I A bit confused about Schrödinger's cat

[AlexCaledin]

I am reading Heisenberg, "Physics and Beyond", - there is a remarkable discussion with Werner's friend Robert, a philosopher; it happened in 1920.

Robert: "I should expect that atoms would, in any case, behave quite differently from the objects of everyday experience. I could imagine that attempts to divide matter even further might lead us to fluctuations and discontinuities from which it would be quite possible to conclude that matter has a grainy structure. But I also believe that the new structures will elude all our attempts to construct tangible images, that they will prove to be abstract expressions of natural laws rather than things."
Werner: "But what if we could see them?"
Robert: "We shall never be able to see atoms themselves, only their effects."
Werner: "That's a poor excuse of an answer. For the same remark applies to things in general. In the case of a cat, too, all you can see is the reflection of light rays, i.e., the effects of the cat, and not the cat itself. And when you stroke its fur, the situation is much the same!"
Robert: "I'm afraid I can't agree with you. I can see a cat directly, for when I look at it, I can—indeed, I must—transform my sense impressions into a coherent idea. In the case of the cat we come face to face with two aspects: an objective and a subjective one— the cat as a thing and as a notion. But atoms are quite a different matter. Here notion and thing can no longer be separated, simply because the atom is neither the one nor the other."

So, a cat is a real part(icipant) of Nature while the atom/particle and QM are merely Nature's answers to Man's questions "what's it made of" and "how to calculate it"...
The Copenhagen QM seems to be the most honest approach because it simply stops where it encounters something not very physical, like that atom-cat difference.

 

[bhobba]

So, a cat is a real part(icipant) of Nature while the atom/particle and QM are merely Nature's answers to Man's questions "what's it made of" and "how to calculate it"...

Since the cat is just as much described by QM as an atom it's as real or not real as an atom.

Be very careful what these early pioneers say - things have moved on a LOT since then.

As Wienberg says - both Bohr an Einstein were wrong (not that I 100% agree with Weinberg - but that is another story - I think if you query him about it its likely more along the lines of the issue I mention later):
http://scitation.aip.org/content/aip/magazine/physicstoday/article/58/11/10.1063/1.2155755

Likely the same with the other pioneers as well - with the probable exception of Dirac.

The real issue with QM, the rock bottom problem is its a theory about observations that appear in a classical world. How does a theory that assumes such a world from the start explain that world.

Thanks
Bill

 

[Flyx]

If you subscribe to the multiverse theory, the cat is dead and alive at the same time, but in different universes. By opening the box, you just see which universe you are in.

 

[Feeble Wonk]

If you subscribe to the multiverse theory, the cat is dead and alive at the same time, but in different universes. By opening the box, you just see which universe you are in.

I understand what you're trying to say here, but you've got to be careful with your semantics.
When considering the cat in the box from the perspective of the "Many Worlds" interpretation, there is only one cat "per world" ( in different universes, to use your words). It's either dead OR alive, because the Geiger counter either triggered the cyanide gas release OR it didn't... though, as you said, the external observer won't know which until the box is opened.

There are other QT interpretations (mostly "collapse" theories) that might describe this situation less definitively, but with MW the cat is never "dead AND alive".

 

[Feeble Wonk]

Since the cat is just as much described by QM as an atom it's as real or not real as an atom.

...as is the box, the person that opens the box and the universe in which the event takes place. (The system being observed, the physicist doing the observing, the apparatus AND the environment in which the observation is made).

 

[Paul Colby]

I guess I don't understand the cat issue either. The state vector applies to a (potentially infinite) ensemble of similarly prepared cats. It's not about a single cat, never was.

 

[Feeble Wonk]

I guess I don't understand the cat issue either. The state vector applies to a (potentially infinite) ensemble of similarly prepared cats. It's not about a single cat, never was.

If your talking about the mathematical formalism, then I suppose that's true... in that the state vector describes the "potential" states of the cat within the isolated system. But, if you're considering what is "really" happening with the cat in the box before an observation is made by opening the box, then that is an interpretational question. As Bill is fond of saying, the formalism is silent with respect to what is happening "between observations".

 

[Paul Colby]

But, if you're considering what is "really" happening with the cat in the box before an observation is made by opening the box, then that is an interpretational question.

I guess you have expectations of what the theory is or "should be" that are not supported by observation. (I've figured out how to do the attribution in the Quote interface, yeah!)​

 

[Feeble Wonk]

I guess you have expectations of what the theory is or "should be" that are not supported by observation.

I'm not quite sure what you mean here. The theory can only tell us what it can tell us, and I have no "expectations" beyond that. The struggle is understanding what it means at the fundamental level. That's where interpretation comes into play.

 

[Paul Colby]

The struggle is understanding what it means at the fundamental level.

Unless your struggle ends in yet a more refined theory which includes thing not qualitatively described by the current theory I have little to say. Interpretation is a pretty wide open endeavor. "God made it that way" is an interpretation, granted not one I'm particularly enamored with. On the other hand, decoherence theory is a theory aimed at addressing the measurement problem using QM. My problem is I don't think there is a problem.

 

[Feeble Wonk]

Unless your struggle ends in yet a more refined theory which includes thing not qualitatively described by the current theory I have little to say.

That is, of course, the hope.

On the other hand, decoherence theory is a theory aimed at addressing the measurement problem using QM. My problem is I don't think there is a problem.

My understanding is that it's not generally accepted within the physics community that decoherence alone resolves the "measurement problem".

 

[Mentz114]

That is, of course, the hope.

My understanding is that it's not generally accepted within the physics community that decoherence alone resolves the "measurement problem".

There is no measurement 'problem' unless you choose an 'interpretation' that carries that baggage. You've succeeded in keeping this post going much longer than is justified.

I suspect you have a bet to see how long you can keep the dead/alive nonsense going

 

[Feeble Wonk]

There is no measurement 'problem'...

Would you say that this is the consensus opinion of physicists in general?

 

[houlahound]

Give my dog a minute with the cat and problem disappears.

 

[Mentz114]

Would you say that this is the consensus opinion of physicists in general?

This is your standard reply. I'm not qualified to comment.

 

[Feeble Wonk]

This is your standard reply. I'm not qualified to comment.

That is my reply. It's a simple question. Would you say that "There is no measurement problem" is the consensus opinion within the physicist community?
If not, then why?

 

[Mentz114]

That is my reply. It's a simple question. Would you say that "There is no measurement problem" is the consensus opinion within the physicist community?
If not, then why?

To find if there is a consensus one needs a poll and some numbers. I'm unable to answer this.

My view of the measurement problem comes from studying textbooks by Ballentine, Peres, Neumaier~Westra, Holland and numerous learned articles.
If you want an argument - take it up with them.

 

[stevendaryl]

Would you say that this is the consensus opinion of physicists in general?

There is no consensus as to whether there is a consensus.

 

[Ken G]

My understanding is that it's not generally accepted within the physics community that decoherence alone resolves the "measurement problem".

The problem is that there is a lot of misunderstanding about what the measurement problem is. One problem is certainly resolved by decoherence-- the problem of how the coherences in the wave function disappear. But all that gives you is what is formally known as a "mixed state." There remains a problem-- so if the problem that remains is what one means by the "measurement problem," then there is still a measurement problem in quantum mechanics.

The problem is this. A mixed state comes up all the time in classical physics-- it's like what happens when you flip a coin and you haven't looked yet. The difference in quantum mechanics is that in classical physics that we are well aware that whatever forces cause the coin to come up heads or tails are simply being overlooked, we are choosing not to include them but we expect they are there-- the mixed state is merely a statistical representation of the actual situation. However, in quantum mechanics, people tend to want to take the formalism more seriously-- the mixed state is supposed to be the projection of the actual state onto the measurement subspace. That's not just a statistical representation of the true state. This is the measurement problem-- how seriously do we take the mixed state that appear after decoherence is all over with. This is what the different interpretations disagree on-- they all agree just fine with the process of decoherence.

 

[StevieTNZ]

However, in quantum mechanics, people tend to want to take the formalism more seriously-- the mixed state is supposed to be the projection of the actual state onto the measurement subspace. That's not just a statistical representation of the true state. This is the measurement problem-- how seriously do we take the mixed state that appear after decoherence is all over with. This is what the different interpretations disagree on-- they all agree just fine with the process of decoherence.

Yes! See pg 209 of 'Quantum Enigma', pg 241 of 'Modern Physics and Ancient Faith'

 

[bhobba]

The problem is that there is a lot of misunderstanding about what the measurement problem is.

IMHO the definitive analysis and definition of the measurement problem is found in my bible on decoherene:
https://www.amazon.com/dp/3540357734/?tag=pfamazon01-20

It has three parts.

1. The problem of a preferred basis. Why for example in the classical world around us do things by an large have a definite position. Decoherence solves that because of an elegant argument that shows it follows from interactions with radial symmetry which most are.
2. The problem of no interference. Why in the world around us we don't generally observe interference terms. Decoherence solves that by converting superpositions to mixed states.
3. The problem of outcomes - ie why do we get outcomes at all. Technically its how an improper mixed state becomes a proper on.e That is the central mystery.

In Schroedinger's cat you can do an analysis with the cat entangled with the radioactive source and show its in a mixed state of alive and dead so can't be in a superposition. Personally I take an even simpler view - the cat has a definite position because its a classical object so never can be in a superposition of alive and dead.

Thanks
Bill

 

[Ken G]

Yes, exactly, many times people are talking about your #1 or #2 when they talk about a measurement problem, and those are not problems in any interpretation-- they're resolved by a more careful application of the decoherence formalism. It is #3 that persists as the actual measurement problem, and the reason it persists is that it is not testable by experiment-- it relates to why we imagine we get the result we get, and all we test is the result we do get.

 

[Paul Colby]

The #3 above I think is a basic misconception. How can QM be applied to an individual measurement? One can't do statistics with one measurement. It's always and for ever a limiting process of measurements.

 

[Ken G]

In Schroedinger's cat you can do an analysis with the cat entangled with the radioactive source and show its in a mixed state of alive and dead so can't be in a superposition. Personally I take an even simpler view - the cat has a definite position because its a classical object so never can be in a superposition of alive and dead.

Yet the irony here is that we generally regard systems as more and more classical the greater their complexity, yet the quantum formalism is supposed to apply to a closed system, which means including everything including the kitchen sink. So I don't think it's the system itself that makes it classical or quantum mechanical, it is the nature of the projection that is necessary in order to talk about that system as an individual entity that decides its classical nature. It certainly requires a very complex projection to talk about a cat as if it was an individual object, when in fact it is strongly entangled with its environment. The entanglements are incoherent, but present. I would define a classical system as one that requires projecting out a complex variety of entanglements, even decohered ones. So the cat is classical, but the cat-and-its-environment is not so clear, and requires an interpretation to say if that is quantum mechanical or classical. The ultimate problem appears when the physicist themself is included in the environment of the cat-- is that whole closed system classical or quantum mechanical? That informs our choice of interpretation, and is a very hard question to answer.

 

[Ken G]

The #3 above I think is a basic misconception. How can QM be applied to an individual measurement? One can't do statistics with one measurement. It's always and for ever a limiting process of measurements.

Yet you are choosing an interpretation to say this. You are focusing on how we test quantum mechanics-- which requires statistical comparisons. So you are taking an empirical stance-- quantum mechanics is a means of making empirically testable predictions, and nothing more. That is what we might regard as a "minimal" view of what science does, the most defensible stance. But many scientists are not content with staying with what they can test-- they want to use science to say what is. That desire motivates less minimal interpretations, where the quantum state is in some sense the actual state of the system. I'm not saying I think that is defensible, merely that many want to think that is the goal of science.

 

[Paul Colby]

When did the Born rule become a matter of interpretation? I must have missed the memo. For me to get beyond the Born rule would take additional science that just isn't evident. Much (all?) of this discussion is generated by what people think science should be and not what it clearly is. Just my opinion but what the hay...

 

[Ken G]

When did the Born rule become a matter of interpretation? I must have missed the memo.

Certainly the Born rule is not a matter of interpretation, which is why every interpretation of quantum mechanics carefully respects the Born rule. Interpretations interpret the Born rule.

For me to get beyond the Born rule would take additional science that just isn't evident.

That's what I mean by a minimal interpretation-- not going past the evidence.

Much (all?) of this discussion is generated by what people think science should be and not what it clearly is. Just my opinion but what the hay...

I would say that's the thinking behind the minimalist interpretations (like Copenhagen, or the ensemble interpretation).

 

[Feeble Wonk]

When did the Born rule become a matter of interpretation? I must have missed the memo. For me to get beyond the Born rule would take additional science that just isn't evident. Much (all?) of this discussion is generated by what people think science should be and not what it clearly is. Just my opinion but what the hay...

I concede that "shut up and calculate" is easier, and more productive for the purposes of "industrial science" because you don't have to get bogged down in questions of "why?". But it is also the legitimate (and more challenging) role of science to explain why.

 

[Paul Colby]

Actually, I don't consider myself in the "shut up and calculate" camp at all, although a little more doing and measuring might teach some. Ask yourself does classical mechanics require interpretation? Classical EM? If you really want to ask why, why is it using real numbers is so effective in classical theories and models? For classical EM people used to go to great lengths to "explain" the fields as some form of mechanical model in some form of medium so they could render it more palatable to their intuition. With the advent of QM the a whole new type of theory was born. Assigning operators to observables and state vectors to ensembles is a very different activity than saying this particle has mass m as a real number. People keep trying to assign state vectors to individual measurements when it's simply a non-sensical thing to do from any "interpretational" point of view.

 

[Feeble Wonk]

People keep trying to assign state vectors to individual measurements when it's simply a non-sensical thing to do from any "interpretational" point of view.

So, is it your contention that "individual" physical events, as described by quantum mechanics, are not fundamentally probabilistic? If so, then aren't you really assuming a hidden variable interpretation while claiming that you aren't using an interpretation?

 

[Paul Colby]

It's not either or situation. QM is not a description of individual events no matter how it's interpreted. As I said before, you can't do statistics on single events. Yes, until QM is superseded (and even when it is because typically newer theories completely encompass the old ones) events are random. People have tried "hidden variables" to "explain" the randomness just like they tried "ether" theories for EM and relativity. These activities have not born fruit. If they do, I'd listen with interest because new physics will be exposed. If they only lead to some involved song and dance which must be performed at each usage of the theory to render it understandable to you, well, not so much.

 

[Paul Colby]

So, is it your contention that "individual" physical events, as described by quantum mechanics, are not fundamentally probabilistic?

Well, it not my contention but rather the way QM is structured as a theory. QM is NOT a description of individual events in the sense you are using.

If so, then aren't you really assuming a hidden variable interpretation while claiming that you aren't using an interpretation?

I don't honestly see this following from anything I've said. You seem to carry a lot of classical baggage, assumption and inferences that are just not supported by the facts. Hence your assertions that I must be assuming events are "not fundamentally probabilistic" or that I'm hiding variables when in fact neither are the case. Yes, I'm a minimalist if labels must be applied. I just don't have the hubris it takes to tell nature how it must behave or be structured.

 

[bahamagreen]

Robert: "I'm afraid I can't agree with you. I can see a cat directly, for when I look at it, I can—indeed, I must—transform my sense impressions into a coherent idea. In the case of the cat we come face to face with two aspects: an objective and a subjective one— the cat as a thing and as a notion. But atoms are quite a different matter. Here notion and thing can no longer be separated, simply because the atom is neither the one nor the other."

Werner's friend Robert was not a sufficient philosopher here, and if he had known physiology he would have known that the critical issue here is that ALL experience is of ultimately just one thing - the firing rates of nerves... that is all the central nervous system ( and brain / mind / consciousness) has to work with - depolarization rates of nerves in the peripheral nervous system, period. There is no sensing of anything "directly"... we only know the boundaries of our nervous system, everything "outside" is an inference. It may seem consistent and somewhat understandable, but it is still all inference, none the less.

In Robert's terms, there are no "things" for the mind, just "notions", all notions, only notions... no "objective" aspect other than an internal subjective inference with respect to other subjective objects.

Cats, dials, light displays, digital counters, tapes, chips, documents, etc. are all subjective inferences based ultimately on firing rates of nerves, as are atoms and their constituents, (and strangely, as are also the result of inference the inferred and alleged central and peripheral nerves themselves... (!)...).

 

[bhobba]

The #3 above I think is a basic misconception. How can QM be applied to an individual measurement? One can't do statistics with one measurement. It's always and for ever a limiting process of measurements.

Of course it can be applied to an individual measurement - you just can't predict the outcome - but it still has an outcome - that is the mystery - why do we get an outcome.

Thanks
Bill

 

[bhobba]

When did the Born rule become a matter of interpretation?

When interpretations like BM and MW came about. You get the same result - but the why is different.

Thanks
Bill

 

[Paul Colby]

Of course it can be applied to an individual measurement - you just can't predict the outcome - but it still has an outcome - that is the mystery - why do we get an outcome.

The only thing QM says about individual measurements is an eigenvalue of the observable being measured will result. As to why one gets an outcome? Do you have a reason question why we wouldn't? Is it possible to formulate your question in terms of QM? I expect this might difficult for all the reasons I've been talking about. The state vector just doesn't have the same connection/relation with phenomena that qualities like mass and position have in classical mechanics and it's reasonable to expect it never will.

 

[AlexCaledin]

... ALL experience is of ultimately just one thing - the firing rates of nerves...

I have browsed http://quantum-mind.co.uk/ a bit and learned that such materialistic theories have to assume that the (remarkable characteristic synchronism of) neurons firing is controlled by the brain's weak EM field...
But, after that, I read in Wikipedia,
"No serious researcher ... believes in an electromagnetic theory of consciousness."
https://en.wikipedia.org/wiki/Electromagnetic_theories_of_consciousness

 

[bhobba]

The only thing QM says about individual measurements is an eigenvalue of the observable being measured will result. As to why one gets an outcome? Do you have a reason question why we wouldn't?

Do you have any reason why you would? Some interpretations do - but the formalism is silent on it. That's why its a mystery. Technically its how an improper mixed state becomes a proper one. If you want to pursue it further best to be specific and examine how an improper state becomes a proper one.

The connection the state vector has with reality (ie if its objective like mass etc or merely phenomenological like probabilities) is interpretation dependant - the formalism is silent on it.

Be very careful about making assumptions not directly implied by the formalism.

Thanks
Bill

 

[bhobba]

learned that such materialistic theories have to assume that the (remarkable characteristic synchronism of)

Its a philosophical minefield and while a legit area of enquiry its not what we discuss here.

For our purposes QM is a theory of observations that appear in an objective common-sense world existing independent of observers. Only very fringe interpretations don't explicitly or implicitly assume that. While they are certainly on topic here, it's very easy to become mired in philosophy discussing them which is not on topic.

Thanks
Bill

 

[stevendaryl]

It's not either or situation. QM is not a description of individual events no matter how it's interpreted.

No, there are certain cases where QM predicts outcomes that are certain, so I don't see any fundamental reason to say that QM isn't about individual events. For example, EPR predicts perfect correlations or anti-correlations for certain measurements.

 

[Paul Colby]

No, there are certain cases where QM predicts outcomes that are certain

This is where I feel the confusion regarding the role of the state vector begins. Having a probability of 1 is a statement a single measurement can't verify any more than it can verify a 0.376 chance. The probabilistic nature of the formalism isn't effected by these limiting cases which are actually not particularly special from an experimental point of view.

 

[bhobba]

Having a probability of 1 is a statement a single measurement can't verify any more than it can verify a 0.376 chance.

The point is its what the theory predicts - verifying its validity is another matter.

My issue is reading more into the formalism than it implies. It is silent on the ontological status of the state.

Thanks
Bill

 

[Tollendal]

Dear Flyx and AlexCaledin,

As I observed in my April 7th commentary, a conscient observer's influence doesn't matter: the coupling with anything does the trick. And multiple universes is a desnecessary idea. Ours is enough!

 

[Ken G]

The point is its what the theory predicts - verifying its validity is another matter.

My issue is reading more into the formalism than it implies. It is silent on the ontological status of the state.

Indeed this was just as true in classical physics as in quantum mechanics, so it is curious in a way that interpretations are such a dicey topic in the latter but not the former. I think it is merely that we all learned our interpretations of classical physics when we were very young, and have kind of forgotten the choices that we had to make. But when you discuss the topic of realism vs. antirealism, which did already come up a bit in the above, that's when you discover that interpretations of classical physics are just as ontologically wide open as interpretations of quantum physics. In short, there is a cat paradox in classical physics too, because of deterministic chaos. There the issue is not whether the cat could be both alive and dead at the same time, it is if you put the cat in the box with the intention of releasing it after 5 minutes, did you decide the fate of the cat when you put it in that box, or did god roll dice over the 5 minute period? Nothing in classical physics answers that, the ontologies there are just as raw as any in quantum mechanics. We have simply stopped asking those questions, so the takeaway message is that ontology is always separate from the formalisms of any physics theory.

 

[Paul Colby]

The point is its what the theory predicts - verifying its validity is another matter.

This may express your point. Mine was more along the line, "The probabilistic nature of the formalism isn't effected by these limiting cases which are actually not particularly special from an experimental point of view."

 

[Gerinski]

Sorry to come into the discussion as I'm not an expert physicist, but I don't find the quantum "weirdness" so weird, at least not when we strip it down to the bones.

The universe demands consistency, that's what I see as the unbreakable rule. Take the infamous double slit experiment, even its modern versions such as the "delayed choice" or the "quantum eraser". If our experimental setup generates and records in any physical way which-path information, we do not need to look at that record, we may send it right away to outer space without looking at it. But the information exists somewhere in the universe, with the potential to influence the future.
If that record will ever, for whatever reason (no consciousness needed!) influence the future, perhaps millions of years from now and millions of light years from earth, the future will be different when the record says that the electron passed through slit A than when it says that it passed through slit B. The future will be contingent on the experiment we performed today in a lab on earth. This means that the present can not consist of a superposition of A and B (an interference pattern), that would mean that present and future are inconsistent. The present experiment outcome, even if we never looked at it, must have been either A or B, but not an interference pattern.

I insist, no consciousness is required at all, any physical interaction with our piece of information in the far future will do, but just to illustrate it better, let's imagine that a civilization in the far future millions of light years away from us does one day find our record and look at it. The future where they look at the record and it says that the electron in our experiment passed through slit A is definitely not the same future where they look and find a record which says that it passed through slit B. So as a demand of consistency, the outcome of our experiment in the present must have been either A or B, but it can not be an interference pattern (a superposition of both A and B).

On the other hand if our experiment does not generate and record which-path information, the future can never be contingent on our experiment. We may as well send the result also to outer space without looking at it. But in this case the future can never depend on the result of our experiment, on whether the electron passed through slit A or B, because the information A or B simply does not exist, our experiment did not generate that information. The future can never depend on a piece of information which does not exist. Any hypothetical civilization finding our record millions of years from now and millions of light years away from earth, will look at the record and find an interference pattern which does not tell them anything regarding whether our electron passed through slit A or B, so that future will be just one and the same one, not 2 possible different futures contingent on the result of our experiment. Since in that case, that future can never be contingent on our experiment, the outcome of our experiment in the present, even when we never looked at it, must have been a superposition of both A and B, i.e. an interference pattern, and that's what the hypothetical future observers will find.

 

[Nugatory]

But the information exists somewhere in the universe, with the potential to influence the future.

You are describing an explanation known as "superdeterminism". If you google for "t'hooft superdeterminism" you will find some of Gerard t' Hooft's writing on the subject (as well as a fair amount of criticism).

You are right that this line of thinking completely eliminates quantum weirdness, and it may be the only way remaining to preserve both locality and counterfactual definiteness. However, superdeterminism brings in its own rather indigestible weirdnesses. For example... Superdeterminism eliminates the quantum weirdness from Bell-type experiments, but at the cost of believing that random atomic decays that happen 4+ years ago (long enough for light to get here from there, so no relativity problems) on Alpha Centauri can influence the random number generator in a lab on Earth so strongly that the generated sequence is not in fact random.

 

[Nugatory]

This thread has met the fate of so many interpretation threads: As the question is not resolvable by experiment or logic, there is no natural way of bringing the discussion to an end... Instead it eventually reaches the point where everything that has been said in the thread has been said before and there's nothing new.

We're there, and the thread is closed.