Changing the Question

I’ve recently been reading Why Does the World Exist?, a book by the journalist Jim Holt. In it he interviews a range of physicists and philosophers, asking each the question in the title. As the book goes on, he concludes that physicists can’t possibly give him the answer he’s looking for: even if physicists explain the entire universe from simple physical laws, they still would need to explain why those laws exist. A bit disappointed, he turns back to the philosophers.

Something about Holt’s account rubs me the wrong way. Yes, it’s true that physics can’t answer this kind of philosophical problem, at least not in a logically rigorous way. But I think we do have a chance of answering the question nonetheless…by eclipsing it with a better question.

How would that work? Let’s consider a historical example.

Does the Earth go around the Sun, or does the Sun go around the Earth? We learn in school that this is a solved question: Copernicus was right, the Earth goes around the Sun.

The details are a bit more subtle, though. The Sun and the Earth both attract each other: while it is a good approximation to treat the Sun as fixed, in reality it and the Earth both move in elliptical orbits around the same focus (which is close to, but not exactly, the center of the Sun). Furthermore, this is all dependent on your choice of reference frame: if you wish you can choose coordinates in which the Earth stays still while the Sun moves.

So what stops a modern-day Tycho Brahe from arguing that the Sun and the stars and everything else orbit around the Earth?

The reason we aren’t still debating the Copernican versus the Tychonic system isn’t that we proved Copernicus right. Instead, we replaced the old question with a better one. We don’t actually care which object is the center of the universe. What we care about is whether we can make predictions, and what mathematical laws we need to do so. Newton’s law of universal gravitation lets us calculate the motion of the solar system. It’s easier to teach it by talking about the Earth going around the Sun, so we talk about it that way. The “philosophical” question, about the “center of the universe”, has been explained away by the more interesting practical question.

My suspicion is that other philosophical questions will be solved in this way. Maybe physicists can’t solve the ultimate philosophical question, of why the laws of physics are one way and not another. But if we can predict unexpected laws and match observations of the early universe, then we’re most of the way to making the question irrelevant. Similarly, perhaps neuroscientists will never truly solve the mystery of consciousness, at least the way philosophers frame it today. Nevertheless, if they can describe brains well enough to understand why we act like we’re conscious, if they have something in their explanation that looks sufficiently “consciousness-like”, then it won’t matter if they meet the philosophical requirements, people simply won’t care. The question will have been eaten by a more interesting question.

This can happen in physics by itself, without reference to philosophy. Indeed, it may happen again soon. In the New Yorker this week, Natalie Wolchover has an article in which she talks to Nima Arkani-Hamed about the search for better principles to describe the universe. In it, Nima talks about looking for a deep mathematical question that the laws of physics answer. Peter Woit has expressed confusion that Nima can both believe this and pursue various complicated, far-fetched, and at times frankly ugly ideas for new physics.

I think the way to reconcile these two perspectives is to know that Nima takes naturalness seriously. The naturalness argument in physics states that physics as we currently see it is “unnatural”, in particular, that we can’t get it cleanly from the kinds of physical theories we understand. If you accept the argument as stated, then you get driven down a rabbit hole of increasingly strange solutions: versions of supersymmetry that cleverly hide from all experiments, hundreds of copies of the Standard Model, or even a multiverse.

Taking naturalness seriously doesn’t just mean accepting the argument as stated though. It can also mean believing the argument is wrong, but wrong in an interesting way.

One interesting way naturalness could be wrong would be if our reductionist picture of the world, where the ultimate laws live on the smallest scales, breaks down. I’ve heard vague hints from physicists over the years that this might be the case, usually based on the way that gravity seems to mix small and large scales. (Wolchover’s article also hints at this.) In that case, you’d want to find not just a new physical theory, but a new question to ask, something that could eclipse the old question with something more interesting and powerful.

Nima’s search for better questions seems to drive most of his research now. But I don’t think he’s 100% certain that the old questions are wrong, so you can still occasionally see him talking about multiverses and the like.

Ultimately, we can’t predict when a new question will take over. It’s a mix of the social and the empirical, of new predictions and observations but also of which ideas are compelling and beautiful enough to get people to dismiss the old question as irrelevant. It feels like we’re due for another change…but we might not be, and even if we are it might be a long time coming.

10 thoughts on “Changing the Question

  1. Robert Kraft

    I like the idea of physics changing the question. I wonder if I could get you to comment on two ways in which it seems to me that physics has changed the way we think about the “Big Why” question.

    First, in the classical regime, we had to separately postulate (a) laws and (b) matter fields, because without both there wouldn’t be anything in the universe. With QFT, it seems that laws themselves are enough as they allow for transitions from nothing (vacuum) to something.

    Second, you do occasionally hear physicists talking about gauge fields as somehow being necessary for a theory to have a certain kind of symmetry (gauge invariance, sometimes understood as a kind of locality). So this is at least a hint that there might be a reason why we have certain fields.

    Of course, like a toddler, we can always ask “but why do we have X [local gauge invariance, etc.]?” But that always seemed to me a kind of naive point to make when compared to the thought that we might actually get information about how “something” comes from “nothing.”


    1. 4gravitonsandagradstudent Post author

      For your first point, I think QFT definitely brought about a different perspective on these things yeah. It’s quite hard to square a “what stuff exists” account with a field-based theory, when there isn’t a clear distinction between “stuff” and physical law. So you pass the question to “why do these laws exist”, which is quite a different question from the sort of “why does the world exist?” that people used to ask.

      For your second, gauge invariance is weird because in some sense it’s an accident of how we describe things, where the actual fields are equivalence classes and not the sort of vectors we might naively write down. It feels like you want the explanation to go from “reality has some gauge group (like SU(3)xSU(2)xU(1))” to “reality has some force-carrying fields”, but I’m not sure you can separate out the two concepts like that. That is, I don’t know if there’s a sense in which “there is X gauge symmetry” implies “there is X field”, instead of them just being the same statement to begin with. (Gauge symmetry implies the Higgs, but that’s sort of a different thing, about how you can’t consistently make this sort of gauge field massive by itself.)

      (There is another potential caveat. You might say, what if you had SU(3) symmetry or the like but not gauge fields, what would happen? That’s called a global symmetry, an example is the “flavor” symmetry between different generations in the Standard Model. There’s an expectation among string theorists that such symmetries shouldn’t exist in a consistent theory of quantum gravity, that they should always be gauge symmetries in the end. If that’s true, then you could think of groups explaining fields in that way, though it feels like a bit of an odd sort of explanation.)


      1. duffieldjohn

        Gauge invariance should really be called phase invariance. AFAIK it entered QFT in 1926 when Vladimir Fock wrote a paper “on the invariant form of the wave and motion equations for a charged point mass” ( According to Jackson and Okun (, Fock “extended the known freedom of choosing the electromagnetic potentials” to quantum mechanics. See the AIP oral history interview where Fock talked to Thomas Kuhn in 1967. Fock said “we all read De Broglie’s paper in ‘23 or ’24 on waves of matter”. And that there was an attempt by Weyl to introduce Goud’s difference in relativity. Fock said he was inspired by Weyl’s work, but “the right place for Weyl’s differential form was just the exponent – the face of the wave function and not the Goud’s”. Fock had a thick Russian accent, you should read Goud’s as gauge, and face as phase.


  2. Giotis

    It seems that Wolchover and Nima are reinventing the wheel.

    These things have already been considered within String theory.

    Excerpt of Maldacena’s interview:

    “IL: When did this idea of emergent spacetime take form? Was it because of the AdS/ CFT correspondence?

    JM: There were different indications. But I would say mainly through research in string theory. The gauge/gravity duality was another example of this.

    My opinion is that spacetime looks emergent when you start from quantum particle theory, but it could be that there is an alternative description where spacetime is also fundamental, and the two descriptions are just two different ways of describing the same system. That is the idea of duality. It means that the same theory can be described in terms of different building blocks. Another analogy for this is that you could describe a novel, let’s say, in English, or you could describe it in French, but it is the same novel. “


    1. 4gravitonsandagradstudent Post author

      Nima is consciously building off the successes of AdS/CFT (recall his initial testing lab for a lot of this is N=4 super Yang-Mills, the place AdS/CFT was discovered). The context is a bit different, but what he’s going for is at least in his mind part of the overall string project: now that we have this powerful, duality-infused view of the world, how do we actually cash that out? What kinds of alternate mathematical descriptions can we build, and what sort of flexibility do they have? Are there objects/theories/properties that are easier to see in one language than another?


      1. Giotis

        Yet String theory is nowhere mentioned and the title is “A Different Kind of Theory of Everything” like this is a newly conceived alternative to String theory.


        1. 4gravitonsandagradstudent Post author

          I think the “Different Kind of Theory of Everything” being motivated there is something that string theorists don’t really know how to do yet.

          Much of the time, when people try to use string theory to generate models of the real world, it’s a top-down, UV first description. You start with a compactification, and go down to some EFT. And from that perspective naturalness is a pretty serious problem: if you want the SM Higgs and nothing else that’s LHC-detectable, you likely need a pretty strange-looking compactification.

          String theory also motivates a different approach, the one Maldacena is alluding to, the idea that holography and duality may mean that the correct “ultimate description” is at a different level than we would normally think. But I don’t think the string community at the moment actually knows how to set this up. You see some progress, holography-motivated solutions to naturalness and the like, but there’s no widely agreed-upon solution.

          (It’s also worth pointing out that this motivation existed prior to/independent of string theory…before AdS/CFT, there was a holographic principle, motivated by the area law for black hole entropy. AdS/CFT is the most explicit example of holography available, but even non-string QG people expect some sort of holographic principle to hold.)

          In Nima’s case, he isn’t really using string theory in any direct way. As such, it isn’t particularly relevant to what he’s talking about. That doesn’t mean he doesn’t expect it to link back to string theory in the end. (Honestly I think he’s the most pro-string amplitudes person except for the actual string amplitudes people, and he’s more pro-string than several of them!) Nor does it negate that the broader context of this sort of speculation is often string-theoretic, that string theory has made powerful contributions to it. But it still isn’t directly relevant: Wolchover’s question is broader, Nima’s work is narrower, I don’t think string theory needed to be mentioned in either case.


  3. Shea Levy

    One interesting way naturalness could be wrong would be if our reductionist picture of the world, where the ultimate laws live on the smallest scales, breaks down. I’ve heard vague hints from physicists over the years that this might be the case, usually based on the way that gravity seems to mix small and large scales.

    I recommend Bohm’s “Causality and Chance in Modern Physics” for an in-depth discussion of this possibility. Its guiding examples are out of date (it was written before von Neumann’s proof of the impossibility of hidden variables was accepted to be proving the wrong thing, and before we had any framework to capture the “particle zoo”), but the core principles still apply (and are the reason I’m not a reductionist)



Leave a Reply to Giotis Cancel reply

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

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

Facebook photo

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

Connecting to %s