I want to see if any of the alternative theories/interpretations of QM have any merit to them(Pilot wave, superdeterminism, etc). I think the motivations behind these theories include(I don't think all of those are very good motivation but still.):

  1. Eliminating fundamental randomness as in Born rule
  2. Eliminating uncertainty, superpositions non-commuting observables
  3. Making physical particles 'imaginable' again. (relates to 2nd point)
  4. Finding a 'deeper structure' behind QM where the principles appear less aribitrary.

Bell's theorem does a good job of restricting the possible alternative theories. It says if you want a hidden variable theory you have to make it non local(e.g. pilot wave). superdeterministic(particle 'knows' that it only has to carry the information that you will want to measure.) or retrocausality...(?)

I can find a lot of back and forth on how good/ungood retrocausality is, but nowhere clearly stated what perceived or real problems in QM it aims to solve and at what cost(i.e. what is the theory?).

So how exactly retrocausal QM looks like?

I guess something goes back in time to affect the past. But what? Why? How will RC diminish on the macroscopic level? Or will it? What is the reason some people believe the promise that it will become a simpler yet still predictive physics?

Is it just empty words, or is there some outline of a theory? In case there is something, please point us to an introduction that answers the above questions or walk us through a known QM thought experiment with the retrocausality mindset and show what may be the merit to it.

  • $\begingroup$ Unitary evolution is time-reversible, so if your interpretation of QM is deterministic then "retrocausality" means the same thing as "causality" --- the future causes the past in exactly the same sense that the past causes the future. $\endgroup$
    – WillO
    Nov 23, 2019 at 22:14
  • $\begingroup$ You have a good point. But the notion of "something going back in time and causes something in the past" is so very vague to me that I cannot exclude that someone has a good case for some theory like that. I think when you say unitary evolution, you restrict time to be a single real-valued global parameter. It is mostly clear that time indeed can not do any funny business from that limited role. $\endgroup$ Nov 23, 2019 at 23:02
  • $\begingroup$ I'd like to know if there are serious ideas where important parts of QM are replaced by something else that resembles retrocausality(even if it is just some subjective retrocausality for certain observers, maybe), or is all of it just empty words and 'interpretations'? $\endgroup$ Nov 23, 2019 at 23:07
  • 2
    $\begingroup$ the notion of "something going back in time and causes something in the past" is so very vague to me.... I'd contend that the notion of "something going forward in time and causes something in the future" is exactly equally vague. $\endgroup$
    – WillO
    Nov 23, 2019 at 23:14
  • $\begingroup$ Sure but we have a conceptually precise(non-vague) theory(QM,QFT) where everything goes forward in time(even antiparticles really) and it is not vague anymore. But is there a good argument out there to give 'time' a very different role than we currently give? $\endgroup$ Nov 23, 2019 at 23:26

2 Answers 2


It is important to understand that in this context the term "retrocausal" is referring to a modern account of causation with which many physicists are still unfamiliar. In this "interventionist" account of causation, a "cause" is simply an external intervention on a system, such as a controllable setting in an experiment. (See, for example, Judea Pearl: https://www.amazon.com/gp/product/052189560X ). Retrocausal models are simply those for which future externally imposed settings are correlated with past model parameters. These models allow hidden parameters to be correlated with future settings -- say, some parameter hiding in the true state of two entangled particles that is correlated with the future measurement settings on those particles. It is trivial to use such parameters to explain Bell correlations if you consider models of this sort; they just normally are not considered.

Explaining how such correlations are generated in a consistent manner would probably have to use a Lagrangian/Action-style analysis, where the whole history is solved "all at once", constrained in part by future boundary conditions (as is normally done in action extremization). It's probably a mistake to think of things "flowing" or "going" forward or backward in time; instead, think in terms of static spacetime-diagrams.

There is currently no retrocausal model of all entanglement phenomena, but there is also no reason why one could not be developed. The best current models can handle two maximally-entangled qubits (the case normally discussed with regards to Bell's theorem). Some examples, and related models are discussed here: https://arxiv.org/abs/1906.04313 .


Employee 1223,

The main motivation behind the search for deeper theories behind QM is the preservation of locality (no physical effect can propagate faster than the speed of light). Locality appears to be a fundamental physical principle. It is a consequence of Einstein's theory of relativity, so Einstein was the first to notice that if QM is assumed to be a complete physical theory (no additional variable are added) it must be non-local. The argument is presented in this paper:

Can Quantum-Mechanical Description of Physical Reality Be Considered Complete? A. Einstein, B. Podolsky, and N. Rosen, Phys. Rev. 47, 777 https://journals.aps.org/pr/abstract/10.1103/PhysRev.47.777

Central to the argument is the EPR (from Einstein, Podolsky, Rosen) reality criterion:

"If, without in any way disturbing a system, we can predict with certainty (i.e., with probability equal to unity) the value of a physical quantity, then there exists an element of reality corresponding to that quantity."

In QM is it possible to prepare a pair of particles (by splitting a diatomic molecule for example) so that the result of a spin measurement on particle A is always opposite to the result of a spin measurement on particle B (as long as the measurements are performed on the same direction). This phenomenon is called entanglement.

Let's formulate the argument in the context of such an experiment with spin 1/2 particles where the measurements are performed in such a way that a light signal cannot get from A to B (they are performed at the same time, at a large distance from one another):

  1. It is possible to predict with certainty the spin of particle B by measuring particle A (QM prediction - the results are always opposite).

  2. The measurement of particle A does not disturb particle B (locality).

  3. From 1 and 2 it follows that the state of particle B after A is measured is the same as the one before A is measured (definition of the word "disturb")

  4. After A is measured B is in a state of defined spin (QM prediction)

  5. From 3 and 4 it follows that B was in state of defined spin all along.

  6. The spin of A is always found to be opposite from the spin of B (QM prediction)

  7. From 5 and 6 it follows that A was in a state of defined spin all along.

Conclusion: QM + locality implies that the the true state of A and B was a state of defined spin. The superposed, entangled state is a consequence of our lack of knowledge in regard to the true state. So, QM is either an incomplete (statistical) description of a local deterministic hidden variable theory or it is non-local.

So, the four "motivations" you list are actually consequences of the assumption of locality. The reason Einstein opposed Bohr's "Copenhagen" interpretation was not his desire to make the theory deterministic, but local.

Let's take a look at Bell's theorem now.

Bell devised an experiment to check which of the remaining options after EPR is true. Do we live in a non-local world (which can be either deterministic or non-deterministic) or in a local and deterministic one?

Now, there are two assumptions in Bell's theorem. One is locality, the other is independence (the hidden variables do not depend on the measurements' settings).

The theorem says:

No physical theory that fulfills the above assumptions can reproduce QM's predictions.

Many experimental tests have been performed and each time the QM's prediction was confirmed, so we either can accept that physics is non-local, or that the independence assumption is false. The choice should be easy now. As stated above, locality is a central physical principle. Independence is not. In fact, it is trivial to find distant non-independent systems, two stars orbiting each other being such an example. So, after Bell, superdeterminism results as a consequence of maintaining that QM is correct and locality holds.

Just like non-locality, retrocausality has never been observed to take place in an experiment and no known theory implies it. So, I think that the main motivation behind the introduction of this concept is a wrong understanding of EPR and Bell arguments. Once you realize that superdeterminism does not require any departure from any known physical principle it makes no sense to go for other options that do make such requirements.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.