I’ve posted more than once regarding my view of the Many Worlds Interpretation (MWI) of quantum physics. I find its rise in modern popularity genuinely inexplicable. (I can’t help but think it’s exactly the sort of thing Dr. Sabine Hossenfelder is talking about in her book, Lost in Math.)
Hoping to find the logic that apparently appeals to so many, I read Something Deeply Hidden: Quantum Worlds and the Emergence of Spacetime (2019), by Sean Carroll. It is, in large part, his argument favoring the MWI. Carroll is a leading voice in promoting the view, so I figured his book would address my concerns.
But as far as I can tell, “there is no there there.”
If anything, the book may have helped me crystallize my concerns.
- If the wave-function is real, what implements it?
- What about energy?
- Scale. A quantum measurement affects the universe?
- Still non-local; split has to be universal.
- Vagueness on exactly what causes a split.
- Probability is schizophrenic under MWI.
The first three (especially #2) involve issues that seem to come close to falsifying MWI (if, in fact, they don’t succeed). I see them as serious obstacles. The latter three are objections that don’t have the same argument power as the first three.
More to the point, Carroll does not effectively address any of these objections. I read the book explicitly looking to understand these issues, and I’m no closer to seeing what the attraction is than I ever have been.
As a persuasive argument, I’d have to judge the book a fail.
In most popular physics books, the author spends time bringing the reader up to speed on quantum physics basics (I can’t count how many times I’ve read these basics now).
Those parts are always fine, no complaints. It’s what comes after that makes the book. When an author is promoting a view — be it birthed or adopted — I pay a lot of attention to tone.
Warped Passages (2005), by Lisa Randall, Cycles of Time (2010), by Roger Penrose, and Our Mathematical Universe (2014), by Max Tegmark, are three examples of books promoting an author-birthed view. All three gain credence with me, and respect, by framing their view as clearly speculative. (Tegmark goes so far as to call his crazy.)
Carroll, who adopted MWI, implies it is the “courageous” view (so if one doesn’t agree with it, one must be a coward). He actually titled the second chapter “The Courageous Formulation” — but then the prologue is titled “Don’t Be Afraid” (I wasn’t).
I’ll just say it’s a huge red flag with me, and that flag contains the words: “Snake Oil” I’m not saying Carroll is necessarily selling snake oil — I’m saying his tactics of presentation are indistinguishable. Carroll comes off as an evangelist to me, and I do not respond well to evangelists.
I even sometimes got a whiff of persecution complex. If one doesn’t embrace MWI, one is part of an oppressive group opposed to truth. The thing is, big claims require big evidence, and being almost entirely a metaphysical claim, MWI has nothing to offer as evidence. (That should all but put it on the same footing as a religion.)
Everything depends on whether one buys the arguments. (Which makes the chapter title “Why Would Anybody Think This?” entirely appropriate.).
The central argument of MWI is a claim to parsimony. The idea is that by removing the vexing thorn of measurement, a big problem of quantum physics goes away.
At the cost of infinite (or nearly infinite) worlds being created at the drop of a photon. So much for parsimony.
In our (single) world duplication of anything concrete has a cost that must be paid — conservation of energy is fundamental to physics. The (single) universe we inhabit, we believe, was created in a Big Bang — an event comprised of infinite density and energy.
But under MWI, an entire universe springs instantly into being when a photon does, or does not, go through a polarizing filter.
It seems to me that what MWI actually does is remove an obvious observational fact: we can measure things. There are aspects of that we don’t fully understand, but I’ve never seen the infamous measurement problem as insurmountable. (On some level it almost seems like whining by physicists that they don’t have the right math to describe what we experience.)
MWI seems to give up on understanding it by throwing it away. To me, that’s throwing out the baby with the bathwater. It’s a mistake!
In chapter eight, “Does This Ontological Commitment Make Me Look Fat?” (Yes, Sean, it really, really does), he has a fictional conversation between smart (and courageous!) fictional Alice and her slow-witted (much less courageous) fictional father. The poor guy just can’t wrap his head around this MWI stuff.
(At one point her father says it all sounds like sophistry. Indeed it does!)
Alice points out that our theory of integers is simple, but that it produces an infinite number of integers.
If that’s the best argument Carroll can offer, he’s got zip.
Because integers are abstract — they are nothing but information. (There is also that a theory of integers ignores the rationals and reals, so if MWI really was like a theory of integers, it would be theory that misses a lot.)
Not only can I have an infinite number of different integers, I can have an infinite number of any particular integer.
So if MWI really was like a theory of integers, then it’s a theory of abstractions and math. Which, frankly, is the only way MWI makes any sense at all — if it’s essentially Tegmark’s view that everything is math.
Because then, as with integers, an infinite number of casually and instantly created worlds would be no problem. Neither would be the idea that the wave-function is everything. It almost follows naturally.
But if the world is physical, then MWI is the least parsimonious theory out there. The claim to parsimony is false.
Going back to my list of objections to MWI, the first one is about this apparent straddling of the line between concrete and abstract.
MWI focuses tunnel vision on the Schrödinger equation:
It equates the equation with reality and further says that’s all reality is. Which seems identical to Max Tegmark’s Mathematical Universe Hypothesis.
If MWI was such an ontological statement, I’d have no problem with it, and my own brief bout of buying into MWI (long ago) involved such a stance.
But Carroll insists the infinite worlds are “real” which raises the question of what implements the wave-function. Math is abstract. To be useful, it has to be implemented by something.
Integers are abstract until we apply them to count physical objects. In the Schrödinger equation, Ĥ (H-hat) stands for the Hamiltonian operator that describes the physical system in question.
When physicists use the Schrödinger equation, of course they compute it on computers. But if the equation is real, and reality is physical, what computes the wave-function?
The situation with energy seems even more dire, and Carroll doesn’t offer much. What he says is through Alice schooling her dad:
“But what about the extra worlds?” her father insisted. “I could measure the energy contained in this world I see around me, and you say it’s being duplicated all the time.”
Alice felt she was on firm ground with this one [ed: Ha!]. “Not all worlds are created equal. Think about the wave function. When it describes multiple branched worlds, we can calculate the total amount of energy by adding up the amount of energy in each world, times the weight (the amplitude squared) for that world. When one world divides in two, the energy in each world is basically the same as it previously was in the single world (as far as anyone living inside is concerned), but their contributions to the total energy of the wave function of the universe have divided in half since their amplitudes have decreased. Each world got a bit thinner [ed: “thinner”?], although its inhabitants can’t tell any difference.”
So, again, how does E=mc2 remain valid in this world? I assume c doesn’t change, so is our mass getting thinner? A kilogram isn’t a kilogram?
Alice’s answer doesn’t just seem like hand-waving; it seems dead wrong (at the very least, it’s hand-waving).
Without a good answer to this issue, I think MWI is falsified.
Scale is another issue I see as a show-stopper. It required a Big Bang to create this universe — a violent event of unimaginable energy and matter density.
But if a photon goes through a half-silvered mirror (or not), poof, a whole new universe just springs into being? Two methods of creating a universe? That is one serious bolt-on to a supposedly parsimonious theory (but I think we’ve put the parsimony claim to bed without dinner already).
Alice and her dad talk about locality. Alice claims the casually created instant new universe can be seen as spreading from the point of creation (the lowly photon that did or didn’t) or as happening everywhere at once.
I think this is false. It has to happen everywhere at once, based on what happens in experiments testing Bell’s Inequality.
These experiments make it clear reality is non-local. More importantly, in an Alice-and-Bob experiment, if Alice measures her photon, according to MWI she creates two worlds — one where she measures spin-up and one where she measures spin-down.
Because the photons are entangled, she necessarily drags Bob into both worlds. He is required to measure spin-down where Alice measures spin-up and vice versa. So Alice has to create two copies of Bob no matter how far away he is.
We know quantum physics seems to be non-local, so this isn’t anything other than the observation that MWI proponents pretty much need to stick with the ontological claim that measuring that little photon instantly creates entire new universes.
If MWI is somewhat vague about the nature of the splitting, it’s even more vague when it comes to exactly what causes a split.
Carroll denies that our decisions result in splitting (because our brains are classical). My question is, when I walk around with polarizing sunglasses, am I creating zillions of new worlds depending on which photons pass through versus which don’t?
If I shine a flashlight at the wall, where the individual photons all hit is random and quantum. Does that create zillions of new worlds?
It isn’t just that MWI seems false and vague — it’s kind of half-assed. Everyone bowed down to “it’s the Schrödinger equation!” and called it a day. It’s not a theory so much as a fantasy.
And then, finally, there’s the probability thing. Which is also an observation. (Both Everett, in his paper, and Carroll spend a fair bit of time justifying probability.)
The truth is, from our point of view, probability works similarly in both single- and many-worlds interpretations. Under MWI, we’re more likely to find ourselves in a more probable branch than not. Under SWI rare events happen rarely.
But under MWI, long-shots always happen. Carroll used his random app to create a random string of 64 bits. Under MWI, the overwhelming number of Sean Carrolls all got reasonable results.
But there is a version that got all 1s and another that got all 0s. That would cause surprise and doubt about the “random” device working. Most of the surprised Carrolls tried again and got reasonable results. But again, there’s one that got all 1s and another that got all 0s.
In fact, there are versions that got 0s (or 1s) no matter how long they tried. There’s a version that replaced the device and still got the same bits. There are also versions that got other kinds of decidedly unexpected results.
They are vastly overwhelmed by the versions with reasonable results, but under MWI those versions always must exist. (No matter how “thin” their realities are.)
This one is a judgement call, but I think that’s beyond the pale.
The only place left for me to explore is Everett’s paper that started all this, perhaps along with Bryce DeWitt’s extension of it using decoherence (which Everett didn’t know about).
So I suppose there might someday be a MWI: Everett post… or maybe MWI: Origins would be appropriate. 🙂
Stay safe, my friends! Wear your masks — COVID-19 is airborne!