Problems with Many Worlds Interpretation

[BruceW]

There isn't any collapse, there's just decohering islands. Basically, there isn't enough information in the initial state to maintain all the coherences that would be required as more and more subsystems interact, so coherences are restricted to tiny subspaces called "worlds." When there's no coherences, there's no crosstalk, so there's no influences either, the branches don't affect each other. But they also cannot perceive each other, they are angels on each other's pins. And yes, the number of such branches is truly spectacular.

Yep, that's mostly right. But there is always cross-talk and different branches always affect each other. The important thing is that once decoherence has happened, this cross-talk is extremely small, and the effect of other branches also becomes extremely small.

So what kind of experiment do we need to distinguish MWI from CI? This depends on what causes the non-unitary collapse in CI. If you define non-unitary collapse to happen when a 'classical measuring apparatus' is entangled, then to disprove CI we simply need to get classical objects in two different branches to affect each other. (In other words, interference between classical objects). This is technologically incredibly difficult.

This same argument goes if we define non-unitary collapse to happen when any human makes a measurement. In this case, we would need to make an interference experiment involving a human (even harder to do, and we may never have the technology to do this).

 

[nazarbaz]

I wouldn't say a new universe is spawned, it's just that the one universe fragments more and more with time (by "fragment" I mean "spawns incoherent islands that have no further influence on each other"). So it's rather like a river forking over and over-- it still all traces back to the same upstream watershed. We are like an ant floating on a log that takes, seemingly at random, a certain fork each time, but the rest of the river is still there, and all branches share a consistent history up until they forked off.
There wouldn't necessarily be any "me" in many of the forks, if I died in some of them, or if I changed so much that it was no longer recognizable as me. Indeed, it seems to me that "I" would only ever be in one of the worlds anyway, but that depends on what is meant by "me"-- this is one criticism I have of MWI, it requires a "model of me" that sounds an awful lot like CI-type collapse if you're not careful. MWI goes to a lot of trouble to maintain unitarity in the universe, but it can never maintain the unitarity of "me", unless "I" am just a projection of something else. How can a projection think for itself? What happens to "I think therefore I am"? There's very little left in MWI of anything I can call "me" without invoking the very collapse that MWI seeks to avoid. This certainly doesn't make MWI wrong, but it shows why I can call it a "radical" form of rationalism.

Indeed, my latest insight into all this interpretation business is that interpretations are kind of like gedankexperiments. Many people seem to think gedankenexperiments are a way to achieve understanding of reality without leaving their armchairs, but you can't understand reality that way, you can only understand a particular theory that way. Gedankenexperiments are ways of probing theories, only real experiments can probe reality. All the same, they are very valuable for understanding theories, and no one would ever "object" to any gedanken on the grounds that a different one is "better." When we look at our theories, we should look at them from all angles, and so we should address all the gedankenexperiments, as well as all the viable interpretations. We just shouldn't imagine we are learning about reality when we do that-- we are learning about our theories.

To the extent that the theories work well, however, we must be accessing some kind of truth about reality via the theories, but it's not a world view we are getting, it's just some kind of taste of the truth, or insight that fits in our heads rather than the actual truth. Maybe every interpretation has some kind of valid lesson about reality to teach us, even though they only help us understand what our theories are saying. Most likely only one (or zero) will survive the next theory, but even so the lesson of each is learned all the same, and captures some shadow of the reality or it wouldn't have been a valid interpretation of a good theory in the first place.

That sounds a bit like a postmodernist view of science with some ideas close to Thomas Kuhn concept of paradigm and Michel Foucault's (or Bachelard's) epistemological nick. These people freaked out when they hear about "progress" and "truth" in science.
I use these words with precaution too but I can't take these guys seriously since accumulation is an effective modality in the constitution of knowledge. To contest that, you have per exemple to prove that the structure of DNA is an "interpretation" with no links to the previous achievements of biology. They urge science to give them a "theory of everything" right now, which is a pretty violent demand, or they deny it the right to seek for truth. Curious, because they think of themselves as anarchists and libertarians. What they are trying to do is to split definitively the scientific experience, that relates subjects to objects, and to think of it as a mere creation of a theoretical subjectivity, since, for them, the world is unknowable per se, by nature.
These views are useful in some fields but their pretention to cover all the realm of thinking, and especially science, is dangerous. They deal the same way with truth, morality and aesthetics. Which is intellectually very suspicious.

 

[mitchell porter]

A few comments:

1) CI does not have to involve "collapses" caused by "measurement devices". In fact the whole idea of wavefunction collapse as anything other than an epistemic process, just like the "collapse of a probability distribution" when you acquire new information, is alien to CI as originally intended - although many people now speak casually about wavefunction collapse as an actual process, a side effect of the de facto realism regarding wavefunctions that has arisen.

Nonetheless, if you refer e.g. to Schrodinger's original article in which he introduces the cat thought-experiment, he clearly distinguishes between two interpretations of the wavefunction's meaning: one as physical state, the other as "expectation-catalog", i.e. something that tells you probabilities about reality, not something that is reality itself.

If we define CI as the view that observables are what's real and that wavefunctions are not real things, then interference effects involving macroscopic devices will not falsify it. The wavefunction of a macroscopic device will still just have been the mathematical construct you use to compute its probable behaviors, and not the object itself, and you will still have zero empirical evidence of there being anything more than one world.

2) Most of these discussions of MWI overlook the fact that there is no derivation of the Born probabilities within MWI, and in fact if you count the worlds you get the wrong probabilities. I suppose it is a sign of the low level at which these discussions occur. They revolve around fuzzy intuitive ideas of "worlds splitting" and thereby increasing in number, or of "wavefunctions collapsing" when someone looks at them.

The fact is that if you do assume a wavefunction to be reality, and then chop it up into worlds in the obvious way, then all worlds are equally real, they all occur once, all outcomes are equally probable, and this is in conflict with experiment! Under other circumstances, MWI would be regarded as falsified by this observation. Casual advocates of MWI just assume that the usual quantum prescription for probability (the Born rule) also applies to their interpretation, but there's no reason why it should, unless worlds are actually duplicated in proportion to the associated probability, and this is a position that almost no-one (perhaps actually no-one) seriously advocates.

The reason is simple. If all outcomes are realized, then probabilities of possible events refer to their actual frequencies of occurrence in the multiverse. If all worlds occur once, then all possible events occur only once, and they are all equally probable, in contradiction to experiment. So if A empirically occurs twice as often as B, it had better be realized in the multiverse twice as often. This is why a really effective version of MWI would involve duplication of worlds: it's the obvious way to get the frequencies right.

Among actual physicists, advocacy of MWI usually means advocacy of a "no collapse" interpretation which doesn't involve a particular division of the wavefunction into worlds. Weeks ago in this thread I had some discussions with Hurkyl and another no-collapse advocate, in which I endeavored to show that this no-collapse interpretation is actually vacuous. It's just a slogan; it doesn't actually describe a picture of objective reality in which we can say which part of the wavefunction corresponds to observed reality. If you try to do that, it inevitably leads back to the naive version of MWI, which has problems, not just with obtaining the correct empirical frequencies, but also with relativity. It's not that it makes the wrong relativistic predictions (as I was just explaining, it already makes the wrong predictions if you just count the worlds), but no-one has yet managed to describe how to divide up a wavefunction in a way that doesn't involve a preferred set of spacelike hypersurfaces, i.e. an objective notion of simultaneity.

3) The debate over the meaning of quantum mechanics is therefore just full of confusions and illogic. Once, everyone believed in CI, but it seems that a lot of people had a simple "uncertainty" interpretation in their minds, according to which the real ontology is just like classical physics, except that the fields are fluctuating randomly and their statistics happen to match up with the quantum predictions. But from Bell's theorem forwards, work on "quantum foundations" has shown that this would require highly unusual correlations in the random fluctuations.

Also, Schrodinger's distinction between the two interpretations of the wavefunction has been lost on many physicists, who advocated CI while also speaking about wavefunctions as if they were real, and so the tendency to say that collapse of the wavefunction is a real physical process sprang up. This has encouraged the rise of the Many Worlds interpretation, but as I have attempted to indicate, that interpretation has a lot less going for it than people imagine. Forget the "weirdness" of multiple worlds, and so on. The important point is that it does not explain the probabilities, and the probabilities are the whole of the empirical content of quantum mechanics! But this point is hardly appreciated except among the few people who work seriously on trying to make MWI work as a physical theory - and the avantgarde there are pursuing the Deutsch-Wallace dodge of trying to derive probabilities from "multiverse decision theory" rather than from the counting of the worlds. In other words, unable to explain how their multiverse would produce the probabilities that we actually experience in the world, they have decided to redefine the very concept of probability in terms of "rational behavior in the multiverse". Then they confuse themselves with baroque exercises in quantum game theory, eventually surreptitiously slip in an assumption equivalent to the Born rule, and announce that they have managed to justify MWI.

All of that is in any case a sideshow; the majority of physicists who believe in MWI adopt the no-collapse interpretation, which is also only tenable if you avoid thinking about it too clearly. Basically, it involves using the Schrodinger equation for everything, treating wavefunctions as physical states rather than sources of probabilities for observables, and relying on decoherence to make sense of everything. But decoherence is simply a shift in the pattern of probabilities. The world we see is always particular - one specific outcome. If you want to say that this world is part of an uncollapsed wavefunction, then you must say that the various parts of the decohered wavefunction are all just as real as what we see, and this leads inevitably back to "naive MWI" and its problems in justifying the Born rule.

4) For people who want to know what's really going on: the answer is that we don't yet know. It could be that quantum probabilities are what you get in a universe in which causal chains go from future to past as well as from past to future, and perhaps quantum mechanics results from subatomic closed timelike curves, or from the self-consistency of a physics in which particles can zigzag in time as well as in space. But no-one has actually proven that this is where it comes from. The transactional interpretation, which is the best known of these zigzag-in-time interpretations, is also more concept or hype than mathematical substance, just like many worlds. It's just that Many Worlds has about 100 times as many believers as the transactional interpretation. But this is due only to sociology and historical contingency; many worlds has zero advantage over transactional interpretation in terms of evidence.

In my opinion, the true path to explaining quantum mechanics lies through arcane technical developments in the foundations of the most advanced physical theories we actually use. For example, the unfolding motivic and twistorial revolutions in quantum field theory, which have recently been promoted by Nima Arkani-Hamed among others. These are initially formal changes, a new calculus in which to perform the calculations, but they are deconstructing and reconstructing the whole conceptual framework, in a way which looks highly nonlocal from the ordinary space-time perspective. It is still impossible to say what the natural concepts to accompany the new formulation of QFT will be, but I will point out that when Penrose developed his twistors he was hoping to explain QM - they were supposed to be an extension of spin networks, which offer a combinatorial derivation of quantum angular momentum, that would also include spatial extension.

But they languished neglected for 30 years until Witten combined them with string theory and showed that the twistor string could explain some mysterious formulae (the Parke-Taylor amplitude) from gauge field theory. That was the start of the "twistor uprising" and it is still unfolding, largely hidden from public view. That is where the real explanation of QM is most likely to be found. Perhaps it will draw upon some concepts found in one of the already discussed interpretations (Bohm, Everett, Bohr...), but as I have been emphasizing, many of those interpretations are not conceptually self-sufficient; they borrow the Born rule from ordinary QM, unable to justify it within their own framework.

The new QFT, since it is starting life as an alternative calculational framework, does not have that problem, and what remains to be done is to round it out and to then find its natural interpretation.

 

[Hurkyl]

This could easily lead someone to eschew the task of specifying THE trajectory, and simply take the equations themselves as the entire theory, which entails ALL possible trajectories within the phase space. This certainly yields a more symmetrical and less ad hoc interpretation... but it's also sterile and pointless.

I wanted to make a comment on this specifically. We have a few hundred more years of understanding about logic and reason -- and ironically the 'sterile and pointless' better models how we actually do things.

The main thing that we can do today that we couldn't do back then is to consider the idea "Okay, we have this mathematical structure -- what sort of reasoning can be done internally to the structure?"

A corresponding philosophy is that that is what is actually important from a scientific theory: it tells us how to make inferences. It tells us things like "From X and Y, you can infer Z". Talking about "THE trajectory" was never the important thing; it was simply a means to the end of determining what inferences could be made.

 

[Hurkyl]

2) Most of these discussions of MWI overlook the fact that there is no derivation of the Born probabilities within MWI,

I've seen some and derived one myself -- e.g. that the weights on the mixture correspond to frequentist probabilities in the limit of infinitely many iid trials. But that aside, there doesn't have to be a derivation anyways.

and in fact if you count the worlds you get the wrong probabilities.

As well you should, since the counting measure has nothing to do with the probability measure.

The reason is simple. If all outcomes are realized, then probabilities of possible events refer to their actual frequencies of occurrence in the multiverse. If all worlds occur once, then all possible events occur only once, and they are all equally probable,

This is not an argument, this is somewhere between wordplay and willful ignorance.

 

[Ken G]

So what kind of experiment do we need to distinguish MWI from CI? This depends on what causes the non-unitary collapse in CI. If you define non-unitary collapse to happen when a 'classical measuring apparatus' is entangled, then to disprove CI we simply need to get classical objects in two different branches to affect each other. (In other words, interference between classical objects).

Actually, I don't think that would suffice. If QM predicts such interference exists, no matter how small or hard to detect, I find it pretty likely that such interference would be found in any experiment sensitive enough to detect it (though other new physics might appear also, at that level of sensitivity). In other words, if we go into the "middle domain" of "quasi-classical" systems, do any of us doubt that we will see clearly how decoherence happens, and when decoherence is incomplete? The classical realm is a realm that is understood quite well via the process of very-nearly-complete decoherence. All CI does is say that "reality" begins once decoherence is fully complete, and reality only exists to the extent that complete decoherence is meaningful. ("There is no quantum world"-- Bohr.)

The reason it does that is not because there must be complete decoherence, it is because complete decoherence is all we can perceive-- our brains are simply not built to "grok" superpositions of outcomes. It is very likely a limit of what we even mean by "perception" that it involves complete decoherence, just as it is very likely a limit of what we even mean by an "experiment" that it involves a subject/object separation. It is not the absence of a tiny bit of incomplete decoherence in a classical measurement that forces the CI interpretation onto us, just as it is not the complete absence of subject/object interactions that forces the scientific method onto us. It is instead the presence of the need for the axioms of empiricism that do both those things-- we must have something demonstrably definite to anchor our conclusions to or it's more like opinion than like science.

So an experiment to distinguish CI from MWI must be an experiment that proves quantum mechanics wrong, which we are motivated to look for regardless of what we think of those interpretations. The differences in CI and MWI are just useful for thinking of ways to find the cracks in QM itself, but watching the decohence occur the way QM predicts it would is not going to falsify CI, we still have the same basic issue that CI is empiricist, and empiricism must follow a direction of logic that starts at the fully decohered result. If one experiment shows only partial decoherence, then CI says that experiment is simply not finished yet.

Perhaps a different example will help see what I'm arguing. Take tunneling. What if there was a debate between people who felt wave functions were real, and people who thought they were just instructions for making predictions. The first camp could point to tunneling, and say, "well there's no classical analog to tunneling, so if we can just find an example where it happens in reality, then we would be falsifying the perspective that wavefunctions aren't real." Then they find tunneling, and claim victory. Trouble is, the other camp just says, "how does that prove wavefunctions are real? All you've shown is they can be used to make tunneling predictions. We already thought of them that way, you haven't told us anything we didn't already know."

Same for an experiment that shows "influences" between the "worlds"-- the empiricist CI person just says "so what, it just means quantum mechanics is making a correct prediction. That doesn't mean those other worlds are real, I would have to be perceiving them for them to be real, and I already know that I do not perceive them, unless you can build an apparatus that allows me to perceive them." It's not just that it's technologically impossible, it's that it isn't perception in the first place-- it is not the way our brains evolved to function, and the empiricist believes we are always stuck inside the world that our brains give us, no matter what postulates you like. Physics is what we can say about nature, not about nature herself-- Bohr again.

 

[Ken G]

That sounds a bit like a postmodernist view of science with some ideas close to Thomas Kuhn concept of paradigm and Michel Foucault's (or Bachelard's) epistemological nick. These people freaked out when they hear about "progress" and "truth" in science.

Not necessarily. We don't have to choose between either saying that there is no "progress" in science, or that the ultimate goal is a "TOE." I don't think either of those views have much of a clue about the progress of science as it has been, and I don't see any reason to expect the progress of science to change suddenly now. Instead, I'd say these lessons of the history of science are pretty hard to deny:
1) scientific progress looks like gradually improving accuracy achieved by radically different ontological pictures, and
2) every generation thinks their own ontological picture is "on the right track", regardless of how completely it differs from previous ontological pictures.
Given these two seemingly uncontroversial facts, I think we can conclude that scientific accuracy converges, but scientific ontology does not. The main problem with postmodernist science is that it is generally not done by scientists, but it does have some useful "cautionary" remarks to make, as you say.

 

[Ken G]

If we define CI as the view that observables are what's real and that wavefunctions are not real things, then interference effects involving macroscopic devices will not falsify it. The wavefunction of a macroscopic device will still just have been the mathematical construct you use to compute its probable behaviors, and not the object itself, and you will still have zero empirical evidence of there being anything more than one world.

Exactly, CI is automatically the interpretation used by empiricism, and is not falsifiable in any context where quantum mechanics works. Even then, it would only be the quantum mechanics that was falsified, not the basic tenets of CI. We would have to find observations that cannot be consistent with empiricism, and that is paradoxical. Thus CI-type thinking will never be falsifiable within anything we now consider to be science. All the same, what we consider to be science today could eventually be returned to its more rationalistic roots someday.

2) Most of these discussions of MWI overlook the fact that there is no derivation of the Born probabilities within MWI, and in fact if you count the worlds you get the wrong probabilities. I suppose it is a sign of the low level at which these discussions occur. They revolve around fuzzy intuitive ideas of "worlds splitting" and thereby increasing in number, or of "wavefunctions collapsing" when someone looks at them.

I think you are building a bit of a straw man here, which may have been Hurkyl's objection too. The crux of MWI is the unitarity of the time evolution of a closed system, not the splitting of worlds (despite its name!). MWI is not actually responsible for the Born rule, that rule must be viewed as part of QM proper, and MWI is merely a way to interpret QM proper such that the unitarity of closed systems is taken as inviolate regardless of what observations suggest when the observer is regarded as part of the closed system. That's why it is a rationalist approach rather than empiricist, it is a fundamentally different way to think about what science is doing. The weights of the splitting, given by the Born rule, is something very different-- we can't ask MWI to supply those any more than we ask CI to do that, the Born rule is just part of the meaning of the wave function regardless of which direction one takes the logic (from wave functions to the observations, as MWI does, or the other way around, as CI does).

3) The debate over the meaning of quantum mechanics is therefore just full of confusions and illogic.

I don't see any illogic, I see just the age-old debate between rationalism and empiricism. You understand that old debate, I presume, so I don't see why you think the modern debate is anything different, even if the core issues do sometimes get confused with verifiable predictions the theory is making.

Also, Schrodinger's distinction between the two interpretations of the wavefunction has been lost on many physicists, who advocated CI while also speaking about wavefunctions as if they were real, and so the tendency to say that collapse of the wavefunction is a real physical process sprang up.

Yes, I think there is a lot of confusion about what CI is really saying, and indeed many who criticize it criticize a naive version of it, rather than its empiricist roots. But I think the same is true of MWI-- many who criticize it don't recognize that it is not empiricism, so using empiricist thinking will always create issues with MWI. One must first adopt the rationalist credo that reality is what we can make sense of by thinking about it-- that the map is the territory because maps are all that we do. The empiricist counters that perceptions are all that we do, and maps are informed by perceptions.

I think I would have to say that maps and perceptions are together all that we do, so we should probably think of interpretations simply as ways of looking at our theories from different angles. No one would suggest there is a "right angle" to see an elephant from, so if taken in this spirit, we should understand all the interpretations. My issue is simply with not recognizing that when we marry a particular one, we are marrying a set of philosophical choices about science that are not shared by all scientists. That is the only important thing to remember.

But they languished neglected for 30 years until Witten combined them with string theory and showed that the twistor string could explain some mysterious formulae (the Parke-Taylor amplitude) from gauge field theory. That was the start of the "twistor uprising" and it is still unfolding, largely hidden from public view.

This is indeed well out of the public view, or even the view among physicists quite frankly, so thank you for drawing attention to it, even if its significance is currently far from clear.

 

[Samshorn]

If unitarity were broken in observations, I think the rationalists in the MWI camp would jump ship immediately…. CI already thinks unitarity is broken in the act of perception …

Doesn’t your second statement cast doubt on your first? I think we agree that, at least from an empirical standpoint, unitarity is already abundantly falsified, and yet this doesn't cause anyone to jump ship. It’s not easy to imagine what additional observations could possibly “break unitarity” for those who - in the face of all the empirical evidence to the contrary - have adopted unitarity as a first principle, and who are willing to posit an infinity of unobservable worlds in order to preserve it.

Maybe you have in mind the idea that the part of quantum mechanics presently believed by everyone to be unitary – namely, the Schrodinger evolution - might be found to fail, and need to be modified in such a way that the state evolution was no longer precisely unitary, and you think such a “change of postulates” threatens the motivation for MWI. Without knowing the details of the hypothetical modification, it’s hard to comment on how it would affect people’s enthusiasm for MWI. However, bear in mind that the quo ante postulates of quantum mechanics already included the non-unitary projection postulate, which was included to match observations, but this didn’t prevent MWI advocates from simply claiming we can do without it – even though from an empirical standpoint we obviously can’t.

Regardless of our observations, one can always bifurcate our theory into a unitary part and a non-unitary part (as we do in traditional quantum mechanics), and then someone who adopts unitarity as a first principle can rationalize away the non-unitary part (as the MWI advocate does for quantum mechanics), with the understanding that the “unitarity” applies only to some posited unobservable meta-world in which the world of our experience is embedded as just an infinitessimal and non-unitary part. But this is silly. Suppose the conservation of momentum was found to be violated in physics, but nevertheless we really really like the principle of momentum conservation, so we ‘preserve it’ by saying that objects sometimes exchange momentum with unobservable objects from other parallel universes, so in the overall meta-universe, including these other parallel universes, we have conservation of momentum! Yay! Obviously no observation can “break conservation of momentum” for us if we are willing to invoke other worlds like this, but just as obviously our satisfaction with having maintained momentum conservation in this way would be absurd. The same applies to the “unitarity” of MWI. (As an aside, in the early 1900s Poincare and Lorentz actually did discuss the prospect of needing to invoke interactions with an unobservable entity in order to maintain conservation of momentum according to the ether interpretation of electrodynamics.)

But I would say this is all somewhat of a red herring - see below.

I don't think MWI-enthusiasts like MWI because of its many-worlds element, which you are quite right they could preserve in any theory, they only like it because of the simplicity and generality of the unitarity postulate, and the many worlds just come along for the ride.

I think you were nearer the truth in your previous message, where we agreed the main motivation (at least technically) for MWI in QM is the uncertainty principle and its consequences, not a yearning for unitarity per se. If some new phenomena came to light, violating existing QM, and people decided the best way fix the theory was to modify the Schrodinger equation to make it non-unitary (rather than modifying the Born rule or the projection postulate, which I think is what would actually happen, but never mind), would this necessarily invalidate the uncertainty principle, eliminate the fundamental quantum of action (h=0), or make all observables commute? Unless we do all these things, we will still have a theory that entails some kind of “state reduction” or "collapse" in the mapping from the variables in the Schrodinger equation to the measures of our experience, and it is the desire to rationalize the palpable non-classical features of this "collapse" - rather than a yearning for unitarity as an abstract niceity - that motivates MWI (over and above the motivation that already exists for non-quantum theories).

Admittedly I have no idea what a revised quantum mechanics with a non-unitary Schrodinger equation would look like. Maybe it really would eliminate uncertainty and imply that all observables commute (perhaps on some deep level that isn’t readily perceivable), in which case we're just back to something like the Newtonian theory… but I have a hard time imagining such a theory accounting for all observations, and even in that case there would still be motive and opportunity to commit MWI – just as there is for Newtonian mechanics.

So, I still think over-exposure to falsification is not a problem of MWI. In fact, quite the opposite, on both counts. Exposure to falsification is good, but MWI doesn’t have any. The main problem with MWI is that it would be sterile and pointless, even it succeeded in giving a satisfactory account of observations – which it doesn’t. To illustrate that, look at your recent comment in another post:

…one universe fragments more and more with time (by "fragment" I mean "spawns incoherent islands that have no further influence on each other"). So it's rather like a river forking over and over-- it still all traces back to the same upstream watershed.

This is a fairly typical prosaic description of MWI, but there are problems. To give just one example, remember that Schrodinger’s equation is time-symmetrical, and yet you say that (according to MWI) the universe evolving according to this equation “fragments more and more with time”. In which direction does this fragmenting increase? Shouldn’t we actually have worlds fusing back together just as much as we have them fragmenting? But then a given stream would be fed by infinitely many different pasts, as well as fragmenting into infinitely many different futures. This has led some prominent MWI advocates to adopt the view that the number of self-coherent worlds is actually constant, and they just are partitioned into shifting equivalence classes. Our lack of awareness of the many futures of our current equivalence class of worlds may be explained away by noting that we aren’t aware of our futures (for some reason), but surely we are aware of our past. Indeed, some MWI advocates (e.g., Deutsch) contend that MWI implies it actually IS possible for an observer to gain awareness of their multiple, mutually exclusive, histories.

One might argue that it’s unfair to expect MWI to resolve the “arrow of time” problem, because it is also an open problem for other interpretations, but I would answer that it is an especially acute problem for MWI, because without a clear and explicit resolution of the arrow problem, we can’t even really say that MWI gives any intelligible account of observations at all. MWI isn’t really an interpretation of quantum mechanics, it is an idea for an interpretation of quantum mechanics.

We are like an ant floating on a log that takes, seemingly at random, a certain fork each time, but the rest of the river is still there, and all branches share a consistent history up until they forked off.

Careful with the ant and the log there... you are smuggling in the very dualism that MWI seeks to exclude on principle. According to MWI, there is no log or ant, there is only water. You are nothing but one particular streamline in the flow. I realize this is inconsistent with your concept of empiricism, which is wedded to dualism, but that's precisely what MWI most vehemently denies. The main motivation for MWI isn't a yearning for unitarity, it is an abhorance of dualism (which of course underlies the kind of empiricism that you've described).

 

[BruceW]

If we define CI as the view that observables are what's real and that wavefunctions are not real things, then interference effects involving macroscopic devices will not falsify it. The wavefunction of a macroscopic device will still just have been the mathematical construct you use to compute its probable behaviors, and not the object itself, and you will still have zero empirical evidence of there being anything more than one world.

Exactly, CI is automatically the interpretation used by empiricism, and is not falsifiable in any context where quantum mechanics works. Even then, it would only be the quantum mechanics that was falsified, not the basic tenets of CI. We would have to find observations that cannot be consistent with empiricism, and that is paradoxical. Thus CI-type thinking will never be falsifiable within anything we now consider to be science. All the same, what we consider to be science today could eventually be returned to its more rationalistic roots someday.

Good, this is how I would define CI as well. In other words, we only use the non-unitary collapse after all other calculations have been done, to show us what the end-user sees. (and I agree it would then make all the same predictions as MWI).

But in so many places on the web and in books, CI is defined differently, so that non-unitary collapse happens whenever any classical measuring apparatus makes a measurement.

In am now wondering if your 'CI' and my 'MWI' are the exact same thing...

 

[Ken G]

But in so many places on the web and in books, CI is defined differently, so that non-unitary collapse happens whenever any classical measuring apparatus makes a measurement.

The way I would put it, a measuring apparatus achieves a certain type of mixed state on purpose, and then our perception is where the collapse comes in. So we have learned to create mixed states in a certain type of way, such that the perception that "collapses" the mixed state will allow us to do science on the outcome. Mixed states (and the decoherence that produces them) happen naturally too, without any measurement, the measurement is just how we've learned to take advantage of this. But none of that is different in CI or MWI, the difference only appears when we account for the perception of the mixed state, and decide how important we are going to regard that perception as.

In am now wondering if your 'CI' and my 'MWI' are the exact same thing...

I would say the difference is in which we view as "the reality"-- the perception of collapse, or the mathematical description of the unitarity of closed-system evolution.

 

[Ken G]

Doesn’t your second statement cast doubt on your first? I think we agree that, at least from an empirical standpoint, unitarity is already abundantly falsified, and yet this doesn't cause anyone to jump ship.

That depends on what we mean by "falsified." To an empiricist, it is indeed falsified, because the reality is what we perceive it to be (when consistent), so if we don't perceive unitarity, it ain't there. But to a rationalist, unitarity is not falsified, because the observations are consistent with unitarity if we simply say that the observer is an open substate of the closed system, so the observer cannot be objective about the state of the whole system. To the latter group, the only way to falsify unitarity is to show that unitary evolution of a closed system doesn't work, when the system is later opened. That would require falsifying quantum mechanics. There shouldn't be any way to falsify either MWI or CI without doing that.

Maybe you have in mind the idea that the part of quantum mechanics presently believed by everyone to be unitary – namely, the Schrodinger evolution - might be found to fail, and need to be modified in such a way that the state evolution was no longer precisely unitary, and you think such a “change of postulates” threatens the motivation for MWI.

Yes exactly-- the theory of quantum mechanics as it stands today is essentially the Schroedinger equation for closed systems, and the Born rule for observations that open them. How we regard that "opening" of the system is the crux of the difference between MWI and CI, but this difference is not in the outcome of the opening, it is in the perceived significance of the opening. But if the closed part wasn't unitary, and there were no other simple postulates to point to (this may be the key addition), no MWI enthusiast would hold to it any more, I feel. Hurkyl can speak for himself, but I'd be amazed if he didn't agree.

I think you were nearer the truth in your previous message, where we agreed the main motivation (at least technically) for MWI in QM is the uncertainty principle and its consequences, not a yearning for unitarity per se.

I think the HUP is the reason we must choose an interpretation like MWI or CI, but the "many worlds" of MWI don't come from the HUP, which is empirical, they come from whatever postulates account for the dynamics. In my "quasi-classical" version where we have a HUP but no Schroedinger, we could still generate a "no collapse" interpretation, and it would still be unitary-- it would just be the postulates of classical mechanics. If we didn't have classical mechanics, it would be like not having Schroedinger-- with no simple and well-unified set of postulates to hold to, there'd be no motivation to retain the unitarity of many worlds. Without a working set of postulates, there just wouldn't be any anchors for the rationalists to hang their hats on. They'd have to tolerate the "collapse" interpretation, they couldn't object if it wasn't violating any other simple set of seemingly valid postulates.

This has led some prominent MWI advocates to adopt the view that the number of self-coherent worlds is actually constant, and they just are partitioned into shifting equivalence classes.

Good point about the time reversibility, although it just means we must modify the analogy to allow the "river" to re-merge. Can we recall two separate pasts? It sounds like the same "perception game" that they play with mixed states-- they would say we do perceive multiple outcomes, and we do remember multiple histories, we just don't realize it somehow. It's as though our brains can only hold one coherent thought at a time, skewing our view of both the present and the past. There's nothing exactly wrong with it, it just seems like an awful lot to hang on the structure of the simplest postulates.

Indeed, some MWI advocates (e.g., Deutsch) contend that MWI implies it actually IS possible for an observer to gain awareness of their multiple, mutually exclusive, histories.

That sounds like a falsifiable claim that MWI proponents had better find a way to avoid! Any theory in which we can recall multiple pasts, or perceive multiple outcomes to a single experiment, represents a radical detour not only for quantum mechanics, but for science itself. One had better have good reason for such a violent departure from the scientific method, it really seems like opening the door to "woo-woo" claims (like past-life regression, etc.). I'm sure Deutsch doesn't want to do that!

One might argue that it’s unfair to expect MWI to resolve the “arrow of time” problem, because it is also an open problem for other interpretations, but I would answer that it is an especially acute problem for MWI, because without a clear and explicit resolution of the arrow problem, we can’t even really say that MWI gives any intelligible account of observations at all. MWI isn’t really an interpretation of quantum mechanics, it is an idea for an interpretation of quantum mechanics.

This is not necessarily a bad thing for MWI-- it does seem quite possible that the arrow of time is indeed a kind of illusion that comes from how we process information rather than what is "really happening." My problem there is, as primarily an empiricist, I feel that a consistent perception cannot be regarded as an illusion-- it is the definition of what is "really happening". Rationalistic approaches to what is really happening seem so futile in view of how much farther science has yet to go.

But I see that both empiricism and rationalism face inherent paradoxes: all empiricists must engage their brains to interpret, rationally, their perceptions, and all rationalists must engage their perceptions of information processing to be able to create a mathematical argument in the first place. This may very well be the reason that we need both rationalistic and empiricist elements to make science work.

Careful with the ant and the log there... you are smuggling in the very dualism that MWI seeks to exclude on principle. According to MWI, there is no log or ant, there is only water. You are nothing but one particular streamline in the flow.

Yes, in effect what I was doing was interpreting MWI from an empiricist standpoint, like "this is what rationalism looks like to the empiricist." It really should be characterized like a rationalist would.

The main motivation for MWI isn't a yearning for unitarity, it is an abhorance of dualism (which of course underlies the kind of empiricism that you've described).

Unification is certainly the rationalist dream, so any form of dualism would be anathema to this goal. Unitarity is just nicely unified, it's always the simplest postulate. What could be simpler than Parmenides' "change is logically impossible"? Ironically, the "many worlds" approach is actually the way to have just one universe, and it never changes, rather than one universe that changes along a time stream. I agree this will always be the goal of rationalism, regardless of whether or not the postulates of QM are falsified-- I just think they wouldn't feel they had a leg to stand on if they couldn't point to some set of currently viable unifying postulates.

 

[Hurkyl]

If you mean by that a one-to-one correspondence with the state space of the measuring device, that is exactly what I am saying an empiricist would never allow.

Empiricism does not assert 'what cannot be observed does not exist'. Instead, what is observed certainly^* exists; about what is not observed we are still free to make suitable assumptions." (I'm paraphrasing von Weizsäcker who was talking about CI, although I don't know the context)

Von Weizsäcker was interested in avoiding paradoxes. I'm interested in understanding the theory, and in avoiding assumptions that unnecessarily limit the scope of applicability of QM.

*: I really don't like absolutes, but I didn't feel it worth rephrasing this bit.

I put no emphasis on the label "eigenvalue," I put emphasis on two things:
1) we perceive definite outcomes of observations
2) the theory of quantum mechanics has a name, and a prescription, for referring to those definite outcomes, and that name is eigenvalue and that prescription involves associating measurements with operators. So #1 is what we perceive, and #2 is how we attempt to understand and unify those perceptions. No interpretations yet.

#1 has already strayed into the realm of interpretation -- it's taking the elements of the mathematical theory of quantum mechanics and interpreting them as saying something about something else -- in this case, about 'reality' and what we 'perceive'.

This approach to interpretation makes the observers god-like. To make observers frog-like, quantum frogs have to be quantum systems and make their observation by quantum interactions -- you need to interpret the the measurement theory as describing the quantum system. (plus some sort of interpretation that says this setup reflects reality)

Yes exactly-- the theory of quantum mechanics as it stands today is essentially the Schroedinger equation for closed systems, and the Born rule for observations that open them.

Which has the problem that observers are gods, and you need multiple sets of rules.

The other main way is Schroedinger equation for closed systems. Subsystems of the system are open, and the mathematical aspect of the Born rule pops out as merely being the operation for extracting the state of the subsystem from the state of the whole.

 

[Hurkyl]

Or if MWI considers that there are multiple worlds that are all sort of similar sharing their part of each outcome (each observed eigenvalue), how on Earth could the worlds be coordinated over time? How could each have a "me" flipping a coin or observing a collapsed value, when after some n number of generations of collapses each world would have vastly differing observers and observation events? This view seems impossible.

I wanted to give a short response from another bent.

How are they coordinated? The Schrödinger equation. Multiple "worlds" and "splitting" aren't fundamental parts of MWI. Instead, they are derived descriptions -- attempts to organize, quantify, and describe the behavior of a quantum state obeying the Schrödinger equation, and somewhat geared to make the connection with other views more plainly evident. (e.g. someone in any particular world sees things as if the universe had collapsed down to that particular component of the quantum state)

 

[Hurkyl]

Gedankenexperiments are ways of probing theories, only real experiments can probe reality. All the same, they are very valuable for understanding theories, and no one would ever "object" to any gedanken on the grounds that a different one is "better." When we look at our theories, we should look at them from all angles, and so we should address all the gedankenexperiments, as well as all the viable interpretations. We just shouldn't imagine we are learning about reality when we do that-- we are learning about our theories.

Agreed! Except for the penultimate phrase.

Real experiments probe reality, but all that gives us is raw data. Learning requires us to analyze the data, and from the analysis, synthesize an understanding.

Physical theories are synthesized from the analysis of that data. Learning and understanding such theories are thus part of the "synthesize an understanding" part of learning about reality.

 

[Ken G]

Empiricism does not assert 'what cannot be observed does not exist'. Instead, what is observed certainly^* exists; about what is not observed we are still free to make suitable assumptions."

But no doubt he is talking about concepts like "electrons" or some such thing that we don't directly observe. He is not talking about things that we should by all rights be able to observe, but don't-- like other outcomes to our observations. That's a very different category of "what is not observed", and one that I would certainly feel empiricism would rule out as part of reality. Indeed, it would have to-- for if we allow other outcomes that fit the theory to occur, how can we rule out other outcomes that do not fit the theory? It is just plain not empiricist to assert that outcomes that fit the theory, but are not perceived, should be regarded as real, simply because they fit the theory. In empiricism, theories are tailored to fit data, data does not just check theories.

This approach to interpretation makes the observers god-like. To make observers frog-like, quantum frogs have to be quantum systems and make their observation by quantum interactions -- you need to interpret the the measurement theory as describing the quantum system.

This is the crux of the matter. To assert that "observers are quantum systems" is a fundamentally rationalist approach, you are saying that observers are ruled by a theory. That's not empiricism, observers are never "ruled" by anything, they are the definers of truth. No theory ever "rules" nature in empiricism, theories are invented by physicists to understand observations. Thus, no observer can be ruled by a theory, all you can do is ask if observers can be treated with the same theory that other things are, but treating anything with a theory is just an analytical choice, not a way to establish what is true.

I wonder why you are so loathe to accept that your approach places what you can think about, and make postulates about, above what you perceive. I would say that is almost the clearest defining characteristic of both MWI, and most everything you say about quantum mechanics. And it is classic rationalism. I really don't think there's any crime in it, but we should see it for what it is.

Which has the problem that observers are gods, and you need multiple sets of rules.

It's not that they are gods in the way we used the "god's eye" view before, but they are certainly gods in that their consistent perceptions are definitely what is real (in empiricism). You object to the multiple sets of rules, but that is your rationalism coming out-- Occam's Razor has no objection to multiple sets of rules if they are needed, and the empiricist says they are needed in quantum mechanics.

The other main way is Schroedinger equation for closed systems. Subsystems of the system are open, and the mathematical aspect of the Born rule pops out as merely being the operation for extracting the state of the subsystem from the state of the whole.

Yes, exactly, that is how the Born rule is regarded in any system in which the laws are the reality. But in any system where the observers are the reality, there is no such thing as an observer that is a "subsystem"-- the observer is there to tell you what is happening, not to be a part of what is happening that you are determining via some laws.

 

[Ken G]

Real experiments probe reality, but all that gives us is raw data. Learning requires us to analyze the data, and from the analysis, synthesize an understanding.

Here we must distinguish two types of "learning"-- there is a more "raw" type, which is like a child "learning" that things fall by repeated experience of the perception of things falling, versus the "learning" of a student in a physics class, who is using abstract reasoning to develop mathematical skills. The child learning that things fall is only noticing a consistency of experience, that will later be compared with the consistencies of other people's experiences, to develop a concept of what is real (versus what happens in dreams, etc). This does require mental effort, but not rationalism. The second type of learning, the learning of the postulates of a theory, is more like how we might learn the rules of chess, or the Peano axioms of arithmetic, or the capitals of the countries in Europe. These are all abstracted from the more "raw" elements of experience.

Now, I will grant you that "levels of abstraction" is an awkward concept. The rationalist, who prefers unified concepts, might be tempted to say that any abstraction is an abstraction, there are no "levels." Empiricism must allow levels of abstraction though, it is central to their axioms. It is certainly the core paradox of empiricism that a perception be regarded as more reliable than a postulate of a theory, when interpretation of perceptions themselves involve some kind of identifiable axiomatic system. But the axioms are simpler and more straightforward, more accessible by anyone, they look like "a consistent perception is real", not "systems evolve via the Schroedinger equation."

On the other hand, the rationalist is not free of pesky paradoxes either-- for when they follow a logical thought process, are they not perceiving their thoughts in their minds? How then can you talk about a "closed system", is not your brain an open one? If the closed system includes perceptions that you are not having, why can it not also include thoughts that you are not having, that seem just as valid to the perceiver of those thoughts as yours do to you? So any logical proof that the postulates are working that you could offer might be countered by a logical proof that they are not working, offered by some other brain in the closed system. You end up having to treat yourself as a god one way or the other-- either you assert that your perceptions decide what is real, or you assert that your reasoning decides what is true. But you are just a subsystem, so neither way can you really say that you speak for the closed whole, unless you simply take it as an axiom that you do.

 

[Eqblaauw]

poll update, I asked the following question to eminent physics , none of which I knew their ideas in advance:

a: I subscribe to the MWI that contains many parallel universes that differentiate during every 'quantum event' (meaning that there are many perhaps an infinite number of copies of everyone that inhibits earth)

b: I subscribe to the MWI that contains many parallel universes that split during every 'quantum event' (meaning that there are many perhaps an infinite number of copies of everyone that inhibits Earth constantly being created)

c: I don't subscribe to a or b, because I think they are both false

d: I don't prefer any of the above

First some remarks
-2 said c/d, those where counted c 1/2, d1/2,
-one said I can't do anything with the poll, cause this has no empirical evidence and therefore has nothing to do with physics,
to avoid all accusations of being biased I didn't count him,
-one said c with a minor change, I don't subscribe to a or b,
to once again avoid all acusations of being biased I didn't count him in,
-one said I don't like both descriptions a or b, to me mwi just means unitary evolution but I chose b,
to avoid all accusations of being biased I did count him as b,
-one sais a,but with a very weak meaning of parallel meanings 'existing' (I don't really know what he means)
but once again to avoid being depicted as biased I counted him as a
- one said d, he didn't thought that a,b captured the good things about mwi, I will count him as d since this poll is about the versions that Max Tegmark and David Deutsch and the likes subscribe to, but if you want to count him as a mwi-er, then you're free to do that, although that isn't really clear from his post . I think it's save to say that no-other (c's) or (d's) where subscribing to any mwi with parallel universes that wasn't mentioned in the poll, because they where free to say so. But there is an iny miny chance These where the answers

C: 15
D: 6
A: 1
b: 1

the reactions came from (I promised to not make their names publicly, but since I don't mention who voted what, I don't think there is any harm in it):
David spergel, Carlo Rovelli, David polizer, David Finkelstein, Richard Muller, M.J. Rees, George F Smoot the Third, Goldreich, James Daniel Bjorken, Richter Burton, John Preskill, Leon N. Cooper, Robert Wald, James Binney, Yakir Aharanov, Andy Fabian, Ulrich Becker, Jim Al-Khalili, Frank Close, Frank Wilczek, Rodolfo Gambini, Jorge Pullin, John Baez.

 

[faen]

I haven't had time to read through this thread, but I am surprised that not everybody realize that MWI is bullocks. There is simply no reason to assume that many new universes are created, more than believing in Santa Clause.

It seems as if the theory is created in order to explain why other quantum states weren't chosen. It says that the other quantum states are the ones which are chosen in another universe. However what really happens with the non chosen quantum states is that those which aren't entangled with the measurement instrument, become entangled with other parts of the environment. Thus no more dimensions, or universes are required to explain what happened to the other quantum states, since they become entangled to the environment and not other universes.. Anybody disagree??

 

[kith]

It seems as if the theory is created in order to explain why other quantum states weren't chosen. It says that the other quantum states are the ones which are chosen in another universe. However what really happens with the non chosen quantum states is that those which aren't entangled with the measurement instrument, become entangled with other parts of the environment. Thus no more dimensions, or universes are required to explain what happened to the other quantum states, since they become entangled to the environment and not other universes.. Anybody disagree??

Yes. What does get entangled is the state of the system with the state of the measurement device (you don't need an additional environment). How is this superposition separated? Is the corresponding porcess unitary? Please provide a reference or at least a mathematical sketch of your idea.

In the framework of open quantum systems and decoherence, you start with a pure superposition state for your system. Due to interactions with the measurement device, it evolves into a mixed state. This by itself doesn't explain why exactly one eigenstate is chosen in a measurement. It just shifts the measurement problem from superpositions to mixed states.

 

[faen]

Yes. What does get entangled is the state of the system with the state of the measurement device (you don't need an additional environment). How is this superposition separated? Is the corresponding porcess unitary? Please provide a reference or at least a mathematical sketch of your idea.

In the framework of open quantum systems and decoherence, you start with a pure superposition state for your system. Due to interactions with the measurement device, it evolves into a mixed state. This by itself doesn't explain why exactly one eigenstate is chosen in a measurement. It just shifts the measurement problem from superpositions to mixed states.

Ok i realized now why the MWI is supposed to solve the measurement problem. I didn't see that that was the intention of MWI when i wrote my previous post..

However i still disagree with the MWI. Let's say you have a system that can get entangled with the environment.. Then a system of n eigenstates can become entangled to the environment in n! ways or something. I'd say this complication reduce the probability of a system that creates new worlds, since it must in addition be aware of all the n! possibilities. Or why would it not take into account the n! possibilities...? Then it would again be random which combination of distribution within the environment it chooses.

Anyway, my belief is that quantum mechanics is an incomplete theory and that the basic steps of actions in this universe is self contained.

 

[Ken G]

Let's see if we can get general agreement that MWI is the only possibility if the Schroedinger equation is the sole description of physical time evolution, and the measurement problem is interpreted as an issue about how to characterize open substates (like observers) as something separate from the time evolution of the closed whole. Can we agree with this? That was never my problem with MWI, I accept that MWI is the way to go if you base everything on that single postulate about closed systems.

My issue is that the way we arrive at this postulate about closed systems is by opening them, by measuring them. We'd have no idea the postulate was true if we couldn't do that. So trying to relegate the act of measurement into some dusty corner of the full evolution of the closed system doesn't ring true to me-- physics is what we can say about nature by doing science, so is fundamentally about opening systems and making a subject/object separation when we do that. There is no such thing as physics without that, so it is not allowed to be viewed as some minor detail living inside some more prevailing rationalistically described time evolution from the "god's eye" perspective.

For this reason, CI adds another postulate to the Schroedinger-controlled evolution of closed systems-- it says that the concept of a closed system is just a concept, and fundamentally it is information about closed systems, obtained by opening them, that is the heart of physics. As such, we also need the Born rule, at the same level of priority as the Schroedinger equation, because neither has any empirical meaning without the other. This approach requires that we regard definite outcomes as real, and forces us to adopt whatever additional postulates allow us to do just that. This is CI.

Now, for completeness, we can make a different choice for the added postulate, as done in deBroglie-Bohm. Here, the Born rule is not viewed as being at the same level of priority as the Schroedinger equation, but something else is-- the hidden variables locked up in the initial conditions. These hidden variables mean that the wave function represents incomplete information, and so does the unitary evolution of the wave function, because there is information in the initial condition that actually determines what happens, completely. The Born rule is then seen as our own statistical treatment of what we don't know about the initial condition, just like how you'd treat your opponents' cards in a poker game.

So the differences in MWI, CI, and deBB-B are none other than different interpretations of the meaning of the Born rule. They all use the Born rule and the Schroedinger equation in exactly the same ways, so none of them are any more falsifiable within QM, but each one might be the way to go with the next theory. Personally, I'm fine with having all three at one's disposal, we should just view each as seeing QM from a different angle, like looking at an elephant from all sides. My only objection is taking seriously the "world view" that these interpretations can suggest.

Since CI really doesn't do that, I prefer CI-- MWI seems to suggest a world view where there really are this dizzying number of other worlds, and deBB-B seems to suggest a world view where the past really does completely specify the future in a deterministic way. I say, physics theories are not intended to spawn world views in the first place, because none of them are likely to be correct enough for their literal interpretations to be authoritative about such matters.

 

[kith]

Let's see if we can get general agreement that MWI is the only possibility if the Schroedinger equation is the sole description of physical time evolution, and the measurement problem is interpreted as an issue about how to characterize open substates (like observers) as something separate from the time evolution of the closed whole. Can we agree with this?

What about the ensemble interpretation? There, the time evolution is also exlusively governed by the Schrödinger equation.

 

[Fra]

To assert that "observers are quantum systems" is a fundamentally rationalist approach, you are saying that observers are ruled by a theory. That's not empiricism, observers are never "ruled" by anything, they are the definers of truth. No theory ever "rules" nature in empiricism, theories are invented by physicists to understand observations. Thus, no observer can be ruled by a theory

In here lies the same paradox I think was mentioned before, and I think it's clear what we have here.

The paradox is

1) on one hand to insist on the empirist observer perspective, in line with the sound science

2) and on the other hand to require that the laws of physics are observer are seen the same to all observers, thus we seek to formulate the covariant transformations.

Both seem reasonable at first sight.

The paradox is that there is no empirical status on the information about the covariant transformation rules that are generally valid. It is valid ONLY for subsystem.

Either you adopty a form of structural realism, that denies (1) when it comes to "laws of nature", but it applies to observations (according to laws) - thus you can imagine that it applies to the universe as a whole (except of course there is no empiricla support!), or you rephrase (2) into a more empirist formulation like Ithink I did ealier in this thread that (2) is replaced by negotiated democratic observers and the laws of physics are not a constraint but rather a result of negotiation. This is another way to resolve the paradox.

/Fredrik

 

[Ken G]

What about the ensemble interpretation? There, the time evolution is also exlusively governed by the Schrödinger equation.

Yes, that's another option, but there it assumes incompleteness right from the start. So we should say that MWI is the only interpretation that claims to have a complete description that is based entirely on the Schroedinger equation as the law of all time evolution of closed systems, including those containing observers.

 

[Ken G]

or you rephrase (2) into a more empirist formulation like Ithink I did ealier in this thread that (2) is replaced by negotiated democratic observers and the laws of physics are not a constraint but rather a result of negotiation.

Or what may have a similar spirit-- we say that laws of physics are examples of whatever keys we can find under the streetlights.

 

[Fra]

we say that laws of physics are examples of whatever keys we can find under the streetlights.

Note sure if I got that metaphor.

To me the key here is interactions in between observers. Laws are somehow the consensus that observers can agree upon.

The difference is wether they HAVE to agree at all times, or wether the agreement is emergent. Before it has emerged, the state of law itself is dismissed into the same "black box" as is the system for which the theory was formulated.

What is the key for me is the democratic process or the negotiation, not possible equilibrium points for such process.

/Fredrik

 

[Eqblaauw]

Einstein has (in my eyes) a wonderful quote:
I don't believe in math, As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality
It doesn't in itself reject the mwi but I think it's a good medicine against some of the thinking of mwi-ers. It surely is a good medicine against 'mathematical universe' of tegmark, and of course all possible universes do exist, just like it is impossible that I ate a large apple pie this morning, the question scientists should be asking is which universe(s) are possible, and you don't do that by making some formulas that are consistent and saying: hey look at this, this is all possible.
Also, I don't imagine that the brightest mind in the world, that's for some reason death blind mute and only read (with his/her fingers) all the pure math books in the world, could ever come up with e = mc2

 

[Eqblaauw]

Tegmark says all consistent mathematically correct universes exist
to know if this has any meaning you can ask yourself the question: consistent with what?
-with itself
that really hasn't got any meaning at all, if it hasn't got any observation to verify it,
when there are observations, then you can see if it's consistent with the universe (or multiverse for that matter).
This mainly is against mathematical universes by the way.

R.I.P. Popper

 

[Eqblaauw]

poll update, (sorry to copy the earlier message with changes, but I somehow couldn't edit it) I asked the following question to eminent physics , none of which I knew their ideas in advance:

a: I subscribe to the MWI that contains many parallel universes that differentiate during every 'quantum event' (meaning that there are many perhaps an infinite number of copies of everyone that inhibits earth)

b: I subscribe to the MWI that contains many parallel universes that split during every 'quantum event' (meaning that there are many perhaps an infinite number of copies of everyone that inhibits Earth constantly being created)

c: I don't subscribe to a or b, because I think they are both false

d: I don't prefer any of the above

First some remarks
-2 said c/d, those where counted c 1/2, d1/2,
-one said I can't do anything with the poll, cause this has no empirical evidence and therefore has nothing to do with physics,
to avoid all accusations of being biased I didn't count him,
-one said c with a minor change, I don't subscribe to a or b,
to once again avoid all acusations of being biased I didn't count him in,
-one said I don't like both descriptions a or b, to me mwi just means unitary evolution but I chose b,
to avoid all accusations of being biased I did count him as b,
-one sais a,but with a very weak meaning of parallel meanings 'existing' (I don't really know what he means)
but once again to avoid being depicted as biased I counted him as a
-one didn't see a difference in a or b, but chose b because he thought the word 'split' was better (I now have added that in b you multiply as do universes, and c you already exist in many universes, but I think the description itself wasn't incredibly vague)
- one said d, He does however subscribe to the Everett relative state interpretation, and does think that all possible universes exist in 'some sense'
maybe some tiny percent of the (c's) or (d's) where subscribing to any mwi with parallel universes that wasn't mentioned in the poll,
But I think it's very unlikely (since I think the logical thing would be to make notice of that) These where the answers

C: 18
D: 7
A: 1
b: 2

the reactions came from (I promised to not make their names publicly, but since I don't mention who voted what, I don't think there is any harm in it):
David spergel, Carlo Rovelli, David polizer, David Finkelstein, Richard Muller, M.J. Rees, George F Smoot the Third, Goldreich, James Daniel Bjorken, Richter Burton, John Preskill, Leon N. Cooper, Robert Wald, James Binney, Yakir Aharanov, Andy Fabian, Ulrich Becker, Jim Al-Khalili, Frank Close, Frank Wilczek, Rodolfo Gambini, Jorge Pullin, John Baez. Donald Lynden-Bell, Rafael D. Sorkin, Mark Sredniki, Warren Siegel

the people that had extra comment and chose c said the following things:X1:

I think the many worlds interpretation is nonsence.
It does not survive Occham's Razor.

!me! when asked what his response was to mwi-ers answer <Occam's razor actually is a constraint on the complexity of physical theory, not on the number of universes. MWI is a simpler theory since it has fewer postulates> he said (I don't understand it, but I think some will) !me!

X1:
Any theory that predicts enormous numbers of copies of the universe
none of which are directly observable is introducing a huge
redundancy for explaining one small effect. That to me is quite wrong
and disobeys Occham's razor. It is easier to imagine that our concept
of a particle or a quantum are wrong in some way. There are already hints
of this from such matters as the spectra of diatomic molecules which
are radically affected by whether the nuclei are identical or not.
If as in Oxygen the nuclei are the same and of spin zero then every
other line in the rotational spectrum of the molecule is missing!
How does the molecule know its tiny nuclei are identical?

X2
Sure, my answer is definitely "c". I do not think that MW is good or
useful, in any variant.

X3
The many-worlds theory is a silly mistake, mistaking possibilities for actualities. It comes from the attempt to deny the statistical meaning of psi vectors
and regard them as real things, present in the individual system; and nevertheless to avoid the mistaken theory of "collapse" that such reification originally led to.
The theory that people actually use has one world, no collapse, transition probabilities, and incomplete descriptions.

X4
NO TO MWI

X5
My best answer is c. The theory is false because it is inconsistent.
a c/d vote said he considered it a argumentum absurdum, that we lack knowledge and he would have voted c if it not contained the phrase 'both false'

one d said: If a parallel universe is unobservable even in principle, then it's meaningless, according to established scientific method.
"Meaningless" means it's neither true nor false, since it's untestable

 

[Ken G]

Note sure if I got that metaphor.

The idea that the laws should be the same for all observers is not a law, it is a standard for laws. It is instructions for where to look for them, which makes them easier to find-- like a streetlight. It also says to look for laws that have a particularly "democratic" element, and to not consider as a law anything less. Note it is not falsifiable, so it is not itself a law.

Now wait, you say, of course relativity is falsifiable. But that's just any particular brand of relativity, like c being constant. Saying that the laws must be the same for all observers just means that if they only work for a subset, look for different laws that are the same for all-- that isn't falsiable. Physics went 2 centuries being content with laws that only work for inertial observers (it's obvious that F=ma is only a law for them), so it's a rather new standard, and we only adopt it because Einstein showed us we could. Einstein "found the keys in the streetlight", and asserted that we should always look for them there because it is easier to find them when you know where to look, and they are the kind of keys we are looking for.

To me the key here is interactions in between observers. Laws are somehow the consensus that observers can agree upon.

Yes, I agree this is the key. There would be no such thing as laws without observers, that's the part that rationalists don't like, and it's what MWI must reject.

What is the key for me is the democratic process or the negotiation, not possible equilibrium points for such process.

The road rather than the destination? Certainly sounds like an apt metaphor for science.

 

[Eqblaauw]

Dear mwi-forum participants,
the poll is closed, I've deleted my account, and there by any direct evidence of the poll, though everything I mentioned can be verified by the people mentioned,
if you want to use this poll for wikipedia or other sources you're free to do that,
I will vouch for that this poll is fully honest, and I also vouch for the statement david raubs poll most likely isn't.
I conducted the poll under the name 'Gertjan Kouweheuve' because I didn't wanted that any of my posts elsewhere would influence the poll. That's the only lie I permitted myself to, with this poll.
This also will be, if I can resist to login somehow, the last post I'm going to submit on this forum (physicsworld).
and as we say in some parts of holland 'houdoe!
Edo Blaauw

 

[nazarbaz]

Not necessarily. We don't have to choose between either saying that there is no "progress" in science, or that the ultimate goal is a "TOE." I don't think either of those views have much of a clue about the progress of science as it has been, and I don't see any reason to expect the progress of science to change suddenly now. Instead, I'd say these lessons of the history of science are pretty hard to deny:
1) scientific progress looks like gradually improving accuracy achieved by radically different ontological pictures, and
2) every generation thinks their own ontological picture is "on the right track", regardless of how completely it differs from previous ontological pictures.
Given these two seemingly uncontroversial facts, I think we can conclude that scientific accuracy converges, but scientific ontology does not. The main problem with postmodernist science is that it is generally not done by scientists, but it does have some useful "cautionary" remarks to make, as you say.

If we assume that scientific progress does exist, we must infer that our actual theories are effectively closer to the right picture than the previous ones (how close is a different question). In the history of science, there is both nicks and continuities.
The continuity is not an inner quality of the scientific effort, it's always super-imposed by rationalizations. The sociology and the factual development of science is misleading in that respect. We need to do the job and never expect to find it ready to use.

 

[Ken G]

If we assume that scientific progress does exist, we must infer that our actual theories are effectively closer to the right picture than the previous ones (how close is a different question).

Actually I would argue, as pertinent to MWI, that this is precisely what we can not infer. Making scientific progress does not require, and probably does not even suggest, that we are getting "closer to the right picture." I say this for three reasons:
1) scientific progress is always measured in terms of the increasing power it gives us over our environment and the increasing accuracy in our predictions, never by how right our picture is,
2) the pictures change so dramatically with each increment of accuracy and predictive power that no convergence of the "pictures" is in evidence, and there is also no direct connection between how much the picture changes and how much the accuracy improves,
3) there is no scientific evidence whatever that there is any such thing as "the right picture" that we could be moving "toward" in the first place.
These are important facts to bear in mind, because otherwise we fall victim to the same kinds of unjustifiable conclusions we turn up our noses at in pseudo-scientific endeavors.

 

[eqblaauw2]

If we assume that scientific progress does exist, we must infer that our actual theories are effectively closer to the right picture than the previous ones (how close is a different question). In the history of science, there is both nicks and continuities.
The continuity is not an inner quality of the scientific effort, it's always super-imposed by rationalizations. The sociology and the factual development of science is misleading in that respect. We need to do the job and never expect to find it ready to use.

I'm sorry but this must be one of the biggest pieces of crap I have ever read.
The newest theory (because at this moment I came up with it) is that the universe is a game played by Crocodile Harry, who has an Oedipuscomplex, and bad teeth.

I do however think a poll can indicate what the sciense community generally thinks at a certain moment, and that what the sciense community thinks generally is more correct then what the sciense community thinked in the past. However at this point the newer theory (MWI) isn't a populair theory at all (as I think I succesfully demonstraded). If and this is a big if, because I think if you do some reliable research you will find that there are other interpretations still more populair then MWI, MWI is the most popular theory that also doesn't mean it is a populair theory since the majority clearly rejects it. Then the most populair idea would be we don't know. But again this is an if, because the MWI theory isn't that populair at all (more populair then you would think at first notice when you read the rather extreme point of view, but still not close to being populair).

P.s. sorry that I couldn't resist login in again, but I really hate ********

 

[eqblaauw2]

I must say I quess the only positive thing of mwi for me is it really made me appreciate more my believe that everyone is unique, and that I'm mortal (if you really think about it being immortal is far more worse then being immortal, although that isn't nice either). And of course that I have made a reliable poll that may have helped people.

 

[BruceW]

However at this point the newer theory (MWI) isn't a populair theory at all (as I think I succesfully demonstraded). If and this is a big if, because I think if you do some reliable research you will find that there are other interpretations still more populair then MWI, MWI is the most popular theory that also doesn't mean it is a populair theory since the majority clearly rejects it. Then the most populair idea would be we don't know. But again this is an if, because the MWI theory isn't that populair at all (more populair then you would think at first notice when you read the rather extreme point of view, but still not close to being populair).

Many of the interpretations have several different possible meanings (MWI in particular). So a survey of the preferred interpretation is not very useful.
In most textbooks, CI is stated as the most standard interpretation, and in most situations CI is the easiest to understand. So for these reasons, I would guess CI is the most popular.

 

[Ken G]

The poll did refute claims that MWI has transitioned into the most popular interpretation, often cited as a consequence of better understanding of decoherence. I've tried to stress that this argument stems from an incorrect placing of "collapse" into opposition with "decoherence", when in fact the correct relationship is "decoherence sets the stage to interpreting collapse, whichever interpretation of collapse is used." CI works just fine in a "decoherence leads to collapse" framework, it was never one or the other.

Also, I still say that bashing interpretations is kind of silly, because all interpretations do is give us a new angle with which to look at our theory. We can bash the taking of an interpretation literally, to the point of using it to draw conclusions about how reality actually works, but that's not bashing the interpretation-- it's bashing what the interpretation is used to do. That's what I mean by creating a "world view" from MWI, and I think the history of science is all too clear on the pitfalls of doing that, even for interpretations not nearly as bizarre as MWI.

 

[eqblaauw2]

Many of the interpretations have several different possible meanings (MWI in particular). So a survey of the preferred interpretation is not very useful.
In most textbooks, CI is stated as the most standard interpretation, and in most situations CI is the easiest to understand. So for these reasons, I would guess CI is the most popular.

read the poll (it's on this forum), read it again if nessasary: it's offering a clear insight that probably the mwi's that are mostly advocated (including Bryce de Witt's, who really for the first time actually 'made' the mwi, and I think thereforethat his version would be the best way to refer to mwi) (with it's consequenses) aren't that populair at all among scientists. Also I got 6 reactions (one I didn't mention in the poll) which also got clear refutations of ANY mwi at all (and perhaps it isn't really relevant this includes a recent nobel laurate, and if I'm not mistaken one older nobel laurate, but I deleted the mail adress so I can't vouch for that) . And I'm sorry but I assume these aren't the only ones in the poll who voted c who rejected any mwi, because the people that did mention this did this while this wasn't itself being asked , actually I think it's most likely the majority did. And very most likely almost everyone rejected the copies idea. That being said 6 already is already more percentage wise then david raub's poll, but that I think isn't really relevant because of the very logical conclusion that in the randomly (only because of there reputation selected) participants in the poll reject MWI.
So this poll is in my opinion (and this opinion is so logical I don't think you can succesfully refuted) is relevant if you want to know what phycisists who think a lot about quantum mechanics think.
So please stop putting mist on very clear things. Also I must say I agree with Ken G's beginning of the post.
And I'm serious, it may sound a bit harsh, but will you (and I mean everyone on this forum) please stop posting things that just don't make sense (not because it's against my own personal opinion, but just because it doesn't make sense, and confuses people, you can of course advocate every opinion you like) so I don't have to come back here and post obvious posts again and again.

 

[kith]

This poll is nonsense.

First of all, your options are very poorly defined. An interpretation of a physical theory is always an interpretation of mathematical objects and their relations. So in order to get useful options for a survey, the first step is to present a generally accepted mathematical formulation of the theory. The second step is to choose different -well-known to be consistent- ways to interpret the mathematics in physical terms. And the third step is to draw conclusions from these interpretations.

You only did the third step, which of course is not do-able in a satisfactory manner if you skip the first two. So you chose two popcultural statements which have something to do with the MWI, and added the options to contradict them or be indifferent. How shall such a survey be able to produce reasonable results?

Plus, your survey is not at all representative. Why do you think these 30 people represent the physics community?

 

[eqblaauw2]

This poll is nonsense.

First of all, your options are very poorly defined. An interpretation of a physical theory is always an interpretation of mathematical objects and their relations. So in order to get useful options for a survey, the first step is to present a generally accepted mathematical formulation of the theory. The second step is to choose different -well-known to be consistent- ways to interpret the mathematics in physical terms. And the third step is to draw conclusions from these interpretations.

You only did the third step, which of course is not do-able in a satisfactory manner if you skip the first two. So you chose two popcultural statements which have something to do with the MWI, and added the options to contradict them or be indifferent. How shall such a survey be able to produce reasonable results?

Plus, your survey is not at all representative. Why do you think these 30 people represent the physics community?

I don't think they represent it, I think it's only a very good indication because they where randomly selected and all where very much highly regarded, and the most likely thing is that these participants where in fact more in favor in respect to 'normal' physicists, because they where all theoretical physicists and cosmologists. Maybe according to your definition a good poll must have a formula, but I think the physicists where aware of the formula of bryce, and if not they knew that A and B contained copies and parallel universes and that's the main thing, and if they did subscribe to this general notion they WOULD mention it (think, I didn't do a poll course, but this is **** logic, maybe it does not meet your scientific standards, but it does meet the standards of sense and that's what matters and should be the main reason you should have scientific standards in the first placee) for the rest your post is clearly a reaction of someone who has in this case only thinks in general rules and not by himself (a bit like someone who stops at a red light, when you are the only person in the world) cause this poll no matter how you interpret it, does refute raub. And I'm not going to repeat the obvious reasons besides what I said in this post. And if you like you can of course do step 3 yourself, though I suggest you do it with some logic.
byebye

 

[Hlafordlaes]

... And I'm serious, it may sound a bit harsh, but will you (and I mean everyone on this forum) please stop posting things that just don't make sense (not because it's against my own personal opinion, but just because it doesn't make sense, and confuses people, you can of course advocate every opinion you like) so I don't have to come back here and post obvious posts again and again.

I liked your poll as a contribution to the discussion, but I like the discussion even more. There are many useful insights apart from strict MWI-or-not, not least of which relates to the relative value, and role, of theory. Fra's musings on some deeper work he is doing, and Ken G's trademark cautionary reasoning are lots of fun! So, no need to come back and post anything over and over again; let's just enjoy the journey, whatever the destination.

 

[eqblaauw2]

I liked your poll as a contribution to the discussion, but I like the discussion even more. There are many useful insights apart from strict MWI-or-not, not least of which relates to the relative value, and role, of theory. Fra's musings on some deeper work he is doing, and Ken G's trademark cautionary reasoning are lots of fun! So, no need to come back and post anything over and over again; let's just enjoy the journey, whatever the destination.

thank you for your nice comment. I just wanted to say that I find it insulting (and I normally dislike that word) that someone said that the poll I conducted is nonsense , especially when someone (me in this case) has really put energy in it. And the statement is also utterly ridiculous (and for this I have given plenty enough arguments). I hope most will agree, and I hope this will be the last post. I still got some evidence (a mail I sent to all the participants to inform them about the results, so if you really doubt my honesty you can send me a message and I will post it to you).
I can also say this poll is conducted because I really wanted to know what highly regarded physicists thought about this subject, and that I used it also for my own information and I wouldn't let myself to be easily convinced, because I really wanted to get a true reliable result to get the uncertainty out of my head. I also will say that I would be depressed if everyone voted a or b, but that would be the price I had to pay. Luckily for me, and some certain person with a psychosis, this is not the case.

 

[nazarbaz]

Actually I would argue, as pertinent to MWI, that this is precisely what we can not infer. Making scientific progress does not require, and probably does not even suggest, that we are getting "closer to the right picture." I say this for three reasons:
1) scientific progress is always measured in terms of the increasing power it gives us over our environment and the increasing accuracy in our predictions, never by how right our picture is,
2) the pictures change so dramatically with each increment of accuracy and predictive power that no convergence of the "pictures" is in evidence, and there is also no direct connection between how much the picture changes and how much the accuracy improves,
3) there is no scientific evidence whatever that there is any such thing as "the right picture" that we could be moving "toward" in the first place.
These are important facts to bear in mind, because otherwise we fall victim to the same kinds of unjustifiable conclusions we turn up our noses at in pseudo-scientific endeavors.

That's exactly what I'm criticizing. The "dichotomy" of the "scientific accuracy" and the "ontological claims" behind it is not obvious to me. It is meaningless to talk about accuracy and progress if our well known theories doesn't tell us something about the world. What accuracy do we talk about otherwise ? Consistency with its own premise ? Aren't confusing science and circular thinking ?
I'm not pretending that our actual theories are fully describing the whole process, I'm not even assuming that they will one day, I'm just asserting that our actual knowledge is describing correctly some aspects of nature. No way to save the scientific point of view without this assumption, and no way to explain the efficiency of technology.
But science is neither reducible to philosophy nor technology even if it needs them to think it's own status and direct its practice. It is all about linking propositions and objects consistently despite what the diffuse postmodernist culture tells us. In that sense, the most dangerous threat science must face is the epistemological nihilism that some esthetes instilled in our understanding of it.

 

[nazarbaz]

I'm sorry but this must be one of the biggest pieces of crap I have ever read.
The newest theory (because at this moment I came up with it) is that the universe is a game played by Crocodile Harry, who has an Oedipuscomplex, and bad teeth.

I do however think a poll can indicate what the sciense community generally thinks at a certain moment, and that what the sciense community thinks generally is more correct then what the sciense community thinked in the past. However at this point the newer theory (MWI) isn't a populair theory at all (as I think I succesfully demonstraded). If and this is a big if, because I think if you do some reliable research you will find that there are other interpretations still more populair then MWI, MWI is the most popular theory that also doesn't mean it is a populair theory since the majority clearly rejects it. Then the most populair idea would be we don't know. But again this is an if, because the MWI theory isn't that populair at all (more populair then you would think at first notice when you read the rather extreme point of view, but still not close to being populair).

P.s. sorry that I couldn't resist login in again, but I really hate ********

Wonderful theory... Stop thinking, that's the climax of science... Crocodile harry...

 

[Ken G]

The "dichotomy" of the "scientific accuracy" and the "ontological claims" behind it is not obvious to me.

Then I'll give you some examples:
1) Aristotelian vs. Newtonian vs. Einsteinian gravity. The represents a clear sequence in converging accuracy-- Aristotelian is not quantitative at all, Newtonian is magnificently accurate for celestial motion. In most applications, Einstenian gravity adds little in the way of increased accuracy, yet holds that gravity is not a force but a bending of spacetime-- two more different ontologies would be hard to find. Einstein's theory essentially widens the domain of high accuracy outcomes, so is clearly moving toward some kind of convergence in accuracy, but the ontology borrows nothing at all from Newton other than some dependence on mass.

2) Classical vs. Quantum mechanics-- again we have two theories that agree essentially completely on some scales of phenomena, but QM extends the domain of high-accuracy outcomes to smaller systems. So there is clearly a convergence of accuracy happening when you pass from CM to QM, but the ontology is completely different once again-- we go from motion being described by position as a function of time to motion being ruled by inobservable mathematical entities that only pass into the real domain when they are in some sense forced to choose a definite outcome by their environment.

3) Thermodynamics vs. Statistical Mechanics-- here we have general laws about how heat and temperature behave, accurate for large systems in thermodynamic equilibrium, transitioning into statistical rules that apply to systems of particles. The latter extends the accuracy of the predictions to systems with fewer particles and different kinds of interparticle forces. The ontology is again completely different-- thermodynamics is a continuous description, statistical mechanics is an atomic description.

And the list goes on. In each case, we find advances that extend the domain of accuracy from one theory to the next, exhibiting a clear convergence of predictive reliability, so show a clear "arrow of progress" in regard to predictive power. Yet, these advances are accomplished via radical shifts in the basic ontology, exhibiting no tendency at all for the ontologies to "converge" other than the most basic consistency requirement. And today, we have research into "string theory", another complete change in ontology, targeted at achieving accuracy improvements that are so tiny they as yet cannot even be probed by our technology.

I'm not even assuming that they will one day, I'm just asserting that our actual knowledge is describing correctly some aspects of nature.

All well and good to say, but what does this really mean? When gravity was a force, was that "actual knowledge" of some aspect of nature, even when it became a curvature of spacetime? Or when it is something that strings are doing? All we ever get to know is the accuracy of our predictions.

 

[eqblaauw2]

STRAWMAN ALERT

Wonderful theory... Stop thinking, that's the climax of science... Crocodile harry...

-

stop thinking? where did I exactly promote this idea? you seem to suggest that newer theories are better then older ones, I don't agree with that, and I gave an example where that idea could logically lead to. This of course doesn't mean that newer theories not in the majority couldn't be true, it just means that you cannot say that they are most likely better because they are newer.If you meant to say that newer theories close to being generally accepted theories or at least have a majority are more likely to better then the older theories I fully agree with you. This standpoint of course doesn't mean that newer theories not hold by the majority of physicists couldn't be true, it just means that you cannot say that they are most likely better because they are newer.
P.s. sorry for being a bit rude in the previous post. But I stay with my point, without the rudeness.
I also wanted to say that I also got a email from an eminent physicist (I don't now who it was anymore, so you may just ignore it if you don't think it's useful, because I've got serious lack of evidence to back it up, I know it's true, but I can understand that you won't believe I got this email, this is a really long sentence ) that said that mwi isn't inconsistent. That's another argument then usually is being brought foreword , but you might look into that, if you want, if you just think it's evidentially bull it would be a little pointless.

 

[Hlafordlaes]

... And the list goes on. In each case, we find advances that extend the domain of accuracy from one theory to the next, exhibiting a clear convergence of predictive reliability, so show a clear "arrow of progress" in regard to predictive power. Yet, these advances are accomplished via radical shifts in the basic ontology, exhibiting no tendency at all for the ontologies to "converge" other than the most basic consistency requirement...

All well and good to say, but what does this really mean? When gravity was a force, was that "actual knowledge" of some aspect of nature, even when it became a curvature of spacetime? Or when it is something that strings are doing? All we ever get to know is the accuracy of our predictions.

I think the confusion I often have is, on one hand, the obvious value of our everyday mental models, based on unexpressed "theories" about how the world works, and which by their practical results seem to indicate we are in close touch with reality. OTOH we have scientific theories that delve into modeling that which is beyond the everyday and shift radically in ontology, as you point out in 1-2-3.

I tend to transfer the idea that we are capable of mentally modeling our environment accurately enough to survive and prosper, to the misguided idea that scientific theories converge on a true description in similar fashion. Still, it does seem that at least in part we do uncover new realities, such as, say, a predicted particle that is found and measured only as a consequence of a theory. That isn't only predictive power, is it? I mean, aren't we also uncovering something "real" and converging in some way on a more accurate description?

 

[Ken G]

Still, it does seem that at least in part we do uncover new realities, such as, say, a predicted particle that is found and measured only as a consequence of a theory. That isn't only predictive power, is it? I mean, aren't we also uncovering something "real" and converging in some way on a more accurate description?

I think what you're basically saying is, if we drop a block, a coin, a doll, a horse, a jack, and a toy car, and they all fall, have we not learned something about our reality that allows us to expect that when we drop a book it will fall too? I think the answer is yes-- we learn consistencies of experience, which we can then reliably generalize to "similar" situations via some sort of unifying theory, even if we can't ever be exactly sure what is going to prove similar and what isn't (like, when we drop a leaf in the wind and it blows away instead of falling). But consistencies of experience that we give a name to, like "gravity" or "conservation of energy" are not quite the same thing as an ontological description. What is it about gravity that requires that things fall? What is it about energy that requires it be conserved? Those are the kinds of questions, the "what's really going on here" kinds of questions (similar to interpretive questions about quantum mechanics or any other theory), that never seem to reach any kind of convergence in the progress of science. I think it is easy to overlook that, and tell ourselves things that simply aren't in evidence, when we imagine some kind of "destination" for scientific understanding rather than the simple process of doing science. That's why I'm skeptical of using science to create worlds views, such as a belief in the actual existence of "many worlds."

 

[Hlafordlaes]

... and a toy car...

gosh, Ken G, gimme a break

... That's why I'm skeptical of using science to create worlds views, such as a belief in the actual existence of "many worlds."

and he deftly puts the thread back on track...