What If Physics IS NOT Describing Reality? (SCRIPT) (Patreon)

Neils Bohr said, “It is wrong to think that the task of physics is to find out how Nature is. Physics concerns what we can say about Nature.” Well it turns out that if we pay attention to this subtle difference, some of the most mysterious aspects of nature make a lot more sense.

What is physics really trying to do? Is it to find the mathematical laws that govern the universe? Not quite - no one has to solve the Schrodinger equation in order for an electron to be able to do its thing. Our laws of nature are just models. So maybe the job of physics is one step removed - to come up with laws that do the best job at predicting how the universe works, and then hope that we can infer truths about the universe based on which laws work best. But actually, some of the founders of quantum theory were convinced that the role of physics was one step further removed still. Neils Bohr insisted that what we actually model is the results of observation, not the world itself. His student and closest colleague,  Werner Heisenberg, put it well “The laws of nature which we formulate mathematically in quantum theory deal no longer with the particles themselves but with our knowledge of the elementary particles.” In other words, the mathematical laws of physics don’t govern reality, they’re not even direct models of reality. Rather, the laws of physics are models of our experience of reality. Physics models our information about the world.

In a recent episode we started talking about informational interpretations of quantum mechanics. Then we discussed one of the more radical interpretations: John Archibald Wheeler’s idea that information is really the most fundamental thing, and it gives rise to the physical - a notion he pithily summarized with the expression “it from bit”. Quantum mechanics tells us that asking questions of the universe radically changes how it behaves. Wheeler followed that simple fact down the rabbit hole to what he saw as the logical conclusion - the most fundamental existence of is in the answers elicited to yes-or-no questions

Wheeler may have been right, but we can explore the power of informational quantum mechanics without committing to quite such a radical interpretation.  Today we’re going to see how a lot of the weirdness of quantum mechanics can make sense if we think about it as a model of our information about the world.

Normally in physics we try to break up the world into its most elementary components. We learn how those components behave, and see if we can rebuild the world from those parts. In quantum mechanics, we have things like particles and fields which can only take on discrete or quantized values. These quantum components also have weird properties like fundamental uncertainty in their values and strange correlations between components that we call entanglement. But as weird as quantum particles and fields are, this still feels like a very physical way to define the building blocks of reality.

The physicist Anton Zeilinger has proposed an informational approach to quantum mechanics in which the world is broken up not into physical parts, but into informational parts. But what does it mean to use information as our building block? Information represents our knowledge about the world. So our new building block becomes a statement about the information we have — for example, about the location or speed or mass of a particle. Zeilinger calls such a statement a proposition - it’s an answer to a question we could ask about the world. He says that a quantum system is a collection of propositions. A quantum system represents our knowledge of the world, not the world itself.

So how do you break up a “world” made of knowledge or information? Well, an informational building block is the answer to a question. So the smallest informational building block is the answer to a question with the fewest possible outcomes. Literally any answer can be reached by a series of well-chosen yes-no questions. So Zeilinger says that  any quantum system can be broken into the results of binary questions. By insisting that the most elementary informational building block contains only one bit - one yes-no answer - a surprising number of the weird results of quantum mechanics suddenly make sense. Things like quantum indeterminacy, entanglement, and the uncertainty principle turn out to be the expected behaviors of this sort of information system.

The simplest way to start is to  look at a quantum system where the answer to a single binary question seems to give a meaningful “physical” answer. Consider quantum spin. From a physical point of view, think of it as a particle’s orientation - a spin axis that can point either up or down. But really the quantum system is the answer to a question: is the spin up or down relative to the direction of measurement? We ask this question with a  Stern Gerlach apparatus, where the magnetic moment of the particles interact with a magnetic field gradient to deflect the particles either up or down, depending on the direction of the spin.

Let’s say we prepare an electron’s spin to all point up relative to our apparatus. The spin contains one bit of information, the “up” result of the spin measurement.

But what if you asked the electron spin, “are you pointing left or right?” We can do this by rotating the Stern Gerlach apparatus 90 degrees. You started out with one bit of knowledge about the particle’s up-down alignment. According to Zeilinger, by definition, that’s all the information that the elementary quantum system of spin can contain. That means the left-right orientation is undefined. It’s in a superposition state of both left and right until it’s measured. Run it  Stern-Gerlach through are horizontally aligned Stern-Gerlach divide and the electron has an even chance of being deflected left or right - its undefined left-right spin chooses randomly between the two because it contains no information about its spin in that direction. And following that measurement, the left-right alignment of the spin has become defined - after all, the electron has to come up with an answer to the question you asked. But now with the left-right alignment defined, our single bit of spin information is taken up, and the up-down alignment becomes undefined. So by thinking of quantum systems as being made of these elementary “quanta of information”, we see the indeterminacy of quantum theory arises naturally. Zeilinger even managed to derive the equivalent of the Schrodinger equation by asking how quantum information should evolve over time.

Quantum entanglement also fits this picture. When we prepared our electrons to be spin-up, that spin was relative to a chosen direction - the vertical in this case. But we could also prepare an electron to have a spin direction that’s defined relative to another particle’s spin. For example, a pair of electrons could be prepared that have opposite spin to each other. Now remember our requirement that the spin of each electron can only contain one bit of information, and that bit is taken up to describe its relationship to its partner electron. Now the information is no longer isolated in the single electron’s spin, but rather spread between two electrons. Two electron spins contain two bits, but those bits are distributed non-locally. If you ask the spin direction of either of those electrons relative to your Stern-Gerlach apparatus, you’ll learn the spin direction of both particles. And in making that measurement, you’ve forced the single bit of spin information in both electrons to become defined relative to your apparatus - whichever way you chose to align it. Our distributed information becomes local to each electron, and in the process appears to force an instantaneous communication between the particles  - a spooky action at a distance. This is entanglement.

So we’ve seen how an elementary quantum system’s information content has to exist with respect to a certain type of question that you ask of it. Spin direction in the last example. This idea leads us naturally to some other staples of quantum theory - like the Heisenberg uncertainty principle. In a previous episode we saw how the uncertainty principle arises from the limited knowledge that we can extract from a quantum wavefunction. For example, that the product of the measurement error in a particle’s position and momentum has to be greater than the Planck constant divided by 4-pi.

In fact, by bringing to bear the tools of information theory we can derive an even tighter uncertainty relation - something known as entropic uncertainty. This approach uses the informational definition of entropy - Shannon entropy - which is a measure of the number of yes-no questions needed to extract all the information from a system. The resulting “entropic uncertainty” has been used to explain one of the original and most mysterious features of quantum mechanics - wave-particle duality.

This was demonstrated in by a team of physicists in 2014. They applied entropic uncertainty to analyze Wheeler’s  “delayed-choice” experiment. This experiment causes a photon to behave like a wave OR a particle depending on the question asked of it. And that question could be asked after it passes through the experimental apparatus. If it’s a wave it travels both paths of the device, if a particle it must only travel one.

Applying the notion of entropic uncertainty to the situation, the team said that the wavefunction contained only one answer to two complementary questions: just as our particles could only tell you if they were up or down, but not left or right, the wavefunction of the photon they considered could tell you either which path the photon took, or the phase of the photon by looking at the interference at the end. Because of the finite information content of the wavefunction, it didn’t have answers to both questions. So, they found that the wave-particle duality of quantum mechanics arises from the limited information, the inability to answer two complementary questions (“are you a particle?” And “are you a wave?”) at the same time.

It’s one thing to use quantum information theory as a mathematical tool, but quite another to claim that information is somehow a more fundamental substance than the tangible stuff that most people believe to be the building blocks of reality. In quantum mechanics, we tend to think of the quantum wavefunction as pretty fundamental. It describes the evolving distribution of probabilities for the results you might get if you tried to measure the properties of a quantum system. Zeilinger and collaborators would say that the wavefunction does not have a physical existence independent of the observer. Rather, the wavefunction and the math that governs it describe our information about the universe, not the universe itself.

But, if the wavefunction is about the information content of a system, again, we come back to the same question that plagued Wheeler: “whose information?” This same question haunted Einstein. Einstein famously once asked whether proponents of these observer-centric interpretations truly believe the moon isn’t there when nobody looks. To this, Bohr and Wheeler and Zeilinger and their supporters might say, “You can’t prove that it is.” To learn about something necessarily involves an observer who is acquiring this knowledge, so all we can ever know about the world is how we interact with it. Whether there is an observer-independent world out there, we, being observers, can never prove. And it turns out that a lot of the weirdness we see in the quantum world make more sense if we pay attention to the fact that our only direct experience is with an entirely informational space time.