I’ve posted a lot over the years on interpretations of quantum mechanics. My writing has tended to focus on comparing the big three: Copenhagen, pilot-wave, and many-worlds. But there are a lot of others. One that has been gaining converts among physicists and others is Carlo Rovelli’s relational quantum mechanics (RQM) interpretation. This is an interpretation that comes up enough in conversation that I’ve always wanted to learn more. So when Rovelli’s book on it was announced, I decided I needed to read it. But Helgoland: Making Sense of the Quantum Revolution took a while to be available in the US, at least in Kindle format. My preorder finally came through last week, so I spent the last few days going through it.
Rovelli is clear at the beginning of the book that this is a partisan work, and he’s not kidding, although this type of partisanship is common in books on quantum physics. This book is about his particular interpretation. He does discuss many of the other major interpretations: many-worlds, pilot-wave, QBism, and physical collapse theories, but he makes clear that his coverage is cursory, and mentions multiple times that the reader can skip these if they want. (I read them anyway, just to see how he’d treat them.)
In another move that I’m starting to see as too common in these types of books, Rovelli’s partisanship includes his description of historical scientists. He sees his interpretation as fitting squarely within the tradition started by Werner Heisenberg, and his descriptions of Heisenberg seem pretty reverent. His view of Erwin Schrödinger, on the other hand, seems hostile, both intellectually and personally. As many authors have done, he describes Schrödinger’s polyamorous lifestyle, but goes a bit further by implying that Schrödinger had pedophilic tendencies. In contrast, in his biographical remarks about Heisenberg, he downplays Heisenberg’s collaboration with the Nazis. (He does mention another scientist who likely didn’t receive a Nobel prize because of Nazi affiliations. Heisenberg had the good luck to receive his Nobel before the Nazis came to power.)
Anyway, Rovelli’s sees Heisenberg’s chief contribution as focusing on observables and then building a theory of the relations between those observables. In his view, Schrödinger’s focus on real waves was a distraction, and the Copenhagen team were right to interpret his wavefunction as a probabilistic mathematical mechanism, a move Schrödinger himself was never happy with. (Although he did grudgingly come to admit the practical benefits.)
The main role of an interpretation is to explain what happens during the measurement process. Quantum objects move like waves, until they’re measured, then they behave like particles. In the classic Copenhagen interpretation, this is usually referred to as the wavefunction collapse. In the strong version of Copenhagen, involving a physical collapse, this was seen as problematic by Albert Einstein, because it involves an instantaneous collapse across all of time and space, leading to nonlocal “spooky action at a distance”, an issue made particularly vivid by quantum entanglement. Weaker versions of Copenhagen only have an epistemic collapse, resembling QBism, and so don’t consider themselves to have this issue.
The big question with Copenhagen is, when does the collapse occur? Niels Bohr’s answer was interaction with macroscopic systems, such as lab equipment, implying that there were different rules for microscopic and macroscopic phenomena. However, no one has managed to find any threshold where a collapse happens. Over the decades, scientists have managed to observe quantum effects in ever larger collections of quantum particles, molecules, and even tiny macroscopic objects. It looks increasingly unlikely that there is any such threshold.
This doesn’t represent an issue for non-collapse interpretations such as pilot-wave or many-worlds, but it does for most collapse interpretations. RQM is a collapse interpretation, but its innovation is to make the collapse a relative event. In RQM, what causes the collapse is an interaction with another physical system. However, the collapse only happens relative to the system interacted with, not with any other system. In other words, a quantum particle can be in superposition relative to one physical system while being collapsed relative to another.
So, if two quantum particles interact, they collapse relative to each other. But to the rest of the world, they remain in a superposition, and the interaction has left them entangled in some fashion. It also pertains to a quantum computing circuit. For each particle in the circuit, once it receives interactions, the circuit has collapsed. However, for the outside world, until there are interactions with the environment, the circuit remains in a superposition of all its possible states.
Making the collapse relative solves the question of when it occurs. It occurs on any interaction, but only relative to the particles involved in that interaction. Similar to many-worlds, this interpretation sees the entire universe as being quantum in nature. But also like many-worlds, it has radical implications. Physical reality exists in the relations, and only in the relations. This by itself isn’t too radical. It’s compatible with the ontic version of structural realism that we recently discussed. But it also implies that properties of physical systems don’t exist for another system at all until the interaction.
This is highlighted when considering RQM’s claim to local dynamics. Consider a couple of entangled particles, one held by Alice and one by Bob. Even if Alice and Bob are separated by light years, when they measure their particles at the same time, the particles collapse into compatible states. With an absolute collapse, this is a problem, because it implies faster than light communication.
But with a relative collapse, the most relevant collapse doesn’t happen until a comparison event, when the results of the measurements have been transmitted (at light speed or slower) to some party, say Charles, who does the comparison. Relative to Charles, the particles haven’t collapsed until he receives the results, even though relative to Alice and Bob their respective particles have collapsed. When Charles does receive the results, that’s when the collapse happens for him. Now we have a completely local interaction. But this only works because under RQM, the reality of the measurement outcomes don’t exist for Charles before he receives them.
So, while many-worlds implies a surplus ontology many find far too extravagant, RQM posits a radically sparse ontology that almost seems like separate interacting solipsistic realities, although centered on physical systems rather than just minds. In the latter parts of the book, Rovelli explores philosophy ranging from Marxist thought to eastern Buddhist thinking that resonates with this view.
Rovelli also veers into a discussion of consciousness. He dismisses ideas about the mind having anything to do with the collapse, or that mental processes are quantum, at least any more so than any other physical process, as well as a host of other quantum mystical notions. But in his view, seeing reality as being composed of interacting viewpoints, as RQM does, helps to close the gap between physics and the mind, eliminating a necessity to reconcile an objective view (which doesn’t exist) with subjective perspectival views. The idea is that both sides of the divide are now perspectival. It’s an interesting idea, but I suspect few troubled by the hard problem of consciousness will be convinced.
This is an interesting interpretation, but in my view it has a couple of drawbacks. One is that, as noted above, Rovelli takes a mostly anti-real stance toward the wavefunction. He notes that we never see a quantum wave, only the interference from it. That’s true but we also never see a quantum particle, only the effects it leaves in measuring equipment. And something causes the observed interference effects. The idea that the wavefunction can predict those effects with the accuracy it does, without modeling reality in some manner, seems implausible. But if we let that realism in, then RQM seems in danger of becoming many-worlds with blinders on. (It’s worth noting that an early name for many-worlds was “the relative state formulation”.)
I also see keeping the collapse postulate as a drawback. RQM does defang one of its worst implications, the instantaneous change in reality that concerned Einstein. But it also leaves in a level of indeterminism. Many will see this as a plus, preferring a physics where everything isn’t determined, and might argue that it’s a matter of taste. But I’m in the camp that sees determinism as something that works well everywhere else in science, and worked well overall for centuries before quantum physics. To me, it doesn’t seem like we should dispense with it lightly, particularly while there are options. (This doesn’t mean that quantum physics would ever be operationally deterministic.)
Finally, it shouldn’t be underestimated just how radical the sparse ontology proposed here is. The interpretation takes general relativity as an inspiration. But in the case of general and special relativity, the conclusions are a necessity driven by observation and mathematics. RQM requires a specific type of collapse postulate, a major assumption, albeit one many will consider justified given the alternatives.
But this is quantum physics. We won’t get by unscathed. Interpretations juggle things like determinism, locality, realism, the arrow of time, a single reality versus multiple realities, and now a sparse versus full reality. Every interpretation requires throwing one or more aspects of common sense reality under the bus.
What do you think of relational quantum mechanics? Do you feel like the sparse ontology is worth it?