The Everett theory of quantum mechanics is testable in ways most people don’t realize.
Before getting into how or why, I think it’s important to deal with a long standing issue. Everettian theory is more commonly known as the “many worlds interpretation”, a name I use myself all the time. But what’s often lost in the discussion is that “world” here is a metaphor for describing something very complex. Bryce DeWitt knew what he was doing when he gave it that name. It quickly and vividly conveys an approximation of the overall idea.
But that comes at a cost. People lose track of the idea that “worlds”, or “universes” in some descriptions, are just a conceptual crutch, a metaphor. This leads them to glom onto details of the metaphor and have questions and concerns that are really more about the metaphor than the actual theory. For example, concern about whole universes “springing into being” is really an issue with the metaphor.
Chad Orzel actually wrote an article about this some years ago when discussing Sean Carroll’s book on the subject. At the time I misunderstood his point, then later thought he was wrong, that the metaphor could be clarified sufficiently. Well, I’ve come full circle and now fully agree with him. So I’m going to try to do the rest of this post without evoking the metaphor. Of course, it’s difficult to talk about stuff like this without using some metaphors, but hopefully avoiding ones loaded with baggage will help.
So what then is Hugh Everett’s theory really about? It’s about trying to understand the ontology of quantum mechanics, with a motivation, at the time it was formulated, of getting closer to a reconciliation with general relativity.
The conventional understanding in collapse interpretations, is that there are two processes at play.
One is what happens for an isolated quantum system. It’s the continuous, linear, and deterministic evolution of pure wave mechanics, including interference between the various states in superposition, which is tracked by a mathematical tool called the wave function. Think of it as an accounting of everything that happens in the double slit experiment up until the location of the particle is known.
The second is what happens on measurement. It’s an abrupt, instant, discontinuous change, where all but one of the states disappear, resulting in a particular location, spin state, or whatever is being measured. The result is random and unpredictable. The wave function can be used to derive the probability of each possible value, but not what the actual answer will be in any individual measurement. Today this is typically called the collapse of the wave function, since all but one of the states it’s tracking, and their interference effects, disappear.
The second process is very mysterious. Like any mysterious process, there are people who insist it’s just fundamental and we need to get over it. And a hard core instrumentalist might insist it gives us what we need. But a laboratory recipe isn’t always helpful when attempting to understand the implications for gravity and cosmology. Everett wanted to get at an improved ontology.
His solution is counter-intuitive. He saw the mistake as assuming that the second process, the wave function collapse, is real, instead of just being a gap in our accounting. Everett advocated removing it from the ontology, to only rely on the first process, the continuous and deterministic evolution of the wave function. His argument is that doing so explains the same observations, but with a leaner, more parsimonious set of rules.
To see how requires a simple understanding of quantum entanglement. Consider if we have two particles, both in a superposition of spin up and spin down. We might write the state of each particle (in a very simplified manner omitting amplitudes and other formal notation) as:
particle state = (up) + (down)
The plus sign just indicates that in a wave function, we’d add the two states together, with any overlap leading to interference. Now, what happens if we have these two particles interact in the right manner? If we do, they become correlated in certain ways, that is, entangled. (Quantum computing leans heavily on this effect.) So they now have an overall combined wave function state, an overall superposition with four elements.
combined state = (up)(up) + (up)(down) + (down)(down) + (down)(up)
If we add a third particle into the mix, we end up with eight elements in the overall superposition:
combined state = (up)(up)(up) + (down)(up)(up) + (down)(down)(up) + (down)(down)(down) + (up)(down)(up) + (up)(up)(down) + (down)(up)(down) + (up)(down)(down)
Notice that each addition into the entanglement multiplies the states of the overall entangled set by the number states brought in by the new particle. Again, nothing controversial here. This is used heavily by quantum computing. If we conduct a measurement on any of the three entangled particles above, we see the entire group apparently collapse into just one of those eight states.
But Everett is saying to do away with the wave function collapse as part of the ontology. So let’s back up to just one particle again and look at this. The conventional collapse interpretation, using the second process above, looks something like this when we introduce interaction with an observer.
combined state = ( (up) + (down) ) (observer)
…which collapses to…
combined state = (up) (observer-sees-up)
or
combined state = (down) (observer-sees-down)
In other words, interaction with the observer has collapsed the states down to one, either spin up or spin down. However, if we do as Everett advises and do away with the second process, then we have to depend on the first process above, the wave function dynamics, to figure out what happens. So instead, we get something like:
combined state = ( (up) + (down) ) (observer)
…leading to…
combined state = (up)(observer-sees-up) + (down)(observer-sees-down)
In other words, the observer, as a quantum system themselves, has become entangled with the particle, and so their state now includes seeing the particle spin up and seeing it spin down. Each element of the observer only sees one state because both they and the particle are also entangled with the surrounding environment. (The entropic jostling from that environment fragments any wave effects and makes them very hard to detect in a process called decoherence.)
So, under Everett, the appearance of the wave function collapse is what a quantum system looks like to an observer that just became entangled with it. In other words, collapse can be thought of as entanglement from the inside.
This implies that the observer and their environment are in a superposition of an ever increasing number of states. Again, we get this by just applying the same rules we used for the individual particles above.
You might object that using the theory for something as large and complex as an observer is a big assumption. And it would be, if it didn’t lead to the same observations as the (now discarded) second process above.
So what does that mean for testing Everettian theory? Remember Everett advocates dropping the second process above for understanding the ontology, and only relying on the first. So any falsification of the first process, of pure wave mechanics, would falsify Everettian theory. This might involve discovering the right hidden variables, including any kind of an actual physical state collapse. And a successful reconciliation with general relativity could falsify it as well, particularly the proposal that just recently came out.
Everett himself also saw the other unseen states of the environment as detectable in principle. Although an understanding of modern decoherence theory shows just how challenging it would be. Still, “challenging” is different from “impossible”. This could someday adjudicate between Everett and Carlos Rovelli’s relational quantum mechanics.
So, some aspects of Everettian theory, arguably the most pivotal ones, are testable. Of course, some aren’t, at least not currently, but that’s true of just about any scientific theory. Under Popperian philosophy, theories are judged by their testable predictions, not their untestable ones, nor by any metaphysical implications we may find disturbing.
Unless of course I’m missing something?
(This post is a vast simplification (probably oversimplified). If you’re interested in the gory details, check out Hugh Everett’s original thesis online, or a more contemporary synthesis in a SEP article about it that distinguishes it from many of the later many-world variants.)
SelfAwarePatterns 
So, my learning from you today is that ‘glom’ is a word!
LikeLiked by 1 person
Ha! You made me look it up just now to make sure I didn’t just make it up. Totally a word. Considered slang by some, but definitely a word!
LikeLike
As to the “collapse of the wave function” as in “The second is what happens on measurement. It’s an abrupt, instant, discontinuous change, where all but one of the states disappear, resulting in a particular location, spin state, or whatever is being measured. The result is random and unpredictable. The wave function can be used to derive the probability of each possible value, but not what the actual answer will be in any individual measurement. Today this is typically called the collapse of the wave function, since all but one of the states it’s tracking, and their interference effects, disappear.”
No thing is collapsing. The wave function is a predictatory description, not a real thing. Since squaring the wave function creates the probability of location/momentum, it is a square root of a probability, whatever the Hell that is.
Consider the flipping of a coin. Before the flip, the probability of the result being “heads” or “tails” is 50% each. Then the flip occurs and a concrete result is acquired. Do we say that the disappearance or collapse of the 50% probability of the other outcome is mysterious?
Too often people talk about wave functions as if they are material objects. Since wave mechanics could just as easily been matric mechanics, I think this is a source of confusion.
LikeLiked by 2 people
Yep, our standard disagreement. I tried to explain in my post last week why the anti-real interpretation of quantum states doesn’t work for me. It’s why I try to mention interference every time I bring it up.
Given the PBR Theorem, saying those interference patterns just happen, with no ontology leading to them, is, to me, bordering on idealism, or semi-idealism. I’m not convinced yet that we need to give up on an external world.
But I suspect we won’t agree on this anytime soon. 🙂
LikeLike
I’m constantly in awe of your ability to actually get your head around this stuff, Mike.
LikeLiked by 1 person
Thanks John. But you just don’t see all the failed attempts where my head bounced rather than wrapped. And I’m sure someone will be along to instruct me on everything I got wrong here.
LikeLiked by 1 person
Thanks for this explanation! I’ve always believed the many worlds interpretation had to be a metaphor, but never had even the slightest clue what it meant. From the outside it looks like science fiction (why have a theory positing worlds we couldn’t possibly know?) Many people find these metaphors captivating, but I find them off putting, at least when they’re not taken as metaphor or explained as metaphor.
LikeLiked by 1 person
Thanks Tina! I hope it was clear, or at least clear enough for the main point.
Everett’s graduate advisor, John Wheeler, who supported his theory early on, actually turned against it when the multiverse versions started showing up in science fiction magazine articles. It can seem too much like justifying old sci-fi / fantasy tropes. So you’re not alone in being put off by the metaphor.
Everett himself always discussed it as happening in this universe. (Although he’d talk in terms of worlds or universes if someone else started the conversation that way.) In some ways, the actual theory is more disorienting and stranger than the notion of other cleanly separated worlds. Shakespeare may have been more right than he could know with the words he had Hamlet say about “more things in heaven and earth” to Horatio.
LikeLike
Interesting story about Wheeler. I wonder if the science fiction stories brought out implications about the theory that he hadn’t quite noticed before?
More things in heaven and earth indeed.
LikeLike
Could be. Not that Wheeler was above considering radical ideas. He once said speculatively to Richard Feynman that maybe all electrons in the universe are just one electron at different points in time. And in later years he came up with the “it from bit” concept, a sort of participatory principle where reality is set by the questions we ask. Which isn’t really compatible with Everett’s theory, so maybe his views just shifted over the decades.
LikeLike
I think I’ve just read about him then in Chalmer’s book. (I guess just as in real life, names don’t register for me.) as I recall Chalmers didn’t like the participatory aspect, but is clearly on board with “it from bit” (or “bit from it from bit from…” I lost track of the direction of its and bits at some point.)
LikeLike
You must be reading Reality+. His discussion about “it from bit” made me feel better about the phrase. When I first heard it years ago, I took it similar to the way Chalmers uses it in the book. But later I read about Wheeler’s full concept with the participatory reality thing. Since then, I’ve never known what someone means when they use that phrase, something like Chalmers’ meaning, or Wheeler’s original one.
LikeLiked by 1 person
Yep, Reality + is the one. Not quite what I was expecting. I liked his famous essay “Facing up the hard problem…” and I was thinking the writing would be more like that, though maybe slightly breezier in the book. I won’t say much here since I’ll probably blog about it (or whatever it is a Substack is supposed to be called) when I finish the book. The ‘it from bit’ thing is something I find problematic, at least in the way Chalmers talks about it, though I’ll have to go back and read that section again to be sure I understand what he meant. I couldn’t really get a clear idea of Wheeler’s use of the phrase from the book, but I did get the sense Chalmers didn’t find the participatory thing appealing.
LikeLiked by 1 person
You moved to Substack? What’s the new URL? (No worries if you’re not ready for traffic yet.)
I recall Chalmers taking it as reality being binary as its base layers. Which of course fits with the whole simulation thing. At least he admits that physics doesn’t really offer much support for it right now.
Wheeler’s version, as I understand it, is that reality depends on which yes or no (binary) questions we ask it. I think it’s his way of trying to make sense of quantum uncertainty. But it strikes me as semi-idealist, which I think was Chalmers’ take too.
LikeLiked by 1 person
That’s strange that you didn’t get the first post on Substack. I imported all of my contacts/subscribers there (or so I thought). Hm. Weird. But to answer your question, yes, I have moved to Substack, tentatively. I’m not sure that I like it, but I have noticed that WordPress just doesn’t seem the same as it did when I first began in terms of discovering new blogs. I’m not sure why that it is. In any case, I like that Substack doesn’t have advertising. We’ll see how long that lasts.
Anywhere, here it is:
https://philosophyandfiction.substack.com/
I think you’re right about Wheeler being some sort of idealist. Chalmers doesn’t like idealism, and I thought he (like pretty much everyone) gave Berkeley an unfair assessment, which is part of my issue with his book, though a somewhat minor one.
LikeLiked by 1 person
Thanks! I found it and left a comment on the first post, mainly to make sure I could. (Some people restrict comments to subscribers.) Success!
I do have an account and placeholder site on Substack, but rarely login. And I’m not wild about email subscriptions. But I did add you to my RSS reader, so I should see new entries moving forward.
I have to admit I’ve never read Berkeley directly, or much on idealism in general. If it’s a view you find promising, might be worth some posts (stacks?) on the strongest arguments for it.
LikeLike
Great! I saw that you were able to comment, which is a relief. I am a little worried about people from outside of Substack finding it too cumbersome, but then again, people outside of WP don’t comment either and maybe they don’t want to.
It’s not so much that I love Berkeley as I find people who haven’t actually read him (or at least don’t seem to have read him) can be a bit too dismissive of him. He’s the guy everybody likes to make fun of, even back in his own day, and unfairly, I think. I suspect part of that may come from the word “idealism” which in ordinary usage means something like “wishful thinking”. Of course in philosophy the term means something else, but I have to wonder if even philosophers who are aware of the distinction are nevertheless biased against anything or anyone labeled with that term. They tend to focus on his bishop status and his uninventive argument for the existence of God, but those are not the things that made him important in the history of philosophy. What’s interesting about Berkeley are his criticisms of empiricists of his time for not being sufficiently empirical—his challenges the idea of matter (which he calls an “abstract idea”) for which the senses give no evidence.
I felt the same way about Kant when a professor made fun of his categorical imperative.
LikeLike
I’m not sure how easy it is to comment on Substack without an account. I can’t recall if I ever did before I had one. But definitely not an issue logged in.
I see where you’re coming from on Berkeley and idealism. It reminds me of my reaction to the withering scorn panpsychists often get. I’m not a panpsychist, but I feel like their arguments are too often brushed aside with little more than vehement incredulity. Since that applies to many views I find more plausible (like the one in this post), I have some sympathy.
I wonder if idealism wouldn’t benefit from a new name. Few people today self label as “materialist”, partly because they’re more physicalist (including energy, fields, and spacetime in their ontology) than Hobbesian materialist, but also because of the negative connotations the word has today for greed and shallowness. One alternative might be “antirealism”, but it’s a bit negative, and while it might cover Berkeley’s view, I’m not sure how well it fits for others.
LikeLike
I hope I wasn’t one of those heaping withering scorn on the panpsychists since I now know I don’t know enough about it. I think of it vaguely as a reductionism in the opposite direction from physicalism, but I haven’t looked into it in any detail.
I hadn’t understood the switch from “materialism” to “physicalism”, but the way you’ve explained it, that change of terms makes a lot more sense.
I was thinking about what I would call idealism, and I haven’t been able to come up with anything really great. My first thought was “radical empiricism” but that’s already too closely associated with William James.
LikeLike
I don’t ever recall you heaping scorn on panpsychism. Most of what I’m thinking comes from scientists on social media, particularly Twitter (even before Musk Xified it).
One name I’ve seen thrown around is “phenomenalism”, which seems to have overlap with idealism, although it seems like a pretty distinct school.
https://en.wikipedia.org/wiki/Phenomenalism
And with “phenomenology”, “phenomena”, and “phenomenal properties” all out there, adding yet another “phenom” seems in danger of further increasing the amount of confusion in the world.
James’ radical empiricism seemed like a type of idealism when I read his writing about it, although I later saw it more characterized as neutral monism, or at least a precursor to it. Particularly with 19th century language mixed in, these outlooks can seem to blur into each other. (Naturally that means they’re often bitter enemies.)
LikeLike
Yeah, it’s a tough one isn’t it? I like “phenomenalism” but as you say, it looks a lot like phenomenology and all the other phenoms in the world.
James is a tough one to nail down. Sometimes this and other times that, at least as I read him. I get a very different version of him in “Varieties” than from “A Pluralistic Universe”, where at times he talked about a kind of pre-theoretical experience which he describes as a kind of drunkenness or whir of sensations, and I thought that violated his own radical empiricism.
LikeLiked by 1 person
The way I prefer to think of MW (and to explain it to others) is that “worlds” do not spring into existence, but on the contrary future “possible worlds” fail to cease existing — they just become inaccessible to each other. That seems to me to be both accurate and concise at the cost of sweeping under the rug the problem of time (of the future constantly sliding through “now” into the past), which really is a different problem altogether.
LikeLiked by 1 person
That’s an interesting word choice: “fail to cease existing”. I think I get what you’re saying, but my experience has been that once we mention the word “world” or “universe”, people drag in a lot of conceptual baggage. Attempts to convince them not to, by emphasizing that a world isn’t a fundamental concept in the theory, don’t seem to be very effective, at least other than with those who already grasp the idea.
Maybe the real lesson is that no one way of explaining anything this complex is ever going to be the only solution.
LikeLike
You should try it my way sometime. In my experience that explanation is readily accepted. After all, the notion of multiple possible futures is pretty intuitive and the point of MW is that in the QM context, such possible futures actually exist and do not vanish, but are simply no longer accessible once their “entry point” is in the past.
BTW, that reminds me of another frequent misunderstanding… It is generally assumed that MW means that all possible futures will come to pass — just elsewhere. But that confuses futures we count possible, with futures that are physically possible starting from the world/universe/whatever as it is. E.g. it is psychologically possible that Donald Trump is about to experience satori and retire to a Buddhist monastery — stranger things have happened. 🙂 But I see no reason to assume that QM could gift us with such an outcome, making such a future actually possible.
LikeLike
I think it would depend on the audience. Talking in terms of possible futures would work for people already primed to think that way. It reminds me of David Deutsch’s approach. He starts with the multiverse and approaches the evidence from that direction. It obviously works for a lot of people. But I was initially repelled by it. Only after I’d found it plausible through an explanation starting with the measurement problem could I seriously consider his version.
On Trump becoming a monk, a lot depends on just how low probabilities can get and still be meaningful. As I’m sure you know, the numbers are infinitely divisible. But who knows to what degree reality is. Based purely on the math, it doesn’t seem like we can rule out scenarios where entropy reverses itself for billions of years. Not that anyone should hold out for a Trump religious conversion.
LikeLike
It seems to me that you may be over-thinking it. I am not talking about any actually existing future worlds. I am talking about a perfectly intuitive notion of multiple possible future outcomes, which is surely, universal. If I am about to toss a coin, anybody would accept that there are two possibilities for the future: I’ll get heads or I get tails. Suppose I get tails. For quantum events, MW says that the heads possibility has not vanished — there is now a separate reality in which I got heads. But I am in this reality and the other one is not accessible. So both used to be possible in my reality as it was, and both are still extant but in parallel worlds — one does not need to imagine future worlds already existing (to some value of “already” :-)). Hence my original comment that this explanation sweeps the question of time under the rug.
I am not clear about your response on distinction between logical and physical possibilities. Do you think that in MW anything that does not contradict physical laws actually happens on macro-scale too? I’ve never come across any argument to support this — it just seems to be accepted by default, which seems odd to me.
LikeLike
On possible future outcomes, maybe I am overthinking it. I do think it’s clear. What I’m not sure about is how convincing it is to a skeptic. Maybe I’m too worried about that, but as a general skeptic myself, it’s where my mind goes.
My understanding is that, under the theory, the possibility exists that everything physically possible, no matter how improbable, happens in the wave function, including macro scale scenarios. It’s allowed by the current math, but physics may someday force us to add news things to the math that are currently below the precision of our instruments. So I definitely wouldn’t say it should be accepted by default.
LikeLike
OK, we disagree on that, then. I cannot see any way of showing that any macro-scale scenario that looks possible to us (however unlikely) is actually reachable by operation of physical laws (known or unknown), given the boundary condition of the world as it is.
LikeLike
I kind of wonder if we’re talking about the same thing here. If boundary conditions prevent something from happening, then in what sense was it possible? For example, it might be physically possible for me to be on Mars under the right combination of conceivable circumstances, but it’s impossible for me to be there two minutes from now, since that’s outside my light cone. Or am I missing your point entirely?
LikeLike
I am pointing out the distinction between what looks to us possible at the macro-level and what is actually possible given the boundary condition of the actual state of the world. E.g. some years ago I had a narrow escape in what could have been a definitely fatal accident. It is generally assumed that MW implies that there is a branch splitting off at that point in which I got killed. But how do we know that the boundary conditions at the time made this a physicalist possibility, however likely or unlikely? The response of “why not?” is easily countered with “why yes?”. I can offer no argument either way and I’ve never seen one offered.
My gut feeling is that macro-level possibilities may be very much more constrained than is often assumed, despite the stochastic way in which our adequate determinism emerges from quantum micro-banging. (Not to be confused with the “everything that happens does happen” of Feynman’s path integral approach, because there that “everything” destructively interferes with itself well below our macro-scale.)
LikeLike
Well, I would definitely agree that what might seem possible and what is actually possible, particularly from a specific point in time, may be very different. And a lot of quick and loose talk about it overlooks this distinction. This is similar to the idea that our choices cause branching. Again, a lot of loose talk from some that more careful scientists try to clarify isn’t a necessary relationship. All of that is before acknowledging that if there is any ultimate discreteness to reality, the most infinitesimal mathematical probabilities become non-physical anyway.
And to your gut feeling, we do have to acknowledge that classical physics exists for a reason. Most of the time quantum randomness averages out into deterministic outcomes. It’s only when the effects of an isolated quantum event get amplified into the environment (in a lab or naturally) that under Everett macroscopic events go off in different directions. Although how often that happens seems like an unsettled issue.
LikeLike
OK, we don’t disagree. 🙂 But I don’t think an appeal to macro-level adequate determinism emerging from the quantum level actually helps. It only says that non-classical outcomes are exceedingly unlikely, not that they are impossible.
Incidentally, there was some really interesting work done by (now Sir) Mike Berry at Bristol on relationship between QM level goings-on and those on macro-level. Curiously enough, Riemann hypothesis appeared to be involved. That was back at the end of 20C — I haven’t followed more recent developments on “quantum chaology”, if there have been any.
LikeLiked by 1 person
It seems like the “testing” isn’t really direct testing at all, It all involves disapproving something or proving something else. Is there a more direct testing? Are there predictions unique to the theory that we can test?
LikeLiked by 1 person
The main one that’s unique to Everett and Bohmian mechanics, is that states in a decohered system should still have interference, which in principle should be detectable. But as I noted in the post, the entropic nature of decoherence makes that very hard. It’s easier if the system is only allowed to be decohered to a limited degree (with no standard on what “full” decoherence means), but then people will say it’s just an elaborate quantum eraser experiment.
The prediction that quantum theory applies at all scales is unique to non-collapse interpretations and RQM. So pushing the size of systems that retain quantum effects strengthens all of those at the cost of collapse theories.
But it’s inevitably going to be a mix of falsifiability and logical deduction, which is actually true in a lot of science. For example, we can adjudicate between Everett and Bohmian mechanics by the difficulties in reconciling Bohm with special relativity.
LikeLiked by 1 person
“pushing the size of systems that retain quantum”
Wouldn’t it need to be the size of the world to prove the theory? And that would include the measuring device too. Can a quantum system measure itself? Wouldn’t that result in a sort of infinity mirror effect?
LikeLiked by 1 person
Focusing on the world is fixating on the metaphor and broader metaphysical implications. Those are untestable.
But they aren’t the theory. The theory is quantum mechanics without an objective collapse. The physical rules are falsifiable.
Of course, we never know what we don’t know, and so can never say with certainty that the rules lead to the broader metaphysical implications. We might discover something a century from now that changes the whole picture. Although the history of science doesn’t provide much precedent for such discoveries bringing things closer to our pre-scientific intuitions about the world. Much of science is probing beyond the boundaries where those intuitions are useful.
LikeLiked by 1 person
Per Wikipedia:
David Deutsch also proposed that MWI will be testable (at least against “naive” Copenhagenism) when reversible computers become conscious via the reversible observation of spin.
Conscious reversible computers? Any clue what that is?
I tried to read the link but it was gibberish to me. I understand reversibility but this is regarding conscious reversible computers.
He also explains the double slit interference pattern with “photons in multiple universes”. It would seem to me more simple to explain it as an artifact of measurement and that the photons don’t really exist.
LikeLiked by 1 person
Yeah, I have to admit I don’t understand what Deutsch is proposing with the conscious quantum AI. My understanding of consciousness is our access to our own mental processes is very limited and unreliable anyway, so I’m leery a quantum AI consciousness would be able to tell us much. I think he’s overestimating what consciousness can do. But I haven’t really tried to parse his proposal.
And his way of explaining things is deeply enmeshed in the metaphor. It might work for people primed by science fiction to buy it that way. I prefer the approach of just talking about the physics and then going from there to the broader implications. But as I noted to someone else in this thread, different approaches work with different audiences.
That said, saying the “shadow photons” (aka the rest of the wave) are an artifact of measurement is trivially easy. Actually coming up with what may be the artifact is a different matter entirely, particularly across the many different types of experiments that have been performed over the last century. Scientists have tried to disprove quantum mechanics from every angle they could. Maybe someone will succeed some day, but any new theory has to fit the same data.
LikeLiked by 1 person
Ok, I think I see what Deutsch is getting at with the conscious AI. It’s a version of the Wigner’s friend thought experiment (itself a variation of Schrodinger’s cat, but escalated to a human being.) https://en.wikipedia.org/wiki/Wigner%27s_friend
If we have a conscious AI running on a quantum computer, and have it measure the quantum state of one of the qubits inside the computer (presumably one outside of the circuits instantiating its consciousness), that’s essentially a Wigner’s friend scenario. From the AI’s perspective, it’ll see the measured qubit/particle as collapsed. But at that time, we’ll still see the entire system, the AI and what it’s measuring, as still in an entangled superposition. When we then ask the AI for a report, it and its measured qubit will collapse for us. Us and the AI will have a different accounting of when the measured qubit collapsed.
Of course, for this to have the psychological weight of an actual Wigner’s friend experiment, there will have to be acceptance that the AI really is conscious. Otherwise it’s probably only as convincing as quantum computing in general.
BTW, the Wikipedia article on many-worlds, which used to be decent, seems to have degraded.
LikeLiked by 1 person
I don’t know how that has to do with reversibility, but the account of observers seeing different things would also argue for RQM, wouldn’t it?
I’m not seeing anything that elevates it above interpretation and borderline metaphysics. Even if it’s right, so what? It doesn’t tell us anything new, does it?
LikeLiked by 1 person
RQM hadn’t been developed when Deutsch came up with this. Both Everett and RQM are relational. Everett refers to apparently collapsed outcomes as “relative states”, and denies there are any absolute determinate ones, which is similar to RQM’s stance that all determinate values are relational.
The difference is that RQM denies any quantum state realism, which means under it, there is no objective reality of the system at all prior to an interaction, and no reality once the system goes back into superposition. So under RQM, the AI and qubit in question don’t exist for us until we ask it for a report. It’s the metaphysical commitment RQM requires to avoid Everett’s ontological implications. If you buy that stance, then the experiment can’t adjudicate between them.
I saw that there’s a more recent proposal for more elaborate version. I haven’t looked it up yet, but I wonder if it narrows things any further. Although I’m not sure how we establish that reality exists when we’re not interacting with it. It’s similar to the fact that there’s no way to convince a determined solipsist that the rest of the world exists.
On what it tells us, just that there is no absolute state collapse. In the 1980s, that left Bohm and Everett. Today it includes RQM and maybe QBism, if you buy either of their metaphysics. It rules out Copenhagen and objective collapse models.
LikeLike
It seems to me that metaphors are actually wonderful ways of helping us relate causal things that are easy to conceptualize with causal things that are difficult to conceptualize. My favorite is relating accessible water pipe function with more theoretical electrical wire function to thus get a better grasp of how electricity works. Here we can observe that water pressure is quite like electrical voltage, and pipe radius for water flow is quite like wire thickness for amperage flow. But this is mainly helpful because nothing seems at all magical here.
Quantum mechanics begins with superposition, tunneling, and entanglement, which seem quite magical to us, and then the “many worlds” metaphor is suppose to explain it by means of an equally magical idea of observers effectively splitting off into other worlds of existence. So for someone like me who’d instead like a causal proposal, it doesn’t seem to explain anything.
If we get rid of the “many worlds” metaphor however given how literally magical that seems, what are we left with? Just a failure to state other observers splitting off into other worlds of existence? This doesn’t seem to be so much of a solution but rather a lack of acknowledgement for the implications of what’s being proposed.
It isn’t politically wise for people like Sean Carrol to say “I believe in magic”. Instead they must say “Magic? If you were as smart and educated as I am, you wouldn’t see this as magic either”. Perhaps he’s right. Or perhaps he’s so gifted that lots of smart people today are trying to justify a theory that tries to explain away apparent magic, by means of more magic. In any case until any of this can be dumbed down to something that makes sense to me, I’ll continue to say that there is either a causal explanation as I presume, or a magical one.
LikeLiked by 1 person
Superposition is a principle from classical wave mechanics. There’s nothing magical about it in and of itself. https://en.wikipedia.org/w/index.php?title=Superposition_principle&oldid=1187692122#Wave_superposition
Maybe you’ll argue that wave-particle duality is magic. But if we see a particle as just an isolated wave fragment, nothing particularly magical there. At least until we get to the wave function collapse, but that’s what Everett’s arguing against.
I’m not a fan of the way tunneling is usually presented, although Sabine Hossenfelder does a good job with it in terms of waves. As is often the case, the wave story is far less spooky than the particle one, a benefit of sticking with wave mechanics. https://backreaction.blogspot.com/2020/11/understanding-quantum-mechanics-8.html
Entanglement is a fancy word for correlations. Most of the spookiness from it comes from the collapse postulate. Again, that’s what Everett is trying to remove. Granted, non-separability is still an issue with Everett, but that doesn’t involve any action at a distance. It does preserve the correlations, but even that isn’t an issue if we can accept some form of configuration space realism.
I suppose you could argue that the 3N dimension configuration space itself is magic. All I can say is that the math works and has been tested for almost a century, with quantum computing taking it to new levels.
On what we have without the worlds metaphor, I went through it in the post. If you have a question or issue about any part of it, or what specifically might be missing, I’m here to discuss it.
I would have some sympathy with the word “magic” for stances assuming we can get all the success of quantum theory without it representing reality in some form.
LikeLike
I get what Chad Orzel is complaining about with the “Many Worlds” talk. Still, I think there’s a simple solution – don’t use that metaphor on laypeople. For those who get the basic Everettian concept, there’s nothing wrong with the “worlds”, you just need to treat them as emergent (“weakly emergent” in philosopher-speak) phenomena. Just like tables and chairs, which don’t need mentioning in physics, but really do exist insofar as there are reliable phenomena (if I put this book on the table, it won’t fall to the floor) that furniture explains.
I really like Orzel’s analogy involving your household budget and the world financial system. Indeed the world financial system does depend, subtly, on you; yet this can be ignored for all practical purposes. Well, most of the time. At a certain point in 2007, maybe not.
LikeLiked by 1 person
I wonder, if laypeople can be introduced to the concept without the worlds, how much they’ll really need them later. Orzel seems to conclude it’s alright to use the worlds, but be more explicit about what they are. That might be right, although in my experience it’s hard to pull people off of irrelevant questions, like how and when worlds split, how many there are, energy conservation, etc.
Similar to Orzel, I needed to get past the multiverse notions, focusing instead on the idea of austere quantum mechanics to take it seriously. And that’s long been the way I’ve tried to describe it, although usually tying it to the worlds at the very end. I often wondered if the last part was a mistake, because it’s often what people end up fixating on in the subsequent discussion.
LikeLiked by 1 person
I just heard Tim Maudlin (on Robinson’s Podcast) say that – and this is a paraphrase – if you wait long enough, interference from those “other worlds” will make a difference in the classical-like behavior of our “world”. If that’s true, it might be a good reason to avoid “worlds” talk altogether.
LikeLiked by 1 person
Ok, now I need to go queue that episode up. Sounds interesting. Thanks!
LikeLike
I have to admit, when I first heard about the many worlds interpretation, I totally got the wrong idea about it. This was many, many years ago, obviously. The idea that all these parallel universe kept popping into existence just seemed absurd, and I also thought it sounded like a violation of thermodynamics. A whole universe worth of matter and energy, created out of basically nothing? That can’t be right.
Based on that first impression of the many worlds interpretation, it took a long time before I was willing to give the idea a second chance.
LikeLiked by 1 person
That matches my experience. Part of the problem is that what Bryce DeWitt used as an effective sales pitch to get fellow physicists’ attention for the physics explanation, too often has become the only presentation provided at all. And by itself it’s flighty and ludicrous.
The first time I saw it in the context of trying to learn about quantum mechanics, I just immediately bypassed it, interested in more “sane” explanations. Even after learning that none of the explanations achieve common sense status, it still seemed the silliest option. When I finally did read about it more closely, the parallel universes presentation was an obstacle that had to be climbed over.
Of course, there are plenty of people primed by science fiction to accept those presentations. But it helps to have a skeptics version of the explanation focusing on just the physics involved.
LikeLiked by 1 person