This new finding by Minev et al. seems to suggest that transitions between atomic states are not instantaneous, but continuous processes wherein a superposition smoothly adjusts from favoring one state to another (if I understand it correctly). The authors also claim to be able to catch a system "mid-jump" and reverse it. Popular articles are here and *here.

I am curious if this finding rules out any interpretations of QM. It seems to generally go against the Copenhagen attitude, which describes measurements as collapsing physical systems into a definite classical state. The popular articles indeed claim that the founders of QM would have been surprised by the new finding.

The link with the asterisk mentions that something called "quantum trajectories theory" predicts what was observed. Is this an interpretation, or a theory? And are they implying that other interpretations/theories don't work?


No. All news stories about this result are extremely misleading.

The "quantum jump" paper demonstrates an interesting and novel experimental technique. However, it says absolutely nothing about the interpretation of quantum mechanics. It agrees with all proper interpretations, including the Copenhagen interpretation.

What the researchers actually did

When a quantum system transitions between two states, say $|0 \rangle$ to $|1 \rangle$, the full time-dependence of the quantum state looks like $$|\psi(t) \rangle = c_0(t) |0 \rangle + c_1(t) |1 \rangle.$$ The amplitude $c_0(t)$ to be in $|0 \rangle$ smoothly and gradually decreases, while the amplitude $c_1(t)$ to be in $|1 \rangle$ smoothly and gradually increases. You can read this off right from the Schrodinger equation, and it has been known for a hundred years. It is completely standard textbook material. The researches essentially observed this amplitude changing in the middle of a transition, in a context where nobody had done so before.

The authors themselves emphasize in their paper that what they found is in complete agreement with standard quantum mechanics. Yet countless news articles are describing the paper as a refutation of "quantum jumps", which proves the Copenhagen interpretation wrong and Bohmian mechanics right. Absolutely nothing about this is true.

Why all news articles got it wrong

The core problem is that popsci starts from a notion of "quantum jumps", which itself is wrong. As the popular articles and books would have it, quantum mechanics is just like classical mechanics, but particles can mysteriously, randomly, and instantly teleport around. Quantum mechanics says no such thing. This story is just a crutch to help explain how quantum particles can behave differently from classical ones, and a rather poor one at that. (I try to give some better intuition here.) No physicist actually believes that quantum jumps in this sense are a thing. The experiment indeed shows this picture is wrong, but so do thousands of existing experiments.

The reason that even good popsci outlets used this crutch is two-fold. First off, the founders of quantum mechanics really did have a notion of quantum jumps. However, they were talking about something different: the fact that there is no quantum state "in between" $|0 \rangle$ and $|1 \rangle$ (which, e.g. could be atomic energy levels) such as $|1/2 \rangle$. The interpolating states are just superpositions of $|0 \rangle$ and $|1 \rangle$. This is standard textbook material: the states are discrete, but the time evolution is continuous because the coefficients $c_0(t)/c_1(t)$ can vary continuously. But the distinction is rarely made in popsci.

(To be fair, there was an incredibly short period in the tumultuous beginning of "old quantum theory" where some people did think of quantum transitions as discontinuous. However, that view has been irrelevant for a century. Not every early quote from the founders of QM should be taken seriously; we know better now.)

Second off, the original press release from the research group had the same language about quantum jumps. Now, I understand what they were trying to do. They wanted to give their paper, about a rather technical aspect of experimental measurement, a compelling narrative. And they didn't say anything technically wrong in their press release. But they should've known that their framing was basically begging to be misinterpreted to make their work look more revolutionary than it actually is.

Interpretations of quantum mechanics

There's a very naive interpretation of quantum mechanics, which I'll call "dumb Copenhagen". In dumb Copenhagen, everything evolves nicely by the Schrodinger equation, but when any atomic-scale system interacts with any larger system, its state instantly "collapses". This experiment indeed contradicts dumb Copenhagen, but it's far from the first to; physicists have known that dumb Copenhagen doesn't work for 50 years. (To be fair, it is used as a crutch in introductory textbooks to avoid having to say too much about the measurement process.) We know the process of measurement is intimately tied to decoherence, which is perfectly continuous. Copenhagen and, say, many worlds just differ on how to treat branches of a superposition that have completely decohered.

Another issue is that proponents of Bohmian mechanics seem to latch onto every new experimental result and call it a proof that their interpretation alone is right, even when it's perfectly compatible with standard QM. To physicists, Bohmian mechanics is a series of ugly and complicated hacks, about ten times as bad as the ether, which is why it took last place in a poll of researchers working in quantum foundations. But many others really like it. For instance, philosophers who prefer realist interpretations of quantum mechanics love it because it lets them say that quantum mechanics is "really" classical mechanics underneath (which actually isn't true even in Bohmian mechanics), and hence avoid grappling with the implications of QM proper. (I rant about this a little more here.)

Quantum mechanics is one of the most robust and successful frameworks we have ever devised. If you hear any news article saying that something fundamental about our understanding of it has changed, there is a 99.9% chance it's wrong. Don't believe everything you read!

  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$ – Chris Jun 9 '19 at 9:21
  • 1
    $\begingroup$ Nice answer, but this answer points out a subtle blame on the Yale group, when you say "They wanted to give their paper, about a rather technical aspect of experimental measurement, a compelling narrative." I've seen this happen elsewhere, with more vigorous disapproval (e.g. Lubos Motl's blog). But no one seems to care that they are a quantum computing group, and they care about quantum feedback control (monitor and compensate decoherence). Also, they never mention Bohmian mechanics, or the press did them a great disfavor. $\endgroup$ – wcc Jun 27 '19 at 18:18
  • $\begingroup$ They do mention quantum trajectories calculation, and this has nothing to do with Bohmian mechanics. It is a way of calculating the evolution of an open system, and there are situations where it is more convenient to use (and still accurate) than master equation - it has found nice applications in simulating laser cooling. $\endgroup$ – wcc Jun 27 '19 at 18:19
  • $\begingroup$ @wcc As I said, the authors are completely in the right! It has merely been misinterpreted by everybody else. You can see what I mean even in the comments on this question. $\endgroup$ – knzhou Jun 27 '19 at 18:23
  • 1
    $\begingroup$ Could you clarify what you have in mind for the "smart Copenhagen" interpretation? $\endgroup$ – tparker Nov 3 '19 at 5:02

Yes, very much so.

Quantum theory is established on the two von Neumann postulates: (1) Evolution (that a quantum state evolves in time in accordance with the Schrödinger Equation) and (2) Projection (that the result of measuring a quantity in a quantum state is one of the eigenvalues of that quantity's operator, with the state transforming into the corresponding eigenstate -- all in accordance with the Born Rule). That's the "quantum jump".

There are different interpretations of quantum theory; and they all rest on how they treat, interpret, explain or explain away the Born Rule. Each interpretation may yield different implications on the question of what/when/where the Born Rule occurs (or what/when/where occurs whatever is meant to replace the Born Rule occurs). This is particularly the case for objective collapse versus no-collapse interpretations.

The experiment provides a hook into what may become a general method for probing a Born Rule collapse even in mid-process, at different stages, and possibly even reconstructing a dynamics for it. This will greatly clarify the issue of which interpretation is the correct one.

Two of the most important things that come out of it are the ability to anticipate a collapse and the appearance of what looks like a reversibility-threshold beyond which a Born Rule event is irreversible.

One additional benefit that may come out of mid-collapse probing experiments is that it may clarify the issue sufficiently well to allow one to answer the (still largely-unanswered) question of what the Heisenberg Picture of the Born Rule (or of the alternatives meant to replace it) is? Or, more generally: what the Heisenberg Picture of the different interpretations is. These is a substantial gap in the literature on these two questions!

  • 8
    $\begingroup$ This is what I was referring to as “dumb Copenhagen”, which effectively zero physicists who have thought about measurement believe. $\endgroup$ – knzhou Jun 9 '19 at 5:51
  • $\begingroup$ agree with RB. the new findings are essentially anti-copenhagen. think it is historically inaccurate and intellectually disingenuous to invent a concept of "dumb copenhagen". aka the moving the goalposts fallacy. there is only a copenhagen interpretation that was given by bohr + heisenberg although possibly the latter explicitly rejected calling it the "copenhagen interpretation". this interpretation basically rejects a "deeper reality" other than the QM formalism. new experiments such as this one show unequivocally there is a subquantum realm as specifically sketched/ outlined by Bohm. $\endgroup$ – vzn Jul 21 '19 at 2:40

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.