August 20, 2003


Posted by Arcane Gazebo at August 20, 2003 6:56 PM Link of the day: A Slate piece on alternate universes. The article well summarizes the interesting issues related to this concept.

Naturally, I have a few comments of my own.

First, my own view on alternate universes: I am reluctantly an adherent to the "many worlds" interpretation of quantum mechanics. I say "reluctantly" because I am sympathetic to the Occam's Razor argument. However, I discovered that the way I think about quantum mechanics is entirely equivalent to many-worlds, and I can't escape that without giving up certain essential assumptions that allow me to make any sense at all of the mathematics. So I had to bite the bullet on that one.

I should point out that alternate universes in the quantum mechanics "many world" sense doesn't explain fine-tuning of the universe for intelligent life; all the fundamental parameters like Planck's constant are the same in all the different universes.

The article points out that some multiverse concepts could pose problems for natural theologians who would like to explain fine-tuning with a Creator. But I would think a bigger problem would be that a Creator isn't an explanation for fine-tuning. It just substitutes one question for another; if the answer to "Where did the universe come from?" is "God", my next question will be "Where did God come from?" And this is actually a harder question! Before we had a nice mathematical universe described by a handful of fundamental equations, and now we have to explain some intelligent being with the power and desire to create such universes. And if the reasoning was good in the previous case, it should be good now: the Creator had a Creator... and off we go into infinite regress.

In other words, if we are satisfied stopping at "In the beginning was the Word, and the Word was with God, and the Word was God," we should be even more satisfied to stop at "In the beginning was the Big Bang." The Slate writer goes further, and argues that many universes really is a simpler case than one universe plus a Creator, and in a certain sense I think he's got a point.

The sense I have in mind is this: one doesn't need a creator to explain logic and mathematics; these things just are, and apply in every universe no matter how many there are. There's no way one plus one could not equal two, since this is just a logical consequence of the definitions of "one", "plus", and "two". If I draw a solution of a differential equation on a blackboard, I didn't "create" the solution - it always existed along with all the other solutions and all I created was a visual representation of it. So, if the universe can be described by some mathematical formalism, it and all others like it already exist as part of mathematics. If we think mathematics exists, we get the "fine-tuned" universe plus all the alternates for free, without needing to invoke either a divine creator or some kind of "natural" creation process.

I've always liked this idea, but it has a huge philosophical problem. The reason we can do science at all is that the laws of nature don't change with time (or if they do, it's in a slow and predictable way). But there are many more mathematically conceivable universes that aren't time-translation invariant, and in fact change the rules randomly, violently, and on all timescales. If all the mathematically possible universes exist in the same way as this one, we are far more likely to be living in one where the laws of physics just change one day - which means inductive reasoning doesn't work, and we might as well give up on science. (See also: the grue paradox.)

So, I've had to shelve that particular multiple-universe idea. In the end the only reasonable position on the origin of the universe is agnosticism - it really is something unknowable and beyond the reach of science.

Which brings me to one final comment on the fine-tuning problem: this is not a problem at all. To ask "why is the universe fine-tuned for human existence" presupposes some knowledge about how universes originate that the asker certainly does not have. Specifically, if I claim there is something unlikely about the fact that physical parameters fall within the narrow range necessary to sustain life as we know it, I am making a statement about the probability distribution of physical parameters. Actually, it's worse than that: saying that the range of parameters is "narrow" assumes information about what kind of universes are possible. These statements take for granted that all conceivable universes are in some sense "possible", and that additionally there is some kind of broad distribution over parameters. Of course these assumptions are entirely unjustified.

The real question to ask is, why does our universe have these particular parameters? But this question takes out the fine-tuning element. I don't think one necessarily has to be totally agnostic on this question (as opposed to questions about origin), since a unified theory of physics may reduce some numerical parameters to conversion factors between units, as relativity did for the speed of light. This only gets you so far, of course; even with the simplest possible theory we'd be wondering why that was the right one and not some other. Tags:

On this subject you will not want to miss the piece in today's Science section of the Times concerning the tendency, as I understand it, of string theory to prove too much.

Posted by: Jeremiah Spur | September 2, 2003 9:40 AM
Post a comment