2 | Carlo Rovelli on Quantum Mechanics, Spacetime, and Reality
What this covers
Carlo Rovelli and Sean Carroll discuss the deepest unsolved problem in physics: reconciling quantum mechanics with general relativity into a working quantum gravity theory. Rovelli, a pioneer of loop quantum gravity, argues that the path forward requires taking seriously the quantum nature of spacetime itself. The conversation moves from foundational physics through quantum interpretations, empirical constraints, and the sociology of modern theoretical physics, tracing how string theory and loop quantum gravity diverged in their approaches and what recent observations tell us about which deserves credibility.
The core of Rovelli's position rests on a reading of Einstein's own insight: since general relativity makes spacetime identical to the gravitational field, quantizing gravity must mean quantizing spacetime itself—dissolving the notion of fixed points. String theory, by contrast, treats spacetime as a pre-existing stage where strings move, which Rovelli sees as conceptually backwards for quantum gravity. The discussion covers recent empirical turns: the 2017 neutron-star merger observation that pinned gravitational and electromagnetic wave speeds to extraordinary precision, the non-discovery of supersymmetry at the LHC, and strong tests of Lorentz invariance that have eliminated competing proposals. A proposed nanoparticle experiment to demonstrate quantum entanglement in the gravitational field itself emerges as a concrete near-term test.
Rovelli also sketches his relational interpretation of quantum mechanics, in which properties exist not absolutely but only between interacting systems, and compares it philosophically to other interpretations. He argues that theory evaluation in practice works through accumulated indirect evidence and shifting credence rather than clean falsification, and that the field's allocation of resources and prestige to string theory over the past decades may not have been warranted given the absence of testable predictions. The conversation treats quantum gravity as increasingly empirical rather than purely mathematical—a shift that shapes which research directions might actually resolve the impasse.
Rovelli argues that a quantum theory of gravity requires taking the quantum nature of spacetime itself seriously rather than treating spacetime as a fixed background, which is why loop quantum gravity (not string theory) is the more faithful approach, and that quantum gravity is becoming an increasingly empirical rather than purely mathematical question.
- In general relativity, spacetime IS the gravitational field, so quantizing gravity means quantizing spacetime, dissolving the notion of fixed points
- String theory treats spacetime as a pre-existing background in which strings move, which Rovelli sees as its core conceptual weakness for quantum gravity
- Recent empirical results (Lorentz invariance tests, gravitational/light wave speed equality, non-discovery of supersymmetry) are constraining quantum gravity theories
Quantum mechanics reveals three foundational truths: discreteness, indeterminism, and relational actuality of properties.
- Quantum mechanics implies a fundamental nondeterminism: even if we control a physical system perfectly and know its physics as well as possible, what we observe tomorrow retains an irreducible margin of indeterminacy that can be made small but never zero — this is not due to insufficient measurement but is intrinsic.
“if we make everything we can on a physical system and we control it perfectly, and we know its physics as best as we can today, nevertheless, what we're gonna see tomorrow has a margin of indeterminacy”
- The third and most shocking discovery of quantum mechanics is that properties become concrete not when 'measured' (a nonsense concept for fundamental physics) but when two systems interact, and crucially a property becomes actual with respect to the system it interacts with and not with respect to everything else — this is the relational interpretation of quantum mechanics.
“property become concrete when two systems interact”
- Quantum mechanics reveals three things about the world we did not know before: that many quantities are discrete (granular/quantum), that we cannot make fully deterministic predictions about the future even with perfect knowledge, and that properties become actual only when systems interact.
“quantum mechanics, it tell us three things about the world which we didn't know before”
- Up to quantum mechanics we believed we could describe the state of a system with arbitrary precision, but the Planck constant defines a minimal volume in phase space (the space of all possible states) that forbids us from specifying a system's state more precisely than that, making it a measure of the discreteness or chunkiness of reality.
“the core of quantum mechanics is a constant, which is the Planck constant, which has the dimension of an action, which is a little volume in phase space”
- Some physical quantities are inherently relational rather than intrinsic: the velocity of an object is not a property of the object itself but a property of the object with respect to another object, demonstrating that relational properties existed in physics long before quantum mechanics and have nothing to do with measurement.
“The velocity of an object is not a property of the object. There's no sense in which the velocity of an object is a property of the object”
General relativity's core insight that spacetime is the gravitational field means quantum gravity requires radical revision of space and time.
- Albert Einstein himself was the first to realize that general relativity was incomplete: in 1916, just one year after writing the basic equation of general relativity and a full decade before textbook quantum mechanics, he wrote a paper stating with great clarity that his own theory would have to be corrected to account for the quantum properties of the gravitational field.
“The first who realized there was a problem, not surprisingly, is Albert Einstein himself, who in 1915, wrote what we call today the basic equation of general relativity. And one year later in 1916, he wrote a paper saying... Well, of course, this is an approximation, because gravity should have quantum properties.”
- Matvei Bronstein was the first to figure out that finding the quantum properties of the gravitational field necessarily implies a radical revision of space and time, because there is a sense in which you cannot measure the gravitational field at a point — since points lose their meaning in quantum gravity, as the gravitational field is space itself.
“he's the first one who figured out that quantum gravity, namely finding the quantum property of the gravitational field, had to imply some radical revision of space and time”
- Einstein's stroke of genius in general relativity was realizing that Newtonian spacetime and the gravitational field are the same thing; gravity is not a thing that lives in spacetime but a manifestation of the nature of spacetime itself, which can therefore stretch, bend, and move.
“Newton's spacetime and the gravitational field are actually the same thing, that's a stroke of... That's general relativity as I understand it.”
- Good physicists, including Einstein, make all sorts of mistakes, and Rovelli suggests that being prolific in making mistakes and being prolific in getting results may go together.
“he was the best in doing mistakes and getting results. Maybe the things go together.”
Loop quantum gravity treats spacetime as quantized grains that constitute space itself rather than particles moving in space.
- In loop quantum gravity, since the gravitational field does not live in spacetime but is spacetime itself, the quanta of gravity are not gravitons (little particles that move in space like photons, which are only a first approximation), but are themselves the discrete grains out of which space is built — grains that cannot be cut into smaller pieces.
“Since the gravitational field does not live in spacetime, it's not a field in spacetime, but it's spacetime itself, the quanta do not live in space or in spacetime, they are themselves spacetime”
- In loop quantum gravity, the discrete pixels of space know who is next to whom, which is represented by links between them forming a network called a spin network; following links around and back to the start traces a loop (the origin of the theory's name), and what we experience as smooth continuous space is this pixelated network seen from a large distance, like a high-resolution TV screen.
“if you start from one of the pixel and go around link, link, link, link and loop back, you make a loop. So these are the loops on loop quantum gravity”
- The core difference between loop quantum gravity and string theory is that a string (a closed string is a little loop) moves in a pre-existing spacetime, reflecting string theory's origins in a pre-general-relativistic worldview where things move over a fixed spacetime, whereas loop quantum gravity's loops do not move in space but constitute space itself, faithful to the general relativistic insight that there is no fixed background.
“a string is... A closed string is a little loop, but it's a little loop that moves in space... While these do not move in space, they make space themselves.”
String theory's dominance in research allocation is undeserved given it has produced no tested predictions and experimental evidence has not vindicated its central claims.
- The non-discovery of supersymmetric particles at the LHC, while not front-page news, was as important as the Higgs discovery; since string theory requires supersymmetry and many physicists strongly expected it to appear at LHC energies, its absence does not rule out string theory (supersymmetry could exist at higher energy) but should lower physicists' credence in string theory, just as a discovery would have raised it.
“the non-discovery of supersymmetry at the LHC, which was a no news because it was a non-discovery, but I think was the great news at LHC, as important as the discovery of the Higgs”
- String theory has backtracked from a period of very high hopes when proponents believed they would soon compute all the standard model parameters, see supersymmetry, and explain dark matter as a neutralino; nature proved more difficult, prompting humility, more reciprocal respect between communities, and increased exchange between string theorists and loop quantum gravity researchers.
“There was a moment in which string theory had very high hopes, and was sort of saying we had it. It has backtracked a little bit from that.”
- Over the past couple of decades, the distribution of finite research resources (academic positions, grant money, prizes) has not been fair: string theory has held a dominant position that may not be deserved given it has not predicted anything definite that has been tested, and loop quantum gravity in particular has suffered from this dominance due to a non-linear bandwagon effect where powerful people in powerful positions push the field in one direction.
“if I look back a couple of decades, the distribution of resources has not been fair”
Observable astrophysical signals, not mathematical elegance, should determine which quantum gravity approach is correct.
- Rovelli is hopeful but not certain about convergence in quantum gravity; he expects that one approach will turn out right and the other wrong (as is usual in the history of science) and hopes the resolution may come from computing observable signals, such as a black hole tunneling into a white hole, that we can recognize in the universe — and he feels quantum gravity seems much closer today than it did 20 years ago.
“We can compute a black hole tunnel into a white hole and recognize that some signals are exactly that, and then we have a clear grasp on something which is happening”
- What will ultimately convince physicists that one quantum gravity theory is correct is not its mathematical cleanness but its ability to explain real data; therefore the productive path is applying loop quantum gravity to astrophysics and cosmology — black holes, dark matter, fast radio bursts, gamma ray signals, signatures in cosmic background radiation — rather than pursuing more pure mathematics.
“what we need is not more mathematics. We need to apply this theory to reality.”
- A proposed experiment, suggested independently by two groups, would take two nanoparticles (motes of dust), each placed in a quantum superposition of two positions, and place them near each other so each feels the other's Newtonian gravitational field; if the resulting state of the two particles becomes genuinely quantum-entangled, this would demonstrate that the gravitational field itself was in a superposition, proving that spacetime can be put in a superposition and has quantum properties.
“You take one particle, a sort of nanoparticle... and you quantum split, you put it in a superposition, or two positions. And you take another particle and you also quantum split.”
Failed experimental predictions and indirect evidence credibility shifts, not clean falsification, drive how physics advances.
- When you try to pinpoint a particle's position with arbitrarily small precision, quantum mechanics makes its velocity (and thus momentum and energy) very indeterminate; since energy equals mass via E=MC², a large energy confined to a small area is like a large mass in a small area, which creates a black hole — so trying to localize too small automatically creates a little black hole, setting a minimal observable scale.
“If you try to pinpoint the position of a particle, in arbitrary small precision, you can, but because of quantum mechanics, the velocity of the particle becomes very indeterminate”
- Some attempts at quantum gravity, such as Hořava gravity, predicted that Lorentz invariance is broken, generating excitement and a decade-long program of astrophysical observations; the resulting strong evidence that Lorentz invariance is NOT violated at the scale those theories required has made such attempts much less credible, though not strictly ruled out.
“some attempts to write a quantum theory of gravity, others than strings and loops, constructed some tentative theories of gravity where Lorentz invariance is broken... One was called Hořava gravity”
- SU(5), a 1970s grand unified theory attempting to unify the non-gravitational forces, made an explicit experimental prediction that the proton should decay into lighter particles like positrons and neutrinos; despite long searches this proton decay has not been observed, effectively ruling out the theory at a scale only two or three orders of magnitude from the quantum gravity scale.
“we have experiments that have ruled out SU(5), which are at a scale which is not very different from the quantum gravity scale, maybe two or three orders of magnitude”
- The naive picture that theories are simply falsified by experiments is too simple; in practice scientists assign credibility to theories that rises as good indirect evidence accumulates and falls when predictions fail to materialize, so science advances through hints and indirect evidence rather than only clean falsification.
“We often say that we propose theory and then they are falsified by an experiment... but I think things are a little bit more complicated, because we put credibility in theory, and the more sort of good indirect things happen, the more our credibility goes up”
- The 2017 observation of two neutron stars merging was seen simultaneously by gravitational wave detectors and electromagnetic telescopes; because both signals traveled from very far away and arrived at the same moment, we learned gravitational waves and electromagnetic waves travel at the same speed to within 1 part in 10^14 or 10^15 — improving our knowledge of this fundamental parameter by about a hundred billion times in a single stroke and discrediting modified gravity theories predicting different speeds.
“we have learned that gravitational waves and electromagnetic waves travel at the same speed, and in fact, if you put the numbers, they travel at the same speed in 1 part in 10 to the 14 or 15”
- Scientists get thrilled — in Rovelli's view too thrilled — when they can claim that in some subtle way everything we knew before was wrong, which is exactly what happened with Lorentz-violating gravity theories that attracted large numbers of theorists and observers before the predicted violation failed to materialize.
“scientists get thrilled when they can say that in some subtle way, everything we knew before was wrong”
- Given the strangeness of quantum mechanics, every interpretation comes with a price: the many-worlds interpretation requires accepting an enormous, heavy ontology of all the worlds, while the relational interpretation requires weakening realism so that a system has no single objective state but only states relative to other systems; the debate over interpretations is fundamentally about which price one is willing to pay.
“given the strangeness of quantum mechanics, any way we think it is, comes with a price. So we have options.”
Covariant loop quantum gravity faces an unresolved mathematical convergence problem that could render the theory meaningless.
- A genuine weakness of the covariant version of loop quantum gravity is that, while it has a well-defined set of equations usable order by order to compute, it has not been proven or even well-indicated that the theory converges in any sense; going to the next order of approximation might cause everything to blow up, leaving open the possibility that the theory is meaningless.
“we don't know if this converges in any sense. So even in the weakest possible sense, we don't know if going to next order... The theory doesn't go fully.”