Physics is Broken!!!

Last Thursday, an experiment reported that the magnetic properties of the muon, the electron’s middleweight cousin, are a tiny bit different from what particle physics equations say they should be. All around the world, the headlines screamed: PHYSICS IS BROKEN!!! And indeed, it’s been pretty shocking to physicists everywhere. For instance, my equations are working erratically; many of the calculations I tried this weekend came out upside-down or backwards. Even worse, my stove froze my coffee instead of heating it, I just barely prevented my car from floating out of my garage into the trees, and my desk clock broke and spilled time all over the floor. What a mess!

Broken, eh? When we say a coffee machine or a computer is broken, it means it doesn’t work. It’s unavailable until it’s fixed. When a glass is broken, it’s shattered into pieces. We need a new one. I know it’s cute to say that so-and-so’s video “broke the internet.” But aren’t we going a little too far now? Nothing’s broken about physics; it works just as well today as it did a month ago.

More reasonable headlines have suggested that “the laws of physics have been broken”. That’s better; I know what it means to break a law. (Though the metaphor is imperfect, since if I were to break a state law, I’d be punished, whereas if an object were to break a fundamental law of physics, that law would have to be revised!) But as is true in the legal system, not all physics laws, and not all violations of law, are equally significant.

What’s a physics law, anyway? Crudely, physics is a strategy for making predictions about the behavior of physical objects, based on a set of equations and a conceptual framework for using those equations. Sometimes we refer to the equations as laws; sometimes parts of the conceptual framework are referred to that way.

But that story has layers. Physics has an underlying conceptual foundation, which includes the pillar of quantum physics and its view of reality, and the pillar of Einstein’s relativity and its view of space and time. (There are other pillars too, such as those of statistical mechanics, but let me not complicate the story now.) That foundation supports many research areas of physics. Within particle physics itself, these two pillars are combined into a more detailed framework, with concepts and equations that go by the name of “quantum effective field theory” (“QEFT”). But QEFT is still very general; this framework can describe an enormous number of possible universes, most with completely different particles and forces from the ones we have in our own universe. We can start making predictions for real-world experiments only when we put the electron, the muon, the photon, and all the other familiar particles and forces into our equations, building up a specific example of a QEFT known as “The Standard Model of particle physics.”

All along the way there are equations and rules that you might call “laws.” They too come in layers. The Standard Model itself, as a specific QEFT, has few high-level laws: there are no principles telling us why quarks exist, why there is one type of photon rather than two, or why the weak nuclear force is so weak. The few laws it does have are mostly low-level, true of our universe but not essential to it.

I’m bringing attention to these layers because an experiment might cause a problem for one layer but not another. I think you could only fairly suggest that “physics is broken” if data were putting a foundational pillar of the entire field into question. And to say “the laws of physics have been violated”, emphasis on the word “the“, is a bit melodramatic if the only thing that’s been violated is a low-level, dispensable law.

Has physics, as a whole, ever broken? You could argue that Newton’s 17th century foundation, which underpinned the next two centuries of physics, broke at the turn of the 20th century. Just after 1900, Newton-style equations had to be replaced by equations of a substantially different type; the ways physicists used the equations changed, and the concepts, the language, and even the goals of physics changed. For instance, in Newtonian physics, you can predict the outcome of any experiment, at least in principle; in post-Newtonian quantum physics, you often can only predict the probability for one or another outcome, even in principle. And in Newtonian physics we all agree what time it is; in Einsteinian physics, different observers experience time differently and there is no universal clock that we all agree on. These were immense changes in the foundation of the field.

Conversely, you could also argue that physics didn’t break; it was just remodeled and expanded. No one who’d been studying steam engines or wind erosion or electrical circuit diagrams had to throw out their books and start again from scratch. In fact this “broken” Newtonian physics is still taught in physics classes, and many physicists and engineers never use anything else. If you’re studying the physics of weather, or building a bridge, Newtonian physics is just fine. The fact that Newton-style equations are an incomplete description of the world — that there are phenomena they can’t describe properly — doesn’t invalidate them when they’re applied within their wheelhouse.

No matter which argument you prefer, it’s hard to see how to justify the phrase “physics is broken” without a profound revolution that overthrows foundational concepts. It’s rare for a serious threat to foundations to arise suddenly, because few experiments can single-handedly put fundamental principles at risk. [The infamous case of the “faster-than-light neutrinos” provides an exception. Had that experiment been correct, it would have invalidated Einstein’s relativity principles. But few of us were surprised when a glaring error turned up.]

In the Standard Model, the electron, muon and tau particles (known as the “charged leptons”) are all identical except for their masses. (More fundamentally, they have different interactions with the Higgs field, from which their rest masses arise.) This almost-identity is sometimes stated as a “principle of lepton universality.” Oh, wow, a principle — a law! But here’s the thing. Some principles are enormously important; the principles of Einsteinian relativity determine how cause and effect work in our universe, and you can’t drop them without running into big paradoxes. Other principles are weak, and could easily be discarded without making a mess of any other part of physics. The principle of lepton universality is one of these. In fact, if you extend the Standard Model by adding new particles to its equations, it can be difficult to avoid ruining this fragile principle. [In a sense, the Higgs field has already violated the principle, but we don’t hold that against it.]

All the fuss is about a new experimental result which confirms an older one and slightly disagrees with the latest theoretical predictions, which are made using the Standard Model’s equations. What could be the cause of the discrepancy? One possibility is that it arises from a previously unknown difference between muons and electrons — from a violation of the principle of lepton universality. For those who live and breathe particle physics, breaking lepton universality would be a big deal; there’d be lots of adventure in trying to figure out which of the many possible extensions of the Standard Model could actually explain what broke this law. That’s why the scientists involved sound so excited.

But the failure of lepton universality wouldn’t come as a huge surprise. From certain points of view, the surprise is that the principle has survived this long! Since this low-level law is easily violated, its demise may not lead us to a profound new understanding of the world. It’s way too early for headlines that argue that what’s at stake is the existence of “forms of matter and energy vital to the nature and evolution of the cosmos.” No one can say how much is at stake; it might be a lot, or just a little.

In particular, there’s absolutely no evidence that physics is broken, or even that particle physics is broken. The pillars of physics and QEFT are not (yet) threatened. Even to say that “the Standard Model might be broken” seems a bit melodramatic to me. Does adding a new wing to a house require “breaking” the house? Typically you can still live in the place while it’s being extended. The Standard Model’s many successes suggest that it might survive largely intact as a recognizable part of a larger, more complete set of equations.

In any case, right now it’s still too early to say anything so loudly. The apparent discrepancy may not survive the heavy scrutiny it is coming under. There’s plenty of controversy about the theoretical prediction for muon magnetism; the required calculation is extraordinarily complex, elaborate and difficult.

So, from my perspective, the headlines of the past week are way over the top. The idea that a single measurement of the muon’s magnetism could “shake physics to its core“, as claimed in another headline I happened upon, is amusing at best. Physics, and its older subdisciplines, have over time become very difficult to break, or even shake. That’s the way it should be, when science is working properly. And that’s why we can safely base the modern global economy on scientific knowledge; it’s unlikely that a single surprise could instantly invalidate large chunks of its foundation.

Some readers may view the extreme, click-baiting headlines as harmless. Maybe I’m overly concerned about them. But don’t they implicitly suggest that one day we will suddenly find physics “upended”, and in need of a complete top-to-bottom overhaul? To imply physics can “break” so easily makes a mockery of science’s strengths, and obscures the process by which scientific knowledge is obtained. And how can it be good to claim “physics is broken” and “the laws of physics have been broken” over and over and over again, in stories that almost never merit that level of hype and eventually turn out to have been much ado about nada? The constant manufacturing of scientific crisis cannot possibly be lost on readers, who I suspect are becoming increasingly jaded. At some point readers may become as skeptical of science journalism, and the science it describes, as they are of advertising; it’s all lies, so caveat emptor. That’s not where we want our society to be. As we are seeing in spades during the current pandemic, there can be serious consequences when “TRUST IN SCIENCE IS BROKEN!!!

A final footnote: Ironically, the Standard Model itself poses one of the biggest threats to the framework of QEFT. The discovery of the Higgs boson and nothing else (so far) at the Large Hadron Collider poses a conceptual challenge — the “naturalness” problem. There’s no sharp paradox, which is why I can’t promise you that the framework of QEFT will someday break if it isn’t resolved. But the breakdown of lepton universality might someday help solve the naturalness problem, by requiring a more “natural” extension of the Standard Model, and thus might actually save QEFT instead of “breaking” it.

26 responses to “Physics is Broken!!!

  1. Excellent. Informative. Lucid.

  2. Peter Stiphout

    Perfectly stated and truly exposed the sensationalism of so many media outlets. Well done and thanks for stating the truth as succinctly as possible. Cheers

  3. Claude Deschenes

    Thank you Matt, always a lightning pleasure to read you.

  4. Ray Stefanski

    The physics community needs to hear your message. After WWII, and the discovery of nuclear energy, physics grew into an enterprise, no longer able to maintain the discipline of the Scientific Method. As the enterprise tries to maintain itself, it must attract funding, mostly from government entities. The hyperbola evident in media physics is, I think, a clumsy attempt to influence decision makers with ever diminishing scientific results: After all the more obvious science is done. Thing have gotten much more difficult.
    But the hyperbola are very damaging to scientific integrity. This integrity, the value of discovery of well founded insights into nature, are our only deliverable. The trade of scientific integrity for funding is morally intolerable, and threatens the survival of the enterprise we’re trying to preserve.
    Thanks for a well written article.

    • Torbjörn Larsson

      There is very little statistics that say there is hyperbole one way or another.

      That goes for claiming that modern science is broken – which it obviously isn’t, read Strassler’s article explaining why – as well, but it is a notable and unfortunate phenomena of the new social media industry. To use sophistry, it may quite possibly threaten that industry by its morally intolerable ease of commenting opinion instead of, say, science facts..

  5. This would not be a reading error?, Or data collection or sighting? I’m not an expert but I think I should redo and redo it all again. Let’s not forget the Opera experiments in Gran Sasso. Thats is my humble opinion.

  6. I love your introduction.

  7. Quick question, why are you calling QFT QEFT? Why not just QFT since that’s the underlying framework?

    • The Standard Model is a QEFT, and even more so If you want to include semiclassical gravity with it. The U(1) has a Landau pole and the neutrino masses may be non-renormalizable operators, and of course gravity isn’t understood near and above the Planck scale. So I think it’s more accurate to describe our understanding of the world in this way.

  8. Torbjörn Larsson

    Nice rant. Indeed, since the Standard Model is an effective field theory we know it is should have corrections at different energy scales [ https://en.wikipedia.org/wiki/Renormalization ]. “… distant scales are related to each other through “effective” descriptions. All scales are linked in a broadly systematic way, …”.

    I discovered that Weinberg describes the laws in the Standard Model in a recent review of effective field theories [ https://link.springer.com/content/pdf/10.1140/epjh/s13129-021-00004-x.pdf ]. ” All these were accidental symmetries, imposed by the simplicity of the Standard Model necessary for renormalizability plus other symmetries like gauge symmetries and Lorentz invariance that seem truly exact and fundamental. Chiral symmetry [i.e. Dirac particles with matter # antimatter] itself is such an accidental symmetry, though only approximate.”

    Re “requiring a more “natural” extension of the Standard Model”, it touches on that too, I think. This may be arguable, but I find papers claiming that neutrino oscillations are not part of the Standard Model [ https://fas.org/sgp/othergov/doe/lanl/pubs/00326607.pdf ]. “e. We could introduce a Dirac mass term for the neutrino that would mirror the mass term for the electron. It would have the form m_v v_e^c n_e . (5) But, as we said above, the field v_e^c is not included in the Standard Model because, so far, weak-interaction experiments have not required it.” What Weinberg claim, as I understand it, is that this sterile Majorana correction is easier than for leptons than baryons at higher energies.

    “Wilczek and Zee and I independently did a catalog of the leading terms of this type. Some of them—those involving baryon number non-conservation—give you corrections of O((E/M)^2). They have not been yet been discovered experimentally. But there are other terms that produce corrections of O(E/M) that violate lepton conservation, and they apparently have been discovered, in the form of neutrino masses.”

    Finally, let me give Ethan Siegel the last word [ https://www.forbes.com/sites/startswithabang/2021/04/08/why-you-should-doubt-new-physics-from-the-latest-muon-g-2-results/?sh=719bde1b6c4b ].

    “As the above graph shows, the R-ratio method and the Lattice QCD methods disagree, and they disagree at levels that are significantly greater than the uncertainties between them. The advantage of Lattice QCD is that it’s a purely theory-and-simulation-driven approach to the problem, rather than leveraging experimental inputs to derive a secondary theoretical prediction; the disadvantage is that the errors are still quite large.

    What’s remarkable, compelling, and troubling, however, is that the latest Lattice QCD results favor the experimentally measured value and not the theoretical R-ratio value. … this discrepancy between two different methods of predicting the Standard Model’s expected value — one of which agrees with experiment and one of which does not — needs to be resolved before any conclusions about “new physics” can responsibly be drawn.”

  9. kashyap Vasavada

    I understand the European group (?) claims that their strong interaction data-based calculation is good and the results agree with Fermi lab expt. On the other hand, Fermi lab theorists maintain that there is 4.2 STD disagreement! At this point, you may not want to bet on who is right. But isn’t data-based analysis more reliable?
    Kashyap Vasavada

    • All the calculations use data to a degree. But there’s a lot of theory that goes into how to use the data. So none of it is trivial. Beyond the scope of this article, for sure, and currently beyond my full understanding.

  10. kashyap Vasavada

    I understand the European group (?) claims that their strong interaction data-based calculation is good and the results agree with Fermi lab expt. On the other hand,
    Fermi lab theorists maintain that there is 4.2 STD disagreement! At this point, you may not want to bet on who is right. But isn’t data-based analysis more reliable?
    Kashyap Vasavada

  11. I know this is kind of a tangent, but—If there was more than one kind of photon, how would you tell?

  12. Robert A. Dorrough

    That’s the trouble with denizens of the 24 hour news cycle, they must say something whether or not they understand the subject. If they were cows people would be scratching their heads wondering what they’re mooing on about. Used to be a mu meson but no longer considered a meson more of a lepton.
    It is true that if one breaks a state law one may be punished. If one breaks a physical law and writes their own they’re likely to win a Nobel.

  13. Pamela Collins

    I hated seeing and hearing the “broken” headlines. The reporters have no clue snd are irresponsibly giving people more fodder for not trusting science.

  14. Thanks Professor. Good to know. I am currently unchaining my car in my garage safe in the knowledge that it will not float away…. yet. I will keep the chains just in case.

  15. Dr. Victor Ottaviani

    Not broken ….just plane wrong….There is one truth in science: today’s scientific fact will be tomorrow’s chuckle.

  16. Vernon Wilson

    Hi Prof, excellent (wish I understood). Could our universe be spinning and the centrifical forces created be mistaken for dark energy?

  17. While misrepresenting scientific events for sensationalistic purposes is a mainstay of media reporting, there are actual scientists out there who help to perpetuate such misrepresentation. They regularly feed the media with
    hyperbole, preferring to sacrifice scientific integrity for personal recognition and exposure. What’s especially harmful is that they are so-called “media darlings”, the ones the media constantly targets when there is any sort of scientific discovery that warrants clarification.

  18. Stumbled upon this post at random and I’m glad I did. Well-written and informative. Brilliant points you make regarding the sensationalism of physics.

  19. Problem is not with theories but with what people make with them. Our understanding of physics was thought to be profound but as always the universe finds a way to humble us and let us know that we are working hard but it is not enough. Our theories of the universe are not enough, if we had read about a thing called dark matter in a essay published in the 14th century we wouldnt be surprised for such a classification…. I guess that says most of what I think…. We must be humble regarding our knowledge. Physic laws are not broken, only our ideas.

  20. It’s only a minor problem, and I think Stephen and I have fixed it.

    Not to worry.

  21. It’s only a minor problem, and Stephen and I think we have fixed it.