Is there any quantum interpretation which isn't "crazy" at all? Exponentially many parallel worlds in MWI, superdeterministic conspiracies, and/or nonlocality in hidden variables, idealism and the idea measurements and observations create the outcome in Copenhagen, retrocausality, negative probabilities which can never be observed, etc. .

Is Nature really "stark raving mad"?

  • 4
    $\begingroup$ The "shut up and calculate" interpretation. $\endgroup$ – MBN Sep 14 '12 at 16:58
  • 4
    $\begingroup$ Unless you have a precise definition of "crazy," I don't think this is a constructive question... $\endgroup$ – David Z Sep 14 '12 at 19:28
  • 4
    $\begingroup$ @MBN: calling "shut up and calculate" an interpretation is like calling baldness a hair colour. $\endgroup$ – Niel de Beaudrap Sep 14 '12 at 20:32
  • $\begingroup$ I agree with MBN. QM needs no interpretation it works (my opinion :-P). And the way the question is formulated it looks rather like a complaint about the fact that nature applies QM ... $\endgroup$ – Dilaton Sep 14 '12 at 23:55
  • $\begingroup$ Neil, if you were bald and someone asked what you hair color was, what would you answer? I am guessing "i am bald". $\endgroup$ – MBN Sep 15 '12 at 10:53

Is Nature really "stark raving mad"? No.

Nature is to a large extend rational and intelligible without any craziness.

Quantum physics tends to generate a sense of mystery,
- perhaps for historical reasons,
- perhaps to generate the interest of laymen in physics (''quantum teleportation'' simply sounds much more impressive than ''copying the state of a photon''),
- perhaps because emphasis in layman accounts of quantum mechanics is put on thought experiments rather than real experiments,
- perhaps because it is too often poorly explained.

But there is nothing mysterious about quantum mechanics if it is understood in the way it is actually practiced - rather than in the way it is customarily talked about.

See the entries ''Quantum mechanics without mysteries'' and ''Foundations independent of measurements'' of Chapter A4: ''The interpretation of quantum mechanics'' of my theoretical physics FAQ at


There is one interpretation that has the roots of the right answer and that is the Consistent Histories interpretation. As explained this is a generalization of the Copenhagen Interpretation and removes what many people refer to as the "measurement problem" and replaces the classical notion of measurement, where a classical "apparatus" causes a wave function collapse of a quantum system, with a process called "decoherence".

In this interpretation the wave function never "collapses", however certain states, called "pointer states", which is an allusion to the idea of a needle of a gauge pointing to a particular value, begin to be preferred by the system. These pointer states are analogous to classical states (although they are arguably not equivalent in order to remain consistent with the notion that classical states are effectively a pure fiction in quantum mechanics).

The key conceptual leap is to understand the classical "apparatus" (e.g. measuring device) is really another quantum system where there is sufficient convolution (for lack of a better word) with the environment so it has become more entangled with the environment and has "decohered more" relative to the nearly pure quantum system that is to be "observed".

One of the key concepts to explore and understand further in this context is that of "separability" which is a measure of entanglement between states, particularly "pure states" (or nearly pure mixed states) which are maximally entangled internally but are independent (separable) of other pure states.

This interpretative approach is arguably the most correct, and also lends itself to the idea of mutual exclusive outcomes, where an outcome of an observation must be consistent and definite with respect to a particular pointer position, meaning the pointer of a compass can point north OR east but not north AND east when its observed with respect to the system, although prior to observation, its probability amplitude can evolve in an entangled state where there is a complex phase of north superposed with east...this miracle is achieved mathematically by using complex numbers and their conjugates and enforcing conditions of orthogonality (or more specifically orthonormality) and unitarity.

  • $\begingroup$ How do you propose that decoherence yields mixed states for marginals, which are meant to be probability distributions over pointer states, when extending the system to include environmental degrees of freedom may purify it? How does this differ from pure states on macroscopic systems converging to the subspace in which their Schmidt vectors are pointer states, but still exhibit long-range entrainment in principle (essentially the MWI)? $\endgroup$ – Niel de Beaudrap Sep 14 '12 at 20:36
  • $\begingroup$ Hal, according to you, is an electron in several places at once, or not? $\endgroup$ – Mitchell Porter Sep 14 '12 at 23:45
  • $\begingroup$ My thoughts are that the mechanism for emergence of classical distributions in phase space is the reduction of uncertainty as quantum mutual information with the increase in the number of potentially pure subsystems. Classicality emerges in the form of the liouville equation in the limit of h. There is intrinsic uncertainty because of noncommuting observables one could think in terms of uncertainty flows/currents as one's focus changes. As far as long range entrainment, all observations are local and only admit consistent states. $\endgroup$ – user11547 Sep 14 '12 at 23:57
  • $\begingroup$ @MitchellPorter, the simple answer is no. The classical entity referred to as an electron is not in multiple locations at once. In fact, that is the whole point of consistent histories and decoherence. First, we are talking about observing events that are grouped with properties associated with an electron. The wave function is simply a probability amplitude function and has no real existence in a classical sense. We will only ever observe the electron in one place if we measure its position. However, the wave function tracks all superpositions of potential observations in its evolution $\endgroup$ – user11547 Sep 15 '12 at 0:04

No. The main problem with qantum mechanics is that it predicts different probabilities of the same events for different observers.

So there can be essentially only two solutions to the problem, both having philosophical disadvantages.

  1. Postulate that there is only one true observer.
  2. Postulate that once the observers measure inconsistent results, the universe splits so that the both observe different universes of their own.

From the philosophical point of view the first idea renders scientific method potentially incorrect depending whether a scientist reporting a measurement is that distinguished observer (or in a thermodynamic contact with him), or not. Since most observers reside on Earth and in thermodynamic contact with each other, this did not bring much problems so far. But in a setting like in Wigner's friend paradox this would lead to a trouble.

The second idea renders scientific method inapplicable because if scientists split into parallel universes after a measurement, they would be unable to exchange results.

  • $\begingroup$ While I agree with this in terms of formalism, there are lots of different philosophical positions regarding what the "ontological status" of the different splitting states are, and so it is best to not be dogmatic, as the different philosophies are not positivistically distinguishable from one another. $\endgroup$ – Ron Maimon Sep 14 '12 at 19:03
  • $\begingroup$ @Cristi Stoica the wave function of an observed system can differ for different observers depending on whether they are entangled with the system or its parts. But different wave functions mean different probabilities, and different probabilities produce contradicting results. $\endgroup$ – Anixx Sep 14 '12 at 19:20
  • $\begingroup$ @CristiStoica: You can't measure incompatible observables on the same system, you need two different systems, and hence two different observables. $\endgroup$ – Ron Maimon Sep 14 '12 at 19:28
  • $\begingroup$ @Cristi Stoica I was talking about the same system. $\endgroup$ – Anixx Sep 14 '12 at 20:30
  • $\begingroup$ @CristiStoica: He meant different ways an observer can choose to measure a system, I think it's clear from context. One has to be charitable in interpreting one another to avoid arguing over these little things forever. $\endgroup$ – Ron Maimon Sep 14 '12 at 20:31

In http://arxiv.org/abs/1111.4630 (see references to my peer-reviewed articles there), I give a tentative positive answer to the following question: "Is it possible to offer a ”no drama” quantum theory? Something as simple (in principle) as classical electrodynamics - a local realistic theory described by a system of partial differential equations in 3+1 dimensions, but reproducing unitary evolution of quantum theory in the configuration space?"

  • 2
    $\begingroup$ @akhmeteli: They can be embedded, but your approach will not describe entangled state. For example, the ground state of Helium in a non-entangled quantum mechanics is 30% larger. -1 for nonsense. It is proved that such approaches don't work, using Bell's theorem, but for this specific approach, the proof is much easier, using any entangled wavefunction, and the experimental refutation is immediate. $\endgroup$ – Ron Maimon Sep 14 '12 at 19:02
  • 1
    $\begingroup$ @Ron Maimon: I respectfully disagree. This approach does describe entangled states in some sense, namely, projections of the functional coherent states in the Fock space onto subspaces of the Fock space with a definite number of particles, for example, two-particle subspace, can be entangled. If you disagree, please give your arguments. As for the Bell theorem, I discuss this issue in detail in my arxiv review (Section V), as well as in a peer-reviewed article. If you have specific objections to my reasoning, please let me know. Otherwise I'll consider your critique baseless. $\endgroup$ – akhmeteli Sep 14 '12 at 19:17
  • 2
    $\begingroup$ @akhmeteli: It's not something you can disagree with, and please disagree disrespectfully, there is no other way. I will not even read your argument, as it is obviously wrong. Please think about it more. $\endgroup$ – Ron Maimon Sep 14 '12 at 19:29
  • 1
    $\begingroup$ @Ron Maimon: As you don't want to "even read" my properly published arguments, I respectfully reject your critique as baseless and don't want to "even read" your comments unless you reconsider your position. $\endgroup$ – akhmeteli Sep 14 '12 at 19:36
  • 3
    $\begingroup$ @akhmeteli: My critique is not baseless, as you are wrong. I will not read things which claim 2+2=5. Your arguments should not have been published, this was a lapse on the part of the journal editors, hopefully not to be repeated. $\endgroup$ – Ron Maimon Sep 14 '12 at 19:48

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