What this covers
This June 2021 Ask Me Anything episode fields questions from Patreon supporters across science, philosophy, media, and other domains, with Sean Carroll working from the framework he lays out in his thesis: Many Worlds Interpretation for quantum mechanics, Bayesian reasoning for epistemology, and the view that phenomena like the Arrow of Time, identity, and Emergence are macroscopic patterns rather than fundamental features. The episode ranges widely—from technical quantum mechanics to cosmology to conceptual puzzles—but Carroll's answers consistently circle back to a core claim: most apparent mysteries dissolve once you adopt the right interpretive lens, and the burden falls on those demanding additional ontology or dynamics to specify their claims with rigor.
The specific territory includes quantum mechanics (uncertainty, entanglement, the nature of particles versus fields), cosmology (dark energy, galaxy-scale intelligence), and classical problems in philosophy of physics and metaphysics. On UFOs and alien visitation, Carroll applies Bayesian reasoning to argue that disagreement lies in prior assumptions, not evidence. On identity and the Ship of Theseus, he defends the view that continuity of pattern, not matter, grounds personal identity—a position with sharp implications for many-worlds branching. He revisits why empirical induction is unreliable compared to Bayesian updating, why Laplace's Demon is physically impossible, and why time travel creates genuine conceptual knots. He also clarifies common misunderstandings: the Higgs field's actual role, why reversibility is not a law but a discovered contingency, and why Emergence is real rather than merely epistemic. Throughout, the tone is skeptical of brute-fact explanations and demands for hidden machinery; Carroll argues that "it's just like that" can count as a genuine answer when the alternative is imposing unnecessary structure on nature.
Carroll argues that many apparent puzzles in physics and philosophy dissolve once you adopt the right framework — Everettian quantum mechanics, Bayesian reasoning, and the view that identity, emergence, and the arrow of time are emergent patterns rather than fundamental features — and that the burden is on those who demand more (collapse rules, brute-fact explanations, alien visitations) to specify their claims rigorously.
- Everettian quantum mechanics is the most parsimonious interpretation because it adds no ontology or dynamical rules beyond the wave function and Schrödinger equation.
- Bayesian reasoning subsumes inductivism and Popperian falsificationism as crude limiting cases.
- The arrow of time, identity, and emergence are macroscopic emergent notions, not fundamental, and can be explained without putting asymmetry in by hand.
Penrose's conformal cyclic cosmology lacks acceptance among working cosmologists and CMB experts, and weighted voting schemes are unjust because those setting weights will bias them in their favor, contrary to democracy's foundational equality principle.
- Roger Penrose's conformal cyclic cosmology claims that evaporating black holes from a prior universe leave imprints on the cosmic microwave background, and that he and collaborators have detected these — but essentially no other working cosmologist accepts this claim, and Penrose, though a great mathematical physicist, is not a CMB data analyst and must trust collaborators whom other experts do not trust.
“essentially nobody else believes this, other than Penrose and his collaborators. There's plenty of working cosmologists who would be very, very happy to find evidence of a pre-existing era”
- Weighting votes by some coefficient like budget contribution is a terrible idea ripe for abuse, since whoever decides the weights will weight their own votes higher; moreover democracy's justification is not its problem-solving efficiency (a bonus) but the principle that all people are created equal and deserve an equal voice.
“once you say that not everyone's vote counts equally, but rather some votes are more important than others, someone's going to have to decide what the importance is, and the people making those decisions, guess what, they're going to weight things so that their votes count more”
Old podcast content should not be retroactively edited to correct disagreements, but genuinely insensitive or offensive material may be edited and transparently flagged.
- Carroll will not retroactively edit old podcast content to correct mistakes he later disagrees with, but would consider editing and flagging content that was genuinely insensitive or offensive — and only ever with full transparency, never sneakily.
“I wouldn't sneakily edit it, if I do ever edit or replace an old podcast, I would certainly want to do so with full transparency.”
The hypothesis that aliens conduct interstellar travel yet remain mostly hidden is incoherent, requiring technology simultaneously advanced enough for near-invisibility but not advanced enough for complete invisibility.
- The hypothesis that intelligent aliens fly interstellar distances to buzz us yet remain mostly but imperfectly hidden is incoherent: it requires technology advanced enough for interstellar travel and near-invisibility but somehow not advanced enough to remain completely hidden — analogous to expecting an omnipotent God to leave only ambiguous evidence rather than being either fully manifest or fully hidden.
“they would buzz us in the atmosphere in small space ships that tried in some sense to remain hidden from us, but failed, so they had enough technology to fly interstellar distances, to zoom around, stay mostly hidden from us, but not advanced technology enough to stay completely hidden from us”
The hard problem of consciousness will not be solved by breakthrough insight but will evaporate as better understanding of brain processes leads us to simply identify neural activity as what consciousness is.
- Following Chomsky's analogy to how Newton abandoned mechanical explanation of motion in favor of intelligible mathematical theories, the hard problem of consciousness will not be solved by a breakthrough insight but will evaporate as we gain better understanding of brain processes and behavior and simply identify that activity as what consciousness is.
“I think it is going to evaporate, it's not that there will be a breakthrough where we say now this is the answer, it's just that we get better and better understanding of what happens in the brain and the human behavior and human thought, and we say, that's it, that's what consciousness is.”
Bayesian disagreement about UFOs stems almost entirely from priors rather than likelihoods, since fuzzy unexplained sightings fit both alien and non-alien hypotheses nearly equally, providing no evidential leverage.
- Being a good Bayesian about UFOs means almost all the disagreement comes from priors (which Bayesianism cannot adjudicate), while the likelihood of fuzzy unexplained sightings is near 1 under both alien and non-alien hypotheses, so grainy footage provides essentially no evidential leverage.
“everyone is entitled to their own likelihood, sorry, to their own priors, but not to their own likelihoods, the likelihood of the data given the theory.”
Most physicists in fundamental fields implicitly adopt many-worlds quantum mechanics, while non-Everettian approaches address only pre-1930 non-relativistic quantum mechanics and fail to engage quantum field theory, gravity, and black holes.
- Most physicists working in fundamental physics, particle physics, cosmology, and string theory are implicitly Everettians even if they don't think about quantum foundations, which is why it makes sense to them that evolution should be unitary in the absence of measurement; non-Everettian approaches (collapse, hidden variables, epistemic) almost exclusively address circa-1930 non-relativistic quantum mechanics and don't engage quantum field theory, gravity, or black holes.
“A lot of people in the area of fundamental physics... are more or less implicit Everettians, even if they don't care about the foundations of quantum mechanics”
Reconstructing historical events with Laplace-demon-like precision is essentially impossible because relevant information was radiated away as light-speed photons, and instruments with sufficient resolution would destroy what they measure.
- It is essentially zero probability that humans could ever build a device functioning as Laplace's demon to reconstruct historical events in fine detail, because the relevant information was radiated away as photons moving at light speed, and any instrument with enough resolution (short enough wavelength) to measure all microscopic particle locations would destroy the macroscopic object being measured.
“if you used high enough energy photons, short enough wavelengths to give you resolution on all the microscopic locations of the particles, you would also destroy the thing you're looking at”
Dark energy is poorly named; its defining features are smooth spatial distribution and negative pressure that gravitationally repels, but 'smooth tension' would be more accurate despite scientific labels persisting by precedence.
- 'Dark energy' is a poor name because everything has energy and its key features are that it is spread smoothly (not clumping into galaxies) and has negative pressure (tension) that pulls like internal springs while gravitationally pushing the universe apart; 'smooth tension' would be more accurate, though scientific labels stick by precedence rather than accuracy.
“One is that it is everywhere, it is spread smoothly throughout the universe... And the other important feature is that it has a negative tension... a negative pressure. Rather than pushing, it pulls.”
According to all known equations, photons in the cosmic rest frame will experience indefinite cosmological redshift as the universe expands without limit.
- With respect to the common rest frame defined by all matter in the universe, a photon will continue to be cosmologically redshifted forever as the universe expands, with no limit, according to all known equations.
“with respect to that frame, a photon, according to all the equations that we know about now, will just get red-shifted forever. There's no limit at all.”
Newton's gravitational theory counts as genuine explanation despite involving action at a distance without mechanical mechanism; scientists must accept answers of the form 'it's just like that' rather than demanding pre-existing solution types.
- Newton's gravitational theory should count as a genuine explanation even though it involves action at a distance and provides no mechanical gear-wheel mechanism; scientists must stay open to answers of the form 'it's just like that' rather than demanding explanations that fit a pre-existing notion of what a solution should look like.
“I wouldn't say that it's not an explanation, it's just not the kind that you were hoping for ahead of time, and I do think that that is something that scientists have to be open-minded about”
There is no logical requirement that every fact must have an explanation; the burden of proof lies on those insisting every fact needs one, since multiple possible worlds exist and some brute fact picks out our actual world.
- There is no logical requirement that the universe being one way rather than another must have an explanation; given that multiple possible worlds are easy to imagine, what picks out our actual world comes down to a brute fact, and the burden of proof lies on those who insist every fact must have an explanation.
“the fact that we live in one rather than the other is going to have to come down to a brute fact... And I'm not afraid of brute facts. These questions about why things are are eventually going to bottom out somewhere.”
Chirality is handedness: for massless light-speed particles, rotation around the momentum axis can be clockwise or counter-clockwise, and the weak interaction's coupling only to left-handed particles constitutes a fundamental asymmetry of nature.
- Chirality is handedness: for massless particles moving at light speed (which have no rest frame), the momentum defines an axis and the particle's rotation around it can be clockwise or counter-clockwise, giving left- and right-handed particles — and notably the SU(2) weak interaction couples only to left-handed particles, an interesting asymmetry of nature.
“The SU-2 part of the weak interactions interacts with the left-handed part of particles, not with the right-handed part, that's a fact about nature that is kind of interesting.”
The universal wave function's branching in a finite-dimensional Hilbert space cannot continue forever; all branches eventually resemble each other as the universe empties to maximum entropy, making branch distinction meaningless.
- Because a region of space corresponds to a finite-dimensional Hilbert space (the observable universe being roughly 10^(10^122) dimensions), the branching of the wave function cannot continue forever; dynamically this corresponds to all branches coming to look alike as the universe empties to maximum-entropy empty space, at which point distinguishing branches stops making sense — like an ice cube finishing melting.
“if branching happens all the time, which it does, then it can't happen forever, 'cause there's only a finite number of branches you could possibly fit into Hilbert space.”
Conservation of information merely means the universe's state plus physical laws determine all past and future states; 'information' is a quantitative translation of 'the state of the universe.'
- The principle of conservation of information just means that the state of the universe at any one time, plus the laws of physics, determines the state of the universe at all earlier and later times; the word 'information' is merely a quantitative translation of 'the state of the universe.'
“The principle of conservation of information is just the idea that the state of the universe at any one point in time, plus the laws of physics, determines the state of the universe at all previous times and all subsequent times.”
Carroll abandoned his earlier belief that the federal government must annually balance its budget, now recognizing that deficit spending with modest inflation for growth and debt reduction during expansions is more economically sophisticated.
- Carroll once believed the federal government should balance its budget every year, but now regards that as a naive economic view, recognizing that running a deficit with modest inflation to encourage growth and paying down debt during expansions is a more sophisticated and natural approach.
“it's very natural to have a deficit, to have a small amount of inflation, to let that encourage growth, etcetera. Pay back some of the debt when the economy is growing.”
The cancelled Superconducting Supercollider would have reached higher energies than the LHC and possibly discovered massive particles beyond the LHC's reach, though equally it might have found nothing in that range.
- The cancelled Superconducting Supercollider would have reached higher energy than the LHC, so it is plausible it could have discovered massive particles the LHC cannot reach (since E=mc² means higher energy enables higher-mass particle production) — though it is equally plausible there are no such particles in that energy range.
“energy is everything when you do this kind of particle physics at colliders, because E equals MC squared, and you want to make high mass particles, you need to have a lot of E so you can get a lot of M”
Equal probability assignment to quantum outcomes fails due to diachronic inconsistency—assigned probabilities shift based on unrelated future measurements on other branches—while the Born rule uniquely avoids this inconsistency.
- Assigning equal probability to quantum measurement outcomes fails because it produces diachronic inconsistency — the assigned probability of an outcome changes depending on unrelated future measurements on other branches; the Born rule (probability equals wave function squared) is the unique probability assignment that avoids this inconsistency.
“the only assignment of probabilities that is consistent diachronically, that does not change if you fiddle around with different kinds of experimental apparatuses, is the Born rule, is that the probability is given by the wave function squared.”
Dividing the universal wave function into branches is a convenient human choice rather than uniquely-determined fact; different branching conventions yield identical observational outcomes.
- Dividing the universal wave function into branches is a convenient human choice, not a uniquely-determined fact; one can equally say Alice's measurement instantly splits the whole universe in some frame, or that branching propagates at light speed and branches knit together when light cones overlap — both give identical observational outcomes for the Alice-Bob entanglement scenario.
“it's up to you whether you say that when Alice measures her spin, the wave function of the universe instantly splits everywhere, including where Bob is, or you have some patchwork quilt of branches”
Carroll prefers variety in places rather than a single favorite, enjoying different experiences in different cities.
- The favorite places question assumes a single best place, but Carroll prefers variety, being a big-city person who enjoys different experiences in different places.
“I actually like being in different places at different times”
Saying 'all but one world collapses' does not constitute a quantum interpretation; one must specify which worlds collapse and whether through physical thresholds or spontaneous processes, stated rigorously and quantitatively.
- Saying 'all but one world collapses' does not constitute an interpretation of quantum mechanics; one must specify which worlds collapse, whether randomly or by some measure, and crucially whether collapse occurs upon reaching a physical threshold (like Penrose's objective collapse) or spontaneously and stochastically (like GRW) — and any such modification of the Schrödinger equation must be stated quantitatively and rigorously.
“this is physics, you've got to tell me exactly what that modification is very, very quantitatively and rigorously and objectively. Until then, you don't quite have an interpretation of quantum mechanics.”
Super-organism intelligence might emerge at intermediate scales, but astrophysically large systems like galaxies cannot attain consciousness because light-speed signal timescales across them exceed the universe's age and stellar lifetime.
- Intelligence or agency could emerge at intermediate super-organism scales (a billion human-sized beings), but truly astrophysically large systems like galaxies cannot become intelligent — not because of distance but because of light-speed signal timescales: a signal takes ~50,000 years to cross the galaxy, and scaling that up relative to a galaxy's life would exceed the age of the universe, which is expanding and emptying out (stars stop shining in 10^15 years), leaving insufficient time for such structures to attain consciousness.
“It's not the distance scales that get in the way, it's the speed of light that gets in the way. It takes tens of thousands of years for a signal to travel from one end of the galaxy to the other.”
If the universe is closed, the wave function stays confined to a small Hilbert space region, leading Carroll to tentatively regard only used dimensions as real and the rest as unreal idealizations.
- If the universe is a closed system, the wave function of the universe never evolves through all of Hilbert space but stays confined to a small region, so quantum mechanics uses only a tiny part of Hilbert space; Carroll leans toward regarding only the used portion as real and the unreached parts as an unreal idealization, though he holds this opinion tentatively.
“the wave function of the universe never evolves all through Hilbert space, if the universe is a closed system, which I take it to be. So in some sense, quantum mechanics only uses a tiny part of Hilbert space.”
Complex systems exhibit recurring universal or near-universal principles—scale-free behavior, cross-scale interactions, emergent sub-system behavior, information processing, non-linear responses—despite each system's uniqueness.
- Although complex systems are messy and unique, they exhibit recurring common features — roughly scale-free behavior, interactions across scales, sub-systems combining into emergent behavior, information processing, and non-linear responses to external influences — so universal (or near-universal) principles of complexity are discoverable, just as people share commonalities despite no two being identical.
“You see sub-systems coming together to show emergent behavior in conglomeration of these sub-systems. You see information processing, you see that there are non-linear ways in which influences from the outside world get thought about”
Science will not become computer science; while computers can find equations fitting data, understanding why relations hold requires human interpretation, and quantum measurement interpretation is qualitatively ill-suited to current computers.
- Science will not become merely a branch of computer science because, while computers can find equations that fit data (as Kepler found ellipses and as Max Tegmark's AI attempts), the deeper task is understanding why those relations hold (as Newton explained ellipses via inverse-square gravity); quantum mechanics needs interpretation not because we don't understand the Schrödinger equation but because measurement outcomes aren't described by it, a qualitatively different task ill-suited to current computers.
“science is enormously more than curve fitting, enormously more than looking for an equation that packages the data in a simple and compact form. We need to understand what is going on”
Publish-or-perish arises because evaluators cannot judge specialized papers and resort to flawed proxies like citations, journal prestige, and paper count; the system is inadequate but hard to replace without mostly unproductive long-term evaluation.
- Publish-or-perish arises because evaluators (provosts, deans, even same-department faculty) cannot read or judge the quality of highly specialized papers, so they fall back on proxies: citation counts (a reasonable but flawed proxy, since Weinberg's electroweak paper went years uncited before becoming the most cited in physics), journal prestige, and crudely sheer paper count — and the system is inadequate but hard to replace because pure long-term thinking would yield mostly nothing from most people.
“we turn to proxies, both because of the practical and in principle difficulties of this. So you turn to how many citations did the paper get.”
ʻOumuamua's 26 km/s velocity is anomalously low for galactic objects, and natural acceleration to relativistic speeds is impossible because stellar assists cannot accelerate objects beyond galactic escape velocity without ejecting them.
- Objects in the galaxy typically move at a few hundred km/s, so ʻOumuamua's 26 km/s is anomalously low (Avi Loeb's point) while 3000 km/s would be anomalously high; it is conceivable but very unlikely for an object to be naturally accelerated to high speeds by stellar gravitational assists, and impossible to reach relativistic speeds, because once accelerated to galactic escape velocity an object leaves the galaxy and stops being accelerated — setting an upper bound.
“once they're accelerated to the escape velocity of the galaxy, they escape from the galaxy and they're no longer being accelerated by gravitational assists from nearby stars... there is an upper limit to speeds that you expect for things in our galaxy, namely the escape velocity”
The master equation and Lindblad equation often fail because they describe open systems while ignoring the rest of the world, working only under restrictive assumptions like uniform thermal baths.
- The master equation and the Lindblad equation describe how a probability distribution or density matrix of an open quantum system evolves, but they often fail because, unlike the always-valid Schrödinger equation, they try to describe a subsystem's evolution while ignoring the rest of the world; they only work under restrictive assumptions like a uniform homogeneous thermal bath with no unanticipated external influences.
“there's a huge, huge problem with these equations, which is that they often don't work. That these equations... work in very specific circumstances, and in other circumstances, they just don't work.”
Lab-leak and bioweapon hypotheses must be distinguished; bioweapon ideas received undue credence for political and racist reasons, but now the opposite error occurs where some discount lab-creation to avoid racist framings.
- On COVID-19 origins, the lab-leak hypothesis (a lab developing a coronavirus that leaked) must be kept distinct from the bioweapon hypothesis (intentional development and release); the bioweapon idea was given undue credence early on for political and racist reasons, but now the opposite error is occurring where some discount the lab-creation hypothesis to avoid endorsing racist framings — and origins should be assessed scientifically, not politically.
“early on this bio-terror explanation was given more credence in certain circles than it should have just for political/cultural/racist reasons... But now the opposite mistake is being made, there's some... they're discounting the lab creation hypothesis because they don't want to give credence to the racist hypotheses”
Neutrinos can exist in genuine flavor superpositions because all three types are very light and nearly light-speed with shared quantum numbers, producing observed oscillations analogous to spin uncertainty.
- Unlike electron-vs-muon (whose mass difference quickly separates wave function parts) or electron-vs-photon, neutrinos can genuinely exist in superpositions of types because all three flavors are very light, near light-speed, and share quantum numbers; each mass eigenstate is a superposition of electron, muon, and tau flavors, producing observed neutrino oscillations analogous to quantum spin uncertainty.
“each mass state is a superposition of the different flavors, as we say, of neutrinos. So the lightest neutrino is part electron neutrino, part muon and part tau neutrino.”
In a topologically closed universe, Gauss's law and Stokes's theorem force net electric charge to zero and net gravitational mass-energy to zero.
- In a topologically closed universe (visualized as a sphere), a single positive charge at the north pole forces its electric field lines to converge and end at the south pole, which by definition is a negative charge — so net charge must be zero; the identical argument with gravitational field lines shows net energy/mass must be zero, all following from Gauss's law and Stokes's theorem.
“they go around the sphere and they focus back on the south pole if you put the charge at the north pole. So what you have at the south pole is a bunch of electric field lines pointing inward and then ending. What that means is there is a negative charge there.”
No fundamental principle requires physical laws to be reversible; reversibility is a discovered fact about actual laws, not a requirement, though our best current laws happen to be reversible.
- There is no fundamental principle that the laws of nature must be reversible; reversibility is a fact discovered about the laws after they are invented — Laplace formulated conservation of information a century after Newton's reversible rules — and it could turn out to be wrong, but our best current laws happen to be reversible, which we accept as a brute fact.
“it's not that we start from a presupposition that the laws of nature must be reversible, it's that we invent the best laws we can... and lo and behold, they're reversible, they just are.”
No algorithm exists for choosing scientific problems; ideas arise from interaction—reading, conversation, talks—and the real art is formulating answerable or progress-yielding puzzles.
- There is no algorithm for choosing scientific problems; most ideas for papers come not from sitting alone in an ivory tower contemplating big questions but from interaction — reading others' papers, conversations, and talks — and the real art is formulating a puzzle that is actually answerable or on which progress can be made.
“Mostly the ideas for papers come from not sitting by yourself in your ivory tower, but from reading other people's papers, talking to people one-on one or one-on several, going to talks that other people give”
Procreation is morally neutral under moral constructivism, where we invent rules rather than discovering pre-existing ones; this grounds skepticism of utilitarianism, which absurdly implies endless procreation even of miserable people.
- Procreation is more or less morally neutral, neither an intrinsic good nor bad, because Carroll is a moral constructivist rather than a realist: there are no once-and-for-all moral rules built into the world; we invent the rules, which remain meaningful and subjective. This is also a reason he is skeptical of utilitarianism, whose maximize-utility logic absurdly implies one should procreate endlessly even producing miserable people.
“I am a moral constructivist, I'm not a moral objectivist or a moral realist, so I don't think that there are once and for all rules built into the structure of the world to tell us what is right and what is wrong. We invent the rules”
Humans prefer organization at the edge of chaos rather than maximum order; entropy and the second law are life's friends, enabling far-from-equilibrium conditions with billions of years remaining.
- Humans do not simply yearn for maximum order; following Per Bak's self-organized criticality and Mandelbrot and Kauffman, interesting biological organization lives on the edge of chaos — between perfect rigidity (a checkerboard) and total randomness — and the rising entropy / second law is not life's enemy but its friend, providing the far-from-equilibrium resource (sunlight on a dark sky) that enables life, with billions of years of it remaining.
“Per Bak... characterized complex systems as living on the edge of chaos... Neither one of those limits is interesting in the sense of biological organization. Where you have interestingness is in between, that's the edge of chaos.”
Requiring counterfactuals and possible worlds for modal language creates tension for strict Humeanism; if only the actual universe exists, it is unclear what justifies carrying the same physics to imagined universes, pulling toward mathematical Platonism.
- The need for counterfactuals and possible worlds creates tension for a strictly Humean view (where only the actual universe exists): if mathematical structures and laws of physics have no independent existence, it is unclear what justifies carrying the same physics and math over to other imagined universes — a consideration that pulls Carroll toward taking mathematical Platonism seriously without fully endorsing it.
“if those laws of physics and mathematical structures don't have independent existence from our universe, what gives you the right to do that? How do you know how to carry over the physics and math from one universe to another?”
The US can remain a cohesive democracy without becoming a European welfare state; the primary driver of populism and democracy's decline is people's felt powerlessness and unheard voice, more than income inequality itself.
- Carroll disagrees with the strong claim that the US cannot remain a cohesive democracy unless it becomes a European-style egalitarian welfare state; while inequality is too high and worsened by the pandemic, he thinks the more important driver of populism — and the more important fix for strengthening democracy — is people's feeling of powerlessness and that their voice isn't heard, rather than income or wealth inequality per se.
“the reason why populists resonate is not because some people are poor... it's 'cause people think that they're powerless, that their voice is not being heard, that the government is going and doing its own thing in ways that they don't have any influence over”
Empirical induction is invalid; the proper approach is Bayesian—assigning credences to competing theories and updating on observations, where the same evidence has different strength under different theories.
- Empirical induction (seeing white swans and concluding all swans are white) is invalid because tomorrow's black swan refutes it, lacking the certainty of mathematical induction; the proper approach is Bayesian — assigning prior credences to competing theories and updating on observations, where the same data (twelve white swans) strongly disconfirms a 50-50 theory but gives zero evidence under a hemisphere-clustering theory.
“Inductivism is a sloppy preliminary version of Bayesianism is what I would say.”
Measuring a particle's gravitational or electromagnetic field collapses the superposed wave function identically to measuring the particle, since fields cannot be separated from particles; gravity is a red herring in such thought experiments.
- You cannot separate a particle from its field (electron from electric field, planet from gravitational field), so measuring a particle's gravitational or electromagnetic field is just as good as measuring the particle itself: a superposed particle's field is correspondingly superposed, and measuring it collapses you onto one location and branches the universal wave function — making gravity a red herring in such thought experiments since electromagnetism behaves identically.
“when you say, well, I observe the gravitational field of the particle, but not the particle itself, the answer is, that's just as good as measuring the particle itself.”
The Higgs field gives mass to particles, but once mass exists it is mass; the Higgs has no special gravity connection beyond providing the mass that then gravitates.
- The Higgs field is merely the mechanism that gives mass to particles like electrons and quarks, but once mass exists, mass is mass and the Higgs can be forgotten (as Dirac and Feynman did QED without it); people wrongly attribute an 'essence of massiness' to the Higgs and infer a Higgs-gravity connection, but the Higgs has no special connection to gravity beyond providing the mass that then gravitates.
“they somehow attribute some essence of massiness to the Higgs field... No, it really doesn't. The Higgs field is the mechanism responsible for giving particles mass, and mass has a connection to gravity, but the Higgs has no connection to gravity because of that connection.”
The prior probability that Kuiper Belt clustering comes from a primordial black hole rather than a planet should be very small, since we observe many planets but no primordial black holes.
- The prior probability that the Kuiper Belt clustering attributed to a hypothetical Planet 9 is instead caused by a primordial black hole should be very small, because we know of many planets and planet-like objects but have never observed any primordial black hole — only large stellar-collapse black holes — so the planet explanation is far better supported.
“we know of lots of planets and we know of lots of planet-like things, we don't know of any tiny primordial black holes, any, right?... So I would put the prior on that as pretty darn small.”
Two particles cannot have opposite wave functions that cancel because there is one universal wave function for all particles; interference occurs only between contributions to a single particle's wave function.
- Two particles cannot have opposite wave functions that cancel like noise-cancelling headphones, because there is not a separate wave function per particle — there is only one wave function for all the particles in the universe; interference happens only between two contributions to the wave function of the same particle, as in the double-slit experiment.
“There is only one wave function for all the sets of particles in the universe, so you can't have the wave function of particle 1 and the wave function of particle 2 interfere with each other”
Time travel is conceptually troublesome because the arrow of time suggests the past is fixed while the future is open; closed timelike curves confuse individual future notions with the universe's past notions.
- Time travel is conceptually troublesome because the arrow of time gives the impression that the past is fixed while the future is open to choices; a closed time-like curve mixes up your individual notion of the future with the universe's notion of the past, which is what makes such scenarios hard to make sense of — though Carroll expects entropy would still increase as always in any workable scenario.
“the existence of a time machine or closed time-like curve mixes up your individual notion of the future with the universe's notion of the past, and then it's hard to make sense of these particular things.”
Emergence is a real feature built into physical law rather than a mere human convenience; our laws permit reliable macroscopic predictions from incomplete data, a non-trivial feature that did not have to hold.
- Emergence is a real feature built into the fabric of nature rather than merely a human convenience born of ignorance; with generic laws lacking locality and microscopic structure, nothing could be said about a system without complete microscopic detail, but our actual laws of physics permit reliable macroscopic predictions (like Earth's orbit from its center of mass) from incomplete data — a non-trivial feature that didn't have to hold.
“it is a feature of our laws of physics that it allows these emergent behaviors in ways that human beings have observational access to the data you need... that's a nice feature of the laws of nature that didn't have to be there.”
Objects can abruptly end in space but not in time because physical laws relate states across moments (via Noether's theorem and conservation laws), whereas spatial continuity lacks such constraints.
- Objects can abruptly end in space but not in time because the laws of physics relate stuff at different moments of time, made precise by Noether's theorem (symmetries yield conservation laws like energy conservation over time) combined with locality (effects propagate no faster than light) — so something with a given energy now will have something similar a moment later, whereas spatial continuity would require uniform everywhere.
“if you have conservative energy and you can't change things at rates faster than the speed of light, that if there is something here now, there's something similar to it, the same amount of energy, a moment later”
Planck-mass black hole remnants as dark matter are unpopular because they would have problematic virtual-process effects, be nearly impossible to produce in sufficient numbers, have Planck-scale weak interactions, and be essentially undetectable.
- Planck-mass black hole remnants as dark matter are unpopular because such relics would have many internal states causing problematic virtual-process effects, and producing the enormous numbers needed in the early universe is hard; even if they existed, their Planck-scale (gravitational-strength) interactions and 10^18-proton-mass scale mean very low number density and extremely weak coupling, making detection essentially impossible.
“If those relics existed, they would have a lot of entropy... there'd be a lot of different ways of making them... and therefore, they could have bad effects in virtual processes, Feynman diagrams”
Carroll's podcast strategy allows guests to share expertise while he briefly states his disagreements; the goal is educating listeners by having them understand experts' viewpoints, with his own opinions amply represented across episodes.
- Carroll's podcast strategy is to let guests share their expertise while briefly stating his own disagreements and reasons rather than debating back and forth, because the goal is to educate listeners by having them enter the mindspace of understanding an expert's view — and integrated over many episodes plus solo shows and AMAs, his own opinions are amply represented.
“What I'm trying to do is educate the people who are listening. So I want them to know what my opinion is and the reasoning behind it, but really, I want you to know I'm bringing some expert”
Gravitons must exist if both general relativity and quantum mechanics are true, as quantizing classical wave theory generically yields particle excitations like photons and phonons; gravitons are light-speed ripples in the gravitational field.
- Gravitons must exist if both general relativity and quantum mechanics are true, regardless of having a full theory of quantum gravity, because quantizing a classical wave theory generically yields particle-like excitations — just as phonons are quantized sound waves and photons are quantized electromagnetic waves; a graviton is a tiny ripple propagating at light speed atop a classical gravitational field, and like photons it cannot escape a black hole.
“If you believe in general relativity and you believe in quantum mechanics, you should believe in gravitons. To believe in gravitons does not mean that you need to think you have a full theory of quantum gravity”
In many-worlds, expectations and rules for living should be identical to a single stochastic world; playing a quantum lottery guarantees winning on some branches but losing on far more, yielding net-negative outcomes.
- For actual humans across the many worlds, all expectations, actions, and rules for living should be exactly the same as if living in a truly stochastic single world; playing the lottery with a quantum random number generator would guarantee you win on some branch but lose on far more branches, so the net outcome (since lotteries are net-negative expectation) is that the totality of your selves lose money — just as the expectation value argues against playing in a single world.
“the upshot of many worlds is that for actual human beings in these many sets of worlds, all of their expectations and all of their actions and all of their rules for living life should be exactly the same as if they lived in a truly stochastic single world.”
A single particle's spread wave function doesn't conflict with relativity if interpreted as uncertainty, but two entangled particles and Bell's inequalities prove local predetermined values cannot work.
- A single particle's spread-out wave function does not by itself conflict with relativity, because you could interpret the wave function as mere uncertainty and treat measurement as learning where the particle always was; it is only with two entangled particles that Bell's inequalities prove you cannot have local, classical, predetermined values consistent with experiment — so genuine entanglement is required to create the relativistic puzzle.
“if all you had were one particle, you could always imagine the wave function was somehow just a way of talking about your uncertainty about it... You can get away with that if you just have one particle, but if you have two particles and they're entangled, you can no longer get away with that, that's the point of the Bell inequalities.”
Carroll has no ideal alternate history because tweaking discrete events invokes chaos theory and prediction difficulty; only preventing clear tragedies like the Holocaust seems straightforward, though even there causal forces might produce equivalents.
- Carroll has no ideal alternate history because tweaking discrete influential events to engineer a better world runs into chaos theory and the difficulty of prediction — e.g., a President Robert Kennedy might not have ended Vietnam (which Democrats JFK and Johnson had long run), and stopping Watergate might have been net-negative by forgoing reforms; only preventing clear large tragedies like the Holocaust seems more cut-and-dried, yet even there the social forces might have produced something similar.
“the question of going to certain discrete, influential historical events and tweaking them a little bit so as to bring about the world I would rather see is much more problematic... There you get into chaos theory and the difficulty of making predictions”
Chaos theory limits practical prediction but not in-principle prediction by Laplace's demon; chaos theory never overturned determinism, and pre-chaos no sensible person believed perfect prediction was achievable in practice.
- Chaos theory only limits prediction in practice for real computers and people; Laplace's demon, a thought experiment with perfect information and perfect calculating ability, has no trouble with the three-body problem or chaotic systems, so chaos theory never overturned the in-principle determinism of Laplace's demon — and even before chaos theory, no one sensibly believed perfect prediction was achievable in practice.
“all the discussion about how chaos theory says you can't make predictions is only in practice in the real world... so there's no sense in which chaos, etcetera, etcetera overturned, the idea of Laplace's demon.”
Quantum gravity is hard because quantizing classical spacetime fails; black hole information, holography, and complementarity suggest gravity's lack of perfect locality and fixed background requires a different approach than standard quantization.
- Quantum gravity is hard, in Carroll's view, because people keep trying to quantize a classical theory of spacetime, whereas the lessons of black hole information, holography, and complementarity suggest quantum gravity will not arise that way; gravity's lack of perfect locality and of a fixed background spacetime makes the standard quantize-a-classical-theory procedure (which worked for electromagnetism and the standard model) fail.
“my takeaway from things like black hole information and holography and complementarity... is quantum gravity will not arise from quantizing a classical theory of spacetime.”
Popperian falsificationism is a crude approximation to Bayesian reasoning; refutations are never cut-and-dried and Bayes corrects this by never setting likelihoods exactly to 0 or 1.
- Popperian falsificationism, treating all conjectures as candidates and slicing away the refuted ones, is a crude approximation to Bayesian reasoning that fails because refutations are never cut-and-dried — an anomaly may indicate a new planet (Neptune saving Newtonian gravity) rather than a falsification (Mercury requiring general relativity); Bayes corrects this by never setting likelihoods exactly to 0 or 1, always allowing for experimental error.
“the Popperian approach to truth is a crude approximation to the Bayesian approach to truth.”
A macroscopic flying object is not entangling with particles it meets; entanglement requires different parts of one system's state correlating with different parts of another's, as in superposed Schrödinger's cat.
- A big macroscopic object flying through the air is not entangling with the particles it meets — it is merely interacting; within a single branch of the wave function a photon either hits the object or doesn't, and neither is entanglement. Entanglement requires different parts of one system's state correlating with different parts of another's, as in Schrödinger's cat where the cat is in a superposition of locations.
“once you're classical, once you're big and coherent and in some branch of the wave function, you're not constantly in-tangling with things around you, that's kind of a more special event.”
Electron spin is genuinely spin, not merely a metaphor; the electron as a quantum field excitation can carry real angular momentum, so spin-1/2 is proper angular momentum in the field.
- Electron spin really is spin: the common claim that it isn't arises from treating the electron as a zero-size dot (which can't spin) or a small sphere (which couldn't spin below light speed to match its angular momentum), but the electron is properly the excitation of a quantum field, and fields can carry genuine angular momentum even when stationary, so spin is best understood as the angular momentum contained in the electron field.
“you should and can think of, can and should think of the spin of the electron as amount of angular momentum contained in the electron field, so that is the best way to think about it. It really is something spinning in some very real sense.”
Spin-1/2 for elementary particles is allowed because the rotation group's topology permits objects returning to their starting configuration only after 720-degree rotation, as exemplified by spinning a coffee cup.
- Spin-1/2 is allowed for elementary particles (unlike orbital angular momentum, which comes only in integer multiples of h-bar) because of the topology of the rotation group: there exist objects whose relationship to their surroundings returns to the start only after a 720-degree rather than a 360-degree rotation, exemplified by holding a coffee cup and rotating it without releasing — and electrons, quarks, and neutrinos realize this spin-1/2 behavior.
“if you think about the relationship between the coffee cup and yourself, if you rotate it without moving your hand, without letting go of it, it will come back only when you do a rotation of 720 degrees”
Carroll changed his mind on demoting Pluto; while definitions are arbitrary, those mapping onto natural features aid thinking, and a planet definition extendable to exoplanets helps careful thought more than including Pluto.
- Carroll changed his mind on demoting Pluto: although language is arbitrary and we can define words however we like, definitions that don't map onto features of reality don't help us think — and as we discover more exoplanets, having a 'planet' definition extendable beyond our solar system aids careful thought, and no such good definition includes Pluto.
“if the words we invent and the definitions we have for them don't map onto some feature of reality, then they don't help us think.”
Ship of Theseus shows identity over time is emergent and approximate, inhering in pattern persistence not specific atoms; duplication scenarios have no answer to 'which is real' because both are continuous with the original.
- The ship of Theseus thought experiment shows that continuity of identity over time is a macroscopic, approximate, emergent notion rather than fundamental: identity inheres in the persistence of a pattern over time, not in the specific atoms, which is why people remain themselves despite atomic replacement, and why questions like 'which is the real one' after duplication (Parfit's teletransporter, or many-worlds branching) have the answer 'neither — both are continuous with the original pattern.'
“the notion of continuity of identity over time is a macroscopic, approximate emergent notion. It's not fundamental.”
Explaining time's arrow without cheating requires similar conditions in far past and future; entropy can always increase, and empty space with cosmological constant may be unstable to baby-universe spawning in both temporal directions.
- To explain the arrow of time without cheating (without inserting a past-future asymmetry by hand into time-symmetric laws), the conditions in the far past and far future must be similar; Carroll and Jennifer Chen's two ideas are that entropy can always increase (no maximum equilibrium) so any state evolved in both time directions sees entropy rise, and that empty space with a small cosmological constant is unstable to spawning baby universes in both temporal directions, making our Big Bang the beginning of one baby universe rather than the middle of everything.
“if you wanted to explain the arrow of time without cheating, so without putting in an asymmetry by hand... you need the conditions in the far, far past and the far, far future to be similar to each other”
Many-worlds is the most parsimonious quantum interpretation—not positing multiple worlds but claiming only the evolving wave function exists, with worlds as emergent consequences, providing a working framework for cosmology and field theory.
- Many worlds is the most parsimonious interpretation of quantum mechanics — not a theory positing many worlds, but the claim that all that exists is the wave function evolving solely by the Schrödinger equation, with no additional ontology or dynamical rules; the worlds are an emergent consequence, and Carroll adopts it as his working framework so he can build cosmology and field theory on it rather than perpetually re-examining alternatives.
“many worlds is absolutely the most parsimonious explanation, the fewest equations, the fewest ontological elements.”
Classical mechanics allows precise simultaneous position-velocity knowledge, and frequency-phase uncertainty applies only to finite-interval waves; quantum uncertainty arises because particles are wave functions and position and velocity are incompatible observables.
- In classical mechanics there is no position-velocity uncertainty (Laplace's demon could know both precisely); frequency-phase uncertainty applies only to waves over a finite interval. The Heisenberg uncertainty principle arises in quantum mechanics because particles are wave functions, and position and velocity are not properties of those waves but incompatible observables — reflecting that what is observable is not what exists when unobserved.
“the whole reason why quantum mechanics is hard, is because what is observable is not what exists when you're not observing it.”
The Higgs field value is spatially constant today giving uniform particle masses, but temporally variable; in the early hot universe it sat at zero effective value, giving electrons and quarks zero vacuum mass.
- The Higgs field value is constant everywhere in space today, so particle masses are the same throughout the universe, but the field value is not constant over time: in the hot early universe the effective potential shifted so the Higgs sat at zero, giving electrons and quarks zero vacuum mass at early times (though plasma effects could give effective masses).
“the Higgs field is not constant over time, and indeed, the mass of the electron or the mass of the quarks was zero at early times.”
Idealism is evidently false because physical world objectivity is more evident than anything; early quantum weakness about measurement has passed with multiple purely physical theories reproducing quantum results.
- Idealism — taking the mental rather than physical world as primary — is a laughably bad strategy, since the objective reality of the physical world is more evident than anything; the early quantum-mechanical moment of weakness (where measurement seemed to bring reality into existence) has passed because we now have multiple purely physical mechanistic theories reproducing quantum results, and the fact that all observers derive the same laws of physics squelches idealism.
“the fact that quantum theory drew a distinction between what we directly observe and what really exists opened a door for people to say, well, maybe we're bringing reality into existence... But now we know better”
Black holes are thought to erase baryon and lepton number because gravity couples only to energy and cannot distinguish particles from antiparticles of identical mass, evaporating into 50-50 baryon-antibaryon emissions.
- Black holes are thought to erase baryon and lepton number: a black hole made of baryons mostly evaporates into photons (zero baryon number) and emits baryons versus anti-baryons roughly 50-50, because gravity couples only to energy/mass and cannot distinguish baryons from anti-baryons (which have identical mass) — a generic expected feature of quantum gravity, though never observed since the effects are extremely weak.
“gravity doesn't know the difference between baryons and anti-baryons. Gravity knows how much energy you have, or how much mass you have”
In Everettian mechanics there is no special observation role; decoherence via environmental entanglement branches the wave function, and the observer trivially goes along being already environmentally entangled.
- In Everettian quantum mechanics there is no special role for observation; what matters is decoherence, which is the tiny quantum system becoming entangled with the many untracked degrees of freedom of the environment — and the observer, already entangled with the environment, goes along trivially, so it is environmental entanglement, not observer entanglement, that branches the wave function.
“decoherence is the tiny quantum system becoming entangled with the environment. So in that more modern framework, yes, entanglement is completely central, but it's not entanglement with the observer that matters, it's entanglement with the environment.”
No step beyond neutron degeneracy pressure is known; when gravity exceeds a neutron star's pressure it collapses directly to a black hole with no intermediate stable states.
- There is no known step beyond neutron degeneracy pressure: when gravity exceeds a neutron star's degeneracy pressure, it is thought to collapse directly to a black hole with no intermediate stable states, which is why information seems to dry up beyond quark stars — though this is hard to prove because it sits at the unknown interface of quantum field theory and general relativity.
“what we think happens is it just collapses to a black hole. There's not like one or more or an infinite number of intermediate steps, it just collapses all the way to zero.”
Carroll doesn't deeply study alternative interpretations because he judges Everett correct, and working physicists must adopt a framework to progress rather than perpetually re-examining alternatives in Cartesian skepticism.
- Carroll doesn't deeply study alternative quantum interpretations because he thinks Everett is correct, and a working physicist must adopt some framework as a starting point rather than perpetually re-examining alternatives — otherwise one is stuck in Cartesian skepticism unable to build anything; everybody assumes some theory as provisionally correct and moves forward, remaining open (as a good Bayesian) to revision if strong evidence or arguments arrive.
“Everybody takes some either theories or views of the world as more or less what they assume is correct until other information comes in and they move on from there... Otherwise, you're stuck in some terrible Cartesian skepticism scenario where you can't believe anything”