It remains embarrassing that physicists haven’t settled on the best way of formulating quantum mechanics (or some improved successor to it). I’m partial to Many-Worlds, but there are other smart people out there who go in for alternative formulations: hidden variables, dynamical collapse, epistemic interpretations, or something else. And let no one say that I won’t let alternative voices be heard! (Unless you want to talk about propellantless space drives, which are just crap.)
So let me point you to this guest post by Anton Garrett that Peter Coles just posted at his blog:
It’s quite a nice explanation of how the state of play looks to someone who is sympathetic to a hidden-variables view. (Fans of Bell’s Theorem should remember that what Bell did was to show that such variables must be nonlocal, not that they are totally ruled out.)
As a dialogue, it shares a feature that has been common to that format since the days of Plato: there are two characters, and the character that sympathizes with the author is the one who gets all the good lines. In this case the interlocutors are a modern physicist Neo, and a smart recently-resurrected nineteenth-century physicist Nino. Trained in the miraculous successes of the Newtonian paradigm, Nino is very disappointed that physicists of the present era are so willing to simply accept a theory that can’t do better than predicting probabilistic outcomes for experiments. More in sorrow than in anger, he urges us to do better!
My own takeaway from this is that it’s not a good idea to take advice from nineteenth-century physicists. Of course we should try to do better, since we should alway try that. But we should also feel free to abandon features of our best previous theories when new data and ideas come along.
A nice feature of the dialogue between Nino and Neo is the way in which it illuminates the fact that much of one’s attitude toward formulations of quantum mechanics is driven by which basic assumptions about the world we are most happy to abandon, and which we prefer to cling to at any cost. That’s true for any of us — such is the case when there is legitimate ambiguity about the best way to move forward in science. It’s a feature, not a bug. The hope is that eventually we will be driven, by better data and theories, toward a common conclusion.
What I like about Many-Worlds is that it is perfectly realistic, deterministic, and ontologically minimal, and of course it fits the data perfectly. Equally importantly, it is a robust and flexible framework: you give me your favorite Hamiltonian, and we instantly know what the many-worlds formulation of the theory looks like. You don’t have to think anew and invent new variables for each physical situation, whether it’s a harmonic oscillator or quantum gravity.
Of course, one gives something up: in Many-Worlds, while the underlying theory is deterministic, the experiences of individual observers are not predictable. (In that sense, I would say, it’s a nice compromise between our preferences and our experience.) It’s neither manifestly local nor Lorentz-invariant; those properties should emerge in appropriate situations, as often happens in physics. Of course there are all those worlds, but that doesn’t bother me in the slightest. For Many-Worlds, it’s the technical problems that bother me, not the philosophical ones — deriving classicality, recovering the Born Rule, and so on. One tends to think that technical problems can be solved by hard work, while metaphysical ones might prove intractable, which is why I come down the way I do on this particular question.
But the hidden-variables possibility is still definitely alive and well. And the general program of “trying to invent a better theory than quantum mechanics which would make all these distasteful philosophical implications go away” is certainly a worthwhile one. If anyone wants to suggest their favorite defenses of epistemic or dynamical-collapse approaches, feel free to leave them in comments.
Sean Carroll:
———
It’s not clear in what sense Everett is ontologically simpler than specific versions of De Broglie-Bohm. To start with, I’m not entirely sure what to think about Garrett’s (or similar) general arguments and ideas which don’t seem very clear or committal about what “hidden variables” are. But I mean a theory which posits there are particles (or some form of matter/energy, not to exclude string theory, etc.) which exist in fairly small localized regions and have definite positions and trajectories in spacetime. That’s the ontology: there are things which move around. We do see “macroscopic” objects moving around, so it’s not a fantastic leap to be making, to suppose they are composed of more fundamental objects which do the same. One piece of evidence we do actually have is that we always see definite results of experiments, whereas we don’t see “alternative” results of experiments that allegedly (according to Everett) exist in other worlds.
Okay, now your claim above is apparently “but you also say the wavefunction exists, so that’s two things (particles in physical spacetime AND a 3N-dimensional wavefunction on configuration space) instead of just one thing: the wavefunction.” Two is bigger than one, so it’s not as “ontologically simple,” if that’s how we’re counting things.
But this strikes me as you making commitments for Bohmians which they don’t need to make. You can say the wavefunction is a law of motion, a law of physics. This “thing” is mathematically represented on configuration space, but the simple fact is that there are configurations of particles in ordinary physical spacetime. Do laws of physics need to exist, as something over and above the physical world they “govern”? No, they don’t need to be like that at all. They can be a Humean “best system,” statements which are true that we use for describing/explaining the facts comprehensively/informatively/efficiently (in ways that are comprehensible to the humans which are making these descriptions for themselves), not things that are extra, which we’re adding to the ontology somehow.
Craig Callender has given a talk (~1 hour) giving some more details about this sort of view, specifically how it relates to de Broglie-Bohm. (Plenty of others have also talked about it for a long time now … and it’s not clear what part of the message isn’t getting through to “the other side,” such that they keep saying things like you’re saying.)
So, would you agree, given that, that it’s not at all obvious how we could compare the “ontological simplicity” of them. There are wavefunction-monists (Everettians), and there are monists about particles in spacetime (“Bohmians,” with a somewhat Humean understanding of what laws are). Those are very different sorts of ontologies. You might say “oh, but there are lots of particles, in lots of places, not just one thing” but that’s obviously a very silly mistake. And I could say “but there are supposedly lots of branches of the wavefunction, not just one thing, yet we only ever have empirical evidence of one branch.” But at least, if you took some care to flesh out some of this, we might be able to ask better questions about what’s real, along with what counts as a simpler realistic theory.
For instance, why do you have this “thing” in a high-dimensional configuration space, if that’s NOT a configuration of stuff in the low-dimensional space? What would it even mean to say that’s the “real” thing somehow, when the only experiences we actually do have are of the low-dimensional space, which it would be nice to summarize in the most efficient and informative way as we can? It would be nice to at least have a clear understanding of what your ontology actually is, however simple you think it may be. I’ve never gotten that (to my satisfaction, so I can understand it) from any proponents of Everett.
What do you think of my theory Sean?
@philophysique
@Joseph Marino
Regarding the time-symmetric approach, you say:
“IMO, it’s the best alternative to the Everettian approach currently on the market.”
But actually a time symmetric approach is completely compatible with Everettian MWI. The worlds don’t just ‘split’ in the future, but also in the past. But in the past direction there is less splitting, because there is less entropy in that direction. You can let up lots of experiments, like the double slit, in a reverse way to better understand how this would work. (Emit photons from an array, and detect them at a point,and then add up the points of emission when a photon was detected. It will form a ‘diffraction paytern’ of emission.) Most experiments work the same way backwards, so why wouldn’t worlds split that direction also.
But the use of the word ‘splitting’ and talking about multiple ‘worlds’ is the most confusing part to people. Nothing really splits. The universal wave function is still a single wave function. This does not change no matter how isolated the branches appear to be.
Why would one insist in an underlying determinism when the evidence of QM excludes this? Why would one insist that reality exists before observation when the evidence Is that a local reality cannot exist? Why would one insist that a correlation depends on hidden information when the evidence is that no information can be shared at such distances in space time? Why would one cobble on MWI to cling to determinism and reality in spite that it can never be proven by any experiment and is unneeded to obtain correct results. if yu insists that a reality exists outside of our subjective observations and hidden from any possible observation, I believe you are confusing Religion with Science. You would have been in excellent company a century ago. However, Einstein was either wrong and God does play dice or Einstein was wrong to believe that a God would make a universe that operated deterministiallly like a clock. Or Einstein was wrong to invoke God because there is none at least for science.
God is the correct solution when one doesn’t understand but wants to feel that one has the answers. If that works for you great, but it seems contrary science.
All formulations of QM give the same answer. The answers are those that experiment confirms. After the double sit experiment answer is that the universe is neither deterministic nor does it have to be real in the way we perceive. A global reality of space and time is disproven: a local reality is subjective and requires observation. Neither MWI nor God nor some hidden real thing must be invoked. No evidence for any MWI or underlying determinism and a century of scientific results against such suggests that even Einstein were he to see the result of EPR would have agreed that God, if he were, must play dice – or at least would be watching “Wheel of Fortune” like all the other old white haired men.
If the justification is ultimately pragmatic in the way you suggest, do you think we should also regard the simpler theory as more likely to be *true*? If so, how does that follow?
@Cede
Your statements are ridiculous. Whether you like it or not, there is no doubt that there are scientific realities that are impossible for us to directly observe. This does not mean we should rule them out. Perhaps we can illucidate these realities using math and science, even if there can never be any real proof.
How can you even believe that the past was a ‘real place’, with ‘real things’ in it? You can not communicate with it. You can’t send any information to it. It could be a product of your randomly assembled imagination. The same, of course, goes for the future. You can try to send information into it, but how do you know it goes there. No matter what point in time, it’s all in your head anyway.
Many people get caught up in this kind of thinking, and that just holds us back from truly understanding how it all works.
One aspect of wave propagation appears quite bizarre and difficult to comprehend. As an example, I am sending this comment thru em waves of certain frequency which will be propagated thru internet at all the places in space on Globe subject to restrictions imposed by speed of light. Now millions of people at different millions of places can reproduce and read the comment wholly. It implies an em wave manifest itself WHOLLY and TOTALLY at at all places. This leads me to think over a nos of unresolved issues viz
I) What is in an em wave which manifest itself wholly and totally at different points in space at same instant of timed? To state electric and magnetic field does no amount to solving mystery any more since than same issue remains for electric and magnetic field.
ii) To state that wave has non-local character raises the issue if non-local behavior is bound by laws of Physics viz speed of light. For example, Can two entangled particles can influence each other instantly when they are located 100 light years away
@Vinod
“Can two entangled particles can influence each other instantly when they are located 100 light years away?”
At least 100 years earlier they were in the exact same location. Particles in the exact same location can influence each other, and particles can maintain their properties until they interact with some other particles.
As long as you think of time as going in two directions instead of just one, there is nothing really complicated going on with entanglement. The influence can be conceptualized as traveling back in time with one particle until the time of interaction with the other. Then, that influence travels forward in time with the other entangled particle. Nothing really spooky, or at a distance.
@Rick Extending your arguments a bit further, all the particles of universe were at same location in remote past when universe came into existence from Big Bang. Many of the particles might not have had interaction with another particles implying their entanglement might be intact which had occurred when universe was vey very small. Now many particles located on a galaxy about 1 billion light years away might be having entanglement intact with some particles on earth. If the particle on that galaxy , due to some process, annihilate into energy, corresponding particle on earth should also convert to energy instantly. But this not been observed.
Secondly, your concept of entanglement, unbounded by speed of light, is dependent upon two way movement of arrow of time. But arrow of time is only forward moving. Except, Scific, arrow of time has not been observed in backward direction
@Vinod Sehgal
I suggest you do a little reading. Nearly all of physics has been shown to be symmetric with respect to time. This is actually not a controversial subject. The “arrow of time” is more about probability.
Also, it is quite true that entangled particles traveling through the universe for billions of years should still behave as if they were produced seconds ago.
Cheers!
It’s a good piece overall but ultimately very confusing to tie in Bell’s inequalities with spacetime. Nonlocality is a consequence of contextuality in Minkowski spacetime, the more confusing aspect of QM is contextuality, that the correlations depend on the configuration of the two measurement systems whether they occupy points that lie on a space-like interval or not. It is a strange result that the configuration of logically/statistically independent indicator functions or measurement devices affect the correlations between the independent results of the two experiments. The correlations inherent to the particle preparation should be the only contributing factor to the measured correlations, yet this is demonstrably not the case. Any hidden variable theory would have to resolve this issue before we could even check if such a theory had “local” or “nonlocal” character when embedded in Minkowski spacetime.
I’m aware that contextual hidden variable theories exist as well as modal interpretations of QM that address this, but in this piece, Nino seems to want to push for hidden variable theories that work for isolated systems. Measurement indicators of observables shouldn’t affect the outcome of other independent measurement indicators of observables in the same system. A system with n such measurements should be able to be analyzed with n-1 such measurements, n-2 and so on down to 1 while still being consistent between all such analyses. This is not true in contextual hidden variable theories.
@Rick: [[But actually a time symmetric approach is completely compatible with Everettian MWI. ]]
Sure, no question. E.g., Lev Vaidman combines TSVF with Everettian MWI.
But my question is: Why consider it as merely a supplement to MWI when you can solve all the ‘quantum mysteries’ very cleanly with time-symmetry alone?
BTW, I don’t have any metaphysical objections to a many-worlds ontology per se. Nor do I think the Born rule is a problem for MWI. No, my main objection is that MWI is psi-ontic. It’s not exclusively an issue for MWI (for example, Bohm is also psi-ontic), but it’s an issue.
The advantage of the time-symmetric approach is that you solve all the quantum ‘paradoxes’ with a locally realistic theory in which the wavefunction is simply a calculation device representing states of knowledge, rather than some exotic non-spatio-temporal element of the ontology. In the time-symmetric approach, the ontology consists exclusively of concrete spatio-temporal entities, interacting locally in spacetime.
The ‘price-tag’ is accepting that a certain restricted type of ‘retrocausality’ happens in the quantum world – specifically, entanglement scenarios – where a future (‘macroscopic’) boundary condition constrains the value of a prior hidden variable. But this is not really much of a price at all, since it is already suggested by GR and other core physics. Time-symmetry means that Measurement is just a time-reverse of Preparation: just as Preparation imposes a (initial) boundary condition on future beables, Measurement imposes a (final) boundary condition on past beables. Ken Wharton’s papers go into the details of how this all works out elegantly.
Don’t get me wrong – I think the Everettian program has a lot going for it. But I do think the time-symmetric family of interpretations are worth exploring on their own as complete solutions – not just as supplements to MWI or other approaches.
I know your not going to like to hear this, but there are too many worlds in the MWI! I forget the name of the theory, but there was a story I read about some physicist running into each other on a train once. They ended up developing a theory that said that two universes could collide, and it would end up causing an explosion that would create another universe. You probably know a lot more about this theory than I do.
Okay, lets assume that quantum experiments are being performed all around the world by multiple scientist. That would make a lot of universes, and it would make it more likely that the universes being made could collide. Suddenly, that means running too many quantum experiments is dangerous, and we are well down the path of self destruction. Then we have not observed anything like that occurring, and every day run of the mill quantum experiments are safe, no matter how often we run them.
Then the MWI would have to be wrong, just because we haven’t all exploded into the birth of a new universe yet.
@philophysique
“But my question is: Why consider it as merely a supplement to MWI when you can solve all the ‘quantum mysteries’ very cleanly with time-symmetry alone?”
I am a true believer in the time-symmetry approach, but without a supplement like MWI, I just don’t see how the singular options for initial and final conditions arise.
Where do the probabilities come from? Why is any particular ‘future’ condition selected to be the ‘real’ one that interacts with the ‘real’ past? (And, of course, the time symmetric situations also).
In my opinion, adding MWI seems to tie up most of the loose ends. Usually lots of possible final conditions, usually less possible initial conditions. But they all still occur.
One thing I wonder about is how it relates to special relativity. It is obvious that the temporal order of two independent quantum interactions is relative, and depends on the observer.
But what about a singular quantum interaction? In order for the temporal order to be reversed, the observer would have to be traveling faster than the speed of light.
But then the observer would interact with the time symmetric reality. They would be interacting with a world filled with antimatter (because all electrons are positrons, etc.) and be annihilated.
@ Rick Two straightforward queries and expect straightforward response
I) Can time flow in backward direction? If yes, Has time moved in backward direction in any part of universe?
ii) Does our earth contains some particles which are still entangled with some particles located thousand and millions of light years away?
@Vinod Sehgal
1) Time doesn’t flow. It’s more like a street with two directions. But it is also intermixed with the other 3 directions of space as spacetime. Every particle that is “moving” forward is also “moving” backwards as a symmetrical particle.
Yes. There are certainly particles located thousand and millions of light years away that are entangled with particles on Earth. In some way, all particles are entangled with each other. But you couldn’t describe how unless you set up the experiment where the interaction occured.
Re. Bob D’s remark in connection with Anton’s: “The predictions of QM violate causality in some situations and have been verified experimentally.”, Bob is correct. In the examples that Anton give in reply, there is no acausal transfer of information/energy. Thus, QM does not PHYSICALLY violate causality. You might just as well say that QFT violates causality, as virtual processes do!
Re. MWI, it’s interesting, but you still have to answer the question as to why we are conscious of only one path. If one adopts the view that if something is not observable, it does not exist, as Einstein did for the aether, otherwise it becomes, it would seem, more a religious decision. In a way, going for MWI could then be like adopting von Neumann’s idea that it is one’s consciousness that chooses the one outcome out of all the other possible outcomes.
Really, the fascinating thing about QM is our failure to “humanise” it! The predictions have proved immensely successful, that we have to accept it for what it is. Why should the microscopic world be anything like our classical world? I, myself, just feel very privileged to touch on the world of QM, a world with an alien culture, whose language we still struggle with after almost a century. Clearly, I’m happy with the Copenhagen interpretation. Of course, there may in the future be an even more fundamental and powerful theory, with physical consequences that justify it, that will encompass QM, just as Relativity encompassed Newton. But, until then, most other interpretations, including MWI, can only be regarded as speculation or, at best, philosophical, and not at all well grounded as far as physics is concerned.
@Rick: [[Where do the probabilities come from? Why is any particular ‘future’ condition selected to be the ‘real’ one that interacts with the ‘real’ past? (And, of course, the time symmetric situations also).]]
The time-symmetric approach is psi-epistemic. That means the wavefunction only represents our partial knowledge of a deterministic one-world reality. Thus, once we allow ‘retrocausal’ boundary conditions, QM reduces to traditional statistical mechanics. The probabilities represent our subjective ignorance as epistemically-limited, time-bound agents. The future condition – the measurement setting – is unique and thus uniquely determines the value of the relevant hidden variable in the past. Since we can have no knowledge of that value until we make the measurement later, we cannot predict experimental outcomes perfectly. But considering the whole 4-D situation ‘all at once’ (block-world style), the relevant past and future boundary conditions jointly constrain the values of all beables in the system.
So, on this picture, there is no need for ‘objective collapse’. No need for a physical process to select one ‘actual’ result among a set of all possible worlds. Rather, there is only one real deterministic world and the wavefunction is merely a housekeeping device that summarizes our imperfect state of knowledge about it. As we know, Bell’s theorem and other no-go theorems render such approaches impossible UNLESS we allow events in the future light cone as potential common causes and not confine ourselves to the past light cone.
Again, I still regard MWI as a top contender – especially in the psi-ontic category. But the time-symmetric local-hidden-variables approach is psi-epistemic – which means all puzzles about probabilities and apparent nonlocality evaporate in a puff of Bayesian updating.
Sean,
What you say in the link at the words “ontologically minimal” is exactly right, but why use the word “ontological” there? It’s the complexity of the theory that’s being minimized, not – as you say – the weight of the objects in the wheelbarrow.
Nick,
The simpler theory is more likely to be true because likely is an epistemic term whose meaning is determined in part by simplicity. It’s a hidden semi-tautology. To quote a famous philosopher, “the supreme goal of all theory is to make the irreducible basic elements as simple and as few as possible without having to surrender the adequate representation of a single datum of experience.”
philophysique,
I love the whooshing sound your arguments make as they go over my head. Why is an ontic wavefunction, with electrons and photons mere patterns within it, more troublesome than ontic electrons and photons plus an epistemic wavefunction? But still, from down here in layman land, your retrocausal family of views looks pretty good.
@Rick
“Yes. There are certainly particles located thousand and millions of light years away that are entangled with particles on Earth”
If those entangled particles located millions of light years away annihilate into energy or undergo any change in property or transform to some other particle, will their entangled pairs at earth also undergo corresponding change instantly or vice versa of this? Has this phenomena ever been observed any where in universe?
@
“Time doesn’t flow. It’s more like a street with two directions. But it is also intermixed with the other 3 directions of space as spacetime”
The way there is controversy regarding Psi ontic or psi epistemic aspects
of quantum states similarly an issue arises regarding ontic or epistemic affairs for mixing of time with 3 dimensions of space also. I think intermixing of space with time is epistemic from observational perspective only. If at ontic level, space and time are sought to lose their identity as result of speed vs a vs speed of light or change of reference, space and time shall get deprived of their independent identity. Without an independent identity of space and time, how matter and energy can exist. I am not refuting the relative values of space and time but these values arise due to change in speeds relative to observer from purely observational perspective. In other words, phenomena of relativeness of space and time is true, of course, but against . That the backdrop of the identity of independent space and time. What is that independent space and time is not yet known to Science.
“Every particle that is “moving” forward is also “moving” backwards as a symmetrical particle”
But how particle will move “backwards “in time.? Only when it moves faster than speed of light. Has motion of material particles at speeds faster than speeds of light been reported by Scientific community?
@ Nick
@ Philophysique
What is the significance of time symmetry if no material particle can move in time reversal direction due to restrictions imposed by speed of light?
@ Dick fong
” If one adopts the view that if something is not observable, it does not exist, as Einstein did for the aether, otherwise it becomes, it would seem, more a religious decision”
Extending above arguments a bit further, Einstein treated physicality of space as nil and even today after 100 years of GR, nothing is known about physicality of space. If physicality of space is nil there is no meaning of physicality of ‘curvature’ of space. does it implies curvature of space is only an epistemic concept without any ground level reality? But observational evidence does support ontic aspects of curvature which means space is not physicality free as was presumed by Einstein. Further if there is some physicality of space implying some granular structure, Riemannian manifold of smooth space is gone which was one of the basic assumption of GR
It is “ontologically simplest” to model the electron as an extended distribution of super current held in force balance through the application of known CLASSICAL laws. It does not self-interact modeled as a spherical delta function as there is no internal field from the electron; this is also fully consistent with EXPERIMENTS going back a near century on charged spheres.
The roots of this “non-radiation condition” can be traced back to the early 1900s. This model offers the most explanatory power by far of any competing theory, uses no free parameters, only constants, provides a PHYSICAL explanation for what an electron is, is internally consistent, and exactly predicts properties where all other theories fail miserably.
If one really believes in the veracity of the above statements of “ontologically simpler” models, the claims about MW in this blog are thereby demonstrated to be false. Perhaps it is more satisfying to live in a world of fantasy where all possibilities are true? I must admit that I am a big fan of Star Trek as well.
Paul Torek asked: “I love the whooshing sound your arguments make as they go over my head. Why is an ontic wavefunction, with electrons and photons mere patterns within it, more troublesome than ontic electrons and photons plus an epistemic wavefunction? But still, from down here in layman land, your retrocausal family of views looks pretty good.”
I don’t know what philophysique’s answer would be, but I would say because the particles themselves more closely represent our direct phenomenological experiences while the wavefunction is an emergent pattern from those same experiences. Said more clearly: We can never measure a wavefunction directly, we just infer from the statistics of repeated measurements on similarly prepared systems that each measurement samples from an underlying complex distribution. In such a view the reality of the measurement is necessary as well as the objects directly implicated in the measurement (the particles, though perhaps to a lesser extent). However, the wavefunction is a convenient way of representing this data such that the underlying patterns are in their most tractable form. You don’t measure the wavefunction really at all even with quantum tomography, you use it as a model to frame your data.
The wavefunction could very well be ontic, but one should not confuse the simplicity of the mathematical language chosen to express that pattern with the simplicity of the pattern itself. You could formulate all of quantum mechanics without ever referencing the concept of a “wavefunction” and the patterns would still be just as simple, only the language to encode them would be significantly more complicated. I don’t know how you could reformulate quantum mechanics as a theory that does not make the same references to the measurement objects (electrons and photons). Quantum mechanics is applied to these objects, their interpretation is not a free parameter in the theory. Technically the particles we measure are not even elements of quantum mechanics as a theory. I am ignoring QFT for the sake of this discussion, which would change the status of these objects (electrons and photons) in relationship to measurement.