I Am Of Two Minds

I really have to stop watching the Colbert Report to get my science news! There have been a couple good controversial snippets this week.

While I am very glad that Stephen has come around to believing that the moon landing was real (as he says, the market has spoken), I am not so glad that he has started believing in the reality of multiple universes. While the alternate-universe Stephen was pretty darn funny, the theory itself just doesn't stack up.

First, a little background: As you may know, quantum systems can exist in superposition states, which means they don't have just one value of some experimentally observable quantity, but have a state which reflects a combination of more than one value. For example, if an electron is sent through a double-slit apparatus, the electron may pass, not just through one slit, but through both, producing an interference pattern on a screen on the other side of the slits. Now, one electron by itself cannot produce an observable interference pattern, but the interference pattern tells us the probability that the electron will be found at various places on the screen. If we were to do the experiment a million times, where each of the million electrons is likely to show up is given by the intereference pattern. If we looked at the result of the million experiments together, we would see the full interference pattern that we get when we send a beam through.

One of the reasons that some people find comfort in the many universes theory is that they see probability from a frequentist perspective. That is, they see probability as only describing what fraction of the million electrons will strike at some particular part of the screen. If we want to talk about the probability that the next word I type is "the," then* there seems to be no way to test that probability on a million me's. The many universes theory says there are a million me's in a million different alternate universes, so that each possible outcome can be carried out as frequently as probability predicts.

While probability does indeed tell us what fraction of the million electrons will strike at some particular part of the screen, that aspect of probability is not essential. I do not have to do the experiment a million times for the probability of striking that particular part of the screen to have meaning. I do not have to roll a fair six-sided die a million times for the probability of rolling a one to be 1:6**. The many-universes theory is just a way of visualizing all the possible outcomes that don't turn out to be actual outcomes, and seems quite unnecessary.

In the article*** that Colbert was referencing, Zeeya Merali claims:

According to quantum mechanics, ... particles are described by “wave functions” representing many mutually contradictory properties.

....If, as Everett argued, quantum mechanics is applied to the whole universe, then it too should exist in a multitude of separate states. There would be a “multiverse” of parallel universes – one for every physical possibility.

This argument for multiple universes is also weak. When we talk about a particle being described by “wave functions” (why is wavefunctions in scare quotes?) representing many mutually contradictory properties, we are talking about a single particle being described by a wavefunction representing many mutually contradictory properties. When we apply this to the universe, we are still talking about a single universe, just one being described by a wavefunction representing many mutually contradictory properties of the universe.

The news reported in this article was that someone finally made the multiple-universes theory no longer contradict the results of actual quantum experiments. It is good for a theory not to contradict experiment, but the single-universe interpretation of quantum mechanics does that just fine without inventing infinitely many (uncountably infinitely many) extra universes. Ockham's Razor, CHOP!

^{*The probability just went to zero.
**I may have to roll it a million times to prove to you that it is fair, however.
***The New Scientist, Volume 195, Issue 2622, 22 September 2007, Pages 6-7; doi:10.1016/S0262-4079(07)62372-0}