I find Eliezer’s insistence about Many-Worlds a bit odd, given how much he hammers on “What do you expect differently?”. Your expectations from many-worlds are be identical to those from pilot-wave, so....
I’m probably misunderstanding or simplifying his position, e.g. there are definitely calculational and intuition advantages to using one vs the other, but that seems a bit inconsistent to me.
I take Eliezer’s position on MWI to be pretty well expressed by this quote from David Wallace:
[...] there is no quantum measurement problem.
I do not mean by this that the apparent paradoxes of quantum mechanics arise because we fail to recognize ‘that quantum theory does not represent physical reality’ (Fuchs and Peres 2000a). Quantum theory describes reality just fine, like any other scientific theory worth taking seriously: describing (and explaining) reality is what the scientific enterprise is about...
What I mean is that there is actually no conflict between the dynamics and ontology of (unitary) quantum theory and our empirical observations. We thought there was originally, because the theory is subtle, complicated and highly unintutive, and because our early attempts to understand it and to relate it to empirical data promote high-level concepts like ‘observation’ and ‘measurement’ to the level of basic posits and confused the issue.
The central case for Everettianism is that it is just plain old quantum mechanics, approached with the default realist perspective that most of us have no problem adopting for practically every other physical theory.* Every other “interpretation” out there adds on extra posits—either ontological posits or epistemological posits that one doesn’t usually hear when talking about other theories—in order to solve a problem that doesn’t actually exist, the so-called “measurement problem”. So it’s not just that MWI is simpler than the other theories; it’s that the sole motivation for the added complexity in other theories—the supposed inadequacy of bare quantum theory to account for our observations—turns out to be bunk.
Suppose someone argued that the general theory of relativity all by itself is inadequate. After all, how does the space-time metric know how to change in the presence of matter? There has to be some transcendent intelligent entity responsible for altering space-time whenever the distribution of energy in the universe changes, so we need to supplement the usual equations of GR with this additional theoretical posit in order to solve this problem. The correct response to this is that the supposed “problem” itself is a mistake stemming from unclear thinking, and that there is no need to posit this additional entity. And since the only motivation for positing this entity’s existence was the pseudo-problem we have just rejected, it would be a mistake to believe that the entity exists. Wallace’s (and I think Eliezer’s) position is that the quantum interpretation debates are just sophisticated versions of this.
* This is not to say that an anti-realist or instrumentalist attitude to scientific theories is a mistake, provided this attitude is a general philosophical position and not motivated by the supposed peculiarities of quantum mechanics. However, many people (e.g. Fuchs and Peres) who advocate instrumentalism about QM aren’t motivated by the attractions of instrumentalism per se but rather by a belief that there is something about quantum mechanics specifically that makes realism untenable. This is a mistake, according to Everettians.
I’m probably misunderstanding or simplifying his position
You really aren’t. His logic is literally “it’s simpler, therefore it’s right” and “we don’t need collapse (or anything else), decoherence is enough”. To be fair, plenty of experts in theoretical physics hold the same view, most notably Deutsch and Carroll.
Doesn’t pilot-wave QM imply literally the exact same calculations happening as for MWI, though?
I sort of got that idea, but if that’s true then I’m not sure I see what the difference between them is in the first place. It seems to me that any theory of quantum mechanics which practically differs from MWI must include collapse somewhere.
(Assuming the difference you’re looking for is “People don’t live in the other branches”.)
Pilot-wave QM has all the calculations as MWI for describing the pilot waves. Then there are particles bouncing around the waves. And then there’s the waveform collapse that happens whenever the particle actually does something. And if you want to explain entanglement, you have to deal with higher-dimensional pilot waves somehow controlling different particles in parallel so the location of one depends on the location of the other.
The Copenhagen interpretation is QM with some bizarre useless stuff added on. Pilot-wave QM is the Copenhagen interpretation with some bizarre useless stuff added on.
In Bohmian mechanics there is no wavefunction collapse “when the particle actually does something”. There is something that the Bohmians call “effective wavefunction collapse”, but that is an emergent phenomenon, not a fundamental dynamical process. The math of the theory says that the wavefunction never collapses, but since the particles are always carried on one branch of the wavefunction, you can treat the wavefunction as if it has collapsed to that branch once an observation is made, and the particle position/velocity calculations will still work out. So you can treat the wavefunction as having collapsed for calculational convenience, but in the actual ontology of the theory the wavefunction behaves exactly as it does under MWI.
Right, that’s my point. In that case, it’s doing the same calculations as MWI and the particles are practically epiphenomenal; ~all observers will find themselves somewhere in the pilot wave.
The Bohmian stance is that the “pilot wave” isn’t a real thing, it’s a mathematical tool. The stuff that actually exists in the universe is the particles. The pilot wave is just a construct we use to predict how the particles move. So it’s a little misleading to say that the particles are epiphenomenal. Ordinarily, when we say that X is epiphenomenal in some theory, we mean that X is causally affected by all the other stuff in the universe but does not itself have any causal effect on any other stuff. The Bohmian position is that there is no other stuff in the universe besides the particles, so it doesn’t make really sense to say the particles are epiphenomenal.
Similarly, saying that all observers will find themselves somewhere in the pilot wave is also a bit misleading. It’s true that there are mathematical structures within the pilot wave (including in those parts of it that do not carry particles) that correspond to observers. However, since the pilot wave isn’t a real thing, those observers don’t actually exist. The only observers that exist are the ones made out of particles.
MWI, on the other hand, interprets the wave function as representing a real physical object, so any structures within the wave function correspond to stuff that actually exists in the universe.
It’s true that there are mathematical structures within the pilot wave (including in those parts of it that do not carry particles) that correspond to observers. However, since the pilot wave isn’t a real thing, those observers don’t actually exist. The only observers that exist are the ones made out of particles.
This is the part I don’t get. How can the pilot wave “not be a real thing” if it’s being computed? Is there some sense in which a thing can be real separate from its being computed?
When you talk about the pilot wave “being computed”, you are assuming a conception of laws of nature that contemporary advocates of Bohmianism would most likely reject. In what sense do you think the pilot wave is being computed? If you want to know more about how Bohmians conceive of the status of the laws of QM, read the first paper I linked in this comment, or at least its conclusion. Basically, you’re supposed to think of Schrodinger’s equation as merely an efficient strategy for compressing information about particle interaction.
The Bohmian stance is that the “pilot wave” isn’t a real thing [...] The stuff that actually exists in the universe is the particles.
That isn’t how I’ve heard it. E.g., Wikipedia: “The onyology [...] consists of a configuration [...] and a pilot wave.” Bohm’s book “The undivided universe” says (I’m going off the Amazon look-inside feature so it’s possible that this would be invalidated by more context): “Let us now discuss this ontology in a more systematic way. Its key points are: [...] 2. This particle [sc. an electron—gjm] is never separate from a new type of quantum field that fundamentally affects it.” (The “new type of quantum field” is the wavefunction.) This seems to say in so many words that the wavefunction is as real as the particles in Bohmian mechanics, which seems to me enough to (e.g.) say that “observers” encoded therein are real.
Bohmian mechanics has developed quite a bit since Bohm. Its most significant contemporary defenders are Sheldon Goldstein, Nino Zanghi and Detlef Durr, and they advocate the ontology I described. See, for instance, this paper. From the abstract:
[...] the modern formulation of Bohm’s quantum theory known as Bohmian mechanics is committed only to particles’ positions and a law of motion… this view can avoid the open questions that the traditional view faces according to which Bohm’s theory is committed to a wave function that is a physical entity over and above the particles...
It seems to me to argue not that the Bohmian stance is that the wavefunction isn’t a thing, but that one good Bohmian stance is that the wavefunction isn’t a thing. Of course that might suffice as a rebuttal to the common claim that Bohmian mechanics is basically a less honest version of MWI plus some extra unnecessary bits.
… Though the paper’s approach doesn’t seem perfectly satisfactory to me—the proposal is that the universal wavefunction should be considered a law of nature, which seems to me about as reasonable as considering every fact about the universe a “law of nature” and doing away with contingency altogether. I confess that I don’t have much in the way of actual arguments against doing this, though. It just seems to violate a general pattern I think I see, that it works best to put random-contingent-looking stuff in (so to speak) the data rather than the code.
(… And: even if we deny that the wavefunction is a thing, the usual argument still seems to me to have considerable force: Bohmian mechanics includes all the same stuff as Everettian, even if it reclassifies some bits as laws rather than things, plus extra stuff—all those ontologically basic particles—that seems to serve no purpose beyond making the theory feel a bit more natural to some physicists.)
It might turn out (I suspect only with a complete theory of quantum gravity in hand) that actually there’s a really briefly specifiable universal wavefunction that naturally gets everything we see as one of its branches. In that case, my objection to treating the wavefunction as a law of nature rather than a part of nature would probably go away. I’m not sure it would really do much to make Bohmian QM look better than Everettian, though.
(Disclaimer: I’m not really a physicist or a philosopher of science, and my intuitions on this stuff aren’t worth very much.)
My stance is that whether or not an entity is real is not a meta-level philosophical question, the way it is usually treated in the realism vs. anti-realism debates, but an object-level scientific question. Our scientific theories, interpreted literally, are committed to certain ontologies. MWI is quite clearly committed to the existence of something like a universal wave function, otherwise none of its explanations or purported merits make any sense. I also think (like David Wallace) that basic quantum theory, interpreted literally and without any ontological or epistemological add-ons, is committed to something like a universal wave function.
So I think that our best scientific theories should determine our ontology. As for which scientific theory to believe in, I think there are a number of different considerations that go into that—empirical confirmation, simplicity, concilience with the rest of our theories, feritility (in terms of useful predictions), etc.
My beef with instrumentalism is that it is, in a sense, too philosophical. It treats the reality or unreality of the unobservable entities posited by our best theories as a further question, one not determined by the theories themselves. Even once we have accepted, say, quantum mechanics as true, the instrumentalist says there is a further question about whether to take its claims literally or to treat them as mere calculational tools for making predictions. My take is that there is no further question. Accepting quantum mechanics as true, or believing in quantum mechanics, implies accepting the theory as a guide to reality. If you’re unwilling to do that, you need to tell me why. If your standards for determining what really exists are so high that not even our best-confirmed scientific theories can meet them, then I suspect you’re working with a concept of “reality” that I, as a pragmatist, have no use for (and can’t even really understand).
OK, so your approach is something like “QM has wave function as a basic description, therefore wave function is real if QM is true”? And if QED requires virtual particles to calculate anything useful, then virtual particles “exist”, and are not a mathematical artifact of the perturbation theory?
Even once we have accepted, say, quantum mechanics as true, the instrumentalist says there is a further question about whether to take its claims literally or to treat them as mere calculational tools for making predictions.
If that’s what instrumentalism is, then I am most definitely not an instrumentalist. To me “accepting QM as true” is a meaningless statement, while “QM is accurate (at explaining and predicting) and fertile (making lots and lots of interesting, useful and accurate predictions)” is a meaningful one.
If your standards for determining what really exists are so high that not even our best-confirmed scientific theories can meet them, then I suspect you’re working with a concept of “reality” that I, as a pragmatist, have no use for (and can’t even really understand).
I would agree with that, assuming I had “standards for determining what really exists”, which I don’t. To me everything imaginable exists to the same degree, just in different contexts, be it stars, photons, baseballs, unicorns, thoughts, ghosts or numbers. Which makes the concept of existence so loose as to be meaningless. So I don’t see any use for it. If you say that this (obvious to me) approach does not fit neatly to some existing (heh) ontology, I would be quite surprised,
OK, so your approach is something like “QM has wave function as a basic description, therefore wave function is real if QM is true”? And if QED requires virtual particles to calculate anything useful, then virtual particles “exist”, and are not a mathematical artifact of the perturbation theory?
Yeah, I believe virtual particles exist. I also believe in the existence of things like phonons, the electromagnetic field, organisms, beliefs and prices. These are all ontological posits of the best theories of the particular domain. I don’t think that there is a meaningful sense in which some of these things are more real than others. Unlike you, I don’t think unicorns or ghosts exist to any degree, because they are not part of our best theory of the relevant domain. I’m not even sure how to think about degrees of existence.
To me “accepting QM as true” is a meaningless statement, while “QM is accurate (at explaining and predicting) and fertile (making lots and lots of interesting, useful and accurate predictions)” is a meaningful one.
From a pragmatist point of view, there isn’t much distance between those two statements. Pragmatists (myself included) reject the correspondence theory of truth.
Interesting, thank you. I guess our views are not that far apart. And I also
don’t think that there is a meaningful sense in which some of these things are more real than others.
though if someone comes up with an interesting, accurate and fruitful meta-model of partial existence, I’d be happy to change my mind.
I don’t think unicorns or ghosts exist to any degree, because they are not part of our best theory of the relevant domain.
Could it be because you are trying to apply them to a wrong domain? Would you agree that in a certain setting (a fantasy tale, a horror story) we can predict behavioral and visual features of the creatures inhabiting it with a fair degree of accuracy? Often more accurately than, say, a path and strength of a tropical storm being born in the Atlantic.
Would you agree that in a certain setting (a fantasy tale, a horror story) we can predict behavioral and visual features of the creatures inhabiting it with a fair degree of accuracy? Often more accurately than, say, a path and strength of a tropical storm being born in the Atlantic.
Yeah, but what we’re using there is a theory of literary and mythological tropes. Those tropes certainly exist, and can be used to predict features of various books and movies. But I think it’s misleading to characterize this as unicorns or ghosts existing. When people ordinarily say things like “I believe ghosts exist”, they’re not referring to predictable patterns in horror stories. I can tell you some things about what the world would be like if ghosts existed, and the world isn’t that way.
If all you mean is that ghosts exist in certain fictional universes, then sure, they do. If someone asks me “Do ghosts exist in Middle Earth?” I’d say “Yes”. If someone asks me “Do ghosts exist?” I’d say “No”.
When people ordinarily say things like “I believe ghosts exist”, they’re not referring to predictable patterns in horror stories. I can tell you some things about what the world would be like if ghosts existed, and the world isn’t that way.
Right. When you extrapolate a model beyond its domain of validity, in this case from stories to the physically perceived world, the predictions of ghost models tend to fail pretty badly. So when people argue about what exists and what does not, all I see is “domain confusion”.
I’m not at all sure what you mean when you say that all you see is “domain confusion”. Do you mean that people in these arguments are talking past each other because they are each talking about different domains? Because I’m pretty sure that is not true in general. Or do you mean that people who say, for example, that ghosts exist are saying this because they are illegitimately extrapolating a theory that works in one domain into another? I don’t think this is true in general either. Or do you mean something else?
Just to clarify: When, in ordinary circumstances, you encounter a debate between two people about whether ghosts exist, do you think one of them is right and the other is wrong?
Or do you mean that people who say, for example, that ghosts exist are saying this because they are illegitimately extrapolating a theory that works in one domain into another?
Yes.
When, in ordinary circumstances, you encounter a debate between two people about whether ghosts exist, do you think one of them is right and the other is wrong?
Usually yes, since people rarely argue whether ghosts exist in mythology. But a discussion about whether numbers exist is almost always a confusion about domains, since numbers exist in the mind, just like ghosts.
Not sure what you are saying. My guess is that you are implying that the quotation is not the referent, and unicorns are hypothetical magical creatures, while “unicorns” are vivid and very real descriptions of them in the stories often read and written by the local bronies. If so, then all I have to say that unicorn is not an accurate or fertile theory, while “unicorn” most definitely is. The difference is the domain of validity: can you go outside and find one running around, or can you mostly encounter them in books and movies? But that applies to most theories. If you go slow, Newtonian mechanics is adequate, if you study fast-moving objects, Newton gives bad predictions. Similarly, if you apply the predictions of the “unicorn” model beyond the domain of its validity, you are going to be disappointed, though occasionally you might discover a new applicable domain, such as a cosplay or a SFF convention.
The distinction is that a theory of “unicorns” is a theory that describes how and why other people (and probably you yourself) think about unicorns, while a theory of unicorns would explain actual unicorns. The latter would clearly fail as a theory, because you’re never going to actually see a unicorn.
The same distinction doesn’t apply to Newtonian mechanics, because Newtonian mechanics is a theory of mechanics, not a theory of how people think about mechanics.
On those grounds, I think it’s quite reasonable to say that virtual particles are real, and “unicorns” are real, but unicorns are not real.
The same distinction doesn’t apply to Newtonian mechanics, because Newtonian mechanics is a theory of mechanics, not a theory of how people think about mechanics.
On those grounds, I think it’s quite reasonable to say that virtual particles are real, and “unicorns” are real, but unicorns are not real.
Not sure if you read anything I wrote in this thread. Note that both Newton’s laws and “unicorn” laws are models. You don’t find Newton’s laws in Nature, just like you don’t find “unicorn” laws. You don’t find virtual particles, either, as they are but terms in the perturbative expansion of a particular quantum field theory (which is also a model, and not found in the wild).
Doesn’t pilot-wave QM imply literally the exact same calculations happening as for MWI, though?
No, it doesn’t. Pilot-wave QM postulates an additional fundamental equation (the guiding equation) that doesn’t appear in MWI. It describes how the behavior of the wave function affects the positions of particles.
Yes, it does all the same calculations as MWI plus some more. The only way to empirically distinguish MWI and Bohmianism is through anthropic considerations (like in the quantum suicide experiment discussed elsewhere in this thread).
There is at least one situation in which you might expect something different under MWI than under pilot-wave: quantum suicide. If you rig a gun so that it kills you if a photon passes through a half-silvered mirror, then under MWI (and some possibly reasonable assumptions about consciousness) you would expect the photon to never pass through the mirror no matter how many experiments you perform, but under pilot-wave you would expect to be dead after the first few experiments.
I’m not convinced there’s a real difference there.
In both cases you expect that in no experiment you observe (and survive) will the gun fire and kill you. In both cases you expect that an independent observer will see the gun fire and kill you about half the time. In both cases you expect that there is some chance that you survive through many experiments (and, I repeat, that in all those you will find that the gun didn’t fire or fired in some unintended way or something) -- what actual observable difference is there here?
In the pilot wave theory, the probability that you will witness yourself surviving the experiment after it is performed say 1000 times is really really small. In MWI that probability is close to 1 (provided you consider all future versions of yourself to be “yourself”). So if you witness yourself surviving the experiment after it is performed 1000 times, you should update in favor of MWI over pilot wave theory (if those are the two contenders).
I am skeptical of the existence of any clearly definable sense of “the probability that you will witness yourself surviving the experiment” that (1) yields different answers for Everett and for Bohm, and (2) doesn’t have excessively counterintuitive properties (e.g., probabilities not adding up to 1).
Probability that any you looking at the outcome of the experiment after 1000 runs sees you alive? 1, either way. Probability that someone looking from outside sees you alive after 1000 runs? Pretty much indistinguishable from 0, either way.
You only get the “probability 1 of survival” thing out of MWI by effectively conditionalizing on your survival. But you can do that just as well whatever interpretation of QM you happen to be using.
If I find myself alive after 1000 runs of the experiment … well, what I actually conclude, regardless of preferred interpretation of QM, is that the experiment was set up wrong, or someone sabotaged it, or some hitherto-unsuspected superbeing is messing with things. But if such possibilities are ruled out somehow, I conclude that something staggeringly improbable happened, and I conclude that whether I am using Everett or Bohm. I don’t expect to go on living for ever under MWI; the vast majority of my measure doesn’t. What I expect is that whatever bits of my wavefunction survive, survive. Which is entirely tautological, and is equivalent to “if I survive, I survive” in a collapse-y interpretation.
Anthropomorphically forcing the world to have particular laws of physics by more effectively killing yourself if it doesn’t seems… counter-productive to maximizing how much you know about the world. I’m also not sure how you can avoid disproving MWI by simply going to sleep, if you’re going to accept that sort of evidence.
(Plus quantum suicide only has to keep you on the border of death. You can still end up as an eternally suffering almost-dying mentally broken husk of a being. In fact, those outcomes are probably far more likely than the ones where twenty guns misfire twenty times in a row.)
In fact, those outcomes are probably far more likely than the ones where twenty guns misfire twenty times in a row.
It’s quite a bit less likely, but if quantum immortality changes the past (when you’re on the border of life and death, it’s clear the gun didn’t misfire), then it would just keep you from running the experiment in the first place.
I find Eliezer’s insistence about Many-Worlds a bit odd, given how much he hammers on “What do you expect differently?”. Your expectations from many-worlds are be identical to those from pilot-wave, so....
I’m probably misunderstanding or simplifying his position, e.g. there are definitely calculational and intuition advantages to using one vs the other, but that seems a bit inconsistent to me.
I take Eliezer’s position on MWI to be pretty well expressed by this quote from David Wallace:
The central case for Everettianism is that it is just plain old quantum mechanics, approached with the default realist perspective that most of us have no problem adopting for practically every other physical theory.* Every other “interpretation” out there adds on extra posits—either ontological posits or epistemological posits that one doesn’t usually hear when talking about other theories—in order to solve a problem that doesn’t actually exist, the so-called “measurement problem”. So it’s not just that MWI is simpler than the other theories; it’s that the sole motivation for the added complexity in other theories—the supposed inadequacy of bare quantum theory to account for our observations—turns out to be bunk.
Suppose someone argued that the general theory of relativity all by itself is inadequate. After all, how does the space-time metric know how to change in the presence of matter? There has to be some transcendent intelligent entity responsible for altering space-time whenever the distribution of energy in the universe changes, so we need to supplement the usual equations of GR with this additional theoretical posit in order to solve this problem. The correct response to this is that the supposed “problem” itself is a mistake stemming from unclear thinking, and that there is no need to posit this additional entity. And since the only motivation for positing this entity’s existence was the pseudo-problem we have just rejected, it would be a mistake to believe that the entity exists. Wallace’s (and I think Eliezer’s) position is that the quantum interpretation debates are just sophisticated versions of this.
* This is not to say that an anti-realist or instrumentalist attitude to scientific theories is a mistake, provided this attitude is a general philosophical position and not motivated by the supposed peculiarities of quantum mechanics. However, many people (e.g. Fuchs and Peres) who advocate instrumentalism about QM aren’t motivated by the attractions of instrumentalism per se but rather by a belief that there is something about quantum mechanics specifically that makes realism untenable. This is a mistake, according to Everettians.
You really aren’t. His logic is literally “it’s simpler, therefore it’s right” and “we don’t need collapse (or anything else), decoherence is enough”. To be fair, plenty of experts in theoretical physics hold the same view, most notably Deutsch and Carroll.
Doesn’t pilot-wave QM imply literally the exact same calculations happening as for MWI, though?
I sort of got that idea, but if that’s true then I’m not sure I see what the difference between them is in the first place. It seems to me that any theory of quantum mechanics which practically differs from MWI must include collapse somewhere.
(Assuming the difference you’re looking for is “People don’t live in the other branches”.)
Pilot-wave QM has all the calculations as MWI for describing the pilot waves. Then there are particles bouncing around the waves. And then there’s the waveform collapse that happens whenever the particle actually does something. And if you want to explain entanglement, you have to deal with higher-dimensional pilot waves somehow controlling different particles in parallel so the location of one depends on the location of the other.
The Copenhagen interpretation is QM with some bizarre useless stuff added on. Pilot-wave QM is the Copenhagen interpretation with some bizarre useless stuff added on.
In Bohmian mechanics there is no wavefunction collapse “when the particle actually does something”. There is something that the Bohmians call “effective wavefunction collapse”, but that is an emergent phenomenon, not a fundamental dynamical process. The math of the theory says that the wavefunction never collapses, but since the particles are always carried on one branch of the wavefunction, you can treat the wavefunction as if it has collapsed to that branch once an observation is made, and the particle position/velocity calculations will still work out. So you can treat the wavefunction as having collapsed for calculational convenience, but in the actual ontology of the theory the wavefunction behaves exactly as it does under MWI.
Right, that’s my point. In that case, it’s doing the same calculations as MWI and the particles are practically epiphenomenal; ~all observers will find themselves somewhere in the pilot wave.
The Bohmian stance is that the “pilot wave” isn’t a real thing, it’s a mathematical tool. The stuff that actually exists in the universe is the particles. The pilot wave is just a construct we use to predict how the particles move. So it’s a little misleading to say that the particles are epiphenomenal. Ordinarily, when we say that X is epiphenomenal in some theory, we mean that X is causally affected by all the other stuff in the universe but does not itself have any causal effect on any other stuff. The Bohmian position is that there is no other stuff in the universe besides the particles, so it doesn’t make really sense to say the particles are epiphenomenal.
Similarly, saying that all observers will find themselves somewhere in the pilot wave is also a bit misleading. It’s true that there are mathematical structures within the pilot wave (including in those parts of it that do not carry particles) that correspond to observers. However, since the pilot wave isn’t a real thing, those observers don’t actually exist. The only observers that exist are the ones made out of particles.
MWI, on the other hand, interprets the wave function as representing a real physical object, so any structures within the wave function correspond to stuff that actually exists in the universe.
This is the part I don’t get. How can the pilot wave “not be a real thing” if it’s being computed? Is there some sense in which a thing can be real separate from its being computed?
When you talk about the pilot wave “being computed”, you are assuming a conception of laws of nature that contemporary advocates of Bohmianism would most likely reject. In what sense do you think the pilot wave is being computed? If you want to know more about how Bohmians conceive of the status of the laws of QM, read the first paper I linked in this comment, or at least its conclusion. Basically, you’re supposed to think of Schrodinger’s equation as merely an efficient strategy for compressing information about particle interaction.
The same way that World of Warcraft isn’t real. Computations are in the map not in the territory.
And yet, a WoW account has persistent state and internal dynamics. It seems real to me. It just isn’t the same thing as what it represents.
That isn’t how I’ve heard it. E.g., Wikipedia: “The onyology [...] consists of a configuration [...] and a pilot wave.” Bohm’s book “The undivided universe” says (I’m going off the Amazon look-inside feature so it’s possible that this would be invalidated by more context): “Let us now discuss this ontology in a more systematic way. Its key points are: [...] 2. This particle [sc. an electron—gjm] is never separate from a new type of quantum field that fundamentally affects it.” (The “new type of quantum field” is the wavefunction.) This seems to say in so many words that the wavefunction is as real as the particles in Bohmian mechanics, which seems to me enough to (e.g.) say that “observers” encoded therein are real.
Bohmian mechanics has developed quite a bit since Bohm. Its most significant contemporary defenders are Sheldon Goldstein, Nino Zanghi and Detlef Durr, and they advocate the ontology I described. See, for instance, this paper. From the abstract:
Some other sources for this view.
Interesting paper—thanks!
It seems to me to argue not that the Bohmian stance is that the wavefunction isn’t a thing, but that one good Bohmian stance is that the wavefunction isn’t a thing. Of course that might suffice as a rebuttal to the common claim that Bohmian mechanics is basically a less honest version of MWI plus some extra unnecessary bits.
… Though the paper’s approach doesn’t seem perfectly satisfactory to me—the proposal is that the universal wavefunction should be considered a law of nature, which seems to me about as reasonable as considering every fact about the universe a “law of nature” and doing away with contingency altogether. I confess that I don’t have much in the way of actual arguments against doing this, though. It just seems to violate a general pattern I think I see, that it works best to put random-contingent-looking stuff in (so to speak) the data rather than the code.
(… And: even if we deny that the wavefunction is a thing, the usual argument still seems to me to have considerable force: Bohmian mechanics includes all the same stuff as Everettian, even if it reclassifies some bits as laws rather than things, plus extra stuff—all those ontologically basic particles—that seems to serve no purpose beyond making the theory feel a bit more natural to some physicists.)
It might turn out (I suspect only with a complete theory of quantum gravity in hand) that actually there’s a really briefly specifiable universal wavefunction that naturally gets everything we see as one of its branches. In that case, my objection to treating the wavefunction as a law of nature rather than a part of nature would probably go away. I’m not sure it would really do much to make Bohmian QM look better than Everettian, though.
(Disclaimer: I’m not really a physicist or a philosopher of science, and my intuitions on this stuff aren’t worth very much.)
As a pragmatist, how do you decide if something is or isn’t real?
My stance is that whether or not an entity is real is not a meta-level philosophical question, the way it is usually treated in the realism vs. anti-realism debates, but an object-level scientific question. Our scientific theories, interpreted literally, are committed to certain ontologies. MWI is quite clearly committed to the existence of something like a universal wave function, otherwise none of its explanations or purported merits make any sense. I also think (like David Wallace) that basic quantum theory, interpreted literally and without any ontological or epistemological add-ons, is committed to something like a universal wave function.
So I think that our best scientific theories should determine our ontology. As for which scientific theory to believe in, I think there are a number of different considerations that go into that—empirical confirmation, simplicity, concilience with the rest of our theories, feritility (in terms of useful predictions), etc.
My beef with instrumentalism is that it is, in a sense, too philosophical. It treats the reality or unreality of the unobservable entities posited by our best theories as a further question, one not determined by the theories themselves. Even once we have accepted, say, quantum mechanics as true, the instrumentalist says there is a further question about whether to take its claims literally or to treat them as mere calculational tools for making predictions. My take is that there is no further question. Accepting quantum mechanics as true, or believing in quantum mechanics, implies accepting the theory as a guide to reality. If you’re unwilling to do that, you need to tell me why. If your standards for determining what really exists are so high that not even our best-confirmed scientific theories can meet them, then I suspect you’re working with a concept of “reality” that I, as a pragmatist, have no use for (and can’t even really understand).
OK, so your approach is something like “QM has wave function as a basic description, therefore wave function is real if QM is true”? And if QED requires virtual particles to calculate anything useful, then virtual particles “exist”, and are not a mathematical artifact of the perturbation theory?
If that’s what instrumentalism is, then I am most definitely not an instrumentalist. To me “accepting QM as true” is a meaningless statement, while “QM is accurate (at explaining and predicting) and fertile (making lots and lots of interesting, useful and accurate predictions)” is a meaningful one.
I would agree with that, assuming I had “standards for determining what really exists”, which I don’t. To me everything imaginable exists to the same degree, just in different contexts, be it stars, photons, baseballs, unicorns, thoughts, ghosts or numbers. Which makes the concept of existence so loose as to be meaningless. So I don’t see any use for it. If you say that this (obvious to me) approach does not fit neatly to some existing (heh) ontology, I would be quite surprised,
Yeah, I believe virtual particles exist. I also believe in the existence of things like phonons, the electromagnetic field, organisms, beliefs and prices. These are all ontological posits of the best theories of the particular domain. I don’t think that there is a meaningful sense in which some of these things are more real than others. Unlike you, I don’t think unicorns or ghosts exist to any degree, because they are not part of our best theory of the relevant domain. I’m not even sure how to think about degrees of existence.
From a pragmatist point of view, there isn’t much distance between those two statements. Pragmatists (myself included) reject the correspondence theory of truth.
Interesting, thank you. I guess our views are not that far apart. And I also
though if someone comes up with an interesting, accurate and fruitful meta-model of partial existence, I’d be happy to change my mind.
Could it be because you are trying to apply them to a wrong domain? Would you agree that in a certain setting (a fantasy tale, a horror story) we can predict behavioral and visual features of the creatures inhabiting it with a fair degree of accuracy? Often more accurately than, say, a path and strength of a tropical storm being born in the Atlantic.
Yeah, but what we’re using there is a theory of literary and mythological tropes. Those tropes certainly exist, and can be used to predict features of various books and movies. But I think it’s misleading to characterize this as unicorns or ghosts existing. When people ordinarily say things like “I believe ghosts exist”, they’re not referring to predictable patterns in horror stories. I can tell you some things about what the world would be like if ghosts existed, and the world isn’t that way.
If all you mean is that ghosts exist in certain fictional universes, then sure, they do. If someone asks me “Do ghosts exist in Middle Earth?” I’d say “Yes”. If someone asks me “Do ghosts exist?” I’d say “No”.
Right. When you extrapolate a model beyond its domain of validity, in this case from stories to the physically perceived world, the predictions of ghost models tend to fail pretty badly. So when people argue about what exists and what does not, all I see is “domain confusion”.
I’m not at all sure what you mean when you say that all you see is “domain confusion”. Do you mean that people in these arguments are talking past each other because they are each talking about different domains? Because I’m pretty sure that is not true in general. Or do you mean that people who say, for example, that ghosts exist are saying this because they are illegitimately extrapolating a theory that works in one domain into another? I don’t think this is true in general either. Or do you mean something else?
Just to clarify: When, in ordinary circumstances, you encounter a debate between two people about whether ghosts exist, do you think one of them is right and the other is wrong?
Yes.
Usually yes, since people rarely argue whether ghosts exist in mythology. But a discussion about whether numbers exist is almost always a confusion about domains, since numbers exist in the mind, just like ghosts.
Ah, but then you’re talking about a theory of “unicorns” rather than a theory of unicorns.
Not sure what you are saying. My guess is that you are implying that the quotation is not the referent, and unicorns are hypothetical magical creatures, while “unicorns” are vivid and very real descriptions of them in the stories often read and written by the local bronies. If so, then all I have to say that unicorn is not an accurate or fertile theory, while “unicorn” most definitely is. The difference is the domain of validity: can you go outside and find one running around, or can you mostly encounter them in books and movies? But that applies to most theories. If you go slow, Newtonian mechanics is adequate, if you study fast-moving objects, Newton gives bad predictions. Similarly, if you apply the predictions of the “unicorn” model beyond the domain of its validity, you are going to be disappointed, though occasionally you might discover a new applicable domain, such as a cosplay or a SFF convention.
The distinction is that a theory of “unicorns” is a theory that describes how and why other people (and probably you yourself) think about unicorns, while a theory of unicorns would explain actual unicorns. The latter would clearly fail as a theory, because you’re never going to actually see a unicorn.
The same distinction doesn’t apply to Newtonian mechanics, because Newtonian mechanics is a theory of mechanics, not a theory of how people think about mechanics.
On those grounds, I think it’s quite reasonable to say that virtual particles are real, and “unicorns” are real, but unicorns are not real.
Not sure if you read anything I wrote in this thread. Note that both Newton’s laws and “unicorn” laws are models. You don’t find Newton’s laws in Nature, just like you don’t find “unicorn” laws. You don’t find virtual particles, either, as they are but terms in the perturbative expansion of a particular quantum field theory (which is also a model, and not found in the wild).
Anyway, disengaging now.
No, it doesn’t. Pilot-wave QM postulates an additional fundamental equation (the guiding equation) that doesn’t appear in MWI. It describes how the behavior of the wave function affects the positions of particles.
Okay, so it does some extra calculations. But it still does all the same calculations as MWI?
Yes, it does all the same calculations as MWI plus some more. The only way to empirically distinguish MWI and Bohmianism is through anthropic considerations (like in the quantum suicide experiment discussed elsewhere in this thread).
There is at least one situation in which you might expect something different under MWI than under pilot-wave: quantum suicide. If you rig a gun so that it kills you if a photon passes through a half-silvered mirror, then under MWI (and some possibly reasonable assumptions about consciousness) you would expect the photon to never pass through the mirror no matter how many experiments you perform, but under pilot-wave you would expect to be dead after the first few experiments.
I’m not convinced there’s a real difference there.
In both cases you expect that in no experiment you observe (and survive) will the gun fire and kill you. In both cases you expect that an independent observer will see the gun fire and kill you about half the time. In both cases you expect that there is some chance that you survive through many experiments (and, I repeat, that in all those you will find that the gun didn’t fire or fired in some unintended way or something) -- what actual observable difference is there here?
In the pilot wave theory, the probability that you will witness yourself surviving the experiment after it is performed say 1000 times is really really small. In MWI that probability is close to 1 (provided you consider all future versions of yourself to be “yourself”). So if you witness yourself surviving the experiment after it is performed 1000 times, you should update in favor of MWI over pilot wave theory (if those are the two contenders).
I am skeptical of the existence of any clearly definable sense of “the probability that you will witness yourself surviving the experiment” that (1) yields different answers for Everett and for Bohm, and (2) doesn’t have excessively counterintuitive properties (e.g., probabilities not adding up to 1).
Probability that any you looking at the outcome of the experiment after 1000 runs sees you alive? 1, either way. Probability that someone looking from outside sees you alive after 1000 runs? Pretty much indistinguishable from 0, either way.
You only get the “probability 1 of survival” thing out of MWI by effectively conditionalizing on your survival. But you can do that just as well whatever interpretation of QM you happen to be using.
If I find myself alive after 1000 runs of the experiment … well, what I actually conclude, regardless of preferred interpretation of QM, is that the experiment was set up wrong, or someone sabotaged it, or some hitherto-unsuspected superbeing is messing with things. But if such possibilities are ruled out somehow, I conclude that something staggeringly improbable happened, and I conclude that whether I am using Everett or Bohm. I don’t expect to go on living for ever under MWI; the vast majority of my measure doesn’t. What I expect is that whatever bits of my wavefunction survive, survive. Which is entirely tautological, and is equivalent to “if I survive, I survive” in a collapse-y interpretation.
Anthropomorphically forcing the world to have particular laws of physics by more effectively killing yourself if it doesn’t seems… counter-productive to maximizing how much you know about the world. I’m also not sure how you can avoid disproving MWI by simply going to sleep, if you’re going to accept that sort of evidence.
(Plus quantum suicide only has to keep you on the border of death. You can still end up as an eternally suffering almost-dying mentally broken husk of a being. In fact, those outcomes are probably far more likely than the ones where twenty guns misfire twenty times in a row.)
It’s quite a bit less likely, but if quantum immortality changes the past (when you’re on the border of life and death, it’s clear the gun didn’t misfire), then it would just keep you from running the experiment in the first place.