Physics Stack Exchange is a question and answer site for active researchers, academics and students of physics. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

(Transferred as a separate question from comments in Scott Aaronson’s gravitational decoherence question) Reversing gravitational decoherence

The modern answer seems to be that they never occur, and that therefore nothing causes them. This leads on (or back) to Everett and MWI. A closely related idea is that there is only one collapse, just as there is only one wave function, but it covers the whole universe, all time and all space.

An older answer, at least sometimes called Copenhagen, is that collapses are merely Bayesian updating, and hence do not occur as physical processes.

A third answer, associated with Wigner and von Neumann himself, is that the observer causes collapses by the act of observation.

GRW suggest that collapses occur gradually, due to a nonlinear modification of the Schrodinger equation.

I do not find these answers satisfactory. See related question, “Can you count collapses?” I am interested in answers that do not modify QM, (unlike GRW), but (like GRW) nevertheless regard collapse as a real physical process that can be observed, and counted.
Can someone provide references to papers that discuss collapses that meet this criterion?
Are there other answers to the collapse question not mentioned above?

share|cite|improve this question
Wigner should not be considered separate from MWI, Wigner was extracting the essential aspect of Everett's interpretation and rewriting it in another way, more palatable to the philosophical prejudices of his time. Regarding Copenhagen and "updating", I don't think anyone considered the updating as Bayesian, since it is not a probability distribution, it was the analog of Baysian updating in quantum mechanics. – Ron Maimon Aug 28 '12 at 9:13
Actually a lot of people I know argue exactly like that, stating that the collapse is nothing but a bayesian update. I couldn't disagree more, but it's certainly not an uncommon thought. – A.O.Tell Oct 11 '12 at 9:53

I think you might find a series of blog posts that I wrote recently useful. They also point to a paper that is available on arxiv and is currently in the review pipeline. See

The posts specifically discuss how the collapse postulate with the Born rule can emerge from unitary quantum mechanics without the need of additional postulates.

share|cite|improve this answer
I have taken a brief look at your paper and intend to read it more fully, it is interesting. How do you deal with the nonlocality problem? Say if you have a Bell state and one observer is doing the measurement, how do you ensure correlated outcomes and collapse and still maintain local unitary evolution? – SMeznaric Oct 11 '12 at 1:56
Thank you for your interest. All I do is unitary evolution, so from a global perspective while evolution happens locally, the state space is nonlocal. So even a local dynamic change can reflect in remotely entangled part of the state. This is no different from, say, Copenhagen. So my work accepts the structural non-locality as a fact, and lives with the consequences. If you would want to change that you'd have to think about an alternative for the quantum state space construction. – A.O.Tell Oct 11 '12 at 9:51

I've explained this here.

Unitarity is a relative attribute of physical evolution. Outside observers always see evolution of interaction between separate systems as unitary, but when the observer itself belongs to one of the separate systems, they perceive non-unitary projection. This should not be surprising, since the observer is also a quantum system, and because of that, it can become entangled, and become a superposition

share|cite|improve this answer
While it is true that subjectively entanglement can lead to an apparent collapse of the quantum state, it's not quite clear that this really leads to the right probabilities. In fact, you have to add extra assumptions to make the Born rule come out, and these assumptions are not generally accepted and it's not clear if they're even consistent. And while the entanglement argument is quite clear for simple model systems, it has some issues as branch counting and subjective reality depend on the level of coarse graining. – A.O.Tell Oct 11 '12 at 9:56

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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