Paper Review: Experimental Rejection of Observer-independence in the Quantum World

The MIT Technology Review recently published an article entitled “A quantum experiment suggests there’s no such thing as objective reality“. This sounds rather spicy: if this were true, and there were no objective reality, what would that mean? Would the postmodernists be right? Would it all be purely subjective? How about claims like “there is no objective reality”–is that an objective fact? If not, how does this all hang together?

Maybe you think this is a case of the media blowing something out of proportion, or getting the science wrong. Surely the title of the original paper is something esoteric and boring like “K-alpha particles exhibit decoherence along mesa-keta spectra at 196mmz intervals” or some other Star Trek made up sounding jargon. But then you look at the title of the paper:

“Experimental rejection of observer-independence in the quantum world“

Uh-oh. This sounds like a (rare?) case where maybe the media are actually reporting accurately on the science—the title would certainly suggest that the scientists believe their experiment rejects objective reality, or something like that. Maybe this is just the scientists selling the paper? So then you read the abstract:

“The scientific method relies on facts, established through repeated measurements and agreed upon universally, independently of who observed them. In quantum mechanics, the objectivity of observations is not so clear, most dramatically exposed in Eugene Wigner’s eponymous thought experiment where two observers can experience fundamentally different realities. While observer-independence has long remained inaccessible to empirical investigation, recent no-go-theorems construct an extended Wigner’s friend scenario with four entangled observers that allows us to put it to the test. In a state-of-the-art 6-photon experiment, we here realise this extended Wigner’s friend scenario, experimentally violating the associated Bell-type inequality by 5 standard deviations. This result lends considerable strength to interpretations of quantum theory already set in an observer-dependent framework and demands for revision of those which are not.”

Hmm. Well. The scientists themselves do in fact seem to be claiming that their experiment gives strong support for an observer-dependent, entirely subjective reality. Even if we wanted to read charitably, and think that this kind of subjectivity was something less exotic and threatening to science, like the kind of observer dependent properties of relativity , we reread the first line of the abstract. The kind of objective facts the authors seem to have in mind are the kind that make science possible. This suggests a much deeper rejection of an objective world than the kind found in relativity (which does not in fact reject an object world, but merely has no privileged reference frame).

Let us take a closer look at the actual text of the paper, then. Throughout my analysis I will also reference the latest entry of the “Philosopher’s On” series of the philosophy site Daily Nous. I’ll link to it a few more times throughout the paper. I’d recommend anyone reading my post, which in some ways takes a bit of a deeper dive into the content of the paper, to read the Daily Nous post after. It is very helpful and enlightening.

When talking about the experiment (which I will discuss in a minute) the scientists write, suggestively, “Can one reconcile their different records, or are they fundamentally incompatible—so that they cannot be considered objective, observer-independent ‘facts of the world’?” (p. 2). They will claim later in the paper that the answer is likely “no”. Indeed, they write in their conclusion:

“Modulo the potential loopholes and accepting the photons’ status as observers, the violation of inequality (2) implies that at least one of the three assumptions of free choice, locality, and observer-independent facts must fail…[one] option is to give up observer independence completely by considering facts only relative to observers…This choice, however, requires
us to embrace the possibility that different observers irreconcilably disagree about what happened in an experiment.” (p. 4)

The sections over which I elided describe the other options; options which I will later argue are better. However, it is clear that the conclusion these scientists prefer is to give up an objective, observer independent world. Why is this?

To understand their argument, first we should understand a few things from the previous excerpt: what exactly was the experiment? What are the three conditions–free choice, locality, and observer-independent facts?

The experiment they conducted was a modification of the famous Wigner’s Friend thought experiment.

Wigner asks us to consider a physical system prepared in a superposition. For example, the system might be an electron, and the electron might be in a superposition of being spin-up and spin-down. What exactly is a superposition? I think an example might be helpful: consider the case in which we are interested in the location of an electron. In particular, we want to know if it in a certain room, or outside the room. We can represent the particle and its room-position (which is a binary observable with two possible values: in the room and out of the room) using the mathematics of quantum mechanics. There is a rule in quantum mechanics called the eigenstate-eigenvalue link (see here for an introduction and history of the rule), that says that the particle has a room-position if and only if the mathematical vector describing the physical system is in an eigenstate of the mathematical object describing room-position observable. This sounds technical, so let’s unpack it a little. Suppose we have a clock:

We can think of the clock as being a Hilbert Space — mathematical object physicists use to describe the state of a physical system. How do we represent whether the particle is in or out of the room? We represent such a property using two vectors at right angles to each other. For example, the two vectors very sketchily drawn here:

The vertical line might represent being in the room, and the horizontal one out of the room. Suppose the state of the particle is represented by the minute hand. The eigenstate-eigenvalue link says that the particle has a position if and only if the minute hand lies on one of the blue lines. In this case, since the minute hand does not lie on one of the blue lines, the particle does not have a position! In this case, we say that the particle is in a superposition of being in the room and out of the room. What is important to note is that even though the particle does not have a definite position, it does have a state–the one represented by the minute hand. In principle, we could design a measuring device to see whether that particle is in that state. This means we can measure whether or not the particle is in a superposition.

Thus, like in the above picture, it can be false that the particle is either in the room or out of the room. Note this is not a contradiction. It is true that the particle is not in the room, and it is not out of the room. However, these are not the only two possibilities. Instead, as I said above, we describe the particle as being in a superposition of being in the room and out of the room. This might be rather shocking, since it seems to defy classical logic. However, it does not. It is rather analogous to a case in which someone believes that every object is either black or white. Thus, for a person with such a conception of colour, the statement “every object is either black or white” seems tautologically true. However, one day this person encounters a red rose. For this person, this might initially feel like a violation of logic. However, upon reflection, she comes to the realization that instead of violating logic, this observation instead demonstrated that she had had an incorrect theory of the possible colours an object might be. The case is the same in quantum mechanics. The electron lacking a position does not defy logic; we rather have to enlarge our initially impoverished space of possible properties.

So, for the Wigner’s Friend thought experiment, we have a particle in a superposition of some property–say, X-spin. Furthermore, suppose we have an observer–Wigner. Since Wigner is a physical system, we can describe him using the mathematics of quantum mechanics. Suppose the friend is to measure the particle, and then Wigner is to measure whether or not his friend and the physical system are in a superposition. Here is how the scientists describe the thought experiment (edited a little to remove some of the math; edits marked with “[*]”):

“According to quantum theory, the friend randomly observes one of the two possible outcomes in every run of the experiment. The friend’s record, [up] or [down], can be stored in one of two possible orthogonal states of some physical memory, labeled either [“electron is up”] or [“electron is down”], and constitutes a “fact” from the friend’s point of view. Wigner observes from outside the isolated laboratory and has no information about his friend’s measurement outcome. According to quantum theory Wigner must describe the friend’s measurement as a unitary interaction that leaves the photon and friend’s record [an] the entangled state… Wigner can now perform an interference experiment in this entangled basis to verify that the photon and his friend’s record are indeed in superposition—a “fact” from his point of view, from which he concludes that his friend cannot have recorded a definite outcome. Concurrently however, the friend does always record a definite outcome, which suggests that the original superposition was destroyed and Wigner should not observe any interference. The friend can even tell Wigner that she recorded a definite outcome (without revealing the result), yet Wigner and his friend’s respective descriptions remain unchanged [6]. This calls into question the objective status of the facts established by the two observers” (pp. 1-2)

This is the tension: from the friend’s perspective, the electron has a determinate X-spin. From Wigner’s, however, the friend and the particle are in a superposition. Thus, the reasoning in this paper goes, there is no objective reality (or at the very least this is called into question).

The actual experiment that the scientists performed is not exactly Wigner’s friend, but what they call an extension of Wigner’s friend. There is another famous type of experiment in quantum mechanics that is meant to confirm the violation of something called a Bell-type inequality. Named after John Bell, he used it to prove his famous theorem, which states that any physical theory that reproduces the probabilistic predictions of quantum mechanics will have to reject a locality condition (to be discussed shortly). There is a vast literature on what exactly Bell’s theorem(s) shows. However, what is important to note here is that this is a well-knows result that has been empirically confirmed many times. What I mean by this is that the kind of probabilistic predictions used in Bell’s theorem have been empirically confirmed to be the optimal predictions we could make (of course things are a little more subtle, but this is the gist–quantum mechanics is the most successful physical theory we have ever written down; as Everett put it, “This formulation describes a wealth of experience. No experimental evidence is known which contradicts it.”).

The experiment discussed in the paper is a combination of Wigner’s friend and the kind of experiment used to test whether Bell’s equality is violated (which is what quantum mechanics predicts). However, the key thing that the experiments were after was whether or not there are “observer-independent fact”–this depends much more on the Wignerian aspect of their experiment than the Bell aspect. (Indeed, it seems to me, and others I will point to later, is that combining these two is a kind of red herring, and is uninteresting.) Karen Crowther on Daily Nous gives a good description of the actual experiment:

“Now, what the experiment actually did was to use QM to calculate the probabilities of each of the possible measurement outcomes, and then compare these to the probabilities calculated from the experimental data obtained (1794 six-photon coincidence events, using 64 settings, over a total of 360 hours). The experimenters did this in order to test the violation of a Bell-type inequality, and the experiment was indeed successful in confirming its violation.”

So, we have a decent understanding now of what the experiment was. Whether or not this experiments actually lends support to what the scientists claim it does, I will leave for later. For now let’s get a handle on the three conditions needed for their theoretical result: free choice, locality, and observer-independent facts.

I’ve actually discussed the free choice condition on this blog before, so I won’t go into too much detail (this will be a long enough post already). The key idea is that the observers of the experiment can freely choose which property of the physical system to measure: there are no conspiracies of physics in which the experiments are bound to measure certain properties so that the results come out just so. Again, for an extended discussion, see the previous post.

The locality condition means that distant measurements (and physical systems in general) cannot affect each other. For example, if I measure the property of a physical system over at Alpha Centauri, it should not instantaneously affect a system you want to measure on Earth. This is somewhat motivated by wanting a quantum mechanics to be consistent with special relativity which imposes a kind of locality constraint, and somewhat motivated by intuition. However, almost every theory of quantum mechanics violates locality; it seems to be something we might have to give up. We’ll discuss this point more a little later.

The final condition is “observer independent facts”. The paper is a little vague and unclear about what this means. Thus from the paper itself it is unclear how this requirement interacts with their result. However, in one of the papers that the experimental paper cites, the condition is specified more clearly. Put plainly, this assumption is that “One can jointly assign truth values to the propositions about observed outcomes (“facts”) of different observers” (p. 5 from the paper linked immediately above). The paper also gives a slightly more technical account of this postulate, but this is sufficient for our purposes here.

Now we have a handle on the experiment, and the three conditions needed for the “no-go theorem for observer-independent facts” which the experiments are trying to empirically support. What is their argument, and is it successful?

Recall from the earlier quote from the conclusion of the paper. The theorem says that accepting the conditions of free choice, locality, and observer independent facts is incompatible with their experiments results, since they violate the Bell-type inequality. This much is true. Thus, we must give up one. Which one? They argue for giving up observer-independent facts. However their argument is rather sparse; they briefly mention that “one way to accommodate our result is by proclaiming that ‘facts of the world’ can only be established by a privileged observer—e.g., one that would have access to the ‘global wavefunction’ in the many worlds interpretation or Bohmian mechanics” (p. 4). First of all, this isn’t true–neither Bohmian mechanics nor many worlds needs a privileged observer. Indeed, as Dustin Lazarovici points out the physicist John Bell of the Bell inequality called both Bohmian mechanics and many worlds “quantum theories without an observer.”

After this remark the scientists immediately shift to the thesis of their paper: that we should give up an observer independent reality. However, as we saw, there is no need to do this–we already have theories that account for this! As Wayne Myrvold remarks:

“Here’s a nice fact about claims of this sort: when you see one, you can be sure, without even going through the details of the argument, that any conclusion to the effect that the predictions of quantum mechanics are incompatible with an objective, observer-independent reality, is simply and plainly false. That is because we have a theory that yields all of the predictions of standard quantum mechanics and coherently describes a single, observer-independent world. This is the theory that was presented already in 1927 by Louis de Broglie, and was rediscovered in 1952 by David Bohm, and is either called the de Broglie-Bohm pilot wave theory, or Bohmian mechanics, depending on who you’re talking to. You can be confident that, if you went through the details of any real or imagined experiment, then you would find that the de Broglie-Bohm theory gives a consistent, observer-independent, one-world account of what happens in the experiment, an account that is in complete accord with standard quantum mechanics with regards to predictions of experimental outcomes.

There are other theories, known as dynamical collapse theories, that also yield accounts of a single, observer-independent reality. These theories yield virtually the same predictions as standard quantum mechanics for all experiments that are currently feasible, but differ from the predictions of quantum mechanics for some experiments involving macroscopic objects.”

He doesn’t mention Everttian (many worlds) quantum mechanics in this passage, but it too is compatible with this result. As Sean Carroll in the same Daily Nous piece puts it:

“My own preferred version of quantum mechanics is the Everett, or Many-Worlds formulation. It is a thoroughly realist theory, and is completely compatible with the experimental results obtained here. Thus, we have a proof by construction that this result cannot possibly imply that there is no objective reality.”

What we have to do in order to keep observer-independent facts is to give up a strong form of locality. Pretty much all of our theories of quantum mechanics give up locality. But we already knew this! It is old news. (Subtle side note: even though Everettian quantum mechanics and GRW flashes are non-local in the sense needed to give the standard quantum mechanical predictions, they are still (very likely) compatible with special relativity. This is not essential for our story here, but is worth mentioning.) Indeed, as Tim Maudlin remarks (you already know whence this quote comes!):

“All of this is even spelled out in the article itself: ‘But there are other assumptions too. One is that observers have the freedom to make whatever observations they want. And another is that the choices one observer makes do not influence the choices other observers make—an assumption that physicists call locality.’ That is, in order to account for the outcome of this experiment, one has to deny that physical reality is local in Bell’s sense. (This gloss on the locality condition is not accurate, but leave that aside.) That is something we have known for 50 years.”

Thus we see that the experimental results really show nothing new, and certainly nothing that should make us question objective reality. It really does seem to be an unfortunate case of physicists either being confused, or trying to sell their result too much (inclusive or!).

The most charitable reading of these results I could find is by Karen Crowther whom I referenced earlier:

“The question is what this experiment demonstrates about QM that was not already known from the thought-experiment plus previous experimental results. Plausibly, what it shows is that a scenario analogous to the one imagined by Wigner is in fact physically possible, and in it the observers do record conflicting facts.”

The least charitable (but also very amusing) quote I encountered was from Dustin Lazarovici:

“In my opinion, the paper does indeed raise some important questions, though they are mostly sociological ones. For instance: Why does physics tend to get exposure and attention merely for making outlandish claims, regardless of their scientific substance? And why do even many experts tend to abandon rational and critical standards when it comes to quantum mechanics? Why, in other words, have we gotten so used to quantum physics being crazy that even the most outlandish claims come with a presupposition of plausibility and relevance?

As a matter of fact, quantum mechanics can be as clear and rational as any respectable theory that came before it. You just have to do it right.”

I have to agree with this sentiment. Although, as for his first question, I think we can explain that by noticing that humans tend to be more interested in outlandish claims like “THERE IS NO OBJECTIVE REALITY”. Again, no one wants to read the paper entitled “K-alpha particles exhibit decoherence along mesa-keta spectra at 196mmz intervals”. It’s not sexy enough to rise to the surface of the MIT Technology Review.

I think the message to take away from all of this is that objective reality is fine, and nothing has really changed at all in quantum mechanics because of this result. I’ll give the last words to Wayne Myrvold:

“Headline news! Stop the presses! A group of experimenters did an experiment, and the results came out exactly the way that our best physical theory of such things says it should, just as everyone expected. Quantum Theory Confirmed Again.

That’s what actually happened, though you’d never know it from the clickbait headline: A quantum experiment suggests there’s no such thing as objective reality .”