Friday, January 30, 2015

Von Neumann not QBist

Blake C. Stacey has a new article on Von Neumann Was Not a Quantum Bayesian:
Wikipedia has claimed for over two years now that John von Neumann was the "first quantum Bayesian." In context, this reads as stating that von Neumann inaugurated QBism, the approach to quantum theory promoted by Fuchs, Mermin and Schack. This essay explores how such a claim is, historically speaking, unsupported.
This paper takes Wikipedia way too seriously.

Quantum Bayesianism, or QBism, is a modern defense of the Copenhagen interpretation, which was the mainstay of the XX century, but is often attack by popularist writers. Mermin advocates QBism partially by arguing that it is a modern new interpretation to compete with other modern interpretations, and partially by arguing that it is the same as what Bohr, Heisenberg, and Schroedinger promoted all along. In spite of his objections, articles on interpretations of quantum mechanics treat Copenhagen and QBism as essentially the same thing.

Von Neumann–Wigner interpretation is also a variant of Coperhagen defined by:
also described as "consciousness causes collapse [of the wave function]", is an interpretation of quantum mechanics in which consciousness is postulated to be necessary for the completion of the process of quantum measurement.
This is a useful term, because a lot of people hate Copenhagen because of the role of consciousness. In this Wikipedia terminology, the von Neumann interpretation depends on consciousness, and the Copenhagen interpretation is the same thing with the consciousness stripped out.

Stacey refuses to say von Neumann was a QBist because he was not a Bayesian. There are multiple probability interpretations, and von Neumann was more of a frequentist.

At this point, you may wonder what the relation to physics is. Quantum mechanics successfully predicts atomic phenomena. What difference does it make how it is interpreted? The Feynman "shut up and calculate" school of thought says that interpretations are unnecessary distractions. When interpretations posit proliferating parallel unobservable universes, it is hard to see why there is any point to talking about such nonsense. If two interpretations have the same physical outcomes, it is hard to see how science can say one is better than the other.

Feynman did not actually say "shut up and calculate", but he did advocate having multiple theories for getting the same results, and criticize philosophical arguments about interpretations.

The metaphysical problem is weirder in the above paper, as the von Neumann and QBism interpretations are essentially the same physically and mathematically, but differ only in the interpretation of what a mathematical probability means. And that difference hinges on consciousness, whatever that is. Supposedly Wigner once remarked that a dog was probably sufficiently conscious to cause the collapse of a wave function (and kill Schroedinger's cat), but a rat was not. He must have realized that he was on thin ice with that.

I think that these interpretations add some clarity to quantum mechanics, but it is a mistake to take them too seriously. Some people think that quantum mechanics is a flawed theory, and it needs a new interpretation to save it. Or we need to get everyone on board some version of many-worlds. I disagree. The interesting questions are the scientific ones, and these interpretational issues barely qualify. We cannot use them in experiments, such as asking a dog to watch Schroedinger's cat and checking to see if the wave function collapses. An interpretation is just a way of thinking about the theory.

1 comment:

  1. Roger,
    I was required as a class project to attend a lecture on Quantum Computing by Dr David J Wineland at my college last night. I think it is fair to say that when you use schrodinger's cat as your primary example of why you are right, you aren't on very firm footing with reality. After all the blather, I noticed what was very discreetly NOT said was that an actual qubit gate had been made to function at all. If you believe that probabilities have actual existence sans the math calculations that generated them, and that there is a magical way to observe them using actual some device, you will believe anything.

    Can you see the moon at right this instant? I like to think it is still there even you cant.

    All quantum mechanical malarkey stems from the simple fact that if you use a mathematics of probability to measure or describe something with, you get your answers in (drum roll please) probabilities. This is not an indication that reality is composed of a probability, it is ONLY an indication of what you described it with.

    You have to really be determined to be silly to use a second hand (or third hand, fourth hand etc..) calculation as a primary cause of anything. Probability is a result of calculation which is a result of a measurement which is the result of the action of being measured which is the product of a measurer using some method to measure something. Get your the order of your cart and horse figured out already.

    An interpretation is also worthless if it does not accurately convey what was said. An interpretation better agree closely with someone else's interpretation of the same thing said or it becomes painfully obvious that no one is actually aware of what was said at all.