In some interpretations of Quantum mechanics (e.g. transactional interpretation), the future affects present. Is this a source of unpredictability in such interpretations, which makes them have the same predictions as the orthodox Quantum mechanics?

I mean, if the future affects the present, then will it be impossible to predict the future by having all the information about the present state? and is it this that makes retro-causal interpretations indeterministic (unpredictable), with results identical to the Copenhagen interpretation? In other words, does the probability stem from out attempt to predict the future merely by the present state, while this is not the whole picture?


Your question is interesting and challenging. However, the way you put it, it cannot be answered.

Let me give an example wrt determinism. We could try to define determinism as follows: Suppose we know all the "input" into an experiment we are able to calculate the "output". So we might ask: Is nature deterministic?

The problem we run into is that the question is not accessible for experimentation.

But could we not just make an experiment twice and see whether the output is the same?

Well, the problem is, we never can do an experiment twice, as already the notion of "twice" includes the fact that we can distinguish the first from the second experiment.

So...we have to start to make model assumptions.

Of course, in the model assumptions we can draw mathematical conclusions. So, in a deterministic deBroglie-Bohm-type of theory you might come to the conclusion that nature is deterministic. In a Copenhagen-type of interpretation you will conclude that nature is not deterministic. In a Feynman-path-integral type of interpretation you might even look at paths in space-time where you exceed the speed of light or run from future into the past.

So, at least inside of an interpretation we can provide some answers (but it does not have very much to do with nature). So what about the transactional interpretation? Is it "X" that make it "Y".

Now, we again face a problem. We could, for example, ask ourselves, if a proposition is true in a theory, but we cannot say, what makes it true. For example: Does 2+2=4 make 2+3=5 be true? We cannot answer this. It is true (or it is not true). We have a chance of pushing one part of what we want a theory to be into the axioms and then derive the rest. But still, it is not the axioms which "cause" the propositions to be true but rather the intention of the scientist.

Thus, it is a bit problematic to answer.


Causality in QM is probabilistic, thus you should expect some degree of unpredictability. Retrocausality in turn, does not imply total unpredictability. So there is no contradiction here.

  • $\begingroup$ How would you distinguish causality from correlation in QM? $\endgroup$ – Nobody-Knows-I-am-a-Dog Nov 14 '18 at 19:18
  • $\begingroup$ I think that the answer depends on the interpretation of QM that you choose. But the definition of causality is tricky, and not really well defined , see en.wikipedia.org/wiki/Causality $\endgroup$ – Wolphram jonny Nov 14 '18 at 19:22
  • $\begingroup$ I would settle for an interventional notion of causality as introduced by Judea Pearl for the purposes of this discussion. I do not understand how Pearl's calculus (developed for a Kolmogorovian type of probability) could be transported to QM / lattice type of probability. $\endgroup$ – Nobody-Knows-I-am-a-Dog Nov 14 '18 at 23:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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