Anton Zeilinger's comment about free will being required for science

[msumm21]

TL;DR Summary

Questioning a remark by Anton Zeilinger that free will is required by science

To abandon freedom means to abandon science. In his book “Dance of the photons”, Anton Zeilinger remarks the following:

“The second important property of the world that we always implicitly assume is the freedom of the individual experimentalist. This is the assumption of free will. It is a free decision what measurement one wants to perform. In the experiment on the entangled pair of photons, Alice and Bob are free to choose the position of the switch that determines which measurement is performed on their respective particles. It was a basic assumption in our discussion that that choice is not determined from the outside. This fundamental assumption is essential to doing science. If this were not true, then, I suggest, it would make no sense at all to ask nature questions in an experiment, since then nature could determine what our questions are, and that could guide our questions such that we arrive at a false picture of nature.”

I'm not following the above quote about free will "This fundamental assumption is essential to doing science"

If I perform experiments and find that conditions X are followed by conditions Y and Y=F(X) (or the probabilistic analog for QM), haven't I done science, independent of how much my free will impacted X?

 

[jfizzix]

I think the notion is that if your decisions are completely determined, then you can't take yourself out of the equation when trying to see how changing X affects Y. Instead, there could be some third variable Z that independently determines both how you will change X and how Y will be changed. in that scenario, X and Y need not be causally related at all.

I wouldn't go so far as to say that true free will transcending causality exists, but one can at least do science in figuring out predictable patterns in Nature. Whether Z determines both X and Y, or whether X determines Y, you can still predict the value of Y by knowing the value of X if your model is right.

For a good reference on how we can deduce causal relationships from data, I recommend
"The Book of Why: The New Science of Cause and Effect"
By Judea Pearl (a world authority on statistical inference)
https://www.goodreads.com/book/show/36204378-the-book-of-why

 

[A. Neumaier]

I'm not following the above quote about free will "This fundamental assumption is essential to doing science"

To make sense of Zeilinger's quote, you should add to the word ''science'' the implicit qualifier ''of the kind Anton Zeilinger is doing''.

If I perform experiments and find that conditions X are followed by conditions Y and Y=F(X) (or the probabilistic analog for QM), haven't I done science, independent of how much my free will impacted X?

Of course this is science.

In astronomy, there is no free will - we have hardly any choice what to measure but must use (and do use) all information available from the light of the heavens to deduce whatever we can.

 

[vanhees71]

In some sense there is indeed no free will, because we can only use preparation and/or measurement devices (in a very broad sense) that themselves follow the fundamental natural use investigated by them. So if there is some validity of the natural sciences, all of which rest on the assumption that there exist generally valid natural laws, then our free will is restricted by the natural laws. E.g., all we know about the spacetime structure indicates that we cannot violate the energy-conservation law (at least not locally ;-)).

 

[Demystifier]

Zeilinger seems to be saying that humans are not a part of the nature:
"If this were not true, then, I suggest, it would make no sense at all to ask nature questions in an experiment, since then nature could determine what our questions are."

 

[Lord Jestocost]

The quote is related to superdeterminism.

The implications of superdeterminism, if it is true, would bring into question the value of science itself by destroying falsifiability, as Anton Zeilinger has commented.

From: https://en.wikipedia.org/wiki/SuperdeterminismRichard D. Gill in "Statistics, Causality and Bell’s Theorem"

I think that Occam’s razor tells us to discard this flavour of super-determinism, also known as conspiracy.
In fact, to abandon freedom means to abandon science: we may discard all empirical (observational)
data. Everything is explained but nothing can be predicted.

 

[vanhees71]

So the paradox is that then you have to state: If there's science (unchangeable natural laws), then there is no freedom (because I cannot change these laws). Now if you say that if there is no freedom you can not have science, these two assumptions lead to the contradiction that if there's science, there cannot be science.

I think it's very easy to solve this apparent paradox: Despite that there are natural laws and thus science, there's enough freedom left. Obviously humans have still a lot of freedom to decide about how to act though they cannot break the natural laws, whose existence is the only condition that there is science (another condition of course is that we can figure out these laws, which to a certain extent seems to be the case given that we have pretty successful science).

 

[EPR]

Is the human mind fully constrained by the natural laws?
To some extent - yes. Genes, environment, previous experience all play a big role in decision making. But... we can change our minds if we went wrong. This seems contrary to deterministic theory as some things we do appear completely unrelated to our survival. Or even lead straight up to our death - like self sacrifice. This isn't entirely deterministic. It's higher thought and probably a new state of being(outside plain determinism). So decision making likely is a very complex process of deterministic and emergent processes.

 

[Pony]

It is easy to construct universes where the laws of physics don't hold, but all of our experiments give us the same results.
Like something does not allow us to do some experiments, or someone changes the results, or something messes up with our brains etc. (E.g. superdeterminism is such a theory.)
Yes, if you want to make statements in the name of Science, like "the apple will fall down if I release it" you must add a ton of assumptions to it, that guarantees that reality exists, and your previous experiments were good, they sample the nature and its laws good (our experiments were random enough) etc. etc.
Personally I don't think that free will is necessary but some assumptions definitely are.

Also obviously it has nothing to do with doing Science.

 

[Elias1960]

Free will or not is not the main problem.

The problem is to get rid of that completely stupid idea of superdeterminism. If one would accept superdeterminism seriously, this would be indeed the end of science.

Fortunately, scientists are anyway very inconsistent in their reasoning, so they will not take superdeterminism seriously. They accept it as an excuse to get rid of the violation of the Bell inequalities, but will completely ignore that the same reasoning could be used by the tobacco industry to get rid of studies showing correlations with lung cancer and all the other results of statistical experiments which would be worth nothing if superdeterminism would be an acceptable excuse to ignore such studies.

So, science is not endangered, thanks that inconsistency which will win with certainty. As it actually already does with the rejection of realism and causality. Nobody takes the rejection of realism and causality seriously outside the discussion about Bell's theorem. That would be foolish. Everybody will ask the tobacco industry for other explanations for lung cancer and smoking correlations, if they claim that smoking does not cause lung cancer. Even if the principle which requires this, Reichenbach's common cause principle, has to be rejected by those who don't want to accept a preferred frame. But it is rejected only for Bell inequality discussions, thus, nothing to worry about. Nobody will question it in everyday science.

And nobody will take superdeterminism seriously outside the Bell inequality discussions.

 

[Demystifier]

And nobody will take superdeterminism seriously outside the Bell inequality discussions.

Yes, and the same can be said for non-realism. Nobody takes seriously non-realism outside the Bell inequality (and quantum measurement problem) discussions.

 

[nrqed]

The very concept of free will is incompatible with science as we know it. It requires something beyond the particles we are made of, something not subject to physical laws. Believing in free will is akin to believing in Gods who are not subjected to the laws of physics.

 

[vanhees71]

Fortunately, scientists are anyway very inconsistent in their reasoning, so they will not take superdeterminism seriously. They accept it as an excuse to get rid of the violation of the Bell inequalities, but will completely ignore that the same reasoning could be used by the tobacco industry to get rid of studies showing correlations with lung cancer and all the other results of statistical experiments which would be worth nothing if superdeterminism would be an acceptable excuse to ignore such studies.

I think the "inconsistency" is on the side of prejudiced philosophers rather than scientists. Why should one "get rid of the violation of the Bell inequalities"? It's one of the best established empirical facts that Bell inequalities are violated as predicted by QT. Why should a scientist want to get rid of this very successful prediction of QT?

 

[Elias1960]

I think the "inconsistency" is on the side of prejudiced philosophers rather than scientists. Why should one "get rid of the violation of the Bell inequalities"? It's one of the best established empirical facts that Bell inequalities are violated as predicted by QT. Why should a scientist want to get rid of this very successful prediction of QT?

It is, of course, not the quantum prediction and its experimental verification those scientists want to get rid of, but of the straightforward consequence, namely the rejection of Einstein causality (reducing it to the weak and irrelevant signal causality) and the acceptance of a hidden preferred frame. The preferred frame is something which cannot be accepted, whatever the costs.

You have chosen another way to avoid this - namely to avoid any discussion which could be classified as philosophical, and accepting signal causality as if it would be the same as strong Einstein causality.

 

[vanhees71]

Again: Einstein causality is not rejected by all these experiments, because they are described by relativistic local QFT, which describes the outcome of these experiments with very high accuracy, and this theory obeys Einstein causality.

What is "strong Einstein cauality" vs. signal causality? All you need for a theory to be consistent with Einstein causality is that space-like separated events are cannot be in a cause and effect relation, and that's the case for relativistic local QFTs.

 

[Lord Jestocost]

Yes, and the same can be said for non-realism. Nobody takes seriously non-realism outside the Bell inequality (and quantum measurement problem) discussions.

As a physicist I do, following Bohr's reasoning: Our theories speak not to any underlying reality, but to our experience of it.

 

[Demystifier]

As a physicist I do, following Bohr's reasoning: Our theories speak not to any underlying reality, but to our experience of it.

It is one thing to claim that a given theory does not describe reality, and another to claim that there is no reality (irrespective of our theories). Taking non-realism seriously means the latter, not the former.

 

[Elias1960]

Again: Einstein causality is not rejected by all these experiments, because they are described by relativistic local QFT, which describes the outcome of these experiments with very high accuracy, and this theory obeys Einstein causality.

What is "strong Einstein cauality" vs. signal causality? All you need for a theory to be consistent with Einstein causality is that space-like separated events are cannot be in a cause and effect relation, and that's the case for relativistic local QFTs.

If you don't know the difference, then don't make claims about this. RQFT obeys only signal causality, not Einstein causality. Even if you continue to name signal causality Einstein causality.

If they could not be in cause and effect relation, then you could prove Bell's inequality. But don't worry, in the world of the instrumentalist there are no cause and effect relations, only correlations. And signal causality gives you that there will be no correlations.

Moreover, for instrumentalists correlations don't need any causal explanations. If your fate correlates as predicted by astrology with the positions of the planets at your birthday, then everything is fine with astrology, nothing missed. Those scientists of the past who rejected astrology because it does not give realistic causal explanations for the influence of the stars on the fate of people are reactionaries. Modern science rejects astrology only because the correlations are not exactly as predicted by astrology or so.

Spoiler

(In fact, there are some correlations as predicted by astrology - those reactionary scientists explain them away as influences of those predictions on those who believe in astrology, self-fulfilling prophecies or so. Sorry, links to this not preserved.)

 

[PeterDonis]

The very concept of free will is incompatible with science as we know it.

Not for compatibilists, whose concept of free will is not the one you are implicitly using here. See, for example, the writings of Daniel Dennett on free will.

 

[Demystifier]

Not for compatibilists, whose concept of free will is not the one you are implicitly using here. See, for example, the writings of Daniel Dennett on free will.

I never understood how is compatibilism different from the claim that free will is an illusion.

 

[martinbn]

Yes, and the same can be said for non-realism. Nobody takes seriously non-realism outside the Bell inequality (and quantum measurement problem) discussions.

Well, standard quantum mechanics takes it seriously. It is in fact an integral part of it.

It is one thing to claim that a given theory does not describe reality, and another to claim that there is no reality (irrespective of our theories). Taking non-realism seriously means the latter, not the former.

No, it doesn't. Non-realism doesn't say anything about whether there is reality or not. It is the position that observables do not have values prior to measurement.

 

[Demystifier]

No, it doesn't. Non-realism doesn't say anything about whether there is reality or not. It is the position that observables do not have values prior to measurement.

By that definition, almost all interpretations would be non-realist interpretations, including GRW, many-world and in a certain sense even Bohmian.

 

[martinbn]

By that definition, almost all interpretations would be non-realist interpretations, including GRW, many-world and in a certain sense even Bohmian.

This is the only definition in this context.

 

[Demystifier]

This is the only definition in this context.

Then why do adherents of the interpretations I mentioned above call themselves realists?

 

[martinbn]

Then why do adherents of the interpretations I mentioned above call themselves realists?

Because some observables do have values at all tmes. In BM positon.

Who states that there is no reality?

 

[Demystifier]

Because some observables do have values at all tmes.

What observable has a value all the time in many world and GRW interpretations?

 

[martinbn]

What observable has a value all the time in many world and GRW interpretations?

I don't know. Why do you think they are realist interpretations? And my last question was who states that there is no reality?

 

[PeterDonis]

I never understood how is compatibilism different from the claim that free will is an illusion.

Because saying that free will is real, but is not what you thought it was, is not the same as saying free will is an illusion. Dennett's opinion (with which I tend to agree) is that people who think compatibilism means that free will is an illusion, have not really thought through the implications of the intuitions that underlie that thought. If they did, they would realize that those intuitions do not form a consistent set: it is impossible for anything real to actually have all the properties those intuitions say free will should have. The compatibilist view preserves everything that is worth wanting about free will, while still making free will something that can actually exist. (Dennett has similar views about consciousness.)

 

[Demystifier]

I don't know. Why do you think they are realist interpretations?

Because they say that the wave function is real, i.e. not merely a representation of our subjective incomplete knowledge.

And my last question was who states that there is no reality?

I told you that several times, but when I quoted their statements which show that, each time you would interpret their words differently. But to a certain extent I understand you, because those people write in a quite vague manner, because they don't want to be too explicit about non-realism (because it's philosophically unappealing) and at the same time they need some kind of non-realism to avoid Bell nonlocality.

Being frustrated with vagueness of non-realist interpretations, in one paper I have decided to explain directly and blatantly how a non-realist theory may really look like and what consequences of such a theory really are. That resulted in the paper http://de.arxiv.org/abs/1112.2034 . I would really be interested to see whether you find it a realist or a non-realist interpretation.

 

[Demystifier]

(Dennett has similar views about consciousness.)

But he says that consciousness is an illusion, doesn't he?

(BTW, when it comes to consciousness, I agree with Chalmers.)

 

[nrqed]

Because saying that free will is real, but is not what you thought it was, is not the same as saying free will is an illusion. Dennett's opinion (with which I tend to agree) is that people who think compatibilism means that free will is an illusion, have not really thought through the implications of the intuitions that underlie that thought. If they did, they would realize that those intuitions do not form a consistent set: it is impossible for anything real to actually have all the properties those intuitions say free will should have. The compatibilist view preserves everything that is worth wanting about free will, while still making free will something that can actually exist. (Dennett has similar views about consciousness.)

I am trying to make sense of this. Especially "preserves everything that is worth wanting about free will, while still making free will something that can actually exist."

Sounds like they redefine free will to be something else than what is usually understood. My point what the usual definition of free will is inconsistent with physical laws. But I will certainly look at what Dennett says.

 

[A. Neumaier]

I am trying to make sense of this. Especially "preserves everything that is worth wanting about free will, while still making free will something that can actually exist."

Sounds like they redefine free will to be something else than what is usually understood. My point what the usual definition of free will is inconsistent with physical laws. But I will certainly look at what Dennett says.

There is no 'usual definition'; it depends on whom you ask.

Philosophical notions like 'free will' are always defined by their usage, not by an authoritative definition. This means that the meaning is necessarily somewhat fuzzy.

Attempts to define them in a way that can be used to make logical deductions necessarily idealize the issue, which can be done in many ways.

These attempts are prone to throwing out the baby with the bath water unless one takes care to preserve 'everything that is worth wanting about free will, while still making free will something that can actually exist.'

 

[nrqed]

There is no 'usual definition'; it depends on whom you ask.

Philosophical notions like 'free will' are always defined by their usage, not by an authoritative definition. This means that the meaning is necessarily somewhat fuzzy.

Attempts to define them in a way that can be usded to make logical deductions necessarily idealize the issue, which can be done in many ways.

They are prone to throwing out the baby with the bath water unless one takes care to preserve 'everything that is worth wanting about free will, while still making free will something that can actually exist.'

Thanks. I would be very curious to have an example of something that is "worth wanting about free will" while not being in contradiction with physical laws.

 

[Lord Jestocost]

It is one thing to claim that a given theory does not describe reality, and another to claim that there is no reality (irrespective of our theories). Taking non-realism seriously means the latter, not the former.

In “Niels Bohr and the Philosophy of Physics: Twenty-First-Century Perspectives” (edited by Jan Faye and Henry J. Folse), Arkady Plotnitsky defines realism and the non-realist’s attitude in the following way (chapter 8):

I define "realism" as a specific set of claims concerning what exists and, especially, how it exists. In this definition, any form of realism is more than only a claim concerning the existence, or reality, of something, but rather a claim concerning the character of this existence. Realist theories are sometimes also called ontological theories. The term "ontological" carries additional connotations. They are, however, not important for the present discussion, and these terms will be used interchangeably here. What defines realism most generally is the assumption that a structure of reality, rather than only reality itself, exists independently of our interactions with it, or at least that the concept of structure would apply to reality. In Other words, realism is defined by the assumption that the ultimate constitution of the domain considered possesses attributes and the relationships among them, which may be either (a) known in one degree or another and, hence, represented, at least ideally, by a theory or model; or (b) unknown or even unknowable.

Non-realist interpretations of quantum phenomena and quantum mechanics, at least that of Bohr and others in the spirit of Copenhagen, not only do not make any of these realist assumptions, but also, in Bohr's ([1958] 1987b, 62) language, "in principle [exclude]" them. Quantum Objects exist, are real, but the nature of this existence places them beyond representation, at least by quantum theory, or even conception, although Bohr might not have been willing to go that far. The reality of quantum Objects is a reality without realism.

[Italics in original]

 

[Demystifier]

The reality of quantum Objects is a reality without realism.

This reminds me of Mermin's correlations without correlata.