What this covers

In this conversation, theoretical physicist Lee Smolin and host Sean Carroll explore how explicit philosophical principles should guide the construction of new fundamental theories in physics. Smolin argues that rather than proceeding by intuition or piecemeal adjustment, physicists should build from well-articulated starting points—especially relationalism, background independence, and the principle of sufficient reason. The core disagreement with much contemporary physics concerns the status of time itself: Smolin contends that time should be treated as fundamentally real, not emergent from a timeless mathematical structure, and that the laws and constants of nature should be understood as dynamical and evolving rather than fixed. Carroll and Smolin also discuss consciousness, examining whether panpsychism can coherently assign causal power to mental properties, and Smolin's speculative proposal that precedent-free events—potentially connected to consciousness—could drive the evolution of physical laws themselves.

The conversation ranges across loop quantum gravity, the relationship between string theory and competing approaches to quantum gravity, and the role of cosmological natural selection as an alternative to the anthropic principle. Smolin distinguishes his position from both conventional conservatism in physics and from treating time as merely emergent; he sketches a picture where the universe is fundamentally a process of events being created from prior causal events, with an objective, universal present. The discussion also touches on how good theoretical physics requires judgment calls on plausibility that reasonable people genuinely disagree about—and why that intellectual disagreement is not a sign of failure but of the field's openness to fundamental revision.

Sharpest takeaway

Smolin argues that fundamental physics should be guided by explicit philosophical principles—especially relationalism, background independence, and the principle of sufficient reason—and that taking time as fundamentally real, with laws and constants that evolve, offers a better path forward than treating space, laws, and the arrow of time as fixed or emergent.

  • New physical theories should be built from articulated principles (relationalism, background independence, causal completeness, reciprocity) to give a 'running start.'
  • Time and its arrow should be taken as fundamental rather than emergent, with the universe modeled as a process of created events.
  • Constants and laws of nature should be dynamical and explained by evolution, as in cosmological natural selection.

The argument · threads10 threads · 42 claims
0.85

Space, time, and laws are not fundamental but emergent from deeper causal processes.

11 pointscentrality 5/5
  • Space may not be fundamental but emergent from something deeper; when locality emerges, the residual property that things talk to non-nearby things—nonlocality—would naturally appear, and this emergent nonlocality could be the very nonlocality observed in quantum phenomena, tying quantum foundations to quantum gravity and to how the universe chose its laws.

    “space is not fundamental, and that space is there for emergent out of something more fundamental and that nonlocality... the property that we call locality... would be disordered... To have nonlocality appearing in the universe and that nonlocality could play a role in the nonlocality that we discover in the quantum phenomenon.”

  • Ordinary physics implicitly treats a small subsystem as weakly interacting with a stable rest of the universe used as a reference; but a true theory of the whole universe, treated as a closed self-contained system, requires a fundamentally different starting point, which is a key lesson of taking relational principles seriously.

    “the way that we actually test the theory implicitly assumes that we're discussing a small part of the universe, which is only weakly interacting with the rest of the universe... if you... want to make a theory of the whole universe... then you've gotta find a different starting point”

  • Bringing gravity together with the other forces will change our description of those other forces as much as it changes gravity—so the question is not just whether to quantize relativity but possibly to 'relativise the quantum.'

    “I tend to agree with Einstein, that bringing together gravity and the other forces will change our description of the other forces as much as gravity so whether we'll relativise the quantum or quantise relativity.”

  • The revolution Einstein began in physics around 1905 has two connected parts—quantum theory and the theory of space, time, gravity, and cosmology—and we are still in the midst of it, with three interlinked challenges: making sense of quantum mechanics, describing gravity and spacetime in the same terms as other fields, and explaining how the universe chose its laws.

    “Einstein started this revolution in physics in 1905. And there were two parts to the revolution, quantum theory... and the revolution of thinking about space and time and gravity and cosmology... And my sense is that we're still in the midst of this revolution.”

  • Physics progresses by stripping away elements previously treated as fixed and given—from Newtonian absolute space to relativity and quantum mechanics—so the less an ultimate theory declares to be fixed once and for all, the better; this should extend to making the constants of nature dynamical rather than given.

    “part of your intuition is that we make progress by stripping away given elements... it's a series of stripping away of absolutes.”

  • Smolin distinguishes time being 'real' (true even of emergent things) from time being fundamental; he wants more than emergence—at the deepest level the world should be a process of events created and then passing, with each event caused by prior events, and an objective, universal 'now' such that the continual creation of events and a real present together constitute the fundamentality of time.

    “the world at that level will be a process of events, which are created and which then disappear. And the events have causes which are prior events”

  • Smolin's least popular position is that time is fundamental and goes all the way down, beneath space and even laws, expressed as a process of events being continually created from prior causal events—a view developed through concrete models like energetic causal sets.

    “the one that says that time is fundamental and goes all the way down below space, below laws.”

  • Smolin holds that the arrow of time is built into the fundamental nature of reality rather than being merely an emergent collective phenomenon arising from time-symmetric microphysics; reintroducing it below Newton's laws requires explaining why the observed laws are approximately time-symmetric (via mechanisms like limit cycles) and seeking time-asymmetric extensions of general relativity and quantum field theory.

    “Absolutely, yes.”

  • Smolin models the universe as fundamentally discrete: a set of events, each created by prior parent events and giving rise to children events; energy and momentum are kept fundamental rather than emergent, and each event has a 'view'—what it sees from its place in the causal network—on which the dynamics acts.

    “the universe consists of a number of events which are created by some just previous events. Each event has a number of progenitors or parent events, and each event gives rise to a number of... children events”

  • In Smolin and Verde's newer work, the fundamental distinction is not past/present/future but between properties that are indefinite (incompletely known) and those that are definite; an event is a moment of transition in which some indefinite quantity becomes definite, and a simple picture of the world is built from that.

    “The fundamental distinction is between some properties that are incompletely known or indefinite, and some quantities that are definite, and that when an event is, is a moment of transition of some indefinite quantity becoming definite.”

  • Because this world has no space—no distances, derivatives, or field equations—dynamics is instead defined on the 'views' of events: a measure of how different two events' views are is computed across all pairs (the 'variety' of the system), placed in the action so its absolute value is potential energy and its rate of change is kinetic energy, with the operational 'now' being events created but not yet finished creating their children.

    “this world has no space so there's no forces that can fall off at one over distance squared... so how are we gonna do dynamics? And one of the ideas is that these views are the stuff that dynamics acts upon”

0.77

Physics must be rebuilt on few well-articulated principles: relationalism, background independence, causal completeness, and reciprocity.

8 pointscentrality 5/5
  • New physical theories should be built from a small set of well-articulated principles—as Einstein did with relativity—rather than developed by unreflective intuition; Smolin's list includes realism, relationalism, background independence, causal completeness, and reciprocity.

    “All the lists that have been done start with the idea of relationalism... the principles that Leibniz laid out... about how to make a background independent”

  • Background independence requires that a theory contain no fixed mathematical structures used merely for reference; anything that influences the evolution of some degree of freedom must itself be reciprocally influenced and dynamical—there can be nothing used as coordinates or landmarks that is not also part of the dynamics.

    “there's no fixed mathematical structures which don't have a back and forth relation with other things... there's not anything which we just use for reference, for coordinates, for landmarks, that's not also dynamic”

  • By the principle of the identity of indiscernibles, no two events or objects in the universe can have exactly the same relational properties—if two events see exactly the same surroundings they must be the same event—and Smolin finds this a very useful principle for constructing physical theories.

    “there can be no two events in the world that are identical, described by the relational properties... if you have an event and you look around you and you see a certain deal around you, and you have another event which sees exactly the same view around it, then they must be the same event.”

  • Relationalism holds that there is no absolute position in space or absolute time; location and time exist only relative to landmarks and to actual physical change and clocks, contrary to Newton's belief in an absolute view, so everything comes down to relationships.

    “there's no such thing as absolutely where you are in space because where you are relative to some landmarks... There's no absolute time... It all comes down to relationship.”

  • Mach's principle—that rotation is rotation relative to all the matter in the universe rather than to absolute space—is realized in general relativity only when the universe is conceptualized as closed; with asymptotic boundary conditions tying spacetime to a fixed form far away, the principle is not satisfied.

    “if we put on what are called asymptotic conditions... then this principle is not realised because we're anchoring... but if the universe is conceptualised as closed, then it is satisfied.”

  • Reciprocity, taken from one of Einstein's general relativity papers, requires that if A influences B then B must influence A; this is why pilot-wave/Bohmian mechanics is unsatisfying, since the wave function influences the particle positions but the particles do not influence the wave function.

    “That was the word for what I was saying about, if A influences B and B influences A. And it comes from one of Einstein's papers on general relativity.”

  • Causal completeness asserts that there is nothing outside the universe—every causal explanation, tracing causes back in time, must remain within this one universe, which has a single causal future.

    “That means there ain't nothing outside the universe. That means that every explanation in terms of causes going back in time has to stay within this one universe.”

  • The principle of sufficient reason holds that for every fact about the universe where one can ask 'why is it this way and not otherwise,' either there is a reasonable explanation (by a reasonable-person standard) or the question itself has no meaning—its most important use being to identify pseudo-questions, such as where the whole universe is located.

    “for every fact about the universe... either there's a reasonable explanation for that... Or... you're asking a question that really has no meaning.”

0.52

Why the laws are the laws requires explanation through cosmic evolution analogous to biology rather than axiomatic acceptance.

4 pointscentrality 4/5
  • Following Peirce, it is not enough to explain nature in terms of laws; the most important thing is to explain why those are the laws, and within science the only way to explain why the laws are the laws is if they are the result of some evolution analogous to biology.

    “it's not enough to explain nature in terms of laws. The most important thing is to explain why are those the laws... the only way of explaining why the laws are the laws is if they're the result of some evolution, analogous to biology.”

  • For a multiverse theory to make a real prediction, the predicted feature must be orthogonal to whether life exists; cosmological natural selection's prediction relating the strange quark mass to the upper mass limit of neutron stars qualifies because it is not coupled to our existence, whereas anthropic 'predictions' are not genuine.

    “To be a real prediction, it has to be a feature of the universe that's decoupled from orthogonal to whether there's life or not.”

  • The idea that constants of nature change over cosmological time, going back to Dirac, has been repeatedly searched for—e.g., variation in the fine-structure constant or in G over G—and has not been found down to several orders of magnitude.

    “The idea that the constants of nature change in cosmology... goes back to João... they've been looked for repeatedly and not been found.”

  • Smolin hypothesizes that most events' 'views' have fixed consequences matching similar past views, but a small minority of views are without precedent—the laws of nature don't know what to do with them—and must discover novel responses; these precedent-free events are conjectured to correspond to events involving qualia and consciousness, giving consciousness causal power and a role in how laws change.

    “most views have fixed consequences... But in order to believe that the laws can change, I have to believe that some very small minority of views are not... don't have precedent... the hypothesis that it's those that correspond to events, which have to do with quality of consciousness.”

0.47

Cosmological natural selection via black-hole reproduction makes genuine testable predictions unlike anthropic string-theory explanations.

2 pointscentrality 4/5
  • Cosmological natural selection proposes that the universe reproduces through black holes spawning baby universes, with constants of nature changing only slightly per generation; this gives a search procedure that drives parameter space toward regions producing many black holes, making predictions—unlike the string landscape with the anthropic principle, which Smolin argues makes none.

    “what the universe wanted to do was reproduce itself and reproduce itself through black holes that led to baby universes... all I had to add was a constraint that said... per dimension list constants to change per generation is small. And then you have a search procedure that will discover those regions of parameter space and elite universes that have a lot of black holes.”

  • The original hope that string theory would yield a manageable number of vacua (e.g., ~100,000 Calabi-Yau compactifications) to search for the one matching the standard model collapsed when Strominger realized there were uncountably many more, making phenomenological selection seemingly hopeless.

    “Andy had realized that there were uncountably many more. That was just the tip of the iceberg... We'll never get any phenomenology out of this because how are we going to search this vast, huge space to find the standard model?”

0.47

Quantum mechanics is incomplete; deeper theory awaits discovery of space-time nature and nonlocality.

2 pointscentrality 4/5
  • Smolin's working hope is that quantum mechanics is incomplete and that there is a deeper physical theory to be discovered—revealing the nature of space and time—which is why he resists the many-worlds interpretation, since accepting Everett would imply there is nothing left to discover in the direction he has pursued in quantum foundations.

    “my hopes are and have always been, that quantum mechanics was incomplete. And there was a physical theory that is physical discoveries to be made, which would be enlightening about the nature of space and time.”

  • Including matter in quantum gravity raises the compelling fact that the standard-model particles (especially neutrinos) are chiral—nature is not symmetric under mirror reflection—and this hidden chirality, which also appears in the Ashtekar formulation of gravity and in twistor theory, may be a fundamental clue rather than an accident.

    “matters becomes chiral... looking at the particles in the standard model, especially the neutrinos in a mirror, they're not the same.”

0.33

Consciousness possesses genuine causal power through precedent-free events and plays a role in law evolution.

2 pointscentrality 3/5
  • Smolin long held a Russellian-monist (panpsychist) view: that qualia and conscious experiences are real intrinsic aspects of the matter making up the brain, preserving physical determinism (no functional or causal dualism) while admitting a dualism of property kinds—external relational properties physics describes and inner intrinsic qualities like the experience of color.

    “I believe that qualia and conscious experiences are real, and they other aspects of the matter which makes up our brain... you get to be a dualist as to the kinds of properties that things have”

  • Carroll's remark—that if qualia and consciousness have no causal influence on the world (since physics already explains all causes) it makes no sense for nature to bother having them—led Smolin to conclude that the naive panpsychist view is wrong and that consciousness must instead have genuine causal power in the world.

    “why would nature, whatever qualia and consciousness are, if they have no causal influence in the world because the causes are already explained by physics, why should nature be part of them?”

0.33

Concrete models of variety-based causal-set dynamics recover spacetime, relativity, and quantum mechanics from simple rules.

2 pointscentrality 3/5
  • From the variety-based dynamics, Smolin's models recover an emergent Minkowski spacetime embedding the events, the motion of relativistic particles, the classical trajectory in the limit where a quantity 1/N goes to zero, and—at leading order in 1/√N with a non-relativistic limit and certain initial conditions—a version of quantum mechanics (a Schrödinger-like equation on a complex wave function with nonlinear corrections).

    “what emerges is a Minkowski spacetime with an embedding that takes these events to points of Minkowski space... For one over N goes to zero... we get the classical trajectory... taking a non-relativistic limit... we drive a version of quantum mechanics.”

  • A model is something definable by a small list of rules that can be 'played' like a game and captures in the simplest way the implications of some assumption or phenomenon without including all the details—distinct from a full theory—and good models (like Per Bak's sandpile) demonstrate an idea clearly and can be solved on paper or coded simply.

    “A model is something that can be defined by a small list of rules. You can imagine playing it the way that somebody would play a game... a good model demonstrates or exhibits some idea clearly”

0.22

String theory and loop quantum gravity pursue the same underlying quantization idea despite harmful sociological division.

3 pointscentrality 2/5
  • String theory and loop quantum gravity should be seen as different versions of the same underlying idea—the development of quantized loops of electric flux—rather than as opposing programs, and the artificial sociological barrier between the two communities was harmful.

    “those were different versions of the same idea. And I certainly try to bring them closer.”

  • Both string theory and loop quantum gravity have had brilliant successes but also roadblocks that have persisted for at least 20 years, and it is no longer productive to keep doing what was done 25 years ago—evidenced by the fact that almost no new researchers pursue those old programs.

    “Each of them has roadblocks, which have been there for at least 20 years and longer. And I think it's ridiculous that we continue”

  • Loop quantum gravity arose by applying the idea—drawn from QCD and the quantization of magnetic flux in superconductors—that electric flux gets quantized and carried along strings, to the gravitational field, using a formulation of general relativity (Ashtekar variables) focused on the same parallel-transport gauge variables as Yang-Mills theory.

    “My first paper was the version of lattice quantum gravity, in which the loops of quantised electric flux on the lattice were the main actors.”

0.19

Both radical conservatism and willingness to discard fundamentals are necessary for progress in deep physics.

2 pointscentrality 2/5
  • Carroll describes two contrasting research philosophies for tackling deep physics puzzles: 'Radical Conservatism' (Wheeler's term)—keep the basic ingredients like the Schrödinger equation and statistical mechanics as solid as possible and derive better understanding from them—versus Smolin's willingness to discard fundamentals like the Schrödinger equation entirely; Smolin holds that both approaches are necessary.

    “my first impulse is always, I think what Wheeler called Radical Conservatism, like keep the basic ingredients as solidly as you can”

  • Smolin treats his guiding principles as provisional hypotheses rather than non-negotiable commitments—if pursuing them fails to lead anywhere, he would question the principles themselves—and their value, per David Finkelstein, is that thinking philosophically gives a 'running start' toward making progress.

    “Surely more the second.”

0.08

Functional and physical explanations both illuminate reality even within deterministic physical law.

3 pointscentrality 1/5
  • The Wheeler-DeWitt approach was frustrating because, although one could find solutions almost by inspection, quantum mechanics requires more than wave functions—it needs observables and an inner product—and in conventional loop quantum gravity they never honestly derived quantum-mechanical observables from classical ones nor constructed the physical inner product.

    “quantum mechanics is more than just the wave functions... What we never did in conventional loop quantum gravity was find quantum mechanical observables that were honestly derived... Nor did we ever construct the physical inner product.”

  • Even though the laws of physics are deterministic, functional explanations remain genuinely enlightening: the reason a particular protein (out of ~20^1000 possible) is prevalent in the biosphere is that it performs a function improving fitness—an explanation we cannot derive by brute computation, so we are blessed to have both functional and physical-law accounts that we can check against each other.

    “we are blessed by having both, that is, we can make a functional explanation for why those proteins exist. And we can also check, do they satisfy the laws of physics... at every stage... But we can't hope with current technology to just compute from the way it functions”

  • Many of the most thoughtful physicists—including Einstein, and per Dyson, Heisenberg and Schrödinger—have worried about the apparent disappearance of the 'now' or present moment from physical theory, with some arguing that quantum mechanics is really a description of the future and that confusions like the measurement problem stem from missing this.

    “these very thoughtful people have worried about the absence of the notion of the now in our physical theory”