Which interpretation is your favourite?

[Fredrik]

Ensemble is minimal since it doesn't require one to believe in anything beyond what can be experimentally verified. Instead one has to accept our ignorance - we don't know what exactly happens with single events.

Agreed. Ensemble is at least as simple, but it's an anti-realist interpretation. What I said is that the MWI is the minimal realist interpretation.

1. You say "it [is] the minimal realist interpretation, because it doesn't contain additional axioms which serve no other purpose than to get rid of the many worlds" but this is completely backwards, there is no such thing as "many worlds" beyond MWI interpretation. There is no such thing in the rest of physics, it is precisely MWI that postulates "many worlds."

The MWI is just the assumption that QM, without modifications, tells us what actually happens, and what does QM tell us? It says that a measurement of an observable A changes a pure state $\rho=|\psi\rangle\langle\psi|$ into a mixed state:

$$\rho\rightarrow \sum_i P_i\rho P_i$$

If this is a description of what actually happens, we would need an additional axiom in order to give one of the terms a different meaning than the others. (If you want to argue against that claim, please do it in the other thread).

It's as if I developed an interpretation invoking ghost of ancestors and then claimed that all other interpretations require axioms to get rid of ghost of ancestors while mine doesn't so it is clearly minimal.

If you think so, you haven't understood the MWI at all.

But even Copenhagen interpretation with it's mysterious wavefunction collapse is much better then the mysterious multiverse for which there is no justification whatsoever. I'd rather have one mysterious process then the whole mysterious multiverse.

The assumption of realism can't be justified, but once we have made it, we're stuck with many worlds until we introduce another assumption to get rid of them. The worlds don't need justification. Their elimination does.

To me MWI looks like nothing more then an elaborate rationalization invented in order to save QM from it's problems.

The ensemble interpretation doesn't have any problems. To me the MWI looks like what you're more or less forced to accept if you believe that QM tells us what actually happens. I also think that QM looks like the first thing a mathematician would come up with if asked to find out if it's possible to define a theory of physics that assigns non-trivial probabilities to possible results of experiments. It doesn't look like a description of anything, but it might be.

MWI interpretation requires one to take existence of immense number of unobservable parallel universes on nothing but belief, how is that scientific or minimal? Especially when the second interpretation requires no such thing and gives exactly the same predictions? MWI flies in the face of both scientific skepticism and Ockham razor.

QM describes a single physical system. The states of that system are represented by the unit rays of a Hilbert space. The time evolution of that state is represented by a curve in the Hilbert space. That's it. To identify our world in there, you have to decompose the system into subsystems, and then choose bases for the subsystems. How is it more minimal to do that than to not do that?

Compare e.g. to the amount of information required to specify a single natural number. It can be arbitrarily large. But to specify all of them, you just need to say "Step 1: Start with 0. Step 2: Add 1 to what you've got so far. Step 3: Go back to step 2." More is sometimes less.

I would say that Occam favors the MWI over all the other realist interpretations, since the other ones seem to require additional axioms.

4. How can it not grow? Do some universes get erased or merge?

5. We have one universe before measurement and 2 after if it's not created out of nothing then where did the energy and matter come from for one extra universe?

It's the same matter. The worlds are just correlations between subsystems. At any point on the curve that represents the time evolution of the state of the universe, there's infinitely many decompositions into subsystems, and infinitely many bases to choose from. A "split" between classical worlds is just the observation that a short segment of the curve can be described as an interaction between subsystems that makes their states correlated. (I disagree with that FAQ answer about conservation of energy).

 

[PTM19]

Agreed. Ensemble is at least as simple, but it's an anti-realist interpretation. What I said is that the MWI is the minimal realist interpretation.

I don't understand what you mean by anti-realist, ensemble doesn't have anything to say about realism, why do you think it does? Realism is the belief that objective reality exists independent of measurements, is this the realism you are talking about?

For example if someone were to develop a local realist contextual hidden variable theory which does away with c as maximum signal propagation limit (setting some higher limit) such a theory would be perfectly compatible with ensemble interpretation so this interpretation doesn't rule out realism.

The MWI is just the assumption that QM, without modifications, tells us what actually happens, and what does QM tell us? It says that a measurement of an observable A changes a pure state $\rho=|\psi\rangle\langle\psi|$ into a mixed state:

$$\rho\rightarrow \sum_i P_i\rho P_i$$

If this is a description of what actually happens, we would need an additional axiom in order to give one of the terms a different meaning than the others. (If you want to argue against that claim, please do it in the other thread).

I don't want to argue about additional axiom, I want to argue that what you state doesn't make much sense as a defense of MWI. Here is what one normally should do - one should make a hypothesis, not assumption, that QM without any additional axioms describes what really happens. Then one performs an experiment and sees that the outcome of an observation is not a mixed state therefore the hypothesis is ruled out. End of story.

You on the other hand assume that QM without any additional axioms describes what really happens and then when it doesn't agree with experiments you argue that it is the experiment which is flawed - but for that you have to postulate parallel unobservable universes.

As you yourself stated in another thread the theory is not just mathematics, what is also needed is interpretation which links mathematics with real world. And it is this part that I object to, it's the interpretation of MWI that postulates additional unobservable universes created every second - this is an extreme violation of Ockham's razor since there is absolutely no reason for all those baroque extensions other then to save the assumption which doesn't lead anywhere anyway.

The fact that there is one less axiom is meaningless, the excessive interpretation far outweighs this questionable benefit. Besides you can always reformulate a theory without an axiom if you are willing to expand interpretation, you can probably even create a theory without any axioms at all just a lot of interpretation if you are desperate, that's why the number of axioms is hardly a measure of anything.

One other important thing is that the approach MWI takes to save the above assumption can be used to save any pseudoscience from contradictory experiments, if you can postulate arbitrary unobservable universes any result can be explained.

The assumption of realism can't be justified, but once we have made it, we're stuck with many worlds until we introduce another assumption to get rid of them. The worlds don't need justification. Their elimination does.

Again, what kind of realism are you talking about? It seems to me that by realism you mean the statement that QM math describes real world but why would you call it "realism"? Or why would anyone find this assumption so important that he would rather introduce unobservable parallel universes in order to save it? Why is this assumption so important to you?

1. To me the MWI looks like what you're more or less forced to accept if you believe that QM tells us what actually happens.
2. I also think that QM looks like the first thing a mathematician would come up with if asked to find out if it's possible to define a theory of physics that assigns non-trivial probabilities to possible results of experiments. It doesn't look like a description of anything, but it might be.

QM describes a single physical system. The states of that system are represented by the unit rays of a Hilbert space. The time evolution of that state is represented by a curve in the Hilbert space. That's it. To identify our world in there, you have to decompose the system into subsystems, and then choose bases for the subsystems. How is it more minimal to do that than to not do that?

Compare e.g. to the amount of information required to specify a single natural number. It can be arbitrarily large. But to specify all of them, you just need to say "Step 1: Start with 0. Step 2: Add 1 to what you've got so far. Step 3: Go back to step 2." More is sometimes less.

3. I would say that Occam favors the MWI over all the other realist interpretations, since the other ones seem to require additional axioms.

1. Yes, but to me the fact that you are forced to accept all the excess of MWI should be a proof that QM does not tell us what actually happens

2. From what you say I understand that you like QM and think it is simple, but that is not enough to make it right. Plenty of nice and simple theories have been ruled out already.

3. I am certain Ockham would disagree, the facts are that there are two interpretations - one requires multitude of unobservable universes while the other doesn't yet both agree with experiments. Ockham's Razor is not about axioms: entia non sunt multiplicanda praeter necessitatem, "entities must not be multiplied beyond necessity," how is MWI not in violation of this principle?

It's the same matter. The worlds are just correlations between subsystems. At any point on the curve that represents the time evolution of the state of the universe, there's infinitely many decompositions into subsystems, and infinitely many bases to choose from. A "split" between classical worlds is just the observation that a short segment of the curve can be described as an interaction between subsystems that makes their states correlated. (I disagree with that FAQ answer about conservation of energy).

Worlds are just correlations? Correlation is a concept which only makes sense when applied to two measures of quantities and it is a measure of quantity itself, how can it be a world how can it have material existence? How for example is an electron a correlation and a correlation between what?

 

[WaveJumper]

PTM19, you are raising some very good points that i totally agree with. However, realism in MWI is not the same as the realism found in our observations. I think MWI is about more or less abstract universes, where determinism is preserved and there is only a weak form of realism implied. By weak realism i refer to the idea that there exists some sort of reality outside of measurements, but we are limited in our ability to fully describe it(though we collectively agree on it). Strong realism(the ability to know what the objects of measurements really are) is incompatible with MWI, imo. I fact, don't think strong realism is compatible with QT as it is today, at all.
While i don't agree with the MWI, i think of it as a mere framework model that solves some of the gnarling problems of QM, without further delving into the details of the universe we appear to live in. But on the whole, i agree with you that it seems too extravangant a solution, kind of like killing a mosquito with a nuclear bomb.

 

[Fredrik]

I don't understand what you mean by anti-realist, ensemble doesn't have anything to say about realism, why do you think it does?

It says that neither pure nor mixed states represent the properties of an individual system. That's as anti-realist as anything can get.

Realism is the belief that objective reality exists independent of measurements, is this the realism you are talking about?

No.

I agree that that kind of realism is consistent with the ensemble interpretation.

Again, what kind of realism are you talking about? It seems to me that by realism you mean the statement that QM math describes real world but why would you call it "realism"?

Yes, that's what I mean. It's the term that Isham used in his QM book. Do you have a better one?

Why is this assumption so important to you?

It isn't. It's just interesting. I used to think QM couldn't possibly be a description of anything, not even a fictional universe, and I used to think that the MWI was an ill-defined mess of inconsistent nonsense. Now that I understand the MWI much better than before, I understand that it's neither an ill-defined mess, nor something that can be ruled out. That makes it interesting.

Here is what one normally should do - one should make a hypothesis, not assumption, that QM without any additional axioms describes what really happens. Then one performs an experiment and sees that the outcome of an observation is not a mixed state therefore the hypothesis is ruled out. End of story.

This doesn't make much sense. QM makes the same predictions about results of experiments regardless of whether it also "tells us what actually happens". That doesn't change because you call it a hypothesis instead of an assumption. And if it had, we wouldn't be talking about an interpretation of QM anymore. We would be talking about a similar but different theory.

You on the other hand assume that QM without any additional axioms describes what really happens

Look, either it does or it doesn't. I'm just exploring the possibility that it does. Why does that bother you so much? I'm not used to seeing such an extreme shut-up-and-calculate attitude.

and then when it doesn't agree with experiments you argue that it is the experiment which is flawed

Huh?! Do you think there are experiments that are inconsistent with the MWI?

As you yourself stated in another thread the theory is not just mathematics, what is also needed is interpretation which links mathematics with real world. And it is this part that I object to, it's the interpretation of MWI that postulates additional unobservable universes created every second

Did you mean "it is not this part that I object to"? That would make more sense. The stuff that links the mathematics to results of experiments is a part of the definition of QM. It's not a part of the "interpretations" of QM, which are attempts to interpret QM as a description of what actually happens.

And no new matter gets created in the MWI. That's just a myth that's being spread by people who have misunderstood it. (It's no better than the "First there was nothing. Then it exploded" misrepresentation of the big bang theory).

- this is an extreme violation of Ockham's razor since there is absolutely no reason for all those baroque extensions other then to save the assumption which doesn't lead anywhere anyway.

I don't think you would be saying these things if you understood the MWI. The mathematical model is very simple.

One other important thing is that the approach MWI takes to save the above assumption can be used to save any pseudoscience from contradictory experiments, if you can postulate arbitrary unobservable universes any result can be explained.

The MWI is that assumption (if we're still talking about "QM tells us what actually happens"), and the claim you're making is completely false. The MWI is just an interpretation of QM, so it has nothing to do with crackpot theories that make predictions that disagree with experiments.

1. Yes, but to me the fact that you are forced to accept all the excess of MWI should be a proof that QM does not tell us what actually happens

It's very clear that it isn't.

2. From what you say I understand that you like QM and think it is simple, but that is not enough to make it right.

If you think that I think that QM needs an interpretation, you must have missed the 20 or so posts where I've been saying that it doesn't. (Which is understandable if you're not a regular reader of the QM forum).

3. I am certain Ockham would disagree, the facts are that there are two interpretations - one requires multitude of unobservable universes while the other doesn't yet both agree with experiments. Ockham's Razor is not about axioms: entia non sunt multiplicanda praeter necessitatem, "entities must not be multiplied beyond necessity," how is MWI not in violation of this principle?

Didn't I explain that already?

Worlds are just correlations? Correlation is a concept which only makes sense when applied to two measures of quantities and it is a measure of quantity itself, how can it be a world how can it have material existence? How for example is an electron a correlation and a correlation between what?

Between the states of the subsystems, or equivalently between the "measured" operator on the system and an "einselected" observable on the environment. For example, between the two spin-z eigenstates of a silver atom and two states of the environment that include physicists that remember getting a specific result.

I should add that this is actually just a way to define the interesting worlds, the ones in which the environment can contain stable records (e.g. memories in a physicist's brain) about results of experiments. A "world" (interesting or not) is defined by a basis for the Hilbert space of the universe. The Hilbert space of the universe is the tensor product of the Hilbert spaces of the subsystems, so to specify a basis for the bigger space is to specify a basis for all of the smaller spaces. Decoherence theory tells us that the interaction between a system and its environment will make the density matrix of the universe almost diagonal in a particular basis. So the interaction selects a basis with a special property. This is called "einselection". There are many more worlds than the einselected ones, but decoherence theory tells us that they won't contain stable records of measurement results, and therefore no conscious observers. (Consciousness involves changing states of memory, and a memory is a stable record of a measurement result).

(There's still a lot I don't know about decoherence theory, so I probably won't be able to explain this much better).

 

[PTM19]

Edit: I didn't see your later edits when I replied.

It says that neither pure nor mixed states represent the properties of an individual system. That's as anti-realist as anything can get.

Ok, I now understand what you mean - with ensemble interpretation QM is anti-realist, in that it doesn't describe reality. Yes, I agree with that, the problem is realism to me has a particular meaning - that the physical reality exist independently from observation, so I took your statements to mean that ensemble interpretation somehow precludes the possibility of such realism.

This is roughly what I mean by realism:
http://en.wikipedia.org/wiki/Philosophical_realism#Realism_in_physics

Look, either it does or it doesn't. I'm just exploring the possibility that it does. Why does that bother you so much? I'm not used to seeing such an extreme shut-up-and-calculate attitude.

It doesn't bother me and I am strongly against shut-up-and-calculate attitude, but my goal here is to show the shortcomings of MWI. The main point was that an assumption which does not agree with experiments unless one introduces outlandish postulates to save it should be considered as ruled out.

Huh?! Do you think there are experiments that are inconsistent with the MWI?

Unless one postulates additional unobservable universes experiments are inconsistent with the assumption that "QM tells us what actually happens." (Edit: As you pointed out we should see mixed states but we don't when we do actual measurements)

As I said above to me this is enough to conclude the assumption is wrong.

Imagine for a moment that someone else say a chemist developed a nice hypothesis with beautiful math which explained most but not all experimental results and which could be modified to explain all but at a cost of spoiling the mathematical picture. What would you say if said chemist postulated existence of additional completely unobservable molecules everywhere so that they can provide what is needed to make the hypothesis agree with all experiments?

Or say a doctor developed a very nice symmetry hypothesis which explained the placement of many organs but when confronted with the fact that some organs are not placed symmetrically instead of modifying the hypothesis he insisted on postulating additional invisible organs which perfectly fulfill the symmetry.

Wouldn't you find such attempts at saving the simplicity and beauty of the math involved ridiculous? How do they differ from the MWI case? But please don't just say that I don't understand MWI without pointing out what is it specifically that makes MWI different from those cases.

Did you mean "it is not this part that I object to"? That would make more sense. The stuff that links the mathematics to results of experiments is part of the theory of QM. It's not a part of the "interpretations" of QM, which are attempts to interpret QM as a description of what actually happens.

I certainly don't see the difference, there is only math and interpretation which attempts to link math to what actually happens. "Results of experiments" is "what actually happens."

And no new matter gets created in the MWI. That's just a myth that's being spread by people who have misunderstood it. (It's no better than the "First there was nothing. Then it exploded" misrepresentation of the big bang theory).
I don't think you would be saying these things I you understood the MWI. The mathematical model is very simple.

How am I to understand it if even proponents of MWI cannot agree what it actually means. The math is not a problem - interpretation is.
From what I've read the usual description is that each measurement creates a new branch one universe separates into two, but you seem to believe all possible universes always exist.

Anyway whether they are created or not it is still true that MWI postulates an immense number of unobservable parallel universes.

The MWI is that assumption (if we're still talking about "QM tells us what actually happens"), and the claim you're making is completely false. The MWI is just an interpretation of QM, so it has nothing to do with crackpot theories that make predictions that disagree with experiments.

It has plenty to do with them in that once you admit claiming unobservable parallel universes as an acceptable explanation for contradictory experimental results you can explain pretty much anything. I've personally seen people argue that heaven and hell are not really different from parallel universes postulated by physics. The problem is that there is some truth to such claims.

It's very clear that it isn't.

Well, to me the fact the assumption "QM tells us what actually happens" forces you to accept all the excess of MWI is more then enough proof it is wrong. Obviously to you it isn't. Fine.

Didn't I explain that already?

You explained that the math is simple and works better in MWI - it doesn't require an additional axiom. Ok, but what I am saying is that it's not the math or the axiom that is problematic it's the interpretation postulating additional universes which violates Ockham's razor principle. I don't think you provided a counterargument to that.

You also stated that ensemble interpretation is at least as simple as MWI, that alone should be enough to agree that Ockham's razor favors ensemble interpretation when one considers the later doesn't require parallel universes.

Between the states of the subsystems, or equivalently between the "measured" operator on the system and an "einselected" observable on the environment. For example, between the two spin-z eigenstates of a silver atom and two states of the environment that include physicists that remember getting a specific result.

I have serious doubts if world can be defined as pure correlations - correlations cannot be fundamental since something else has to exist first for correlations to be possible. To me saying that everything is just correlations is like trying to define natural numbers with just addition without any numbers to begin with, but maybe I am wrong here, I don't have enough time to fully think it through.

 

[Fredrik]

(Edit: As you pointed out we should see mixed states but we don't when we do actual measurements)

No, I said that if QM describes what actually happens, the universe will be in a mixed state after a measurement, and none of the terms can be dismissed without reason. But that doesn't mean that we should see mixed states! "We" are states of well-defined memories, and you only have the memory of measuring spin up in the world where the result was "up". See the stuff at the end of my previous post (added in my last edit), but read this correction first: I shouldn't have said that a basis defines a world. I think I should have said that the worlds are the terms in the density matrix, and that every possible way to express the density matrix in terms of a basis is equally valid. So not only are there infinitely many worlds (terms in the density matrix). There are infinitely many ways to describe the universe as consisting of worlds! (One for each basis).

What that stuff at the end of my previous post means is that the basis that decoherence singles out as "special" is precisely the one that describes the universe as consisting of worlds in which the physicist's brain quickly goes into a well-defined state with a memory of only one measurement result. In the MWI, this is a very significant part of the reason why we can't experience superpositions, but it's not the whole story. I think an experience can be described as the formation of a well-defined memory state, and that would imply that your memory is always in a well-defined state of remembering a specific measurement result in the world where that was the result of the measurement.

Edit: See also this post.

Imagine for a moment that someone else say a chemist...

Or say a doctor...

Irrelevant. The MWI is an interpretation of QM, not a new theory, so it doesn't make any predictions that QM doesn't.

I certainly don't see the difference, there is only math and interpretation which attempts to link math to what actually happens. "Results of experiments" is "what actually happens."

The result of the experiment isn't all that actually happens. What about the actual experiment? Didn't that happen? Didn't something happen to the silver atom as it passed through the inhomogeneous magnetic field of the Stern-Gerlach apparatus?

The difference between the two types of interpretations is huge. The first kind is what turns a mathematical model into a theory. (I define a "theory" as "a set of statements that associates a probability with each possible result of each experiment in some set of experiments". Note that a theory doesn't have to describe any aspect of reality between state preparation and measurement, not even approximately. It just has to make predictions about results of experiments). The second kind of interpretation turns the theory into a (possibly completely incorrect) description of what actually happens. The first kind of interpretation is a part of the definition of QM, but an "interpretation of QM" is always of the second kind.

How am I to understand it if even proponents of MWI cannot agree what it actually means.

I agree that it's difficult. I haven't found a definition of the MWI that really makes sense in any of the articles I've read (admittedly not that many), and I certainly haven't been able to extract one from the MWI proponents here. There's also a lot of really bad information about it on the internet.

Well, to me the fact the assumption "QM tells us what actually happens" forces you to accept all the excess of MWI is more then enough proof it is wrong. Obviously to you it isn't. Fine.

Do you often consider something proved wrong the moment you discover that it has features that rubs your emotions the wrong way?

You also stated that ensemble interpretation is at least as simple as MWI, that alone should be enough to agree that Ockham's razor favors ensemble interpretation when one considers the later doesn't require parallel universes.

I'm undecided on that, since the ensemble interpretation strongly suggests that there's an underlying hidden-variable theory, with variables weird enough to avoid the conclusion of Bell's theorem.

I have serious doubts if world can be defined as pure correlations - correlations cannot be fundamental since something else has to exist first for correlations to be possible. To me saying that everything is just correlations is like trying to define natural numbers with just addition without any numbers to begin with, but maybe I am wrong here, I don't have enough time to fully think it through.

When you have some time to spare, you might want to check out David Mermin's "Ithaca" interpretation. That one is really "correlations without correlata".

http://arxiv.org/abs/quant-ph/9801057
http://arxiv.org/abs/quant-ph/9609013

 

[SW VandeCarr]

Yes, but it is not an interpretation. Those who claim that they use only 'Shut up and calculate' are not fair enough - they are using the interpretational things (Born rule for example) to map the number they get into what they observe. When they get 0.5498585 as a result they can say only 'I get 0.549885 after my calculations'. When they say 'I get 0.549885 and hence I expect blah blah blah they DO use interpretation, they just don't admit it.

Do you need an interpretation to do the science? You do the calculation and get a result. The result agrees with experiments. What else to you need?

 

[Dmitry67]

The result NEVER not agrees with the experiment without an interpretation! There are just 2 different things from 2 different worlds: numbers and physical objects. You need an interpretation to map them (and only MUH-compatible TOE will be interpretation-free, because there will be numbers of both sides)

 

[SW VandeCarr]

The result NEVER not agrees with the experiment without an interpretation! There are just 2 different things from 2 different worlds: numbers and physical objects. You need an interpretation to map them (and only MUH-compatible TOE will be interpretation-free, because there will be numbers of both sides)

QM is science. Interpretations of QM are philosophy (not that there's anything wrong with that). There's a plethora of interpretations out there which indicates we have no idea of what quantum "reality" is. However, you can still do the science and I disagree with the idea that you need to choose an interpretation to design experiments. If you believe MWI is "correct", how does that affect the kinds of questions you ask or the way you design the experiments?

EDIT: Ok, I grant you "shut up and calculate" can be construed to be an interpretation, but it makes no claims about what quantum reality "really" is.

 

[Dunebug7]

Personally as I said I accept ensemble interpretation and that we don't know what happens in the case of individual events - we can only make predictions about an ensemble.

But I don't see how decoherence can solve the problems of MWI I mentioned if that answers your questions.

I find decoherence troubling even in MWI.

It seems a little well you know devoid of being able to be proved. I'm not sure natural law gives a toss about our subjective biases anyway. If a woods falls in the trees and there's no one around to see it what colour is it?

I voted for LQG. Btw but what do I know?

 

[Dunebug7]

QM is science. Interpretations of QM are philosophy (not that there's anything wrong with that). There's a plethora of interpretations out there which indicates we have no idea of what quantum "reality" is. However, you can still do the science and I disagree with the idea that you need to choose an interpretation to design experiments. If you believe MWI is "correct", how does that affect the kinds of question you ask or the way you design the experiment?

That is very true. However Copenhagen interpretation I suppose at least has an experiment or two like Bell's and the two slit even if it is depressingly probabilistic; and tends to throw up more questions than answers. It's all a matter of taste though let's face it.

I suppose you are saying shut up and calculate?

________________

"If a woods falls in the trees and there's no one around to see it what colour is it?"

Well really!

 

[Fredrik]

Dmitry67:

You need to distinguish between the axioms of a theory of physics, which "interpret" the mathematics as predictions about results of experiments, and the kind of "interpretations" discussed in these threads. When we say "quantum mechanics", we're referring to a theory of physics, not to an abstract mathematical model, so the kind of interpretation you're talking about now is already included in the definition of "QM". An "interpretation of QM" is something entirely different.

It's an interpretation on top of the one that we use to make predictions about results of experiments. It's an additional set of axioms about relationships between things in the model and things in the real world, which turns the theory into a "description of what actually happens". The description is either approximately correct or completely incorrect, and there's no way to know which.

Every interpretation has its own answer to the question of whether QM is just an algorithm that tells us how to calculate probabilities of possibilities, or also an approximate description of the real world?. "Shut up and calculate" is the idea that the answer is irrelevant. "The ensemble interpretation" is the idea that the first option is correct. All other interpretations are speculations about how the second option might be correct.

 

[Jonathan Scott]

I think that the transactional interpretation is close, but I've found at least one glitch in it as currently defined, which is that if you have two entangled events with a lightlike separation, then in theory the observation you choose to make on the second one could be determined by the result of the first, making it impossible to complete the transaction until after it has completed. This is fairly easy to fix by adding additional rules, since sub-light-speed communication doesn't require any magic in this case, but I don't think nature should need that extra complexity.

My current preferred interpretation is that quantum waves are deterministic but involve two factors - a the usual structured part which gives rise to spinor and vector factors and the primary characteristics of interactions, which propagates at c, and some additional factor associated with relative phase and entanglement which propagates a transactional part in a space-like way, effectively much faster than c. As what is propagated is purely relative, it doesn't have a direction; it is just a mathematical agreement between the ends.

Yes, this is quite vague, but so is the transactional interpretation.

 

[Dmitry67]

But TI has the same issues with the collapse as CI

 

[disregardthat]

But TI has the same issues with the collapse as CI

What exactly are these issues?

 

[Dmitry67]

Measurement device in CI is as badly defined as an 'absorber' in TI. For example, light in glass is moving slower then in vacuum because it is absorbed and re-emitted by atoms with some delay. However, in TI glass, air, mirrors, lens are not 'absorbers' while retina or photocamera is an absorber. What is a difference? There is always a measurement device at the end of 'transaction' in TI.

So TI does not bring anything new to the measurement problem in comparison with CI. It just buys realism in Bells experiments at the price of backward causality.

So the issue is, how can you
1. define what an absorber is and how it is different from non-absorber;
2. can QM be applied to absorber? if yes, then if will not end a transaction (as all elementary QM interactions are T-symmetric), If no, then explain the pure magic associated with an absorber.

 

[sas3]

All right, Shut up and calculate is in the lead.

 

[disregardthat]

Measurement device in CI is as badly defined as an 'absorber' in TI. For example, light in glass is moving slower then in vacuum because it is absorbed and re-emitted by atoms with some delay. However, in TI glass, air, mirrors, lens are not 'absorbers' while retina or photocamera is an absorber. What is a difference? There is always a measurement device at the end of 'transaction' in TI.

So TI does not bring anything new to the measurement problem in comparison with CI. It just buys realism in Bells experiments at the price of backward causality.

So the issue is, how can you
1. define what an absorber is and how it is different from non-absorber;
2. can QM be applied to absorber? if yes, then if will not end a transaction (as all elementary QM interactions are T-symmetric), If no, then explain the pure magic associated with an absorber.

So the problem is that it is not a realist view, because it does not treat the observer as an object?

 

[Jonathan Scott]

QM implies that there are spacelike connections between entangled events, but that those connections are relative, so they cannot be used to send information.

I think the simplest explanation is that those connections are physical; although information can only propagate at c, the two-state phase quantities involved in entanglement literally have a form of spacelike connection such that the events at either end which resolve the entanglement are forced to be in a specific phase relationship. It doesn't matter which end you consider is resolved "first" as the relationship is relative, so there is no causality problem.

The mechanism of this spacelike connection is at present unspecified, but that doesn't affect the general nature of the interpretation.

Is this an established interpretation, and if so is there a name for it? Are there any obvious problems with it (apart from the mechanism for spacelike connection)?

(As I mentioned before, it is loosely related to transactional, but it involves a spacelike connection rather than connections going back and forward in time).

 

[Dmitry67]

So the problem is that it is not a realist view, because it does not treat the observer as an object?

Yes, an observer (or measurement device) can not be described using QM. Because if we try to apply Qm rules to systems of atoms called 'measurement devices', they don't collapse the wavefunction. All collapse interpretations associate some magic property with some configurations of atoms.

 

[Astepintime]

Irrelevant. The MWI is an interpretation of QM, not a new theory, so it doesn't make any predictions that QM doesn't.

Nice discussion guys, but this statement reminds me of why I dropped these discussions twenty years ago! It gets you nowhere! If there are no differences in the predictions then just pick the one you like and be done with it. Until an experiment successfully differentiates between the interpretations it seems that it is just religion or worse metaphysics.

 

[OmCheeto]

String Theory.

But only because I play the violin.

I don't really know anything about any of these theories.

 

[disregardthat]

Yes, an observer (or measurement device) can not be described using QM. Because if we try to apply Qm rules to systems of atoms called 'measurement devices', they don't collapse the wavefunction. All collapse interpretations associate some magic property with some configurations of atoms.

What exactly are these "magic properties" you associate with the collapse phenomena, and why can they not be used?

CI appeals to the non-realist, and one can find it consistent with this view.

 

[PTM19]

No, I said that if QM describes what actually happens, the universe will be in a mixed state after a measurement, and none of the terms can be dismissed without reason. But that doesn't mean that we should see mixed states! "We" are states of well-defined memories, and you only have the memory of measuring spin up in the world where the result was "up". See the stuff at the end of my previous post (added in my last edit), but read this correction first: I shouldn't have said that a basis defines a world. I think I should have said that the worlds are the terms in the density matrix, and that every possible way to express the density matrix in terms of a basis is equally valid. So not only are there infinitely many worlds (terms in the density matrix). There are infinitely many ways to describe the universe as consisting of worlds! (One for each basis).

What that stuff at the end of my previous post means is that the basis that decoherence singles out as "special" is precisely the one that describes the universe as consisting of worlds in which the physicist's brain quickly goes into a well-defined state with a memory of only one measurement result. In the MWI, this is a very significant part of the reason why we can't experience superpositions, but it's not the whole story. I think an experience can be described as the formation of a well-defined memory state, and that would imply that your memory is always in a well-defined state of remembering a specific measurement result in the world where that was the result of the measurement.

Edit: See also this post.

I think I get what you mean, but I don't think you understood my criticism since your objection already assumes many worlds.

I'll try to be more explicit. Imagine the time before MWI was developed - the time when there was no such concept as "many worlds" in physics. If someone were to perform the experiment back then the fact that mixed states were not observed should lead him to conclude that QM without collapse does not agree with experimental results.

This is what I meant and I argue that it would have been the right conclusion and that it is not a sound approach to physics to invent unobservable universes to circumvent contradictory experimental results even if that makes the underlying mathematics much more consistent and appealing.

Irrelevant. The MWI is an interpretation of QM, not a new theory, so it doesn't make any predictions that QM doesn't.

It looks like you dismissed my examples without even considering them, the situation in both cases is analogous to MWI. I'll first quote those examples to make it easier to refer to them and then explain one in more detail.

Imagine for a moment that someone else say a chemist developed a nice hypothesis with beautiful math which explained most but not all experimental results and which could be modified to explain all but at a cost of spoiling the mathematical picture. What would you say if said chemist postulated existence of additional completely unobservable molecules everywhere so that they can provide what is needed to make the hypothesis agree with all experiments?

Or say a doctor developed a very nice symmetry hypothesis which explained the placement of many organs but when confronted with the fact that some organs are not placed symmetrically instead of modifying the hypothesis he insisted on postulating additional invisible organs which perfectly fulfill the symmetry.

I argue that both examples are exactly analogous to MWI. Let's focus on the doctor, I will modify it a bit to further stress the analogy. So, let's say that over the years doctors developed a symmetry theory which correctly explains placement of most organs - this is analogous to QM (both are theories whose predictions agree with experiments in most cases). There is however a problem, some organs like hearth are not placed symmetrically - this is an analogy to measurement problem. One way to deal with wrong placement problem is to amend the symmetry theory but that will spoil the mathematical beauty and consistency by ruining the perfect symmetry - this solution would be analogous to adding wavefunction collapse. Now however one doctor develops a radical solution - he postulates existence of additional unobservable organs which are placed in such a way that symmetry is preserved - for example he claims we all have an unobservable second hearth on the right side of the chest - this novel interpretation of symmetry theory of organ placement is analogous to MWI. Notice that additional postulated organs are unobservable and therefore the theory makes no new experimental predictions - exactly like MWI. Also I want to stress that just as it feels unnatural to add collapse which "artificially" selects one state over the other it is also unnatural to break symmetry and "artificially" place hearth on the left side of the body when the right side would have been just as good from mathematical point of view.

I consider the above example to be a perfect analogy to MWI, the case of a chemist can also be made into a perfect analogy, what's more I believe that by postulating additional unobservable entities almost any problem can be "solved" in similar manner. To me this is not science unless there is a way to experimentally verify such "solutions."

Now, I suspect you would find a proposition that we all have unobservable additional organs, which preserve body symmetry, ridiculous and this is precisely how I see MWI and for exactly the same reasons.

The result of the experiment isn't all that actually happens. What about the actual experiment? Didn't that happen? Didn't something happen to the silver atom as it passed through the inhomogeneous magnetic field of the Stern-Gerlach apparatus?

The difference between the two types of interpretations is huge. The first kind is what turns a mathematical model into a theory. (I define a "theory" as "a set of statements that associates a probability with each possible result of each experiment in some set of experiments". Note that a theory doesn't have to describe any aspect of reality between state preparation and measurement, not even approximately. It just has to make predictions about results of experiments). The second kind of interpretation turns the theory into a (possibly completely incorrect) description of what actually happens. The first kind of interpretation is a part of the definition of QM, but an "interpretation of QM" is always of the second kind.

Ok, I know what you mean, so to clarify I object to the second kind of interpretation, specifically the part which invokes unobservable parallel universes.

Do you often consider something proved wrong the moment you discover that it has features that rubs your emotions the wrong way?

No, I tried to explain my reasons, they don't involve emotions, it's more a matter of skepticism and conviction that ideas which are not falsifiable are not science. Besides extraordinary claims require extraordinary evidence.

I'm undecided on that, since the ensemble interpretation strongly suggests that there's an underlying hidden-variable theory, with variables weird enough to avoid the conclusion of Bell's theorem.

Yes, it does leave room for a better theory but any such theory would not be part of ensemble interpretation and therefore should not affect whether ensemble is preferred based on Ockham's Razor principle.

When you have some time to spare, you might want to check out David Mermin's "Ithaca" interpretation. That one is really "correlations without correlata".

http://arxiv.org/abs/quant-ph/9801057
http://arxiv.org/abs/quant-ph/9609013

Ok, I'll have a look, thanks for the links.

 

[Fredrik]

I think I get what you mean, but I don't think you understood my criticism since your objection already assumes many worlds.

You claimed that the MWI predicts that we should see many worlds. I'm saying it doesn't. The only way to argue for either of these two positions is to examine what the MWI actually says.

Imagine the time before MWI was developed - the time when there was no such concept as "many worlds" in physics. If someone were to perform the experiment back then the fact that mixed states were not observed should lead him to conclude that QM without collapse does not agree with experimental results.

I don't know how you define this "QM without collapse", but if it "doesn't agree with experimental results" then it doesn't make the same predictions as QM, which means that it isn't the MWI. So your objection against the MWI appears to be that QM can be modified to something undefined that makes incorrect predictions. That obviously doesn't qualify as an argument against the MWI.

I argue that both examples are exactly analogous to MWI

And I argue that they're not. To prove that conclusively, I'd have to spend a lot more time extracting information from you about your fictional theory of organ symmetry than I'm willing to do, so I'll just point out one obvious difference: The fictional theory doesn't explain why the unobservable organs are unobservable.

I consider the above example to be a perfect analogy to MWI, the case of a chemist can also be made into a perfect analogy, what's more I believe that by postulating additional unobservable entities almost any problem can be "solved" in similar manner. To me this is not science unless there is a way to experimentally verify such "solutions."

It's possible that it was originally intended to be an explanation of why QM is a probabilistic theory (I haven't read Everett), but it certainly isn't such an explanation. It doesn't solve any problems. It's just the minimal realistic interpretation of QM.

Ok, I know what you mean, so to clarify I object to the second kind of interpretation, specifically the part which invokes unobservable parallel universes.

If we reject all interpretations of the second kind, we can't ever claim that a theory describes reality. We can't even say that "the Earth is spherical" is an approximately correct statement. "The Earth is spherical" would be nothing more than a meaningless string of text that we can use to obtain predictions according to a specific set of rules.

In this particular case, we all feel that it's obvious that the theory not only makes accurate predictions, but also describes reality. In other words, we all feel that it's obvious that the simplest possible interpretation (of the second kind) is correct. (Approximately, of course). Most of us also think that it makes sense to say that Newtonian mechanics describes reality. We can think of it as an approximate description of (some aspect of) the universe we live in, or as an exact description of a fictional universe, but it's certainly a description of something.

Things get a bit more weird when we start talking about special and general relativity. There's a theory involving an aether and a preferred reference frame that makes the same predictions as SR. We can think of these as two different but equivalent theories, or as two formulations of the same theory. They make the same predictions, but describe things differently. There's also a version of GR that treats spacetime as flat, and measurement devices as being distorted by influence from matter. This theory makes the same predictions as GR, but describes things differently.

At this point in time, no one expected that we would ever have to settle for less. Everyone just assumed that a good theory had to be an approximate description of reality. They would even define the word "theory" to mean exactly that. People like Bohr, Heisenberg and Schrödinger, certaintly weren't looking for an algorithm that tells us how to calculate probabilities of possibilities. They were looking for a description of what actually happens at the atomic level.

It must have been quite a shock for the physicists of the time to find a theory that's so weird it doesn't have an obvious interpretation of the second kind. Physicists have always been trying to find the truth about what things really are like, and without an interpretation of the second kind, one can question whether the theory has improved our understanding of how the world works at all! Imagine that the theory "the world is spherical" had been stated in the form of some text that you can't even understand, along with instructions about how to use that text to find predictions. This is how QM must have seemed to the physicists at the time.

Note that a theory that assigns non-trivial probabilities (not always 0 or 1) to possible results of experiments doesn't even satisfy the condition of falsifiability. If the probability of getting a specific result is 99.9999%, you can still get a different result a million times or more. So to bring statistical, non-descriptive theories like QM into the world of science, we even had to drop the requirement of falsifiability from the definition of science, and replace it with a related concept which I think of as "statistical falsifiability". (I think it's obvious what I mean by that, so I won't explain it). Can you imagine changing the definition of science to get a new theory to be considered science? This what dowsers, homeopaths and remote viewers occasionally try to get us to do. (I'm sure that scientists had their share of encounters with those fools back then too).

With all that in mind, it's certainly not surprising that people have been trying to find ways to interpret QM as a description of something. Some people think that it's just a waste of time because experiments can't distinguish between interpretations. "It isn't science". I think this is a bizarre attitude (even though the "it isn't science" bit is technically correct). If we find a bunch of logically consistent interpretations, there's at least a chance that we have found the correct description of the universe. And even if they're all wrong, they might still improve our understanding of the theory, and therefore our understanding of the real world. Even an incorrect description might make it easier to for us to solve problems in QM, by giving us a way to visualize what's actually happening in a fictional universe where experiments have the same results as in ours. There's also the possibility that one of the interpretations will be easier to modify into a new theory than the others, and if that's the case, working on that interpretation has been science all along. It just wasn't possible to see it before the job was done. I don't think many of those people would have that attitude if they understood the things I'm talking about in this post.

There are also people who think that it would be a waste of time to try to interpret QM, because QM can't be interpreted. This I can understand. I was in that camp myself until I understood the MWI well enough to see that it isn't as crazy as it seems. I still don't think that QM can tell us what actually happens in our universe, but I find it fascinating that it might, and even if I knew that it doesn't, I would still be interested in finding out if it can describe anything at all (a possible universe).

There are two more things that I find really annoying when I read what other people are saying about interpretations. The first is the belief that QM obviously can be interpreted. This is just wishful thinking. Mermin displays this belief in the articles I linked to. Penrose does it indirectly in his books by repeatedly claiming that the measurement problem is a huge problem in physics. (It isn't a problem at all for the ensemble interpretation, but it's a problem for some realistic interpretations). The second is the attitude that interpretations don't need proper definitions. It's hard to even find two people who have the same thing in mind when they mention the name of a specific interpretation.

 

[RUTA]

I don't worry about the multiplicity of interpretations for QM, under determination is unavoidable in physics as Fredrik pointed out about SR and GR. I'll happily live with such multiplicity as long as each ontology is consistent with all of our theories of physics. We don't have that situation now because violations of Bell's inequality imply causal and/or constitutive non-locality while GR is local on both counts. This incongruity prompted Smolin to write (The Trouble with Physics, 2006, p 9), "This is probably the most serious problem facing modern science," and (p 10), “The problem of quantum mechanics is unlikely to be solved in isolation; instead, the solution will probably emerge as we make progress on the greater effort to unify physics.” That remains to be seen of course, but those of us in foundations who subscribe to this attitude are hoping to find clues to unification by considering various interpretations/ontologies for QM. The study of QM interpretations serves as a basis for the study of unification in that sense.

 

[PTM19]

Fredrik, I really appreciate your feedback and I understand your position but I don't think we can get much further with this discussion. I think my position should be more or less clear by now, I simply don't find the arguments for MWI convincing for reasons I already stated. To me the doctor analogy holds, you claim it doesn't say why organs are unobservable - does MWI say why parallel universes are unobservable? In any case organs can be in a parallel universe.

I understand that some may prefer to believe QM does indeed describe reality no matter where it leads them in hope they can learn something about reality from interpretations like MWI. Personally however I believe we have a much better chance to understand reality if we accept ensemble interpretation and the fact that QM does not describe reality and instead concentrate on searching for an underlying hidden variable theory which does. Unfortunately it is my impression that not many people are going that route and I suspect this may be the reason why there has been hardly any progress in physics lately.

 

[Fredrik]

does MWI say why parallel universes are unobservable?

Yes, the version of the MWI that I'm talking about does that. (I tried to explain how). But it's hard to find two people who mean the same thing when they say "MWI", so you won't have any problems finding a version that doesn't.

 

[rewebster]

Fredrik, I really appreciate your feedback and I understand your position but I don't think we can get much further with this discussion. I think my position should be more or less clear by now, I simply don't find the arguments for MWI convincing for reasons I already stated. To me the doctor analogy holds, you claim it doesn't say why organs are unobservable - does MWI say why parallel universes are unobservable? In any case organs can be in a parallel universe.

I understand that some may prefer to believe QM does indeed describe reality no matter where it leads them in hope they can learn something about reality from interpretations like MWI. Personally however I believe we have a much better chance to understand reality if we accept ensemble interpretation and the fact that QM does not describe reality and instead concentrate on searching for an underlying hidden variable theory which does. Unfortunately it is my impression that not many people are going that route and I suspect this may be the reason why there has been hardly any progress in physics lately.

I agree

 

[RUTA]

I understand that some may prefer to believe QM does indeed describe reality no matter where it leads them in hope they can learn something about reality from interpretations like MWI. Personally however I believe we have a much better chance to understand reality if we accept ensemble interpretation and the fact that QM does not describe reality and instead concentrate on searching for an underlying hidden variable theory which does. Unfortunately it is my impression that not many people are going that route and I suspect this may be the reason why there has been hardly any progress in physics lately.

Check out where the interpretation called Relational Blockworld (published in 2008 in Foundations of Physics and Studies in History & Philosophy of Modern Physics) suggests physics should go (http://xxx.lanl.gov/abs/0908.4348). That interpretation suggests a definite path for unification and quantum gravity based on constitutive non-locality as the correction for GR. We're trying to solve the equations for (discrete) tensor CFT now to see precisely where the theory differs from GR. This is an example of how the study of QM interpretations can lead to new ideas for physics.

 

[rewebster]

Check out where the interpretation called Relational Blockworld (published in 2008 in Foundations of Physics and Studies in History & Philosophy of Modern Physics) suggests physics should go (http://xxx.lanl.gov/abs/0908.4348). That interpretation suggests a definite path for unification and quantum gravity based on constitutive non-locality as the correction for GR. We're trying to solve the equations for (discrete) tensor CFT now to see precisely where the theory differs from GR. This is an example of how the study of QM interpretations can lead to new ideas for physics.

I just read the thread:

"The Fatal Flaw in Every Techno Show on TV

Let's Enhance!"

https://www.physicsforums.com/showthread.php?t=367648

for some reason, I got the same feeling when reading this post

"Let's Enhance!"

 

[RUTA]

I just read the thread:

"The Fatal Flaw in Every Techno Show on TV

Let's Enhance!"

https://www.physicsforums.com/showthread.php?t=367648

for some reason, I got the same feeling when reading this post

"Let's Enhance!"

We're still waiting for the referee reports, but if you've found a "fatal flaw" in the paper, let us know and we'll withdraw it.