What this covers
This monthly question-and-answer episode features Sean Carroll responding to queries from Patreon supporters on topics spanning physics, philosophy, media, politics, and beyond. The breadth is genuine: questions range from the fundamental structure of spacetime and black hole thermodynamics to moral theory, animal ethics, and climate strategy. Throughout, Carroll's recurring move is to show that apparent binaries—biological sex versus human sexuality, objective morality versus constructed morality, fundamental laws versus emergent descriptions—collapse under scrutiny, and that insisting on simple answers often masks a failure to distinguish what is truly fundamental from what is useful at a particular Level of Description.
On physics, Carroll addresses black holes (their interiors, Hawking radiation, time travel paradoxes), quantum measurement and energy conservation, thermodynamics and the cosmic microwave background, infinities in physical theory, and gravitational effects. On philosophy, he elaborates Moral Constructivism as distinct from relativism, defends caring more for those near you without relativizing all values, and justifies his dietary choices through capacity for abstract fear rather than metaphysical claims about life's worth. On contested ground, he argues that Emergence and entropy are distinct phenomena, that panpsychism and emergence are incoherent when combined, that peer review is a useful filter but not a substitute for personal judgment, and that biological sex has a real binary in gamete size but that conflating it with the multidimensional human concept of sexuality is deliberate equivocation. He addresses methodological puzzles too: why the past hypothesis must be assumed rather than proven, how to justify logic and belief in an external world when Boltzmann brains loom, and when metaphors like rubber sheets aid rather than mislead understanding.
Carroll argues that many contentious questions—about biological sex, morality, free will, and the foundations of physics—are genuinely complicated and resist simple binary answers, and that insisting on simple answers usually reveals a hidden agenda or a failure to distinguish fundamental laws from emergent descriptions.
- Biological sex has a real binary in gamete size but human sexuality is multidimensional, and collapsing them is a deliberate equivocation
- Cause/effect, free will, and morality are emergent higher-level descriptions, not features of the fundamental laws of physics
- Naturalism is well-defined; the burden of definition falls on those positing the supernatural
Naturalism succeeds without defining the supernatural; physical laws make testable predictions while God remains ill-defined.
- Naturalism is not threatened by the inability to define the supernatural; naturalists have a clear picture—a world obeying discoverable rules tested empirically—and the problem of defining what counts as 'non-natural' falls on the non-naturalist, not the naturalist.
“I think naturalism is perfectly obvious. The problem here is defining supernaturalism and that's not really my job.”
- God is a poor hypothesis for explaining physical features of the universe because it is too ill-defined—you cannot say what kind of universe God would create or why—whereas physical laws make specific predictions, giving a prior preference for naturalistic explanations; additionally, an omnipotent God could easily make his existence clear and has not, which counts as evidence against existence.
“The problem with God as a hypothesis to explain features of the physical world is, it's just way too ill-defined... what kind of universe would God create and why? And these are very, very hard questions to answer.”
- Naturalists should avoid the word 'spiritual' because it leads people to think of non-physical spirits; the underlying experiences (oneness, transcendence) are real, valuable, and fully compatible with the laws of nature, but our inherited vocabulary from a thousand years ago hasn't caught up, so we need better words even though Carroll doesn't know what they are.
“when you use the word spiritual, people are going to think of spirits. They're going to think of these spirits as non-physical, non-natural manifestations. And if you don't want them to think about that... don't use the word spiritual.”
- The premises of panpsychism combined with emergence are incoherent: if you are genuinely a panpsychist claiming consciousness is absolutely fundamental, you cannot also say it is emergent and not existing at the level of electrons; moreover these proposals are too ill-defined to answer questions about 'consciousness particles' or wave-function collapse, and if they were specific enough they would probably be ruled out by experiment.
“If you're literally panpsychist or pan-experientialist, then I don't think you can say it's an emergent phenomenon. I think that the whole idea of panpsychism is to say that consciousness is absolutely fundamental.”
- A catastrophic failed physics experiment (e.g. creating a planet-pulverizing particle) is not a credible great filter candidate, because everything we do in particle accelerators happens at far higher energies and far more often naturally as high-energy cosmic rays collide throughout the universe; clever attempts to evade this strong bound exist, but the smart money is that such an event will not happen.
“everything we do here on Earth is done at much higher energies and much more often out there in the universe. High energy cosmic rays bump into each other, bump into the Earth.”
- American constitutional democracy, though over two centuries old, has no guarantee of lasting; a plausible near-term failure mode is a 2024 election in which state legislatures or secretaries of state refuse to respect their voters' will and submit electors contrary to the popular vote, overthrowing the democratic decision—a scenario that is conceivable and plausible in a way it was not 10-20 years ago.
“those states choose to not respect the will of their voters and to hand in a list of electors that is different than what people actually voted for, you could absolutely overthrow the actual democratic decision of our system.”
- You cannot use experiment to rule out a fluctuating universe and prove the past hypothesis, because Bayesian theory choice requires multiplying the likelihood by the prior; in a randomly fluctuating universe the prior probability that current ordered observations just fluctuated into existence is so overwhelmingly large that no amount of new likelihood data can beat it down, so the past hypothesis must be assumed rather than derived.
“you can't actually be a good Bayesian and rule it out by data because under that set of assumptions that all configurations of stuff in the universe are equally likely, the prior probability that you're seeing things randomly fluctuate into existence is so overwhelmingly large that you can't beat it down just by likelihood functions and collecting new data.”
- Mindscape listeners donated $165,000 to Giving Multiplier over the past year, funding over 5000 deworming treatments, over 3500 vitamin A supplements, 400 malaria nets and more, demonstrating the philanthropic impact of the audience.
“We got $165,000 in donations from Mindscape listeners to Giving Multiplier.”
- The most effective critique of effective altruism (going back to Peter Singer) is that it overemphasizes charitable giving, which is good but often doesn't change the world, because some problems are systematic and require systematic upheaval; charity shouldn't be a distraction from more systematic ways to change things that would have a bigger ultimate impact.
“charitable giving is good, but it in many ways doesn't change the world, sometimes we have systematic problems that need some systematic upheaval and giving money to this or that non-government organization that is doing good is not addressing the underlying systematic issues.”
- Carroll favors subjective probability: even in a universe with truly stochastic laws, he would treat probability as subjective—reflecting our lack of knowledge of the future—because as an eternalist about time, all moments exist and we simply don't know what the future holds; presentists who believe in stochastic laws are more likely to be objectivists about probability.
“If you think that what exists is all moments of time, but you just don't know what is going to happen in the future, then you can be someone who believes in truly stochastic laws of physics but still treats probability as fundamentally subjective.”
- High-scoring sports like basketball are preferable to Carroll because the large number of scores averages out random fluctuations—a single lucky shot is rarely decisive—whereas low-scoring sports like soccer let random fluctuations play a larger (though not definitive) role, since there are fewer chances to average over them.
“Over the course of a basketball game, you can always hit a lucky shot here or there, but the chances that single lucky shot will be decisive are very small because there's an integrated effect of the whole effort”
Consciousness and free will are real as emergent higher-level descriptions compatible with deterministic fundamental physics.
- Free will is real but as an emergent, higher-level, compatibilist concept; it has nothing to do with the fundamental laws of physics (which are deterministic in both Newtonian and many-worlds frameworks) and everything to do with how we usefully talk about human beings at the higher emergent level, so it is compatible with whatever the fundamental laws happen to be.
“there's literally no difference in the discussion of free will between this universe with a deterministic equation the Schrodinger equation and many worlds versus a classical universe with a deterministic equation Newton's equation.”
- In the many worlds interpretation, parallel universes do not exist within some space; rather space exists within each of them—each branch is a parallel copy of spacetime. The reason descendant worlds don't influence each other is not about emergent space but is a mathematical consequence of decoherence: when environment states are perpendicular, there is zero overlap and no interference between worlds.
“The motto is that the universes in many worlds do not exist within some space. Space exists within them.”
- Emergence and entropy are related but distinct: emergence is prior, since the usefulness of coarse-graining microstates into macrostates is what makes entropy (the log of microstates per macrostate) definable; entropy specifically tracks thermodynamic directionality of time, whereas emergence is broader—you can have emergence (e.g. Earth's center-of-mass motion around the sun) without generating any entropy.
“I would say that entropy is a special case of the usefulness of coarse graining and emergence.”
- At the fundamental level there are only patterns/laws of physics relating the universe's configuration across moments, and no single moment causes any other; cause-and-effect relationships become a relevant and sensible way to describe the universe only at the higher emergent level where there are macrostates and an arrow of time, so saying 'the cat caused the ball to fall' is completely fine as an emergent description but not as a fundamental claim.
“A web of cause and effect relationships becomes a relevant and important way to think about the universe at the higher emergent level... It makes 100% sense to say the cat caused the ball to fall on the floor. It does not make sense to say that is a deep down feature of the fundamental nature of reality.”
- A spinning mass 'drags' spacetime without expending energy because 'drag' is the wrong word—there is no friction-like force pulling it back; the static, eternal configuration of spacetime around a rotating body simply reflects its angular momentum, just as a non-rotating mass has a static gravitational field, with no energy transferred from the spinning mass to spacetime.
“The statement is simply that the static eternal configuration of the spacetime reflects the angular momentum of the body in it that is spinning. There is no energy being transferred from the spinning mass to spacetime.”
- When you measure a particle's spin and it changes, momentum (or angular momentum) is still conserved because the measurement involves physical interaction with an apparatus whose own angular momentum participates, so the angular momentum goes elsewhere; energy is a subtle exception that behaves differently.
“you have interacted physically when you measure with some apparatus and that interaction involves the spin or the angular momentum of the apparatus as well and everything is conserved.”
- Information conservation does not fit Noether's theorem framework because Noether's theorem only runs one way—each symmetry of the Lagrangian gives a conserved quantity, but not every conserved quantity requires a symmetry; information is conserved as a basic feature of any Lagrangian-based formulation of physics, even absent any symmetry, while non-Lagrangian formulations need not conserve it.
“It's not the other way around. It's not true that for everything conserved there must be a symmetry. The fact that information is conserved is a basic feature of the setup that has Lagrangians in it in the first place”
Morality is constructed by agents and communities, permitting critique grounded in one's own framework rather than objective truth.
- Moral relativism says morality is relative to a community's beliefs so outsiders cannot critique it, whereas constructivism says morality is constructed (not discovered out in the world) but allows the constructor to freely critique others—not on the grounds that they are objectively mistaken, but on the grounds that according to one's own constructed morality they are doing wrong; Kantian constructivists believe one uniquely rational system can be constructed while human constructivists allow plural valid systems.
“constructivism is just saying that morality is constructed. In other words, it's not out there in the world to be discovered. It's something that human beings come together and individually and collectively construct”
- It is morally acceptable to care more about those near and dear to you (e.g. your children) than about distant strangers, and Carroll's moral constructivism supports this since people construct morality from their inclinations; pure consequentialism (a la Singer) demands inhumanly difficult equal concern for all, and taking it too seriously makes people miserable—but the better critique is that most of us probably under-help the distant world rather than over-help those near us, and it is not a zero-sum game.
“I think it's okay to put more effort into caring about your children than to caring about people elsewhere in the world or in other times that you don't know and have never met.”
- Carroll justifies eating cows and pigs but not on metaphysical grounds—he attaches no intrinsic worthiness to lives—but on the crucial difference that humans can abstractly conceptualize their own death and fret about hypothetical counterfactual futures (e.g. becoming afraid merely from words), whereas cows only instinctively fear immediate danger; he still opposes needless animal suffering as a matter of empathy.
“that ability to fear death, to fear not just to instinctively fear that you're in danger, which certainly cows can do... but to mentally visualize the hypothetical prospect of being in danger just for abstract intellectual reasons... that's the thing that other animals just don't have.”
- Whether mathematics is invented or discovered is a genuine, active debate among professional philosophers and mathematicians, not just mass-media fodder; Platonism/realism (math is discovered, real platonic existence) is probably the majority view among mathematicians, while nominalism (math is an invented language for describing the world) is a credible opposing position.
“this is very much a real debate. In fact, we have had here on the podcast people with opposite points of view about this.”
- Hume's claim that 'reason is and ought only to be the slave of the passions' means logic can tell you how to get what you want but cannot tell you what to want; values and goals are not derivable from objective facts (you cannot derive ought from is), so passions supply the goals and logic serves them—and Carroll agrees with this view.
“we have passions that tell us what we want. Logic is going to tell us how to get it... Logic does not tell us what we want. And I do agree with that view.”
Physics discovers empirical laws through toy models; biological and social systems resist decomposition and may admit no discoverable laws.
- Spherical cow thinking—abstracting away complications, analyzing a toy model, then adding complications back—works extremely well in physics but not in biology or sociology; Carroll suspects it is not a necessary feature of anything we'd call laws of physics, conjecturing there could be alternative laws where everything couples to everything else and complications cannot be abstracted away, making them much harder to discover.
“I don't see why there must be a regime of the physical world which is amenable to that kind of, let's abstract away the complications and then let's add it back in later reasoning.”
- Physics reads modern textbooks rather than original writings because the ideas exist independently of their originators and are often understood better by later generations standing on others' shoulders, whereas philosophy more often reads originals because in some areas the ideas are hard to capture in translation and different readers extract different meanings; both fields could learn from each other, with physics needing better historical understanding.
“the reason why in physics we generally read modern textbooks rather than the original writings is because the ideas are there. They exist independently of the person who came up with them. And in fact, we generally understand the ideas better than the people who came up with them”
- Astrology can be discounted because it violates the laws of physics: the only long-range forces are gravity and electromagnetism, and in both cases nearby effects vastly outweigh those of distant celestial bodies, so it is far easier to trust well-established physics than to overthrow it for the flimsy reasons astrology offers; also there is no reason the moment of birth would be privileged over conception or any other moment.
“We know what the long range forces are, they're gravity and electromagnetism. And in both cases, nearby effects are enormously more important than the effects of other celestial bodies.”
- Even if new life were forming via abiogenesis somewhere on Earth today, it would not flourish noticeably because existing life has evolved for billions of years and is extremely good at consuming available resources, so any new protomolecular life would simply be outcompeted and die out.
“the life that already exists is way better than any little protomolecule life at competing for the resources that we have here on Earth.”
Epistemic justification for logic and perception rests on pragmatic success, not foundational proof or certainty.
- The use of logic and the assumption that one is not a Boltzmann brain are justified pragmatically rather than foundationally: if logic is false or you are a Boltzmann brain, things are hopeless and you can make no progress, whereas if you hypothesize your impressions and logic are roughly correct, you can make progress—so that is how you choose to act regardless of ultimate truth.
“if logic is false, if I am just a Boltzmann brain, then things are hopeless... Whereas if I hypothesize that my impressions of the world are roughly correct, my knowledge of logic is roughly on the right track, then I can make progress.”
- People make a reasoning error by treating something being merely 'possible' as implying at least a 10% chance of being true; the better question is always about likelihood and reasonableness, not bare possibility.
“I think that people make a mistake like if something's possible, then there's at least a 10% chance that it's true. And I want to get over that particular option.”
Spacetime structure and quantum fields have precise mathematical definitions despite ongoing interpretive debates.
- Backward time travel in general relativity cannot reach a point before the time machine was built, because a time machine is a wormhole connecting two existing points whose ends must be stretched into the future from the present—you cannot start in the future and send a tendril back—so the absence of densely-packed time machines today is not evidence that backward time machines are impossible.
“You can't use a time machine in general relativity to go back to a point before you built the time machine.”
- The question of a black hole's interior volume is not well-posed because volume is a feature of space, not spacetime, and how you slice spacetime into space and time strongly affects the answer; the event horizon area is well-defined via the black hole's rest frame, but the volume is slicing-dependent and not necessarily unique—it is certainly not zero or infinity.
“The volume inside, in the context of relativity, is going to depend on how you divide space-time up into space and time. Volume is a feature of space, not of space-time.”
- Liouville's theorem (constant phase-space volume) is compatible with chaotic systems whose neighboring trajectories diverge exponentially, because the original volume blob keeps its total volume but gets stretched and pulled into long thin tendrils; this is hard to visualize only because chaotic dynamics require higher-dimensional phase spaces than the two dimensions people typically imagine.
“it gets stretched and pulled into taffy with long little tendrils while keeping the overall volume constant... it's because phase space has a much higher dimensionality than you can visualize very well.”
- Hawking radiation depends on the existence of horizons; ordinary planets and stars have exactly zero Hawking radiation because quantum fields can settle into a static vacuum state, whereas a black hole is like an unstable nucleus whose wave function is not static and gently evolves from 'not decayed' toward 'decayed', which we observe as particle emission until evaporation; extremal (zero-temperature) black holes would be stable but are thought to be physically unreachable.
“Black holes do not have solutions to the equations of quantum mechanics that are perfectly static around them, and so they change over time a little bit. And we observe that time evolution as particles being emitted from the black hole until it evaporates”
- The stress-energy tensor on the right side of Einstein's equation is not just an ad-hoc bin for all forms of energy; it has a precise mathematical definition as the variation of the matter action with respect to changes in the metric, giving confidence the basic picture is solid despite ongoing technical debates about improving it in quantum field theory.
“the best way of thinking about the stress energy tensor is basically the tiny variation of the matter action with respect to changes in the metric.”
Rights and dignity matter beyond consequentialist calculus; voting reform and effective altruism improve outcomes when separated from bad actors.
- Carroll is not a full consequentialist because he wants to live in a world where people have rights that should not be trampled even if doing so would benefit the rest of society; there are reasons to treat people with dignity and respect over and above consequentialist calculus, such as rejecting punishing innocent families to deter crime.
“I don't want to live in a world where you deter bad behavior by punishing innocent people and even if that gives you the best consequentialist answer.”
- A 'diet lottery voting' system—randomly sampling 100 citizens to vote with mandatory turnout—could improve politics by making 'get out the vote' efforts pointless and by changing how money distorts politics, since spending millions to win the affections of a small random sample is much harder than funding mass campaigns.
“The other thing that this thing would do would be to change how money infects politics... if it's just randomly selected people every year, it's harder to see how spending millions of dollars to win their affections would work.”
- The collapse of FTX and fraud by Sam Bankman-Fried does not undermine the idea of effective altruism, because the fact that a bad actor holds an idea does not affect whether the idea is right or wrong; it may undermine confidence in the judgment of the EA community for going along with an obvious fraudster, but that is a story about human psychology and susceptibility to wishful thinking, not about the validity of EA.
“The fact that some bad actor holds an idea, in my view, doesn't affect whether the idea is right or wrong at all. It might undermine one's confidence in the judgment of people in the community”
- Carroll's ability to distill guests' complex ideas comes not from special talent but from two things: Mindscape being a one-person operation where he personally invites only guests whose knowledge genuinely interests him (so he is truly motivated to understand), plus a lifetime of practice at the intersection of research and public communication, where talking to someone outside one's own ruts helps reframe ideas freshly.
“Every single person who comes on has some knowledge that I want to get to. And so I'm motivated to try to understand it better myself.”
Sex involves biological binary at gametes but is multidimensional at human scale; equivocating between levels serves political agendas.
- Biological sex involves a real binary at the level of gamete size (ova vs sperm), but human sexuality is multidimensional (chromosomes, hormones, development, secondary characteristics, psychology, culture); people who insist sex is 'real, immutable and binary' are deliberately equivocating between the narrow biological term of art and the human-scale concept in order to draw political conclusions about rights and bathrooms.
“The reason why people go on and translate that into biological sex is real, immutable and binary in a way that doesn't make clear that they are referring to this term of art within the context of biology is precisely because they want to claim more than that.”
- The 'human beings are bipedal' analogy used to defend the sex binary actually refutes it: casually saying humans have two legs is fine, but adding signalling words like 'necessarily, always, exclusively' makes the statement false—exactly as 'sex is real, immutable and binary' adds adjectives precisely to make a definitive claim that is wrong.
“But what if you say human beings necessarily always and exclusively have two legs? The content of your statement is the same... but your statement has become false because all those little signalling words are precisely doing the work of saying, I'm not just trying to be casual.”
- The rights and dignity of transgender and intersex people do not depend on the developmental biology of sex; people who want to discriminate will do so regardless of the language adopted, so the linguistic battles over 'trans women are women' are less important than securing actual rights and treating people with dignity.
“But the claim that the rights of people of different genders, including transsexual people, do not depend on the developmental biology of sex that I agree with.”
Gravitational waves, antimatter distribution, cosmological uniformity, thermal equilibrium, entropy directionality, and de Sitter space pose distinct puzzles.
- In a duality like AdS/CFT, conscious beings would perceive only one description, not both; the side they find themselves on is likely picked out by weak coupling (where the classical approximation works easily, as with the small fine structure constant) and possibly by which side allows classicality through decoherence and the arrow of time.
“if you're weakly coupled on one side of the duality, you're strongly coupled on the other. In this strong coupling world, it's just harder to make sense of the world... therefore, it's harder to know what any beings would say about the world.”
- Large collections of antimatter galaxies are not realistically possible because in the early universe matter was a smooth homogeneous plasma; if even 10% were antimatter it would have annihilated into super-high-energy radiation, making the cosmic microwave background look completely different from what we observe.
“if half of that plasma or even 10% of it were antimatter, you would know it would annihilate away very quickly into super high energy radiation and the microwave background would look like a very, very different place.”
- Unlike electromagnetic waves, which can be absorbed or reflected because charges come in positive and negative, gravitational waves are extremely hard to manipulate because gravitational 'charge' (mass) is always positive and gravity is attractive and very weak, so to a good approximation gravitational waves propagate essentially undisturbed (aside from negligible energy loss and gravitational lensing).
“gravitational charges are just the masses of the particles that you're making up your system from, and those are always positive. Gravity is always attractive... And so it is enormously harder to manipulate gravitational waves”
- Whether approaching de Sitter space leads to the Boltzmann brain problem depends on one's assumptions about quantum gravity and quantum field theory in curved spacetime; if you take the quantum field theory seriously, it is natural to approach de Sitter space without getting dynamical fluctuations into Boltzmann brains, so the Boltzmann brain problem may not be a problem at all.
“it is very, very natural to imagine that we do approach to de Sitter space, but we do not get dynamical fluctuations into Boltzmann brains. That's a technical point about how the fluctuations work.”
- Thermal equilibrium does not explain the uniformity of the CMB, because in a universe with gravity, thermal equilibrium produces lumpiness (black holes and empty regions), not smoothness; the real mystery is why the CMB has the black-body spectrum of an object in equilibrium without gravity even though it exists in a universe with gravity, and inflation provides one way to thermalize matter degrees of freedom while the gravitational configuration remains out of equilibrium.
“We live in a universe with gravity and in a universe with gravity thermal equilibrium does not look smooth. You get lumpy regions with black holes, empty regions elsewhere.”
- The second law of thermodynamics says entropy increases, but there is no rule requiring entropy to increase as fast as possible or via any particular state (such as through black holes); the specific path a system takes from low to high entropy can be very complex and is situation-dependent, not governed by a simple rule.
“There's no rule that says entropy increases as fast as possible or in this particular way or anything like that. Those rules are all much more situation dependent and complicated.”
Scientific communication should use accessible metaphors while being explicit about limitations and boundaries.
- Picturesque metaphors and visualizations like the ball-on-a-rubber-sheet for gravity or the expanding balloon for the universe are acceptable and useful as long as you are clear about their limitations and help the audience see those limitations, rather than presenting them as the literal truth; you should use all the available metaphors since different ones help different people.
“You have to work your way from the picturesque metaphors to the exact equations, and different metaphors and visualizations and analogies are going to be useful to different people. So I'm always of the opinion that you should use all of them and be very, very clear about what the limitations are.”
- Carroll acknowledges his snarky tweet ('actual science would like a word') was terrible wording that lowers the tone of debate, and that people should not be glib, snarky, and sarcastic on social media even though there is enormous pressure to do so.
“It's snarky, it's dismissive, it lowers the tone of the debate. And I try to do better.”
- The CMB is ever-present because last scattering was a moment throughout all of space, not a single distant surface; the surface of last scattering is just the intersection of that moment with our past light cone, so as time passes we see photons from slightly farther away, and because the CMB is extremely smooth it appears static even though we are always looking at slightly different places.
“We talk about the surface of last scattering, but it's better to think about a moment of last scattering throughout all the universe.”
- Carroll selects Mindscape topics to be more or less universal and still interesting in 10 or 100 years, deliberately avoiding current-event political analysis (like election aftermaths) in favor of political theory and big-picture questions, and has a rough guideline against having currently-serving politicians on because their job is to get things done rather than be scholars.
“I want the questions we talk about here to be more or less universal, to be still interesting a hundred years from now or at least let's say 10 years from now.”
Individual judgment and understanding matter more than peer review status for evaluating scientific claims.
- Quantum mechanics can in principle derive chemistry and materials properties (covalent bonding, hardness, elasticity) but the calculational difficulty is enormous because Hilbert space is huge, so physicists use approximation methods like density functional theory and build higher-level models (like BCS superconductivity) that glue onto lower-level quantum descriptions; the killer app for quantum computers is simulating such quantum systems.
“quantum mechanics lives in Hilbert space. Hilbert space is very, very big. Even when you have a small number of particles, the calculational difficulty of following the quantum equations becomes very, very difficult.”
- Carroll is much less definitive that time is non-fundamental than that space is non-fundamental: he thinks space is almost certainly not fundamental, but whether time is emergent or fundamental is a much dicier, undecided proposition—and even if time is emergent, one can still rationally 'trust an emergent process'.
“I think space is not fundamental, almost certainly. I think that time is a much dicier proposition, and we just don't know.”
- Peer review is a labor-saving device that raises the fraction of worthwhile papers but is not a substitute for personal judgment; if a scientific claim really matters to you, you should read and understand it and judge its likely truth yourself, regardless of whether it appears in a peer-reviewed journal, since unreviewed papers can be true and reviewed papers can be false.
“The peer-reviewed system, like many other things, is a labor-saving device. It gives you a little bit more credence that the result is reliable and respectable, but it is not a substitute for your actual judgment.”
Conservation laws, infinities, and energy behave distinctly in quantum versus classical mechanics, requiring domain-specific rules.
- Energy is not conserved in quantum measurements (it fluctuates, with decreasing fluctuation sizes over time), unlike other conserved quantities; this is because energy is special—it is the Hamiltonian that governs how the wave function evolves in time, so a universe in a definite energy eigenstate would never change. Any interpretation where energy is the expectation value of the Hamiltonian implies non-conservation, but interpretations like QBism (state not physical) or hidden variables avoid it differently.
“Energy is special in quantum mechanics. Energy is the thing that tells you how the wave function evolves with time... if you are in a state with definite well-defined energy and nothing ever changes, literally nothing happens in such a universe.”
- There is a rough consistent distinction in physics regarding infinities: infinities in the setup or arena (size of the universe, Hilbert space, phase space) have no principled reason to be finite and aren't treated as problems, whereas infinities arising as predictions for quantities that are normally finite within the universe (curvature at a singularity, renormalized Feynman diagrams) are worrying because we never observe infinite things within nature.
“The kinds of infinities we tend to worry about are when a quantity within the universe that is usually finite is predicted to become infinite because of some prediction of an equation”
The hierarchy problem differs structurally from the cosmological constant problem in allowing new physics at accessible scales.
- The hierarchy problem (Higgs mass naively predicted to be enormous due to diverging quantum corrections, yet observed small) is structurally parallel to the cosmological constant problem, but differs crucially because the hierarchy problem allowed for new physics at accessible energy scales (supersymmetry, extra dimensions), whereas the cosmological constant's energy scale is so low that no new physics could hide there.
“in the case of the hierarchy problem, there was at least the possibility that we would discover new physics at the right energy scale to explain the hierarchy problem... Whereas with the cosmological constant, the energy scale... is so incredibly low... that there's no room for new physics there.”
- A slowly-changing (dynamical) cosmological constant is very hard to make work because such a dynamical field would interact with other things in the universe in ways we should already have observed, and a constant cosmological constant with possible future phase transitions is also problematic because such a transition probably should have already happened; the most natural solution is a truly constant cosmological constant.
“if you have some thing that is slowly changing, that's a dynamical entity, that's a field and that field can interact with other things in the universe. And we should have seen those interactions already. And we haven't yet.”
Rare technological civilizations and regulatory minimalism for platforms reflect practical constraints over catastrophic inevitability.
- Carroll's preferred solution to the Fermi paradox / great filter is that technological civilizations are simply very rare—either because life itself is rare or because complex eukaryotic/multicellular life is rare—which is much easier to believe than the alternative that catastrophes like climate change, war, or pandemics reliably wipe out civilizations that arise thousands or millions of times.
“my very straightforward solution to the great filter is that technological civilizations are just super duper rare, either because life is rare or because complex eukaryotic/multicellular life is super duper rare.”
- An individual's contribution to the entropy of the universe is negligible and not worth worrying about; humans stay at roughly constant internal entropy and increase the world's entropy only through irreversible processes like eating and sweating, but natural processes like the sun radiating into space do enormously more to increase universal entropy than every human who ever existed combined.
“The sun radiating out into space does enormously more to increase the entropy of the universe than every human being who ever existed in total.”
- Fixing global capitalism to reduce carbon emissions is far easier than physically moving the Earth's orbit or even terraforming the atmosphere; the difficulty hierarchy is a sobering reminder that since we cannot even prevent ourselves from breaking Earth's already-good atmosphere, terraforming another planet like Mars or Venus would be vastly harder.
“It is much much easier to imagine fixing the earth's atmosphere one way or the other... Fixing capitalism is actually even easier than that.”
- Government should keep regulation of platforms like Twitter to a minimum because their assets are intangible—the value is the user base, not a factory—so the people could move tomorrow and the platform could be replaced (as Microsoft Explorer's apparent monopoly was), meaning these platforms are not inevitable; regulation becomes more justified only if a platform becomes as necessary and unique as roads or radio frequencies.
“there's no inevitability to these giant social media platforms. Their assets are rather intangible... The asset of Twitter is the user base. It's not some factory... the people could easily move tomorrow.”
- Current political tension stems partly from increased political polarization (politicians game-theoretically maximizing reelection by stoking dislike of the other party) and partly from modernity leaving many people feeling powerless and left behind—without good jobs, security, or healthcare—which breeds resentment and disrespect for democratic institutions that demagogues then inflame to seize power, echoing the dynamic that ended the Roman Republic.
“There's a lot of people out there who feel powerless, who do not feel that the democratic institutions are on their side... And this builds up resentment... And then other people take advantage of that, right? They inflame those sentiments and use them to get power.”
The holographic principle differs fundamentally from simulation hypothesis in mechanism and empirical grounding.
- The 'cosmos is a hologram' and 'cosmos is a computer program made by a master programmer' are entirely different ideas: the simulation hypothesis posits an intelligent creator and has no supporting evidence, whereas the holographic principle is a mechanistic claim that reality can be equivalently described in one fewer dimension, with good physics evidence behind it as part of our best understanding of quantum gravity.
“The cosmos as a hologram is an entirely different idea... it's just the idea that there's an alternative description of reality that encodes all the information about our world in a space that has one fewer dimension”