Does this QM experiment show that science is doomed?

[francis20520]

TL;DR Summary

Paradox puts objectivity on shaky footing (https://science.sciencemag.org/content/369/6506/889 )

Here is the article: https://www.sciencemagazinedigital....cle.action?articleId=1611119#articleId1611119

If 2 experimenters (or people) cannot agree on a single observation does this mean that science which depends on many observers agreeing on "observing" the same thing (objectivity, reproducibility etc) is meaningless and doomed?

Because science relies on measurements and facts. If according this paper, facts and measurements are relative then how can science continue as "science"?? What will happen to the saying "facts are hallowed"?

Does this mean that science can never find what "reality" is? That objective reality is beyond the possibility of science?

What does this research papers outcome imply to scientific facts like "electrons exist", "electrons mass is so and so" etc?

 

[PeterDonis]

You have put your finger on a key reason why I think all the hype about these results is overblown and misleading. The scientists who do these experiments certainly aren't saying that the results they have obtained in their labs are somehow not real, or that someone else could make some quantum measurement on them and their labs that would somehow invalidate their results. They are treating their results as objective facts, which makes claims that these results somehow threaten objectivity rather dubious.

There have been quite a few previous discussions here at PF on the general topic of "Wigner's friend" type experiments, and I believe there has been at least one previous thread on this particular kind of experiment, since the one referenced in this article is not the first of its kind to be published recently.

An important thing to remember about all of these experiments is that they are not actual "Wigner's friend" experiments in the sense that term is properly used. There are no humans in these experiments who are having quantum operations done on them; there are no humans who are being put in quantum superpositions of, for example, observing result A and observing result B in some lab. There are not even complex quantum systems with many degrees of freedom, but still much simpler than humans (say rocks or microscopic dust particles) being experimented on. The quantum systems being experimented on are qubits, and the experiments themselves are just garden-variety quantum operations on qubits, of the sort that have become commonplace now that quantum information is a popular and fast-growing field. The technical skill required to do these experiments is impressive, and it is certainly not a waste of time to confirm more quantum predictions in this domain; but again, I think the hype about these results is overblown and misleading.

 

[francis20520]

I am also curious about this hidden variable issue.

So, are these researchers saying that 2 observers will make different measurements or establish different facts about the same "thing"?

Because from my understanding, if we rule out solipsism (which is obvious) there has to be "something" out there.

There has to be some fundamental particles. Otherwise how can exist?

So, according to this experiment, are 2 observers measuring the same objective reality differently or is it something else??

I just cannot figure out (as a layman) what that "something else" is.

 

[PeroK]

There has to be some fundamental particles. Otherwise how can exist?

Who says there are no elementary particles?

If scientific experiments produce results at odds with your naive expectations of reality, then it is your naive expectations that are " doomed". The science goes on full steam ahead.

 

[francis20520]

Who says there are no elementary particles?

If scientific experiments produce results at odds with your naive expectations of reality, then it is your naive expectations that are " doomed". The science goes on full steam ahead.

I am ALL for science. I am on the side of science and objective reality.

I am trying to figure out why some quantum physicists say thing like "there is no objective reality".

I am thinking that their experiments could be flawed or just wrong.

But I am not a physicist so that is why I ask for your help.

 

[PeroK]

What does "there is no objective reality" actually mean? In its simplest form physics has two aspects: the results of experiments; and, a mathematical theory that can produce those results.

The trouble starts when you try to explain what's happening "in reality". Isaac Newton identified this issue with his own theory of gravity. He knew that it modeled the solar system but made no physical sense.

 

[Demystifier]

Another hype. AFAICS it's just a new form of quantum nonlocality, slightly stronger than Bell nonlocality. The actual paper is this: https://www.nature.com/articles/s41567-020-0990-x

 

[vanhees71]

In my opinion it's much ado about nothing, because the whole paradox is simply solved by just taking the quantum theoretical probabilities seriously and as "objective facts". There's nothing indicating that anything is wrong with this minimal statistical interpretation, and then there's no paradox and everybody in the Wigner's-friend scenario agrees on the objective facts based on their measurement results, and these results are objective facts and the content of these facts is simply about the statistical properties of outcomes of the measurements of each observer, and there's never a contradiction between the various probabilities for the measurement outcomes of these observers. This view also resolves the apparent paradox between locality of interactions between the measurement devices and the quantum system and "stronger-than-classical" correlations between measurement outcomes far-distant parts of a quantum system in an entangled state wrt. of the measured observables.

All this, of course, belongs to the "interpretation section" of this forum.

 

[DrChinese]

Because science relies on measurements and facts. If according this paper, facts and measurements are relative then how can science continue as "science"?? What will happen to the saying "facts are hallowed"?

Does this mean that science can never find what "reality" is? That objective reality is beyond the possibility of science?

This is a misunderstanding of key elements of modern science.

1. Objective reality is a reality that exists independent of an observer. As vanhees71 says, you can consider the reality described by quantum theory as being such that statistical properties/results are accurate (what he calls a minimal interpretation). Other interpretations handle this differently, and most deny the existence of an observer independent reality - and have for 90+ years. (Such is referred to as "subjective reality".) So in that respect, the cited experiment is not considered exactly "new".

2. Science is based on useful descriptions of patterns and pattern exceptions. Nothing about the concept of "subjective reality" takes anything away from the utility of modern science. The error is thinking that what you know is "true" in some absolute sense. Theories, being useful models, should not be considered as "true". Rather they should be considered as useful in the realm they apply to. Don't confuse the map with the territory it describes.

 

[francis20520]

@DrChinese , I have to disagree here. The main purpose of science IS to figure out how the universe works, what its components are, and how the universe got to be the way it is now.

I find it preposterous and an insult to think that Einstein spent 10 years to "discover" GR hoping that some time later it can be used in correcting coordinates of GPS satellites.

 

[PeterDonis]

The main purpose of science IS to figure out how the universe works

Figuring out how the universe works is not the same as having a theoretical model of how the universe works that is perfect and exactly true. @DrChinese is only saying science is not the latter. He is not saying science is not the former.

I find it preposterous and an insult to think that Einstein spent 10 years to "discover" GR hoping that some time later it can be used in correcting coordinates of GPS satellites.

Nobody has said this so you are attacking a straw man.

 

[vanhees71]

As I said before, I don't think that QT says that there's no objective reality. It only says that nature is not deterministic, i.e., that there is an objective randomness. The quantum state describes the statistics for the outcome of measurements given the situation the system is prepared in. This simply implies that observables have only determined values, if the system is prepared in a corresponding state, and that it is impossible to prepare a system in such a way that all observables are simultaneously determined (uncertainty relation). This doesn't mean that nature is subjective. All observers agree about the probabilities predicted by QT, and there's also no intrinsic contradiction on the probabilities in the Wigner's-friend setup, as long as you don't assume that there should be more than the probabilistic description of QT to have a "complete theory".

 

[francis20520]

Figuring out how the universe works is not the same as having a theoretical model of how the universe works that is perfect and exactly true.

So are you saying that the "model" of the atom created by science is NOT the way it actually is found in nature?

But then what else could it be?

If a scientist say and verify by experiment that atoms exist what else does it mean than the model is modelling exactly how it is in nature?

Or r you saying that models proposed by scientists (atoms, GR, electromagnetism etc) are just "Models" only and we don't know what there is actually?

 

[PeterDonis]

are you saying that the "model" of the atom created by science is NOT the way it actually is found in nature?

I'm saying that our theoretical model of an atom is not the same as the actual atom. Our theoretical model does not capture everything about the atom, and might capture some things wrong. Scientific models are always subject to revision as we gather more evidence.

what else could it be?

Do you mean, what else could an atom be? An atom is what it is. Our theoretical models are not the same as the things they model. This should be obvious.

 

[PeroK]

So are you saying that the "model" of the atom created by science is NOT the way it actually is found in nature?

I think it was Heisenberg who first realized that nature at the fundamental atomic and subatomic level can only be described by mathematics. It's not that there is nothing there; it's that ultimately there is no way to describe what's there other than by a mathematical model. You can't look at an atom the way you can look at the Solar system. Experiments involving atoms tend to involve some sort of inference from data like an emission spectrum. There is an alternative theory to orthodox QM called Bohmian mechanics, where the particles have definite positions and are guided by a pilot wave. If you could look at a hydrogen atom the way you can look at the Solar system you could immediately say whether the electron is "really" being guided round an oribit by a pilot wave; or, is "really" represented by a cloud of probability. This cannot be done, so the two theories persist as viable alternative descriptions of nature.

So when you say "found" in nature, what experiment would determine whether a hydrogen atom is as described by orthodox QM or by Bohmian/Pilot wave theory? You cannot just look and see.

 

[francis20520]

I think it was Heisenberg who first realized that nature at the fundamental atomic and subatomic level can only be described by mathematics. It's not that there is nothing there; it's that ultimately there is no way to describe what's there other than by a mathematical model. You can't look at an atom the way you can look at the Solar system. Experiments involving atoms tend to involve some sort of inference from data like an emission spectrum. There is an alternative theory to orthodox QM called Bohmian mechanics, where the particles have definite positions and are guided by a pilot wave. If you could look at a hydrogen atom the way you can look at the Solar system you could immediately say whether the electron is "really" being guided round an oribit by a pilot wave; or, is "really" represented by a cloud of probability. This cannot be done, so the two theories persist as viable alternative descriptions of nature.

So when you say "found" in nature, what experiment would determine whether a hydrogen atom is as described by orthodox QM or by Bohmian/Pilot wave theory? You cannot just look and see.

But doesn't Bohmian mechanics require faster than speed of light messaging which violates GR?

There is no experimental verification of Bohmian mechanics. So it's just a hypothesis.

But QM is a theory which is verified by experiment.

So I don't know whether your analogy is correct.

 

[PeroK]

So I don't know whether your analogy is correct.

Bohmnian mechanics (BM) predicts everything that QM predicts. (As far as it can.)

Both orthodox QM and BM require non-locality, but not faster-than-light communication.

In any case, the fundamental point is that all we have is mathematics to describe the sub-atomic world. There may be several physical interpretations that are consistent with the mathematics. In which case it's not possible to choose between these physical interpretations simply by looking at nature to see what it is "really" like.

 

[vanhees71]

So do you claim to have a working Bohmian interpretation of relativistic quantum field theory, where nothing propagates faster than the speed of light? That would be interesting. Note that within non-relativistic physics there's no constraint hindering ffaster-than-light communication. To the contrary it's even assumed, e.g., in describing the gravitational interaction a la Newton. It's an interaction at a distance within this model.

The way Einstein causality is realized in both classical and quantum theory is by a description via local (quantum) field theories. For the quantum theory that means that the Hamiltonian density (in the Heisenberg picture) is a field operator depending on one space-time argument and obeys the microcausality condition, i.e., it commutes with any local observable at space-like separated arguments. This implies a Poincare-covariant S-matrix, fulfilling the linked-cluster theorem, which indeed means that there's no faster-than-light signal propagation for observable quantities but the possibility of "inseparability" as described by entanglement, which is often sloppily called "non-locality".

 

[PeroK]

The way Einstein causality is realized in both classical and quantum theory is by a description via local (quantum) field theories. For the quantum theory that means that the Hamiltonian density (in the Heisenberg picture) is a field operator depending on one space-time argument and obeys the microcausality condition, i.e., it commutes with any local observable at space-like separated arguments. This implies a Poincare-covariant S-matrix, fulfilling the linked-cluster theorem, which indeed means that there's no faster-than-light signal propagation for observable quantities but the possibility of "inseparability" as described by entanglement, which is often sloppily called "non-locality".

This exemplifies the point. There is no experiment to say that nature is "really" field operators with commutation properties. We can only say that that mathematics models nature; not that nature really is a field of such operators.

Another mathematical theory (even if it's not BM) might achieve the same description.

And, if QFT itself gets superseded, then where did these field operators disappear to?

Also, this is a B level thread!

 

[DrChinese]

So are you saying that the "model" of the atom created by science is NOT the way it actually is found in nature?

But then what else could it be?

If a scientist say and verify by experiment that atoms exist what else does it mean than the model is modelling exactly how it is in nature?

Or r you saying that models proposed by scientists (atoms, GR, electromagnetism etc) are just "Models" only and we don't know what there is actually?

We know plenty, and we don't know plenty. As pointed out by others in this thread: The most accurate models we have are mathematical and make no attempt at all at physical interpretation. The models that attempt to provide physical explanation are dogged with baggage many find unacceptable.

When you ask "what else could it be?": you can say that the model works well with the experimental questions that have been asked so far. Many scientists want to ask questions for which there is currently no suitable experimental representation. Those future answers, if indeed they are answered in the future, might allow us to know what else it could be.

 

[Sunil]

Who says there are no elementary particles?

If one interprets this as "elementary particles are nothing fundamental", then I think many people think so. All those who consider the fields to be more fundamental. For them, the "elementary particles" are nothing but pseudoparticles, simply energy levels of the quantum fields, similar to the "phonons", the "particles of sound" in quantum condensed matter theory.

And there is a good reason to consider the fields as more fundamental: Semiclassical gravity. The fields are, then, well-defined things, but the particles not. The very notion of a particle changes in time once the gravitational field changes.

So do you claim to have a working Bohmian interpretation of relativistic quantum field theory, where nothing propagates faster than the speed of light? That would be interesting.

Bohm.D., Hiley, B.J., Kaloyerou, P.N. (1987). An ontological basis for the quantum theory, Phys. Reports 144(6), 321-375
Kaloyerou, P.N. (1994). The Causal Interpretation of the Electromagnetic Field. Physics Reports 244, 287-358
Kaloyerou, P.N. (1995). Causal Interpretation of the Modified Klein-Gordon Equation. Found Phys 25(10), 1413-1460

Of course, something propagates immediately, as in Bohmian interpretation in general. But a non-signalling theorem holds.

 

[Sunil]

But doesn't Bohmian mechanics require faster than speed of light messaging which violates GR?

No, it does not require "messaging". It requires faster than light causal influences. This is not in conflict with any experiment supporting GR, all it requires is a (possibly hidden) preferred frame.

There is no experimental verification of Bohmian mechanics. So it's just a hypothesis.
But QM is a theory which is verified by experiment.

No, there is no such thing as experimental verification of general theories. Theories can be tested, falsified (the prediction fails) or confirmed. But confirmed does not mean verified, it only means not yet falsified.

All experiments which have confirmed QM have also confirmed BM. Because BM has predicted the same result. So there is no difference between QM and BM in experimental confirmation.

 

[DrChinese]

If one interprets this as "elementary particles are nothing fundamental", then I think many people think so. All those who consider the fields to be more fundamental. For them, the "elementary particles" are nothing but pseudoparticles, simply energy levels of the quantum fields, similar to the "phonons", the "particles of sound" in quantum condensed matter theory.

And there is a good reason to consider the fields as more fundamental: Semiclassical gravity. The fields are, then, well-defined things, but the particles not. The very notion of a particle changes in time once the gravitational field changes.

Are fields really "more fundamental" than the particles we observe? Clearly this is just a byproduct of the model and in no way can be answered objectively. So again I say: no model is reality. But the usefulness of particular models in particular situations can be compared. That does *not* make one true, and the other false.

For example: In the vast majority of cases, Newtonian gravity is actually a "better" theory than General Relativity. That's because Newtonian gravity is more useful when enough other factors are unknown or insignificant to the desired precision. (And please, don't try to explain to me that GR reduces to NG in some limit - that's the whole point.)

 

[vanhees71]

The question is, what do you mean by "particle". For me the most comprehensive model we have today is the Standard Model, which is based on QFT, and within that model a "particle" is a asymptotic free Fock state, and even this interpretation is problematic at least for charged particles since due to the long-ranged nature of the electromagnetic interaction (Coulomb potential fall only with $1/r$) these are not the right asymptotic free states, giving rise to infrared problems. Formally this is due to the masslessness of the em. field, but that's what implies the long-range properties of the em. interaction. The usual perturbative remedy is to use appropriate "ladder resummations" resumming the soft-photon contributions which all contribute formally in the same order in the expansion in powers of the coupling constant. This leads to "infra particles", and in this more physical picture it's a usual "particle" surrounded by a "cloud of soft (virtual) photons", i.e., surrounded by its own em. field. So the idea of a "particle" is a pretty complicated issue. So whether you consider "particles" or "fields" as the "fundamental entities" is not so easy to answer.

From conceptual clarity even in classical physics, I'd say fields are superior: In electromagnetism there's no selfconsistent description of electromagnetically interacting charged point particles while the field theoretical description in terms of a continuum mechanical (field) model has no such issues. FAPP of course, as far as you can get with the point-particle picture, the Landau-Lifshitz version of the Lorentz-Abraham-Dirac equation is a good approximation to a dynamical model of charged point particles and electromagnetic fields.

 

[Demystifier]

As I said before, I don't think that QT says that there's no objective reality. It only says that nature is not deterministic, i.e., that there is an objective randomness. The quantum state describes the statistics for the outcome of measurements given the situation the system is prepared in.

But some versions of the Bell theorem do not assume determinism. So what those versions of the Bell theorem really show us, in your opinion?

 

[vanhees71]

It's not clear to me what you are referring to. Do you have a (physics rather than philosophy) paper as an example for such versions of Bell's theorem?

 

[Demystifier]

It's not clear to me what you are referring to. Do you have a (physics rather than philosophy) paper as an example for such versions of Bell's theorem?

See e.g.
https://arxiv.org/abs/1802.10119 (especially Sec. VIII.)
https://arxiv.org/abs/1402.0351

 

[francis20520]

I am bit confused.

So you are saying that if say somebody asks you what is the smallest unit of matter you have discovered, you have to really say "We really don't know". Because according to your answers "science" actually does not know anything. It just models what it "thinks" is there, but not sure whether "that" is how actually "it" is.

So, if this is how science is done I don't see how it will be any different 100 or 1000 years from now. Even then scientists will still say "we are not sure. It might be this or that".

Doesn't this make this whole business of "science" nothing much to brag about??

What's the big deal then in giving people Nobel Prizes?

 

[Ibix]

Doesn't this make this whole business of "science" nothing much to brag about??

All right... all right... but apart from better sanitation and medicine and education and irrigation and public health and roads and a freshwater system and baths and public order... what have the Romans scientists done for us?
- Monty Python (nearly)

 

[francis20520]

All right... all right... but apart from better sanitation and medicine and education and irrigation and public health and roads and a freshwater system and baths and public order... what have the Romans scientists done for us?
- Monty Python (nearly)

So what your saying is that the value in science is purely utilitarian.
Science is useless in really figuring out anything about the universe.
Because models of science can never 100% tell us how the universe actually works or what constitutes it.

 

[vanhees71]

Of course not. The scientific method is about what we can learn from observations and building models (and rarely even theories) describing all observations within a consistent mathematical model. If we discover objective deviations between a model (or even a theory) we have to adapt the model (or theory) to the new objective facts. That's not a bug but a feature of science!

 

[Motore]

Because models of science can never 100% tell us how the universe actually works or what constitutes it.

How can they, if they are models. We can refine them with time and knowledge and get them closer (perhaps arbitrarily) to describing objective reality, but the models can never be objective reality.

That's not a bug but a feature of science!

Exactly!

 

[Ibix]

Science is useless in really figuring out anything about the universe.

No - I'm pointing out that it is far and away the best method of figuring out anything about the universe that we have. As evidence, look at all the stuff we can build because of science improving our understanding of how everything works. To say "science is useless because I cannot see how to do one thing I want to do with it" seems rather peculiar to me.

 

[Sunil]

So you are saying that if say somebody asks you what is the smallest unit of matter you have discovered, you have to really say "We really don't know". Because according to your answers "science" actually does not know anything. It just models what it "thinks" is there, but not sure whether "that" is how actually "it" is.

So, if this is how science is done I don't see how it will be any different 100 or 1000 years from now. Even then scientists will still say "we are not sure. It might be this or that".

Doesn't this make this whole business of "science" nothing much to brag about??

What's the big deal then in giving people Nobel Prizes?

The point is that we cannot be sure, certain, that our actually best theories are really true.

Actually, we are even sure that they are not true, given that we have yet open scientific problems - GR cannot be true because it is not a quantum theory, the SM cannot be true because it is not part of a theory which covers gravity, and because it is only an effective field theory, that means, has to be replaced by a better theory below some critical distances (which is guessed to be of order of Planck length).

But even if we would have a theory which does not leave open a single problem of fundamental physics, and which covers everything in fundamental physics, we could not be sure that this theory is the final one. We cannot exclude that tomorrow we will observe something new, not adequately described by that theory.

This is so, and this will remain so forever. But of course we know much more about Nature today. We can make predictions about things we have not even known that they existed at all, and these predictions appear correct if tested. This predictive ability raises. And it gives us a good base for believing that our best existing theories are, even if not true, then at least good approximations of reality.

 

[DennisN]

Doesn't this make this whole business of "science" nothing much to brag about??

In 1969 humans visited another celestial body for the first time in history, the Moon (Apollo 11).
All thanks to science and technology. And if I remember correctly they used Newtonian physics for computing the flight, landing etc. That's quite something to brag about, in my opinion.

What's the big deal then in giving people Nobel Prizes?

In 1956, John Bardeen, Walter Brattain and William Shockley got the Nobel Prize in physics for their work on developing the transistor.

And the thing you are using for posting on this forum (e.g. computer, tablet or mobile phone etc.) are full of - guess what?

Transistors.