Sean Carroll on the Many Worlds Interpretation of quantum mechanics

Related to the post last week on quantum mechanics, here is a talk that got mentioned in the discussion thread.  Warning: Carroll is a passionate advocate for the Many Worlds Interpretation, so don’t expect a fair and balanced discussion.  The video is about an  hour long.

A couple of points.  Carroll notes that we shouldn’t judge Everettian quantum mechanics by its implications.  I think this is right.  Judging it by those implications is a reductio ad absurdum argument.  I honestly don’t know why people bother with reductio ad absurdum arguments anymore since reality has clearly demonstrated to us, time and time again, that it is definitely absurd.

But Carroll also notes that we shouldn’t criticize the MWI (Many Worlds Interpretation) for not being falsifiable.  Here I think he’s wrong.  (Carroll’s relationship with falsifiability is a bit rough.  He questioned the value of the concept a few years ago.)  The problem is that what currently is falsifiable applies to all interpretations of quantum mechanics.  This makes all of them equally susceptible to this criticism (except for the instrumental/epistemic version of the Copenhagen Interpretation).  To avoid this criticism, the MWI would need to be uniquely falsifiable.

Science fundamentally is about building theories that can make accurate predictions.  An instrumental version of quantum mechanics meets this standard.  But while all the interpretations that attempt to go further make their own predictions, none are testable, at least not yet.  Until they are, they’re speculation, fascinating speculation to be sure, but speculation just the same.

Still, the MWI is arguably an elegant solution to the core issue.  If it didn’t imply undetectable branches of the wavefunction (universes) everywhere, there’s a good chance we’d all be on board.  And remember, reality is absurd.

30 thoughts on “Sean Carroll on the Many Worlds Interpretation of quantum mechanics

  1. I meant to comment on your original post about how great it was, but got diverted into some other things.

    I am more or less in the “shut up and calculate” camp but still don’t understand the topic in depth so mostly I refrain from opining on it. Sometimes I think that multiple interpretations of QM are much like Rubin’s vase. They are illusory and a product of our perceptual equipment (eyes and brain) in the case of the vase or equations and measurement techniques in the case of QM.

    Sometimes, I think there might be something relating to time involved. For all I know, this may be fundamental to QM and I have missed it. But one obvious difference between pre- and post-measurement, collapse, or world split is that time has passed. If time really didn’t exist or was illusory there would be no world split or collapse.

    Our psychological sense of time is dependent on memory, which consists of measurements and recording. If we had no memory of prior events, we would have no way to sense that present events are different from prior events. In a way, our sense of time arises from the continual calculation of the differences between the present and the past; however, our past is really the results of measurements or observations that have persisted into the present. Even if we take the subjective sense of time out of the picture, the objective measurement of time requires the either the implicit or explicit recording or measuring of events. Take the primitive example of a castaway on an island marking the passage of time with scratches on a rock. Or the movements of the hands on an analog stopwatch accumulating hours, minutes, and seconds.

    Liked by 2 people

    1. Thanks James!

      On time, I think those are excellent points. It’s funny. The only way we have to objectively measure time is as a ratio of other things whose time we experience subjectively. So we measure days by the rising and setting of the sun, months by the moon’s movement and cycles, years by the seasons and constellations moving through the sky, etc.

      Years ago I read Max Tegmark’s ‘Our Mathematical Universe’. He is a proponent of the MWI and see its deterministic nature as compatible with a static unchanging universe as seen from outside of its four dimensional structure. To understand why, you have to remember that each of the posited MWI “worlds” are not separate spacetimes created after the quantum event, but independent branches of the wavefunction evolving in the same spacetime. (Although I have to admit that the idea that they could all evolve through and around each other seems dubious.)

      Liked by 2 people

  2. I think we covered this topic in the last post as far as anything I can contribute. I do remain skeptical to the point of needing a much better understanding of the proliferation of actual honest-to-gosh realities and where their energy comes from.

    One thing we didn’t touch on is how MWI affects probability. I was never into stats wrt probability, and apparently that’s one more place things get very unintuitive. So I don’t really understand it, but apparently MWI requires some hoop-jumping there, because all the possibilities come true in some branch.

    (Along those lines, I’ve heard of something called “quantum suicide” that suggests you’d still go on living in one (or more) branch(es), so it wouldn’t be a big deal. [shrug])

    From your comment above: “The only way we have to objectively measure time is as a ratio of other things whose time we experience subjectively.”

    Do you include under that umbrella things like vibrating atoms and pulsars? I ask because their periodicity suggests (to me, obviously) something fundamental about time. (Which, as you know, I consider more fundamental than the universe, because the Big Bang happened in time.)

    Liked by 2 people

    1. I’m actually somewhat puzzled why probability is an issue. It seems to come from a conception of probability being absolute. In the ontological versions of the Copenhagen Interpretation, it is. The wave function probabilities are what they are from all perspectives.

      But in all other endeavors, probabilities are epistemic things. The probability of whether it will rain tomorrow isn’t an absolute. It represents what we can say based on our current knowledge. If we could somehow overcome the issues with a complex system and have perfect knowledge of the weather, we would know with 100% certitude whether it was going to rain tomorrow. But with imperfect knowledge, we work with probabilities.

      From that, it stands to reason that probabilities are from the perspective of an observer. One observer, with more information, will have different probabilities than another with less information.

      In my mind, under the MWI, the probabilities just shift from being absolute to being relative. Prior to a measurement, I don’t know which branch of the wave function “I” will find myself in afterward. Yes, a copy of me will be in all of them, but in my personal subjective future timeline, I will only perceive myself to be in one. The probability is essentially the probability that I will find myself in each of those branches and seeing a particular quantum measurement outcome.

      But I assume it’s more complicated and I’m missing something?

      “Do you include under that umbrella things like vibrating atoms and pulsars?”
      I would think so. We measure the duration of each in relation to things we can subjectively experience.

      But I’m not necessarily on the time-is-emergent train. I agree that it does seem fundamental, and haven’t seen a compelling reason to doubt that. Not sure I would see it as more fundamental than the universe though, but this might get into how we’re defining “universe”. And I have to admit I haven’t studied why some physicists think time isn’t fundamental.

      Liked by 2 people

      1. “But I assume it’s more complicated and I’m missing something?”

        Beats me! I have the same outlook on probability that you do, that it’s a matter of knowledge. Apparently there are Bayesian and frequentist interpretations, and I’m suspicious of any science that requires interpretation.

        Which is a big part of why quantum physics really annoys me. 😀

        “We measure the duration of each in relation to things we can subjectively experience.”

        Right (which is why relative time is so hard to refute). It’s how reality seems filled with natural oscillators, from tiny vibrating atoms to giant pulsars, that I take as strong circumstantial evidence that time is fundamental.

        I’ve heard the assertion “there’s no time in physics,” and it always puzzles me, because nearly every important physics equation I’ve ever seen has that t in it somewhere.

        (What’s more common is the assertion those equations work regardless of the sign of t, so there’s no preferred direction of time. The “arrow of time” is another one of those little mysteries we’d really like to solve.)

        “Not sure I would see it as more fundamental than the universe though, but this might get into how we’re defining ‘universe’.”

        I meant the one formed by the Big Bang (or so we assume). There is some context in which the BB occurred. I think of that as the “meta-universe” — one that gives rise to a universe. Or universes, although I’m not big on multi-verse ideas. One is enough for me.

        “And I have to admit I haven’t studied why some physicists think time isn’t fundamental.”

        Apparently something else is, and it gives rise to time? I don’t get it, either. Emergence seems to factor in to many accounts I’ve heard.

        It often starts by pointing out how, for example, the Earth’s surface looks flat close up, and for local purposes can be considered flat, but if you step back and view it from a distance, not so flat anymore. Curvature emerges.

        This is held up as an example of emergence outside of time; just a matter of two perspectives. (I don’t buy it. The description embeds time within it: one must “step back” to the larger view. First there is one thing, then there is another.)

        In any event, the metaphor is intended to show that time could emerge from “something” (necessarily timeless!) even though emergence seems to occur in time, and it’s hard to conceive of any “timeless process” that could result in time emerging.

        Sometimes I think theoretical physics and higher math can lead to an experience not unlike taking LSD…

        Liked by 1 person

        1. I’m actually agnostic on whether there was a time before the big bang. Depending on which account of cosmic inflation you read, it either happened before or just after the beginning of the big bang. If before, then obviously you can’t have space inflating if there isn’t time for it to inflate. And under the eternal inflation model, time has to be there for all of it.

          Some people talk about the overall universe (or meta-universe) being a static structure with four or more dimensions, one of which is time. Stephen Wysong’s block universe comes to mind. Although I don’t think we can coherently speak of a perspective being outside of that structure, since such a perspective would require time to exist.

          Ethan Siegel has a post up about how photons don’t experience time. No massless particle that moves at the speed of light does. This also means that whatever direction their momentum is in doesn’t exist for them. Which seems to imply that mass and time are inescapably tangled up with each other. No mass means no time. I could maybe see this line of reasoning leading to time being some kind of side effect.

          Think about this too long and you might want drugs for other reasons.

          Liked by 1 person

          1. Time does seem a pretty deep mystery. It makes for a weird dimension; we’re all being dragged through it at a constant speed, and it’s strictly a one-way trip. What is the “now”?

            There’s the turtles problem, so at some point you just have to throw your hands up and accept that something just is. Beyond that, it’s all metaphysics.

            Kant thought time was an internal intuition, something so fundamental to us that it structures our thoughts. It’s not clear to me if he thought it was equally fundamental externally; I’m not clear on exactly how he stood on realism.

            (I’ve long associated the phrase “empirical realism” as the flip side to his “transcendental idealism” but recently tried to search for it and didn’t find much. It may be I saw it in some paper I read, but it isn’t common. It does seem to fit with his approach, from what I can tell.)

            What gets me about photons is, what is wavelength like from their perspective?

            Also, when they slow down going through water or glass, are they still timeless? Is c itself really slower in different mediums?

            “No mass means no time.”

            Time arises from mass? I tend to see mass as an emergent property (either from Higgs interactions or from energy). The Higgs part of it seems to assume (ha!) time for the interaction.

            Despite what I said above, E=mc2 contains no t. (In this case, we’d care more about: m=E/c2) Once a particle, even a massless photon, is moving, there’s a momentum term, which includes velocity, which is d/t.

            But the rest mass equation doesn’t reference time, for whatever that might mean…

            Liked by 1 person

          2. Good question on wavelength. In truth, the relation between wavelength and the photon has always been confusing for me. Photons are the quanta of electromagnetic radiation. They’re supposed to be discrete packets. When the wave that is the photon strikes a cone photoreceptor on our retina (or a photographic plate), what exactly is going on? If the wave collapses to a single point, how are the wavelengths then discriminated from each other?

            On what happens when light travels less than c in a medium, my understanding is that what’s actually happening is that individual photons are being absorbed by atoms in the medium and re-emitted, with the absorption and re-emission time adding to its overall phase velocity through the medium. So individual photons never travel at less than c. (Unless I’ve muffed my understanding of this, which is possible.)

            Which makes me think that, from the photon’s perspective, what it really is, is a connection between two charged particles to transfer energy. To the photon this happens instantly and across zero distance. But to the rest of the universe, it looks like is a packet of energy traveling between the particles at c. I changed “looks like” to “is” in the previous sentence because the expansion of space stretches photons, which is why distant galaxies are red shifted and the CMB is stretched into the microwave range.

            Liked by 1 person

          3. “If the wave collapses to a single point, how are the wavelengths then discriminated from each other?”

            For photons, energy, frequency, and wavelength all amount to the same thing, so the interaction is bound by the energy level. Like how materials can be transparent to some frequencies but opaque to others. The transparent ones, the photon energy levels don’t match the available electron jumps of that material, so they just pass through.

            Opaque materials, eye cones, photo film, CCD chips, their electron jumps match visible light (but not, say x-rays or radio).

            I like your explanation of why photons go slower in some materials; makes sense. It’s always struck me as slightly weird how carefully c is qualified as “in a vacuum.” Like somehow it validates in my head the idea of ontologically slower c otherwise.

            But it makes way more sense photons would never experience time always at c.

            “Which makes me think that, from the photon’s perspective, what it really is, is a connection between two charged particles to transfer energy.”

            When you think about it, that really is what all the bosons are about. Not always charged, maybe, but same idea of transferring energy between matter particles.

            I’ve never really thought this through, but I think what it boils down to is that what we perceive as color (wavelength, frequency) is just energy level. We see color because certain energy level photons can interact with molecules in our cone cells. Or molecules in a color filter or color film.

            Liked by 1 person

          4. I can see the frequency aspect matching up with the wavelength, so a shorter wavelength is equivalent to more frequent photons arriving and more energy coming in over a particular period of time. That also matches up with what happens in FM radio (frequency modulation). But what happens in AM (amplitude modulation) in terms of photons?

            I’ve read some answers that the photon itself varies in energy, but that doesn’t seem right. Isn’t the whole point of the photon is that it’s a quanta of electromagnetic energy? If its energy can vary, then we’re in a situation where its wavelength (size), frequency, and energy level can all vary, so what is so discrete and quantized about it?

            (I know Einstein won his Nobel for a good reason and this is just my ignorance.)

            Liked by 1 person

          5. Hmmm. According to this Quora, photons don’t have amplitude. The amplitude is a factor of how many photons are involved.
            But the energy of a single photon still varies, just in correlation with its wavelength, not its frequency of arrival. Which still leaves me wondering what’s so quantized about them.

            Forgot to say that I think you’re right about energy level and cones. A red sensitive cone has a slightly different chemical makeup than a green sensitive one. I assume it comes down to different atoms with different electron orbitals and which level of energy will cause those electrons to jump and induce an electrical flow, initiating a reaction that leads to an action potential.

            Liked by 1 person

          6. “…so a shorter wavelength is equivalent to more frequent photons arriving…”

            My understanding is that an individual photon can have an energy level (aka wavelength, aka frequency; frequency, of course, is just the inverse of wavelength).

            “But what happens in AM (amplitude modulation) in terms of photons?”

            Well, FM would require many photons with their frequency modulated to create an FM signal. AM would also require many photons with the modulation involving the “brightness” of the signal — i.e. the number of photons at any instant.

            In both cases, “modulation” would apply to the stream of photons (even in the case of radio).

            Ah, just read the second comment; will go read the link. You raise a really good question. If a photon is a quantized packet of energy, how can that energy packet have different levels?

            I suspect it will come down to something along the lines of a soliton, that what’s quantized is the soliton wavepacket comprising the “photon” in the EM field.

            “I assume it comes down to different atoms with different electron orbitals and which level of energy will cause those electrons to jump and induce an electrical flow, initiating a reaction that leads to an action potential.”

            That’s my understanding with the caveat that the interaction may be going on at the molecular level more than atomic, just in that most of the atoms are going to be CHON, but the molecular chemistry varies considerably.

            Liked by 1 person

          7. The Quora link doesn’t really answer our question. The closest is maybe from the answer from Dr. Stark:

            “If a photon increases in frequency, it becomes heavier E=hf. Frequency is related to wavelength by c=λf. The higher the frequency (or smaller the wavelength) – the more point-like it becomes. It makes the wavefunction look more like a dirac delta function localized at a point.”

            I’ll see if I can find a reference to how an individual photon can have a frequency (and hence a “color”). Or more directly, how a photon can have different energy levels at all if it’s a quanta.

            Liked by 1 person

      1. “That is, there was no time before the Big Bang.”

        As Stephen Hawking put it, there’s nothing north of the North Pole.

        Except this is wrong if one views “going north” as “going towards the North Star.” The Earth exists in a greater context, and “north” does have meaning in that context.

        Likewise I assume the Big Bang took place in some context, some framework of laws that allowed the Big Bang to occur. One can question whether location means anything in that context, but I think it’s harder to question whether time does.

        For the BB, or anything, to “occur,” time seems necessary to me.


  3. As a fresh thread…

    I think the probability issue in MWI is related to the perceived probability of the world we see around us. I believe it’s a variation of the multiverse anthropic problem.

    Without MWI, the current branch is the only branch, and its current state is the sum of all past events, which all had some probability of occurring. Given a Gaussian distribution, presumably this world lies roughly in the center of the bell curve.

    There have been unlikely events in our past, but far fewer than more likely events. Tautologically, likely events are more likely.

    With MWI, we apparently live on a branch among a vast multitude, and that branch just happens to be one of the more probable ones. Lucky us!

    So I guess the mission for MWI is to explain how it is we find ourselves on such a mundane branch. (Assuming I understand the issue at all.)

    AIUI, they attempt to show that the odds of all past events in non-MWI are the same as the odds of selecting the current branch in MWI.

    Liked by 1 person

    1. My understanding is that we can still see low probability quantum events. So while we’re more likely to be on the most mundane path, it’s not guaranteed. At least some of the time, we’ll be on the low probability path. For an outcome that only has a 5% chance of happening, we’ll see it 5% of the time.

      I think you mentioned the quantum suicide concept. I’m not sure if this is the same thing, but Max Tegmark described the quantum suicide or subjective immortality experiment. You set up a pistol that is set to fire on a quantum event that has a 50% chance of happening each second. You then put your head in front of it.

      In any one universe, you have a 50% chance of dying in the first second, meaning that you die in 50% of all universes. But in the other 50% you go on living. But you then have a 50% chance of dying in the second second, leading you to having been killed in the first two seconds in 75% of all universes but surviving in 25%. You continue sitting in front of the pistol, surviving in an ever smaller share of the universes.

      However, if the MWI is true, then from your subjective perspective, you go on living. With each second that passes, increasingly low probability events intervene to stop you from dying (lightning strike, earthquake, etc). After an hour of sitting with your head in front of the pistol, you could be confident that the MWI is true. Of course, in virtually all other universes, you would leave grieving friends and family behind who would be less convinced.

      But that means under the MWI that we all have quantum immortality. If the MWI is true, each of us will (subjectively) live to the end of the universe despite its improbability, and will know the MWI is true, providing we remember this thought exercise. (Or that there isn’t a logic error somewhere in this proposition.)

      Reportedly Hugh Everett had no fear of dying because he believed in quantum immortality.

      Liked by 1 person

      1. “I’m not sure if this is the same thing, but Max Tegmark described the quantum suicide or subjective immortality experiment.”

        Yep, same thing.

        “You continue sitting in front of the pistol, surviving in an ever smaller share of the universes.”

        With the caveat that, if we’re talking infinite universes, then any smaller share still results in infinite universes.

        “But that means under the MWI that we all have quantum immortality.”

        The mechanism that would allow immortality would seem a separate issue, but if one is possible, then surely it would have to happen.

        Or one might look at it as suggesting infinite copies of you will be born and die normal lives at infinite points in the future.

        It just seems like, if MWI is true, then there should be some account of something like quantum suicide where some very bizarre chain of events occurred.

        But who’s to say it didn’t? I’ve always thought intelligent life was tiny odds.

        Liked by 1 person

      2. However, if the MWI is true, then from your subjective perspective, you go on living

        This glosses over the fact that in many of the futures, you don’t have a perspective. Or you have a very brief very painful perspective. I would be very disturbed if the evidence indicated that in most states of the universe, my wife was grieving. Even if the only state I can directly observe is one where she isn’t.

        Quantum suicide? Don’t try this at home.

        Liked by 1 person

  4. How far have we come; really…. Sean Carrol may as well put a bunch of broken bones in a jar, shake them up and cast them onto the ground. Then he and his crony theoretical physicists can read the bones and tell us the future. Carrol and his cronies get paid big bucks for the type of garbage they feed our young people…… good luck with that one.

    There are two fundament mistakes theorists make when trying to build a model of reality. The first error is asserting that our phenomenal reality is grounded in a model of realism. The second error is the assertion that such a thing as “law”, a paradigm of which the laws of physics are a derivative. Law, that ethereal, nebulous, some “thing” that commands unwavering obedience from its unknowing, unsuspecting subjects. Get real here. It’s time to move past the superstition of the dark ages intrinsic to the twenty-first century. Neither one of those two grounding predicates are true, they are primordial architectures grounded in superstition.


    Liked by 1 person

    1. Hey Lee,
      I think most scientists would characterize the word “law” in this context as being a kind of metaphor in reference to consistent regularities. They’re not so much rules as standard processes by which reality works. Of course, the first people to use that word were thinking it actually was God’s laws, but the scientific usage has long since evolved away from that, at least for most scientists.

      But if you see even regularities as problematic, what would you suggest in its place?


      1. Mike,

        I don’t have a problem with regularities as you perceive the context of law. What I have a problem with is when leaders within the scientific community dogmatically state that some “thing”, whatever that thing is has to “obey” the laws of physics. That type of vocabulary is oppressive and down right abusive. Imagine what type of response a man would get if he asserted that his partner was required to obey the laws which govern their relationship…


        Liked by 1 person

      1. Smythe,

        I read your post….. Bottom line, cognito ego sum is not a foundation for anything. I’m not saying that our experienced reality is not real in its own context. Context is the qualifier here. Kant brought to the West Nagarjuna’s Two Truths doctrine and Kant’s model of transcendental idealism is the ideal model because it clearly and definitively draws a line of distinction between reality and the appearance of reality. Ours is the appearance of reality and as long as we “insist” upon the world of appearances being the “real” reality and not merely the appearance, we will forever be entangled in a maze of duality.



  5. *Currently * verifiable? I don’t see any need to restrict scientific theorizing to what’s currently verifiable. What really matters is what is ultimately verifiable. Of course that’s a lot harder to evaluate, to put it mildly, but tough noogies.

    Liked by 1 person

Your thoughts?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.