I don't know whether the proper term is Physicalism or Physical Reductionism (it's Reductive Physicalism--thank you, Conifold), but what I'm referring to is the position that everything reduces to physics. My question is, for proponents of this doctrine (assuming there actually are any), which branch of physics does everything reduce to? Quantum Mechanics? Particle Physics?

The reason I ask is that not even all branches of physics reduce to one branch of physics. With a few exceptions, different physical theories interlock and agree with one another in a pleasing way that reassures the student that it isn't all just BS. For example, Newtonian mechanics can be seen as an approximation to relativistic mechanics for smallish amounts of energy. This is what you'd hope for, since a theory that gave different results from Newtonian mechanics in everyday circumstances isn't likely to be right. And it is also true that statistical mechanics, both the classical and quantum versions, give some insight into the results of thermodynamics by going down a level--the microscopic level of molecules.

However, it is not the case that thermodynamics reduces to statistical mechanics. It is its own separate discipline based on generalizations from macroscopic empirical observations, and its validity does not depend on a particular microscopic statistical interpretation. Even if some flaw is discovered in the statistical approach, it does not mean that the laws and results of thermodynamics are suddenly in doubt. The second law of thermodynamics doesn't go away just because nobody has figured out how to derive it as a theorem in statistical mechanics (That's probably not the most relevant example, since a lot of people are apparently satisfied that it has been, and it's not central to my point). Also, as Conifold pointed out in the comments below, there is still a lack of conceptual reduction of thermodynamics to statistical mechanics. Thermodynamic concepts have to be imposed on the microscopic picture, rather than falling out naturally.

So when you look at what physical theories are actually like, you see that they are a bit more like a quilt than a perfectly nested set of Russian dolls. Thus I'd like to know a bit more specifically what people have in mind when they say everything reduces to physics.

  • 1
    I think the idea is that the empirical laws we discover in physics will eventually be shown to be derivable from some fundamental theory. (At least as approximations, the way Newtonian gravity turns out to be an approximation to general relativity--wherever their predictions differ, general relativity is always found to be correct. Similarly I believe all the laws of thermodynamics are currently derivable from statistical mechanics, do you know any exceptions?) This idea of physical reductionism doesn't require that we be able to do that in practice yet given our current knowledge.
    – Hypnosifl
    Mar 18, 2020 at 3:26
  • 2
    The proper term is reductive physicalism. As to what kind, there is an ideal of the fundamental "theory of everything" that all other scientific theories (in principle) reduce to. Ideally, on the model of thermodynamics and statistical mechanics. It is understood that this may not be achievable in practice (actually reducing quantum chemistry to quantum mechanics is prohibitively complex, but few doubt that it is possible), and what the "theory of everything" might look like is left open to future discovery, although rough contours of some candidates are sketched.
    – Conifold
    Mar 18, 2020 at 3:59
  • 1
    The fact that a low-entropy state of an isolated system will tend to evolve into a higher-entropy state can be derived from statistical mechanics, though the overall fact that the entropy of the universe has been continually increasing--the "arrow of time" problem--is thought to depend on low-entropy boundary conditions near the Big Bang that aren't yet explained (but I think most physicists would bet that if we do get a final unified theory of fundamental physics and apply it to cosmology, it would explain this fact of low-entropy boundary conditions).
    – Hypnosifl
    Mar 18, 2020 at 6:16
  • 1
    The second law follows from ergodicity assuming low entropy initial state of the universe, see How do you prove the second law of thermodynamics from statistical mechanics? A "theory of everything" is normally expected to account only for the acting laws, not accidents of initial conditions and their consequences. Although there may be a yet unknown law that explains why the Big Bang produced a low entropy state.
    – Conifold
    Mar 18, 2020 at 7:38
  • 1
    I think practical limitations of it, especially for explanatory value, are generally acknowledged. It is more of a proof-of-concept type of aspiration that plays into a particular metaphysical view of how the world is. There is a deeper limitation, that many philosophers pointed out, that even in the prototypical case of thermodynamics there is no conceptual reduction. Thermodynamical concepts (entropy, enthalpy, pressure, etc.) do not jump out of SM. They have to be imposed by hand, and only then can mathematical reduction of the laws be effected. So "everything" isn't everything.
    – Conifold
    Mar 19, 2020 at 3:58

2 Answers 2


They did work out how to express the laws of thermodynamics using statistical mechanics. What you are talking about is unification in physics. Temperature for instance was discovered to be a macroscopic consequence of microscopic degrees of freedom. Shannon while working on information transfer showed entropy is also a result of information change in a system.

Maxwell's equations unified electrical and magnetic phenomena. Special Relativity arose integrating the single speed of light from Maxwell's equations with Newton's. Until electromagnetism, nuclear decay, and nuclear structures, all got neatly wrapped up in one structure, Quantum Field Theory. That is our best theory. And General Relativity is likely going to get integrated through looking at information flow.

John Wheeler made probably the strongest statement how that might look with his 'It From Bit' doctrine:

"Every it – every particle, every field of force, even the space-time continuum itself _ derives its function, its meaning, its very existence entirely... from the apparatus-elected answers to yes or no questions, binary choices, bits." - quoted from here which has far more info

On the other hand, Hawking concluded late in his career a truly unified theory of all physics could be impossible.

Given that structures as important for making predictions as are atoms, namely brains, are emergent we can interpret reductionism as fundamentally limited.

Abstraction and simplification, are simply how minds work. We have to make things tractable. I don't see unification as truly reductionism, but as translatability, interoperability between different modes of description. Very often something complex like a brain is irreducible in practice, it is vastly complex. But we have no reason to think it's made of more than atoms, and when we really 'understand' them we will have efficient abstractions (that's what understanding is). So I expect unification to reach quantum-gravity, probably with deeper insights into information that look like It-From-Bit. But that won't be the end of it anymore than understanding a cell explains the brain. But the things we experience are in one language, information, what Galileo meant when he said mathematics is the language of God. It is experience that unifies, by translating and connecting.

  • Thanks for the answer and the linked articles! Wheeler's claims are a little too radical for me to wrap my head around just yet. Perhaps with time. However, I have been convinced by the work of Landauer and Bennett that information plays a central role in the understanding of the 2nd Law. It is part of the reason why I'm not satisfied with the "proofs" of the 2nd Law from stat mech that I've looked at so far. But there are many of them, and I'm probably not in a position to assess all of them. Mar 20, 2020 at 4:33
  • Worth noting that although quantum field theory provides a general framework for expressing different forces as different quantum fields, it does not actually "unify" them in the sense meant by physicists--unification refers to the idea that high enough energies, what previously looked like two diff. field theories actually turn out to be aspects of a single quantum field. This sort of unification has already been achieved for the electromagnetic and weak nuclear forces, which unite as "electroweak theory", but so far neither the strong nuclear force or gravity have been united this way.
    – Hypnosifl
    Mar 20, 2020 at 17:18
  • @Hypnosifl: It's the electroweak force, described by Quantum Electro Dynamics. Unification with the strong force in an elctronuclear force hasn't been experimentally tested, but the models make sense and are ready to be tested. Gravity is in a different category, where we know only that it can't be unified using current quantum tools
    – CriglCragl
    Mar 21, 2020 at 2:15
  • My understanding is that quantum electrodynamics does not deal with the weak force, electroweak theory is a distinct quantum field theory (see the last paragraph of the 'history' section of the QED article which says QED served as a model for later quantum field theories including electroweak theory). And I think there are many proposed unifications of electroweak with strong (known as Grand Unified Theories or GUTs), there aren't strong theoretical arguments for thinking one specific proposal (if any) is most likely correct.
    – Hypnosifl
    Mar 22, 2020 at 19:36

When reductionists say that everything reduces to physics, they mean that everything can ultimately be explained by the fundamental laws of the physical world, whatever those are.

We happen to have at the moment a small set of independent but complementary theories which still have a few holes and sometimes conflict, but that is irrelevant. What matters is that the study of the physical world and the improvement of those theories is, ultimately, the only game in town.

  • 1
    I guess my main complaint is that the supposed ultimate explanation isn't much of an explanation for most things. To recycle an example I used earlier, what could the hoped for theory of everything tell you about the motion of a spinning top that you don't already know from the classical theory? Seems like it would take quite a bit of doing to get from the ToE even to the idea of an approximately rigid macroscopic body. Mar 21, 2020 at 21:01
  • 1
    Yes it does take quite a bit of doing. But to a reductionist, all that is a consequence of Nature's many layers of complex behaviour, each of which is an emergent property of stuff obeying the layer beneath. A top is an emergent property of its constituent atoms and electromagnetic fields, an atom of its constituent particles and the electroweak field, its nucleons of quarks and gluons. Kinematics is an emergent property of energetic physical objects, and so on and on. I do not accept it 100%, but it is the reductionist position. Mar 22, 2020 at 9:20
  • Depending on how emergent is interpreted, I'm a reductionist according to the account you gave. I would think most everyone is who does not believe in a separate Mind stuff. I accept something like physical determinism. I just don't think it follows that all facts reduce to the most basic physical facts in any meaningful sense. Actually I think most facts of interest to us don't reduce to physics, even though in some (useless) sense everything is physically determined. So it seems that reductionists must mean something extra, or they just aren't being careful with their words. Mar 22, 2020 at 20:21

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .