The Many-Worlds Interpretation of QM

[Quotidian]

Greetings all. I'm a new poster here but have spent some time on philosophy forums previously. I subscribe to New Scientist and try and stay reasonably current with science, on the popular level at least.

This week they have had a feature about physical cosmologies and 'theories of everything' and so on.. They mentioned the Everett 'Many Worlds Interpretation' of quantum physics. I did a bit of reading on it, and was interested to find out that many working physicists regard it as the most plausible way of interpreting the results of quantum physics experiments.

The thing is, it seems completely outlandish to the layman. The idea that reality itself keeps 'splitting' into uncountable versions just seems extraordinarily far-out - at least to me. Yet this seems to be really what it is saying. It is called, after all, 'Many Worlds'.

So the question I have is, what is the problem that this is the solution for? Or, to put it another way, if for some reason, it was declared that the 'Everett Many-Worlds Intepretation' was really untenable, then what would it force those who favour it to accept instead? They would be forced to admit that:

 

[Nugatory]

They mentioned the Everett 'Many Worlds Interpretation' of quantum physics. I did a bit of reading on it, and was interested to find out that many working physicists regard it as the most plausible way of interpreting the results of quantum physics experiments.

Selection bias favors MWI in the press because it's cool-sounding and fun to write about.

Or, to put it another way, if for some reason, it was declared that the 'Everett Many-Worlds Intepretation' was really untenable, then what would it force those who favour it to accept instead? They would be forced to admit that:

Nothing. No problem at all. There are several workable interpretations of QM, they cannot be distinguished experimentally, and you end up doing the same math with all of them.

 

[kith]

So the question I have is, what is the problem that this is the solution for?

The problem is called the "measurement problem". On the one hand, QM has a deterministic law according to which systems evolve as long as no measurements are performed (it's called the Schrödinger equation). But when we perform a measurement, we don't get results according to it. So orthodox QM postulates a set of (indeterministic) rules how to get results which are in agreement with experiments. Part of these rules is the so-called wavefunction collapse which singles out one actual outcome from a number of possible ones. This works incredibly well but is conceptionally unsatisfying for many people.

Everett's initial motivation to question the orthodox view was this: consider you have an observer A and a quantum mechanical system S. When A performs a measurement on S, indeterministic wavefunction collapse happens. But what if we add another observer B? From the point of B, the combined system A+S should undergo deterministic evolution until B himself performs a measurement. This leads to contradictory results for A and B.

Orthodox QM solves this by taking the viewpoint of A as the right one. It simply postulates that A himself must not be described as a quantum system. So QM is not a more fundamental theory which underlies classical mechanics, but only a theory of things which are small in a certain sense. For big things, we have to use classical mechanics.

Everett takes the viewpoint of B. From there, all possible outcomes of A's measurement are present in the calculations. There's nothing what singles out one outcome, so we can interpret each of them as belonging to a different world. Everett and his successors also claim that they don't need to postulate additional rules to connect the theory with measurements, but can derive the rules of orthodox QM. If this is true, the Many Worlds interpretation would also be conceptionally simpler as the orthodox view. However, there is no consensus if it can be done.

 

[Khashishi]

Physics is all about creating models for nature, and I wouldn't put much faith into any portion of a model which is not measurable. I say that two models that give the same predictions for measurements are actually the same theory. The non-measurable parts are just tools used to connect our understanding together, but probably have nothing to do with the real world. So we are free to pick the simpler model. Not because it is more correct, but just to make it easier on us.

(When different models give DIFFERENT predictions, then we still want to pick the simpler model consistent with measurements, but for a different reason.)

 

[kith]

I think the physical relevance of different interpretations is that they may a) lead to a better understanding of the theory and it's implications and b) suggest different ways to approach the known problems of the theory.

 

[wuliheron]

So the question I have is, what is the problem that this is the solution for? Or, to put it another way, if for some reason, it was declared that the 'Everett Many-Worlds Intepretation' was really untenable, then what would it force those who favour it to accept instead? They would be forced to admit that:

The MWI provides an explanation while, understandably for physicists, the more dreaded possibility is that there is no explanation for indeterminacy. As bizarrely interesting a subject it may be to explore the possibility that with a few possible exceptions it might be ultimately ineffable just gets under a lot of scientists' skins. You know how it goes.

 

[Nugatory]

The MWI provides an explanation while, understandably for physicists, the more dreaded possibility is that there is no explanation for indeterminacy. As bizarrely interesting a subject it may be to explore the possibility that with a few possible exceptions it might be ultimately ineffable just gets under a lot of scientists' skins. You know how it goes.

Mmmm... If that were what was going on, I'd expect that the statistical ensemble interpretations would also be upsetting a lot of people. I'm not seeing that happening.

 

[wuliheron]

Mmmm... If that were what was going on, I'd expect that the statistical ensemble interpretations would also be upsetting a lot of people. I'm not seeing that happening.

I thought the term statistical explanation was an oxymoron.

 

[Nugatory]

I thought the term statistical explanation was an oxymoron.

Fair enough, but there is a perfectly respectable statistical interpretation of quantum mechanics: http://www.kevinaylward.co.uk/mwg-internal/de5fs23hu73ds/progress?id=N9yFh2Ri5s

De interpretationes non disputandum est...

 

[wuliheron]

Fair enough, but there is a perfectly respectable statistical interpretation of quantum mechanics: http://www.kevinaylward.co.uk/mwg-internal/de5fs23hu73ds/progress?id=N9yFh2Ri5s

De interpretationes non disputandum est...

Yeah, I've seen the 404 error page before and wondered what it signified.

 

[Quotidian]

The MWI provides an explanation while, understandably for physicists, the more dreaded possibility is that there is no explanation for indeterminacy.

So you're saying that it's an alternative to admitting that indeterminacy - I take it this is another expression for the manifestation of 'uncertainty' - might actually indicate a terminus to the limits of knowledge?

That seems fair.

if for some reason, it was declared that the 'Everett Many-Worlds Intepretation' was really untenable, then what would it force those who favour it to accept instead?

Nothing. No problem at all

Sorry, but I'm really not buying that. I don't think Hugh Everett created his theory on a whim, and I don't think he defended it for no reason. I think that he believed it solved a problem, and, given the outlandishness of his solution, the problem must be a mighty big one.

Anyway, thanks for your replies all. I am not going to stick around. So long.

 

[wuliheron]

So you're saying that it's an alternative to admitting that indeterminacy - I take it this is another expression for the manifestation of 'uncertainty' - might actually indicate a terminus to the limits of knowledge?

That seems fair.

Yeah, the experimental evidence in recent years for fundamental research into indeterminacy has been exciting to say the least. Some had held out hope that entanglement could explain indeterminacy only to have experiments indicate entanglement itself is subject to indeterminacy and contextual, that is, the strength of the entanglement varies with the number of quanta entangled. A recent paper even indicated that if a real cat were in superposition we'd still only see one cat because the accuracy of the measurements required are beyond astronomical. In other words, no need for an explanation of the collapse of the wavefunction because it is only a limitation of our ability to perceive such things.

 

[bhobba]

I thought the term statistical explanation was an oxymoron.

Hmmm. I suspect only if you let certain philosophical biases get in the way.

An explanation is a model that predicts outcomes in accordance with experiment. Its got nothing to do if the predictions are statistical or deterministic.

I for one believe Gleasons Theorem implies a fundamental statistical character of nature:
http://en.wikipedia.org/wiki/Gleason's_theorem
'Gleason's theorem highlights a number of fundamental issues in quantum measurement theory. The fact that the logical structure of quantum events dictates the probability measure of the formalism is taken by some to demonstrate an inherent stochasticity in the very fabric of the world. To some researchers, such as Pitowski, the result is convincing enough to conclude that quantum mechanics represents a new theory of probability. Alternatively, such approaches as relational quantum mechanics make use of Gleason's theorem as an essential step in deriving the quantum formalism from information-theoretic postulates.'

Getting back to the original question different interpretations of QM abound and Many Worlds is just one of them. None can be experimentally distinguished from the others and all lead to the same math. All suck in their own unique way and what you do is pick the one that sucks to you the least. I don't like MW because to me it sounds like mystical nonsense - that's an uncharitable view - a more charitable one would be such a drastic way of viewing reality is simply not required.

Thanks
Bill

 

[Fredrik]

I'm going to describe the way I think about some of the interpretations mentioned above.

QM is undoubtedly a very good theory in the sense that it makes very accurate predictions about results of experiments. But it's not clear if QM is just an assignment of probabilities to possible results of measurements, or if it's also a description of what's actually happening to the physical system that it makes predictions about.

So the key question is this: Does QM describe the physical system?

If we assume that the answer is "yes", then many worlds seem unavoidable. So I would take the answer "yes" as the starting point of a definition of a MWI. A full definition would have to specify things like what exactly we mean by a "world".

Many people feel that it's not even worth thinking about that question, since it can't be answered by science, at least not until we find a better theory to replace QM. This is the "shut up and calculate" approach. It's perfectly valid, but I think most people who claim to support it are making some unscientific and possibly contradictory assumptions without even realizing it.

If we assume that the answer is "no", there are a number of ways to describe how to think about QM. These descriptions go by names like "the Copenhagen interpretation", "the statistical interpretation" and "the ensemble interpretation", but it's questionable if they should be considered "interpretations" at all, since they do not offer an answer to the question of what is really happening to the system. I also think that these three "interpretations" all say the same thing, only in slightly different words. For example, the ensemble interpretation says that a state vector should be thought of as representing not the properties of the system, but the properties of an ensemble of identically prepared systems. This "ensemble" doesn't have to exist at any moment in time. It can consist of the particles that are being detected when we run an experiment that involves only one particle many times. So "the ensemble interpretation" is really just a different way to say that QM is just an assignment of probabilities to possible results of experiments.

The statistical interpretation says the same thing as the ensemble interpretation, and (if we take Ballentine's 1970 article as the definition) it also claims that every particle has a well-defined position at all times. I've been getting the impression that most people who claim to adhere to the statistical interpretation have missed this detail. At first I thought that there must be some "Bell's theorem"-type of argument that rules out the possibility of every particle having a position (even when its wavefunction is spread out), but I don't think there is. So this version of the statistical interpretation seems plausible to me. However, if particles have positions, they would have to behave in a very strange way. So shouldn't an interpretation that makes this claim also explain that behavior? This brings us to Bohmian mechanics. Isn't that what we would use to explain the weird behavior?

So it seems to me that a statistical interpretation with the assumption about particles having positions is just a Bohmian interpretation in disguise, and a statistical interpretation without the assumption about particles having positions is just the ensemble interpretation in disguise.

The Copenhagen interpretation has been so distorted by misinterpretations that it's impossible to write down a definition that wouldn't get half the Copenhagen supporters to yell that it's wrong. But I think it's at least clear that Bohr liked to emphasize that measurements must have results. An interaction that doesn't wouldn't be considered a measurement. (This post elaborates a bit). So I find it hard to interpret his views as saying anything other than what I've already said above (about QM and the ensemble interpretation).

 

[dx]

If we assume that the answer is "no", there are a number of ways to describe how to think about QM. These descriptions go by names like "the Copenhagen interpretation", "the statistical interpretation" and "the ensemble interpretation", but it's questionable if they should be considered "interpretations" at all, since they do not offer an answer to the question of what is really happening to the system.

According to the copenhagen interpretation, systems simply do not have a behavior which is independent of their interaction with the measuring instruments:

It is impossible, in the analysis of quantum effects, of drawing any sharp separation between an independent behaviour of atomic objects and their interaction with the measuring instruments which serve to define the conditions under which the phenomena occur.

In quantum theory, it was argued, the logical comprehension of hitherto unsuspected fundamental regularities governing atomic phenomena has demanded the recognition that no sharp separation can be made between an independent behaviour of the objects
and their interaction with the measuring instruments which define the reference frame.

As regards the quantum-mechanical description of atomic phenomena, emphasis is laid on its consistency within a wide scope and, especially, on the elucidation of the well-known paradoxes as regards the problem of "physical reality" through the recognition that, in the proper quantum effects, we have to do with phenomena where no sharp separation is possible between an independent behaviour of the objects and their interaction with the measuring agencies necessary for the definition of the observable phenomena.

The very fact that in quantum phenomena no sharp separation can be made between an independent behaviour of the objects and their interaction with the measuring instruments, lends indeed to any such phenomenon a novel feature of individuality which evades all attempts at analysis on classical lines, because every imaginable experimental arrangement aiming at a subdivision of the phenomenon will be incompatible with its appearance and give rise, within the latitude indicated by the uncertainty relations, to other phenomena of similar individual character.

 

[bhobba]

The statistical interpretation says the same thing as the ensemble interpretation, and (if we take Ballentine's 1970 article as the definition) it also claims that every particle has a well-defined position at all times. I've been getting the impression that most people who claim to adhere to the statistical interpretation have missed this detail.

That's true and I noted it when I read it as well.

However that is not the view of his textbook QM A Modern Approach.

I think its safe to say he no longer holds that view and indeed taking into account Kochen-Sprecker, Bell etc you really can't maintain it these days without advocating other stuff such as contextuality.

I hold to the Ensemble Interpretation and certainly I don't think of it that way.

Thanks
Bill

 

[wuliheron]

Hmmm. I suspect only if you let certain philosophical biases get in the way.

An explanation is a model that predicts outcomes in accordance with experiment. Its got nothing to do if the predictions are statistical or deterministic.

Dictionaries merely list the most popular definitions of words and are not exactly famous for waxing philosophical. It may be a personal failing or my own philosophical bias, but I really enjoy using widely recognized definitions for terms. It just makes attempts to communicate so much more productive.

Nor are model cars considered "explanations" of cars by most people because they can clarify nothing whatsoever about cars even if you can use them to predict the cat will pounce. Hence, I assume that is the reason we have two words, models and explanations, and don't just conflate the two Willy-Nilly.

 

[bhobba]

Dictionaries merely list the most popular definitions of words and are not exactly famous for waxing philosophical. It may be a personal failing or my own philosophical bias, but I really enjoy using widely recognized definitions for terms. It just makes attempts to communicate so much more productive.

Nor are model cars considered "explanations" of cars by most people because they can clarify nothing whatsoever about cars even if you can use them to predict the cat will pounce. Hence, I assume that is the reason we have two words, models and explanations, and don't just conflate the two Willy-Nilly.

Nor am I not using words in any sense different to that used in science.

In physics an explanation is synonymous with mathematical model:
http://en.wikipedia.org/wiki/Mathematical_model

'Mathematical models can take many forms, including but not limited to dynamical systems, statistical models, differential equations, or game theoretic models. These and other types of models can overlap, with a given model involving a variety of abstract structures. In general, mathematical models may include logical models, as far as logic is taken as a part of mathematics. In many cases, the quality of a scientific field depends on how well the mathematical models developed on the theoretical side agree with results of repeatable experiments. Lack of agreement between theoretical mathematical models and experimental measurements often leads to important advances as better theories are developed.'

'Mathematical models are of great importance in the natural sciences, particularly in physics. Physical theories are almost invariably expressed using mathematical models.'

This is very well known and it really is surprising it needs to be pointed out.

Simply think back to good old Euclidean Geometry which is the basis of how all modern physical theories are expressed. The point here though is such models need not be deterministic.

Thanks
Bill

 

[michael879]

Ok so to answer your actual question, the answer is pretty straight forward. The "problem" that all the crazy interpretations of QM are trying to solve is the illogical results of the "Bell's inequality" experiment. This experiment (which was carried out and showed to be consistent with QM) demonstrates that QM is incapable of being explains by anything remotely classical in nature. In order to explain QM you either need to give up locality, determinism, forward causality, or introduce some strange new concepts like many worlds (or new pilot quantum forces). Many worlds has a significant advantage over the other interpretations in that its deterministic, forward causal, and local in a sense. There are some nonlocal aspects to the interpretation, but as far as physical objects are concerned its a local theory.

Now I'm going to try to explain to you why MWI might not be as crazy as it seems. One of the ways to make MWI more reasonable is by thinking of the universe as being a giant medium for standing waves. If you think of just a string standing still, you can imagine that instead of it having no waves on it, it actually has ALL possible waves on it. The amplitudes all cancel out and still sum to 0, so its a perfectly reasonable thing to say. Similarly, if you think about the universe as some medium for waves, all possibilities are acted out on this null "string", cancelling each other out. The "splitting" we observe is just an illusion, as MWI is completely deterministic (everything is predetermined and time is really just an illusion).

 

[1977ub]

In order to explain QM you either need to give up locality, determinism, forward causality, or introduce some strange new concepts like many worlds (or new pilot quantum forces).

You left out superdeterminism, which is both crazy and classical. :)

 

[michael879]

You left out superdeterminism, which is both crazy and classical. :)

Right lol, I did forget one of the most ridiculous interpretations there is :P

 

[TGlad]

The thing is, it seems completely outlandish to the layman. The idea that reality itself keeps 'splitting' into uncountable versions just seems extraordinarily far-out - at least to me. Yet this seems to be really what it is saying. It is called, after all, 'Many Worlds'.

Physics at the quantum level is the physics of waves. The description of of electrons, photons etc is exactly that of waves, as seen in Schrodinger's wave equation http://en.wikipedia.org/wiki/Schr%C3%B6dinger_equation.
Molecules are just the sum of the waves which are the component particles. Logically, humans, planets and the whole universe is a big and complicated sum of waves.

The funny thing is, everyone accepts that there are many simultaneous states of each elementary particle, giving them their wave behaviour, and everyone accepts that worlds are composed of these particles. But many people don't like to accept that there are therefore many simultaneous states of the world (many worlds theory).

 

[wuliheron]

Nor am I not using words in any sense different to that used in science.

In physics an explanation is synonymous with mathematical model:
http://en.wikipedia.org/wiki/Mathematical_model

That same webpage describes a model saying "A model may help to explain a system" rather than insisting a model is an explanation. Again, we have two words for a reason and I can model explanations and try to explain a model. Unless you happen to work on an official physics dictionary I'll stick with the common definitions.

 

[Maui]

I did a bit of reading on it, and was interested to find out that many working physicists regard it as the most plausible way of interpreting the results of quantum physics experiments.

MWI is ruled out by Occam's razor but at the same time it's the most intuitive explanation for realists. Yes it takes no less than a hundred thousand new universes for a fly to move from the window to table and the MWI is not in the textbooks whereas the CI is.

 

[Fredrik]

MWI is ruled out by Occam's razor

This is not true at all.

 

[Maui]

This is not true at all.

Care to elaborate? If all the other universes are real than Occam's razor fails.

 

[michael879]

Care to elaborate? If all the other universes are real than Occam's razor fails.

lol Occam's razor is a guideline not a law of nature! Anyway I don't really see how MWI goes against it at all...

 

[bhobba]

lol Occam's razor is a guideline not a law of nature! Anyway I don't really see how MWI goes against it at all...

It doesn't.

MWI is a perfectly valid interpretation - just not to my taste - I don't think such a drastic view of the world is required. But that doesn't mean a damn thing - opinions are like bums - everyone has one but it doesn't make it right.

Thanks
Bill

 

[Fredrik]

Care to elaborate?

It's a straightforward interpretation of a theory built up around a simple mathematical structure. If the structure is simple and it contains many worlds, then "many worlds" is a simple idea.

If all the other universes are real than Occam's razor fails.

Would you like to elaborate on why you think so?

 

[1977ub]

In an earlier era before QM would/could/should have MWI possibly been advanced to explain why dice rolls come up as random? Why or why not? Why is MWI better than just throwing up your hands?

 

[bhobba]

In an earlier era before QM would/could/should have MWI possibly been advanced to explain why dice rolls come up as random? Why or why not? Why is MWI better than just throwing up your hands?

Even in an earlier era it was known the reason dice came up random was lack of knowledge about the factors that went into it - such is not the case with QM - or rather if it is no one has been able to find them and certain theorems exist that show such if they existed would operate in a bit of a weird way eg contextually and non locally.

Its better than throwing up your hands because it takes the formalism of QM literally and to its logical conclusion.

Thanks
Bill

 

[1977ub]

Even in an earlier era it was known the reason dice came up random was lack of knowledge about the factors that went into it - such is not the case with QM - or rather if it is no one has been able to find them and certain theorems exist that show such if they existed would operate in a bit of a weird way eg contextually and non locally.

Its better than throwing up your hands because it takes the formalism of QM literally and to its logical conclusion.

Under superdeterminism one does not need to pick between contextual or nonlocal, correct?

 

[bhobba]

Under superdeterminism one does not need to pick between contextual or nonlocal, correct?

Super-determinism bypasses bells theorem and you can preserve locality, but it doesn't help with non-contextuality.

To the best of my knowledge no one has yet exploited the loophole and actually constructed such a theory - in fact from what I can gather many think its highly unlikely such a theory even exists - but I am no expert in the area and may stand corrected.

Thanks
Bill

 

[wuliheron]

In an earlier era before QM would/could/should have MWI possibly been advanced to explain why dice rolls come up as random? Why or why not? Why is MWI better than just throwing up your hands?

No explanation whatsoever means there is no particular place to begin investigating. That might be nice for artists, but doesn't cut it when you talk about building the largest machines in the world. Investors want to know the subject will be systematically analyzed.

 

[Fredrik]

In an earlier era before QM would/could/should have MWI possibly been advanced to explain why dice rolls come up as random? Why or why not? Why is MWI better than just throwing up your hands?

You seem to think that the MWI is just the idea "hey what if all the possibilities actually happen". That idea could certainly have been put forward before QM. Just say that each point of the phase space of the universe is a world that actually exists. This would have been a not entirely unreasonable speculation about what's really going on, but it's not suggested by the theory itself. You can certainly view classical mechanics as a description of what's actually happening to the universe without being led to the conclusion that all possibilities are real.

However, if we assume that QM is a description of what's actually happening to the universe, then we immediately find ourselves in many worlds territory. See e.g. how I argued against what I at the time thought people meant by "the Copenhagen interpretation" in the quote below. (Now I know that there are so many different ideas that are claimed to be "the CI" that the term is hardly useful anymore).

The main assumption of the CI is that state vectors can be identified with physical systems, i.e. that each state vector describes all the properties of the system it represents. Let's label that assumption (1). I said that if we add this on top of QM, we get a contradiction, but that's not quite right. What we get is many worlds. So QM+(1) contradicts the assumption that there's only one world. Let's label that assumption (2). Obviously, (2) should also be considered part of the definition of the CI.

So I'm not going to argue that QM+(1) is logically inconsistent, I'm going to argue that CI=QM+(1)+(2) is. The argument can't be made rigorous, since the assumptions (1) and (2) aren't mathematical statements. An informal argument is the best anyone can do.

The Schrödinger's cat thought experiment has taught us that the linearity of the SE implies that if microscopic systems can be in superpositions, then so can macroscopic systems. The details of this part of the argument are included both in Ballentine's 1970 article and in his more recent textbook. (Section 9.2).

(A calculation that includes decoherence effects would change the argument somewhat, but not enough to solve the problem).

Suppose that we prepare a large and complicated system, e.g. a system that includes you, in a state like |this>+|that>, where |this> and |that> describe two different experiences you can have in there. Now the problem is that (1) says that |this>+|that> is a complete description of the physical system. Clearly this means that neither |this> nor |that> can be a complete description of the physical system, and this means that what you actually experience as a part of that system is no more than half the story. If the complete description includes both of your possible experiences, then so does reality. Otherwise it wouldn't be a complete description.

Therefore QM+(1) implies that there are many worlds. This means that QM+(1)+(2) is inconsistent.