Chapters Eleven and Twelve

The Multiverse


A Physicist’s History of Bad Philosophy

I’m reviewing chapters eleven and twelve together, as they’re really not separable. In chapter eleven, Deutsch describes his view of quantum mechanics, the many universes interpretation. He makes a quite compelling case (at least to me), although of course it’s curious that his view remains a minority stance among physicists today.

Deutsch describes what is essentially a version of the double slit experiment, called the Mach-Zehnder interferometer. (The Mach is Ludwig Mach, son of Ernst Mach, the philosopher who didn’t believe atoms existed because no one could see them.)

In the Mach-Zehnder interferometer, a single photon enters the apparatus through a half-silvered mirror. Ordinary mirrors on each possible path direct the photon toward a second half-silvered mirror, and then into a detector on one of the two possible paths. Here’s the amazing thing. Fifty percent of the time, the photon passes through the half-silvered mirror, the other fifty percent of the time the photon bounces off the half-silvered mirror. But when the photon, from either direction, reaches the second half-silvered mirror, it always does the same thing. If it passed through the first mirror, it bounces off the second. Every time. If it bounced off the first mirror, it passes through the second mirror. Again, every time. The result is that 100% of the photons end up in one of the detectors, and none in the other detector.

Conventional quantum mechanics explains this as the result of interference between photons along each path. How can this be, if the photons are fired one at a time, with any arbitrary amount of time between firings? That’s left mysterious. It just happens. Somehow.

Even more mysterious is what happens if the photon starts on either path of the interferometer past the first half-silvered mirror. In this case, the photon ends up in each detector about half the time. In the language of more conventional quantum mechanics, starting the photon past the first mirror gives “which path” information (because you know for sure which path it is taking). The result is that there’s no possibility of interference.

Deutsch has a quite different description. He says that for any single photon, it actually does take both paths. One path is real in one universe (really far more than just one), the other path is real in another universe (again, far more than just a single alternate universe). Because any one photon is a multiversal object, it doesn’t actually exist fully in any one of these universes until it becomes entangled with the objects in that universe. This is a very different use of the word “entangled” than the more common usage, meaning two photons in a single universe whose properties affect one another. Deutsch means that the photon is entangled with its detector, the observer of the detector, the observer’s colleagues, and so on. It is this entanglement that forces the photon into one particular universe, forever closing it off from all others. It is this entanglement that makes our world the solid entity we experience. But according to the many universes view of quantum mechanics, our world is only one of an enormous of universes, all of them essentially right here with us, yet forever separate. Wow!

The many universes hypothesis is fascinating, challenging, and deeply disturbing in some ways. I’m not yet convinced, but I am intrigued. One of the most intriguing parts is the fact that every time we look at a photon, an electron, anything, we never, ever find it as a wave. We always find it as a particle. Only in bulk, and after the fact, and by analysis, does the wave nature of an object ever appear. We never find part of an electron, but only whole electrons. This might make it seem that the null hypothesis ought to be that these particles really are always particles, whether we’re looking or not. But more below. As I said, I’m intrigued.

More intriguing than chapter eleven, maybe, is the discussion in chapter twelve. Here Deutsch describes what makes the Copenhagen interpretation a bad explanation (or even not an explanation at all), and just how it was, historically, that physics abandoned the idea of explanation when it came to quantum mechanics. Deutsch describes how the Heisenberg and Schroedinger approaches to quantum mechanics were both found to give excellent (and identical) predictions if, and only if, a rule of thumb was added to them. That rule of thumb essentially says, “abandon all possible results but one.” Which one? You can’t know until you measure. If you measure, you discover a single result. If you repeat the experiment again and again, you get a spread of results, exactly the spread predicted by the mathematics of the theory.

Then, Deutsch says, “disaster struck.” The disaster was that scientists became satisfied with the predictions. Instead of trying to explain the spread of results, they decided to “shut up and calculate.”

In retrospect, Deutsch says, we can see why the rule of thumb works. It’s because the measurement we make happens when the quantum particle (the photon in the Mach-Zehnder interferometer, for instance) becomes hopelessly entangled with one particular universe – our universe – making the quantum effects too small to measure. In some cases (like having the photon hit a screen) the measurement is the entanglement. So of course we always get exactly one photon.

Deutsch then describes why the bad philosophy of the Copenhagen interpretation was so disastrously bad. It wasn’t just that it didn’t explain anything; it actually prevented the growth of knowledge. It declared itself complete, and immune to criticism. It embraced a contradiction, that a photon could be a particle and a wave at the same time.

Deutsch points out that in a sense this is true, but the explanation is bad, or missing altogether. The explanation (according to Deutsch) is that the photon is a particle, but exists in multiple universes, and shares space with many, many “fungible” copies of itself. “The entire multiversal photon is indeed an extended object (wave), while instances of it (particles, in histories) are localized.” (p 275) This is what “wave-particle duality” actually means, at least in the many universes interpretation.

Viewed in this way, many universes is not so extreme. It’s very much like Feynman’s “sum over paths” approach. Perhaps sum over paths just doesn’t follow all its consequences completely? Perhaps those consequences are many universes?

I especially like how Deutsch compares the “studied ambiguity” of the Copenhagen interpretation with how we approach almost every other field of science.

“(W)e do not speak of the existence of dinosaurs millions of years ago as being ‘an interpretation of our best theory of fossils’: we claim that it is the explanation of fossils.” (p 281)

The most powerful part of Deutsch’s argument may be his blasting of the consequences of Bohr’s “studied ambiguity” when promulgating the Copenhagen interpretation. Those consequences are the proliferation of claptrap and silliness that surrounds quantum theory among people who don’t know any better (or maybe who should know better), those “who want to defy reason and embrace any number of irrational modes of thought.” (p 276) It comes directly out of the silliness of saying that quantum mechanics is just too weird for us to understand. Why come to that conclusion when an understandable (though admittedly still wild) idea is out there? Why, in short, should Copenhagen, with its impossible-to-grasp duality (which has never actually been observed either) and its mathematically inconsistent collapsing wave functions, be the default position? Particularly when that default position seems, at its very heart, to prevent progress from happening?

The deepest theory we have ought to do better. As I said, I’m not yet convinced of many worlds. But I am listening.