Saturday, June 08, 2019

Book Review: “Beyond Weird” by Philip Ball

Beyond Weird: Why Everything You Thought You Knew about Quantum Physics Is Different
By Philip Ball
University of Chicago Press (October 18, 2018)

I avoid popular science articles about quantum mechanics. It’s not that I am not interested, it’s that I don’t understand them. Give me a Hamiltonian, a tensor-product expansion, and some unitary operators, and I can deal with that. But give me stories about separating a cat from its grin, the many worlds of Wigner’s friend, or suicides in which you both die and not die, and I admit defeat on paragraph two.

Ball is guilty of some of that. I got lost half through his explanation how a machine outputs plush cats and dogs when Alice and Bob put in quantum coins, and still haven’t figured out why the seer’s daughter wanted to be wed to a man evidently more stupid than she.

But then, clearly, I am not the book’s intended audience, so let me instead tell you something more helpful.

Ball knows what he writes about, that’s obvious from page one. For all I can tell the science in his book is flawless. It is also engagingly told, with some history but not too much, with some reference to current research, but not too much, with some philosophical discourse but not too much. Altogether, it is a well-balanced mix that should be understandable for everyone, even those without prior knowledge of the topic. And I entirely agree with Ball that calling quantum mechanics “weird” or “strange” isn’t helpful.

In “Beyond Weird,” Ball does a great job sorting out the most common confusions about quantum mechanics, such as that it is about discretization (it is not), that it defies the speed of light limit (it does not), or that it tells you something about consciousness (huh?). Ball even cleans up with the myth that Einstein hated quantum mechanics (he did not), Feynman dubbed the Copenhagen interpretation “Shut up and calculate” (he did not, also, there isn’t really such a thing as the Copenhagen interpretation), and, best of all, clears out the idea that many worlds solves the measurement problem (it does not).

In Ball’s book, you will learn just what quantum mechanics is (uncertainty, entanglement, superpositions, (de)coherence, measurement, non-locality, contextuality, etc), what the major interpretations of quantum mechanics are (Copenhagen, QBism, Many Worlds, Collapse models, Pilot Waves), and what the currently discussed issues are (epistemic vs ontic, quantum computing, the role of information).

As someone who still likes to read printed books, let me also mention that Ball’s is just a pretty book. It’s a high quality print in a generously spaced and well-readable font, the chapters are short, and the figures are lovely, hand-drawn illustrations. I much enjoyed reading it.

It is also remarkable that “Beyond Weird” has little overlap with two other recent books on quantum mechanics which I reviewed: Chad Orzel’s “Breakfast With Einstein” and Anil Ananthaswamy’s “Through Two Doors At Once.” While Ball focuses on the theory and its interpretation, Orzel’s book is about applications of quantum mechanics, and Ananthaswamy’s is about experimental milestones in the development and understanding of the theory. The three books together make an awesome combination.

And luckily the subtitle of Philip Ball’s book turned out to be wrong. I would have been disturbed indeed had everything I thought I knew about quantum physics been different.

[Disclaimer: Free review copy.]

Related: Check out my list of 10 Essentials of Quantum Mechanics.


  1. What makes quantum mechanics weird is that we observe it from a classical perspective. In one sense what you say about Hamiltonians, matrices, operators and calculate is the case. Quantum mechanics by itself is really rather straight forwards. The problem is that we observe things from a classical perspective, and quantum mechanics is also placed within a classical stage called spacetime. Because of this things now do indeed appear very odd. So the real question is then. “why is there a classical world at all?” We have our usual ideas of large masses, large momenta, large action S >> ħ and the wavelength argument going back to de Broglie that λ = h/p, for p momenta and h = 2πħ. For large enough system with huge action or momentum the quantum stuff is just insignificant. However, we really do not know how classical reality is stable. Zurek's einselection program tells us there can be states stable against any reservoir of states, but as yet beyond some level of hand waving we really do not know why there exist these states that for “large N” are stable and so called classical.

    There is this result that is making the rounds

    which has features similar to a weak measurement. The transition of an atom, here a quantum dot or artificial atom, is found to have measurable influence on an ancillary state as it is about the transition. It does not describe the entire transition, but indicates there is something going on beyond our usual idea of a sudden collapse. The collapse or state reduction works for a classical system with λ_c = h/p << λ_q because the classical system momentum, even if it is p = E/c or mc for mass in a rest frame, is much larger than the wavelength of the quantum system. Therefore the time of this state reduction, T_red << t = 1/ν of the system, and is high frequency information irrelevant to the measurement. However, for a quantum system as an L^½ system transitioning into a classical stochastic system L^1 by decoherence, we can question whether there is a phase where the quantum system is quantum chaotic or exhibits strange behavior. This might occur on a short time scale we are able to probe. We may then address questions this way on how classically stable states exist.

  2. "... what the major interpretations of quantum mechanics are ..." The book "Beyond Weird" contains no mention of Milgrom or MOND. I say that Milgrom should have been awarded a Nobel prize 15 years ago .... no matter what the empirical successes of MOND really might mean in terms of the foundations of physics, I find it difficult to believe that MOND is irrelevant to the foundations of physics.
    "The first step is that we need to revisit the validity of Newton's universal law of gravitation." — Pavel Kroupa
    Pavel Kroupa, Wikiquote

    1. What the heck does MOND have to do with the interpretations of quantum mechanics? MOND is a classical theory.

    2. Dr. Hossenfelder: Your work is mentioned in this post.

      David Brown: Although MOND would modify gravity, it would only do so at galaxy scale distances; even on our solar scale and laboratory scale we need no change to our existing models of gravity. So the modification you are talking about has nothing to say about quantizing gravity, which is the only way a new theory of gravity would have impact on the foundations of physics.

      I'd also mention there has been recent observation of two galaxies that seem to have no dark matter.

      An excerpt from the end of that article:

      "Because the effects of dark matter are evident in every other known galaxy except DF2 and DF4, the team's findings indicate dark matter must be a tangible substance that is separable from normal matter. This means that some alternative theories to dark matter, such as Modified Newtonian Dynamics (MOND) — which introduces extra gravitational forces on galactic scales — fall flat."

      In other words, a new rule of galactic scale gravity cannot be selective and leave some galaxies out.

      However, other things might be selective, a property like superfluidity in dark matter that may requires a certain minimum density.

      Dr. Hossenfelder talks about superfluid dark matter in this video:

      And also in this blog post:

      Under the hypothesis, there would be two phases of Dark Matter; one "normal" and the other, with sufficient gravitational pressure, a superfluid that exerts an additional force that looks like MOND.

      But Both DF2 and DF4 are extremely diffuse galaxies, with about 1/200th the number of stars as our own. Thus they may exert insufficient gravitational pressure to condense their dark matter into a superfluid.

      In either case, MOND would a misnomer; it doesn't actually change the rules of gravity. Like Newton's gravity is an approximation of Einstein's gravity that doesn't account for extreme circumstances, MOND would be an approximation of the behavior of superfluid dark matter that also doesn't account for extreme circumstances (like very diffuse galaxies).

      That said, there is a new job for astronomy; characterizing the threshold of galactic density to narrow down the transition point for dark matter exhibiting "normal" behavior or "superfluid" behavior, which conceivably might tell us something new about dark matter itself.

      To begin with, we have bucket loads of galaxies that DO seem dark-matter-ish, so one boundary on the density is the least-dense of these (i.e. superfluidity prevails at or before that density is reached).

    3. The description of MOND introducing new forces on galactic scales isn't right: it is not about distance/scale at all. It introduces new forces at low accelerations, and can become Newtonian ("dark matterless") under the external field effect which is frequently neglected by those trying to falsify MOND. Although ideally one would really try to falsify a relativistic version: we already know MOND is false because it is only Newtonian.

      See the blog Triton Station for more details: when calculated properly DF2 was actually within MOND's range of error. I don't know about DF4, but as an ultra diffuse I would be surprised if it had significantly different conditions. You must keep in mind that any time someone talks about MOND modifying gravity at scale/distance that they are attacking a straw man version, as this is not what MOND is about at all.

      MOND directly has nothing to do with quantum gravity, but it may well be that a proper theory of quantum gravity will have MOND as a side-effect, as we already know MOND would have to be modified to be right and that gravity must be modified on the quantum scale. Since this is rather an indirect currently purely hypothetical relationship (that would be more about unifying quantum field theory than pure quantum mechanics as well) I think the book was correct to leave out mentioning it. Not all books on quantum mechanics need to be about unifying it with gravity.

  3. Thanks so much for this Sabine. I'm glad it mostly worked for you. It's a funny thing about the PR box (cats and dogs) analogy. When Sandu Popescu explained this to me (without cats & dogs, but same thing), a penny really dropped for me about how the issue was related to sharing of information - and how one could formulate a Bell experiment in these terms in a way that added a fresh perspective. But it seems that lots of people don't really get the analogy. Perhaps it's just the way I've explained it... I don't know. Glad you liked the diagrams too! - I wanted to get away from the usual coldly formal line drawings and create a more "blackboard" feel. Anyway, I'm very grateful!

    1. These cats and dogs here are indeed so cute.
      Maybe QM does not use PR boxes, because it is not only about sharing information but mainly about generating new information. (Shannon entropy, randomness, flow of time, L² spaces and its duals)

    2. I am a big fan of this blog and Dr. Hossenfelder's work, but I'm a first time commenter. This seems like a good time to ask a burning question I have had for many years.

      I read Beyond Weird and was waiting anxiously for a discussion of the relational interpretation of quantum mechanics (as popularized in a recent book by Carlo Rovelli), but it only got a passing mention, which is consistent with other discussions of QM interpretations I have read.

      Does this happen because the relational interpretation has an obvious flaw? If not, why does it not get more attention, since, in my opinion, it is easier to swallow ontologically than Many Worlds?


    3. Tim E,

      I wrote about the relational interpretation a long time ago (one of the first things I wrote on this blog). I don't know of course why it receives so little attention but I would guess it's because really it doesn't solve any problem. At least I don't see how.

  4. Could you help me understand and flesh this out a little? Or is it just superposition?

    "....there isn’t really such a thing as the Copenhagen interpretation), and, best of all, clears out the idea that many worlds solves the measurement problem (it does not).

    In Ball’s book, you will learn just what quantum mechanics is (uncertainty, entanglement, superpositions, (de)coherence, measurement, non-locality, contextuality, etc), what the major interpretations of quantum mechanics are (Copenhagen... "

    1. Peter,

      To say the obvious, the best way to answer your question is to read Ball's book. I don't mean anything deep, just that the Copenhagen people didn't exactly write down a clear definition, and no one really knows what Bohr was talking about anyways. So if you ask three physicists what the Copenhagen interpretation is, you'll get four answers.

  5. I too just finished Ball's book and am still sorting out my thoughts about it, but I'd certainly recommend it to anyone who is already somewhat familiar with the subject. It's refreshingly non-polemical, doesn't shy away from asking all the hard questions, and doesn't get bogged down in historical disputes over exactly what some dead guy meant when he said such-and-such.

    That said, there were many passages that I didn't understand due to what seemed to be careless choice of words. And there were many other passages that I understood only because of prior exposure to the subject, but that someone less experienced could easily misunderstand due to careless choice of words. I'm unsure of who exactly the intended audience for this book is.

    I was also annoyed by the lack of a table of contents, the lack of chapter numbers, and the need to perform lengthy sequential searches when going either forward or backward between the main text and the end notes.

  6. Forgive me, but I feel "there isn’t really such a thing as the Copenhagen interpretation" is one of those true but not very helpful things that experts like to say.

    For example, this is in the Encyclopedia Brittanica: "Bohr's viewpoint, which became known as the Copenhagen interpretation of quantum mechanics, was that reality can be ascribed only to a measurement."

    I've read enough to understand (I think) that there was no single coherent 😁 viewpoint shared by all the founders of QM; but at the same time, I sympathize with a prospective who comes to your blog, runs across a statement like that after perhaps daring to read the Britannica article, and never comes back.

  7. A weird duck in the icehole seems an apt way of putting it.

  8. Good review. I don't usually buy pop-q books either -- I recommend Feynman's QED to folks instead -- but your description makes this one sound (maybe?) worthwhile. The $$-sound is some cash landing in Philip Ball's Amazon account... :)

  9. OK, a quick read of the first 1/3 of the book confirms a suspicion of mine about human nature: We are a very noisy species. Sometimes out of that noise, pristine clarity emerges... and sometimes not. Quantum remains in the "not" column.

  10. off topic, but ethan siegal is crowing the triump of dark matter over mond here, and how the bullet cluster is re-examine to support dark matter, he acknowledge about the speed of collision which is attributed to magnetic fields here

  11. There are 3 views of QM: philosophical, mathematical and physical. The interpretations are philosophical. Hamiltonian, groups, wave function, Hilbert spaces are mathematical. The experimental results are physical. We say we understand QM when we can match observations and calculations. It doesn’t mean we know the Truth because different mathematical formulations can describe the same set of observations.

    Leave the philosophizing to philosophers. This is what Feynman means when he said “nobody understands quantum mechanics.” But then who understands metaphysics? As Wittgenstein said metaphysics is nonsense. It is not nonsensical in itself but anything you say about it must be. Do we know quantum physics? IMO “knowing” is conjuring a simple mental picture that we assume to be true because it is easier to understand according to the scientific method. Schrodinger’s cat is surreal as the Cheshire cat.

  12. My own consideration on books like these is to look inside and see if they have at least a sufficiently complete treatment of sum-over-histories in the mix with their other content, at the level of, for example*, Fay Dowker and Rafael D. Sorkin. If they don't, I don't have any interest in reading them.

    * Hilbert Spaces from Path Integrals
    Fay Dowker, Steven Johnston, Rafael D. Sorkin
    "It is shown that a Hilbert space can be constructed for a quantum system starting from a framework in which histories are fundamental."

  13. Dear Sabine,
    I wonder if you can clarify why you state that "there isn’t really such a thing as the Copenhagen interpretation", but then you recognize that in this book one can learn "what the major interpretations of quantum mechanics are (Copenhagen, QBism, Many Worlds, Collapse models, Pilot Waves)", to me taking for granted that all those interpretations exists.

    May be you mean that there is no "one" official and unique definition of the Copenaghen interpretation? Or what?



  14. I find it quite amusing and right to the point how Claes Johnson, a Professor of Applied Mathematics, classifies QM: “Quantum Mechanics is unphysical because its interpretation is statistical which makes it non-physical, because physics is not an insurance company.”

  15. The comments here seem to me to underestimate the importance of the empirical successes of MOND. MOND's empirical successes seem to suggest that something is profoundly wrong with Newtonian-Einsteinian gravitational theory and at least one of Newtonian's 3 laws of motion — and perhaps the 1st law of thermodynamics. My guess is that the necessity of a new paradigm of the foundations of physics is demonstrated by the empirical successes of MOND.
    Kroupa, Pavel, Marcel Pawlowski, and Mordehai Milgrom. "The failures of the standard model of cosmology require a new paradigm." In THE THIRTEENTH MARCEL GROSSMANN MEETING: On Recent Developments in Theoretical and Experimental General Relativity, Astrophysics and Relativistic Field Theories, pp. 696-707. 2015.


    Is this the part about many worlds and measurement problem? I did not see any problem. It seems like the author did not like the conclusion and therefore decided it was wrong. Maybe I missed a deep point or maybe there is an other argument in the book.

    1. Matti,

      Yes, that is the part. I think it explains the problem very well. I doubt that if Ball's explanation doesn't work for you I can do any better. Though I may write something about the topic in future blogpost.



Comment moderation on this blog is turned on.
Submitted comments will only appear after manual approval, which can take up to 24 hours.
Comments posted as "Unknown" go straight to junk. You may have to click on the orange-white blogger icon next to your name to change to a different account.