A. Neumaier]

With that definition of "free", a human is not more free than a robot.

This is indeed the philosophical position of strong AI.

I don't think it is what most people mean by "free".

But it is consistent with what most people mean by "free". Most people have reasonable motives for their free decisions; the motives together with the external constraints determine the decisions. The minority for which this is not the case are considered to be whimsical or psychotic by their surrounding.

Do you really think that a child is not free in its decisions just because we can predict that it will say yes when it is asked whether it likes to have ice cream?

 

[mattt]

Most people that believe in Free Will, will reject the explanation that everything they do is the result of a deterministic or stochastic process. Their conception of Free Will has no place in science, in my opinion.

 

[mattt]

No. Free only means that it is determined by nothing else than the person whose free will is under discussion, not that it is not determined at all!

That can be an interesting definition, but it is not the way most people think they have free will, at all.

They just believe that their actions are not the results of a deterministic or stochastic (or any combination of them) process.

 

[A. Neumaier]

Most people that believe in Free Will, will reject the explanation that everything they do is the result of a deterministic or stochastic process.

Really? Many of those are atheists who believe that everything is purely the result of the natural laws encoded into quantum mechanics. We don't have interpretations of quantum mechanics other than either stochastic or deterministic.

 

[mattt]

Really? Many of those are atheists who believe that everything is purely the result of the natural laws encoded into quantum mechanics. We don't have interpretations of quantum mechanics other than either stochastic or deterministic.

I know, but that's the way they feel, and that's what they say they believe.

That's what they say, I have had hundreds of conversations along these lines with all kinds of people that believe that they have Free Will, no matter if they are scientist or not, religious or not.