### Scott Glancy

Scott Glancy is a theoretical physicist in the Applied and Computational Mathematics Division in NIST's Information Technology Laboratory. His current research interests are in quantum information...

Just a Standard Blog

For this past Christmas my wife, Rebecca, gave me a T-shirt that says "Quantum mechanics: The dreams stuff is made of." This is an allusion to the book *The Dreams That Stuff Is Made Of: The Most Astounding Papers of Quantum Physics*—*and How They Shook the Scientific World*, edited by the late Stephen Hawking. For some time, I hesitated to wear the T-shirt because I found its message to be problematic. My feelings toward this very thoughtful gift have to do with NIST’s experiments in quantum foundations, the world’s most random numbers, and why I happily wear the T-shirt now.

I am a physicist who does research in quantum information theory. One of the things that other physicists love to do is argue about the interpretation of quantum theory. There are MANY different interpretations. Wikipedia lists only (only!) 13, and you can see 26 of them categorized in "Interpretations of quantum theory: A map of madness." To explain these interpretations in detail is, as we sometimes write in academic papers, "beyond the scope" of this blog, but I will give a few examples to show how they tell wildly different stories about what is happening at the quantum level of our world.

Some interpretations claim that quantum particles can send faster-than-light signals to one another. Some claim that everything from quantum particles to humans to galaxies evolves in an ever-expanding superposition that contains all possible events happening together like images projected on top of one another, but we can only see the one image of the events we experience. Some claim that human consciousness can "collapse the wavefunction." The wavefunction describes all the possible outcomes that may occur when quantum systems are measured. When a human measures a quantum system, they observe one thing out of all the possibilities. This is the "collapse." Some others claim that quantum particles are not "real." That’s a bit of a misnomer: It means that some of their properties are determined only when measured. Before measurement, they exist in a dream-like state of unreality.

The T-shirt Rebecca gave me seems to endorse this last view that stuff is made of dreams, which is a problem for me.

While most physicists working in quantum foundations tend to choose one interpretation and champion it, I like most of the 26, even the dreamy one. When I told Rebecca, "I don't know if I can wear this shirt," my dilemma was that I didn't want to champion just one interpretation. I can’t decide between these many interpretations. How should I choose? We learn in elementary school that when faced with several different hypotheses, we should do an experiment to reject one or more of them.

Quantum theory has been subjected to many experimental tests over the past century, and it keeps passing the tests. It is so reliable that, despite its apparent strangeness, quantum mechanics is now essential for nearly every modern technology. We trust our lives to quantum mechanics every day, in the computers that control our cars, in the lasers that speed our communications around the world, in the atomic clocks that guide the global positioning system (GPS), and in the medical scans that diagnose our diseases.

However, as good scientists, we continue to test quantum theory.

That’s why I was excited to participate in a recent experiment at NIST to prove that quantum particles violate the principle of "local realism." According to local realism, all particles have definite properties for all possible measurements (realism) and communication between particles is no faster than the speed of light. Classical theories of physics, like those that describe how everyday objects, planets and stars move, obey this principle, but quantum theory predicts that entangled quantum systems can violate local realism and that this violation can be observed in an experiment called a "Bell Test," after John S. Bell.

In a typical Bell Test, two particles are entangled with one another, separated and sent to two measurement stations. At each station, a random number generator chooses how each particle will be measured. The experiment is repeated many times, and the relationships between choices and outcomes are analyzed. The relationships between particles that obey local realism are limited by a set of mathematical rules called "Bell's inequalities," but quantum particles can violate Bell’s inequalities. Since Bell discovered the inequalities in 1964, many Bell Test experiments have shown violation of local realism. However, before 2015, all Bell Tests suffered from loopholes—potential mechanisms, however contrived, that would allow systems that obey local realism to violate the inequalities. In 2015, our experiment and two others (at the University of Vienna and the Technical University of Delft) were the first to close all loopholes and show violations of local realism.

Since 2015, we have improved the optical devices we used in our Bell Test experiment and advanced our mathematical analysis tools. Through the heroic effort and skill of my experimental colleagues, violation of local realism has become routine in our laboratories. We participated in The Big Bell Test in which humans provided the random measurement choices. More than 100,000 people entered 0s and 1s into the Big Bell Test website, and those choices were distributed to 13 labs around the world, each of which successfully performed its own version of a Bell Test.

At NIST, we are now using our Bell Test to produce the world’s most random numbers, meaning that our random numbers are the most difficult to predict before they are generated in our laboratory. By proving that the data from a Bell Test violates local realism, we also prove that the data is unpredictable even if a hacker may have secretly compromised the experimental devices. We use new mathematical tools to quantify the amount of secure randomness in the Bell Test data and to extract a string of random bits, each of which is 0 or 1 with 50 percent probability. Eventually, we hope to incorporate a Bell Test randomness source into the NIST Randomness Beacon, which publishes a certified 512-bit random number every minute. These public random numbers can be used for tasks such as choosing which voting machines should be audited, authenticating identity online, and performing secure multiparty computation. The Bell Test is advancing from an exotic test of fundamental physics to a useful technology.

After decades of work, the predictions of quantum theory have been confirmed, and the hypothesis of local realism has been rejected! So, did that allow us to rule out any of the many interpretations? Unfortunately not. The Bell Test tells us that either locality or realism (or both) should be discarded, but it doesn’t tell us which. In fact, the problem goes much deeper than the Bell Test because most of the interpretations make *exactly* the same predictions for all possible experiments. In other words, all the interpretations work. Experiments can only test predictions about things we can measure, and predictions are calculated from the theory’s mathematical structure. All the various interpretations are derived from the same mathematical foundation, which is, in my opinion, the essence of quantum mechanics. The different interpretations are our human attempts to describe the mathematical essence of quantum mechanics with human language. Because the interpretations are built on the same mathematical foundation and make identical predictions, there is no possible experiment that anyone can ever do that can support one interpretation over others.

Because experiments are not helping, many physicists argue for their favorite interpretation of quantum mechanics by emphasizing virtues such as simplicity, aesthetic beauty, appeal to intuition, usefulness as clickbait and others. However, I usually find these arguments unconvincing because, for example, I don’t know how to weigh the simplicity of one theory against the elegance of another. In the past, I despaired over my inability to know the truth because I can’t choose a single interpretation. Now, however, I rejoice because we do have some access to the truth: It is encoded in the mathematical structure of the theory. There are many good ways to interpret or translate from the mathematics to natural language. Maybe no single translation perfectly captures the mathematical source, but they can all be useful tools for understanding.

I am happy that we have a wealth of interpretations, so, after careful consideration, I decided to wear the "Quantum mechanics: The dreams stuff is made of" T-shirt. I like its nonrealist interpretation. But to avoid favoritism, as I told Rebecca, I need more T-shirts with slogans supporting the other 12 (or 25, depending on how you look at it) interpretations. If you have a good quantum T-shirt slogan, or know of one, leave it in the comments below

Scott Glancy is a theoretical physicist in the Applied and Computational Mathematics Division in NIST's Information Technology Laboratory. His current research interests are in quantum information...

Remember that documentary you saw that finally explained metrology and why measurements are critical to practically every aspect of modern life? Yeah, neither

As the digital age moves forward, it’s becoming impossible to avoid interacting with artificial intelligence (AI) systems. Computer assistants and AIs perform

Think back to the last time you did an information search. Did you find what you were looking for? How easy was it to find? Search is now pervasive, and we

"I am found", on front of T-shirt, "I am lost", on back. The observer would interpret the shirt according to how they were positioned to see the shirt.

- “back to the future” on back of shirt, “front to the future” on front of the shirt

- “ You don’t know me from atom”

- “I bought this shirt at a 2-for-1 quantum mechanics clothing sale”

Thanks for those fun suggestions. I especially like "back to the future / front to the future" because some people believe that entangled particles are able to communicate with one another by sending information backwards through time. Also one of the several sources of data used to make measurement choices in our loophole-free Bell Test in 2015 was a mixture of several movies and TV shows, which included *Back to the Future*.

I haven't made an exact count, but it seems to me that the half of the interpretations that involve a "mind" are inconsistent with the fact that the mind arises from the brain, which is itself a quantum object. A properly mentalistic quantum theory would be one that implies its own metatheory. All existing theories are merely first-order and incomplete in that sense. The late Harvard professor Sidney Coleman interpreted QM in a way that included the mind as a quantum object in his 1994 lecture "Quantum mechanics in your face", which he never wrote up as far as I know, but you can find a video of it online.

When you accept that minds and even measurement apparatuses are quantum objects, you end up having to reject the notion of a "particle" as a real, autonomous thing. Conveniently, quantum field theory already does this: it says that particles are just perturbations of the quantum field. (Technically, the 17 coupled fields of the standard model, but for most purposes, the electron and photon fields are all that are needed.) And because each field is continuous throughout spacetime, questions of "locality" just do not arise.

Measurement systems made up of continuum quantum fields also lead you to reject the idea that there are pure binary "yes/no" answers to any question. This gets you into the deep waters of point set topology and suggests that the many worlds or many minds of some interpretations are really just convenient references to imaginary slices through a single extended object.

So I would keep the front of your shirt the same, but put on the back "Quantum fields: the stuff that dreams are made of."

I would be interested to see some theorist's analysis of the NIST experiments in terms of QFT. But I suspect that the math is just too complicated...

My understanding of interpretations that involve a "mind" is that they claim that there is an essentially non-quantum or aspect of human consciousness, which causes wavefunction collapse. Although brain (as distinct from the mind) might or might not be a quantum mechanical object able to exist in superposition, the conscious mind (they claim) is not able to exist in superposition. There are various mechanisms proposed for why or how consciousness is not quantum in spite of the apparently quantum nature of the matter that makes up the brain. That might feel inconsistent, but I don't know that it is logically contradictory. Also, there is no possible experiment that can distinguish between a mind that can cause wavefunction collapse and a mind that becomes entangled with the wavefunction.

I would not say that quantum field theory rejects the notion of particles being real. Instead it deepens our understanding of what particles are: excitations of quantum fields. Even in non-relativistic quantum theory we learned that "particles" have wave-like behavior, so I don't think that quantum field theory's more sophisticated understanding is too radical.

Certainly even with systems made of quantum fields, one can still obtain binary "yes/no" answers to questions. For example, photon fields still have discrete polarizations of horizontal or vertical.

I am not sure what you have in mind when you ask for an analysis of NIST's experiments in terms of quantum field theory. In our Bell Test experiments, the goal is to reject classical theories. Surprisingly even though we use quantum mechanics to design the experiment and exploit quantum entanglement in the experiment, the analysis is classical. We simply ask "Can a classical theory produce data like we saw?". We find that even the best classical theory provides a very poor match to our data, so we conclude that classical physics cannot explain the behavior in our experiment. Of course, quantum field theory can explain the experiment.

Nice article sir

T-shirt idea:

Quantum Mechanics: It may be right, depends how you look at it.

Considering Quantum Mechanics' tremendous success in experiments and technologies, I would say "Quantum Mechanics: It is right, no matter how you look at it.".

Quantum Mechanics says: I'm sexy 50% of the time, depending on how you look. Beers are on me!

So Bell's inequality shows that either realism or locality must be given up? That's cool...

So I'm wondering

1) what happens in the presence of significant spacetime curvature... e.g. near a bh

2) why our notions of time and causality are so wrapped up in preserving the dictates of special relativity (e.g. v_signal<c always) when we know that:

2a) SR (and QFT) is only valid in locally flat spacetimes

2b)spacetime itself has so such restriction on the speed of its expansion -

2c) and now that the order of cause and effect may be reversed in the presence of large masses - see Zych, et al "Bell’s theorem for temporal order", Nature Comm. Aug 2019

I'm not saying we should completely give up on cause and effect, but I'm saying that maybe we should think about re-evaluating the many restrictions imposed to make QM conform to SR.

1) Near the event horizon of a black hole, the experiments would be much more difficult to perform because the space-time curvature will disturb the entanglement in the photons' polarizations, but one should still be able to see violation of local realism in principle. Near the center of a black hole, nobody knows what will happen because we don't have a unified theory of quantum gravity.

2) No one has ever done an experiment in which we have seen the predictions of relativity fail, so these are the best dictates that we know of regarding the structure of space-time.

2a) QFT can be adapted for curved space-times, for example to describe the behavior of quantum particles in a fixed, curved space-time background. However, we do not know how the quantum particles effect the space-time.

2b) I'm not sure what you mean by this.

2c) My interpretation of the paper by Zych, Costa, Pikovsky, and Brukner is that they have highlighted a particular way in which quantum theory and general relativity are not compatible. Their thought experiment does not prove that cause and effect are necessarily reversible -- only that they can be reversible if some relevant features of quantum theory and relativity are preserved during their fusion.

I agree that we must re-evaluate the many restrictions of quantum theory and relativity. Many excellent physicists are working hard developing such theories, and there are plenty to choose from. The main difficulty is the lack of experimental data that we can use to test these theories.

So does Bell's Theorem mean that when 2 entangled particles spin are measured, it's like they were spinning already partially lined up with where the detectors will be?

@idontgetit, when we see that Bell's Inequality is violated in a loophole-free experiment, we could explain what is really happening to the particles by saying that "their spins were already lined up with where the detectors will be". At least that explanation is compatible with the experiment. How are they able to line up? It could be that the particles receive a faster-than-light message telling which way to line up. It could be that the particles had advance information about the detectors' orientations even before the supposedly random choices were made. It is also possible that the particles do not have definite values for their spins until they hit the detectors. All of these different explanations are compatible with experiments, so which one you prefer is up to you.

In support of the pilot wave interpretation: "The wavefunction is my co-pilot".