Thursday, August 17, 2006

String theory Zen

I'm currently entering reviews of our books into the database, working against a six-month backlog. Nobody here can believe that they originally intended to have me on for only three weeks -- there's just so much to do. Anyway, this is meant to be a boring task but I find it really interesting and relaxing, especially compared to the other things I'm working on. For one thing, since I'm in publicity, the only time I get to hear about potential scientific objections or qualifications to our trade books is when expert reviewers bring them up. (This is particularly true since I'm the token Science Nerd on the marketing staff.) Plus, I get to flip through publications I can't afford to subscribe to, and see what's up.

Now, I'm working out of a huge box -- Dan and I moved our dishes in one exactly like it -- full of muddled papers. Like I said, we haven't had the manpower to get to this stuff for months. So occasionally a magazine will show up that turns out not to have any of our books in it, and I just kind of poke through it quickly and see what interests me. This is how I ended up reading a more or less positive though somewhat snarky review of Leonard Susskind's The Cosmic Landscape in a January Times Book Review section. The book is about string theory, and is amusingly illustrated with a collage picture of stars, a measuring tape (?), and dental floss. I'm not saying I understand string theory, but wow, the artist sure doesn't. (And why should s/he, really.)

The review is written by Discover editor Corey Powell, who is reportedly a great guy and certainly knows plenty about science. One of us is wrong on something, though, because I found myself very puzzled by part of the piece. Powell writes:
Furthermore, it is inevitable that we would find ourselves in a universe well-suited to life, since life can arise only in those types of universes. This circular-sounding argument -- that the universe we inhabit is fine-tuned for human biology because otherwise we would not be here to see it -- is known as the Anthropic Principle and is reviled by many cosmologists as a piece of vacuous sophistry.
Now, I think I'm confused. Is this the Anthropic Principle? (Wikipedia and her internet consorts are of little help on this one, I'm afraid.) Because I always thought the Anthropic Principle was something more like "the universe is this way because that's how humans need it to be to survive" -- a clearly absurd argument -- rather than "of course the universe is such that we can survive, else we would not have survived to ask about it." It seems more akin to the argument that humans have their specific traits (including, perhaps, the desire to believe in a designer!), not because they were made that way, but because the traits they have either were more adaptive than other traits, or came along with adaptive traits, or were sufficiently adaptive to remain. The inevitability, in other words, is in hindsight, not in design.

The pro-design argument hinging on the fact that "If gravity were slightly stronger than it is, for instance, stars would burn out quickly and collapse into black holes; if gravity were a touch weaker, stars would never have formed in the first place" (Powell) actually seems absurd to me for exactly the reason Powell is calling the Anthropic Principle -- and I hardly want to go on record saying I support the Anthropic Principle, so I'd like to get this cleared up! Is it, in fact, sophistry to say that the reason the universe operates on such principles that life on earth can exist is that, should any other circumstances have arisen, we'd hardly be any the wiser? Am I being insufficiently philosophically rigorous here? Am I unknowingly espousing a belief that amounts to intelligent design, or -- much less egregiously -- that demonstrates a vast ignorance of tricky cosmological theories?

I guess what I'm wondering is, if a universe can't support life, and no life evolves to notice, does it make a sound? And should it make a universe that can (and does) support life any more or less plausible to the life in question?


Nick said...

I'm not sure I can answer your questions, but I'd point to this article on Everything2 by a fellow who I can vouch has a generally very good understanding of Physics. He gives the one example of the use of the Weak Anthropic Principle that really makes a whole lot of sense to me:

"in 1937 Dirac pointed out that a certain combination of physics constants h-bar/Gcmπ was rather close to the age of the Universe (about 10 billion years). Equivalently, the ratio of the gravitational force to the electrostatic force between two electrons (about 10-40) was close to the ratio of the time for light to cross a hydrogen atom to the age of the Universe (about 10-40). Dirac took this coincidence as a sign that the physical constants should be varying over time to maintain the relationship as the Universe aged. However Dicke later pointed out that because of the physics that determines the lifetime of stars, you expect an average star to burn out after about a time h-bar/Gcmπ. Combine this with the fact that elements heavier than H, He and Li can only be produced in stellar cores and supernovae and you quickly find that life on planets (which are just condensed stardust) is only possible when the age of the Universe is a few times the typical stellar lifetime. It was inevitable that we would observe that the combination of constants Dirac took was close to the age of the Universe: the Dirac "coincidence" wasn't one."

My basic understanding of the Anthropic Principle (specifically I think it's the "Weak Anthropic Principle") is that the idea is based on the simple fact that the probability of X occuring, P(X), is different than the probability of X occuring given that you know Y occured, P(X|Y). What makes it "anthropic" is that Y is generally something about the existence of humans, or human-like entities, who measure or experience X.

If X could have many different values at many different times or places, this is generally a good explanation of why X has a certain value where Y occurs, analogous to saying that you're observing an anomolously high number of restaurants with crabs on the menu because you happen to be in Maryland. We will observe environments with anomolously high amounts of H2O because human life can only occur under those conditions. There are many other places without much H2O, but we aren't there.

A question like, "Why do the physical constants lie in the small range of values that allow human life?" is much more problematic. The essential thing in the logic I just used was the idea that there are in essence many "trials", while it's improbable that any given place will have water, it's probable that water will exist somewhere. Then I reasoned that people will only exist in one of those regions, hence the probability of us observing a water-rich environment is high. If you believe that there have been many "universes" over time or that there are many pockets of space in which the fundamental constants take on different values, then the anthropic principle might still hold, but if you don't beleive those things, then it's not much use. Either way, it doesn't seem to be teleological.

I don't think I know enough to say much more of use about the Anthropic Principle, but I will say this: I think there is another answer to the question, "Why do the laws of nature take on the very improbable forms that allow life?" That answer is that the probability of those particular laws of nature isn't low, it is undefined. I believe that this question of probability is not well posed, it's meaningless.

Imagine that there is a constant of nature, foo. Furthermore, suppose that foo must have a value between 0.1 and 0.2 in order for life as we know it to exist. Now, the local intelligent design guru says, "Look, foo has to be sooo finely tuned for life to exist. The universe must have been designed that way, becuase it's far too improbable to have happened by chance. Naturally, that might seem reasonable, but let's think about actually calculating the probabilty that foo takes on a value that allows the existence of life.

Well, first, what are the possible ranges that foo can take on? Who knows? We only have one universe with one value of foo. Ok, suppose that we can plug any positive value of foo into our equations and still get something that basically makes sense, so we'll say that it could have any value foo > 0. But what's the probability distribution? How likely is foo to lie in a given range of values? Again, we have no way of saying. Perhaps we can assume that all values of foo are equally likely? No, we can't. Because foo can take on values out to infinity, that probability distribution isn't well defined; the probability of lying in any interval would be zero. But let's suppose, for the sake of argument that we know that foo has to lie in the interval 0.1 < foo < 10. Now we can assume that all values are equally likely and calculate a probability for foo having a value that allows life to exist (namely, P(0.1 < foo < 0.2)=0.1/9.9=1/99). But I claim that this reasoning is still faulty.

See, our result depends entirely on the arbitrary choice of how we write our equation. The laws of nature we're discussing were written in terms of the constant foo, but we could have just as easily defined things in terms of a constant bar, where foo = 1/bar. In that case, the allowed set of values (0.1 < foo < 10) becomes 0.1 < bar < 10, but now the set of values that allow life are (0.1 < foo < 0.2) 5 < bar < 10. If we were to take our laws of nature written in terms of bar and use the same logic as before, that is assume that every value of bar is equally likely, we'd find that the probability of a value consistant with life is P(5 < bar < 10) = 5/9.9 ~ 1/2. If our answer depends entirely on the arbitrary choice of the way we write things down, then our answer is meaningless.

So, even being very generous and granting a lot of things, we find no reasonable way to assign a probability to the situation, and this doesn't even take into account the question of more significant changes in the laws of nature than just varying a few constants, Fundamentally, if we don't know anything about what values a are possible and the relatively likelihood of each, there's no way we can say how likely a certain range of values is.

Wow, ok, that turned into a long diatribe. Sorry. :-)

8/18/2006 1:08 AM  
Nick said... your post now that I'm fully awake, it seems I sort of went off on a tangent. The short answer to your question is that there are several things called "The Anthropic Principle". I think the two ideas that you contrast in your post are referred to as the "Strong Anthropic Principle" and the "Weak Anthropic Principle". I've read definitions of the Strong Anthropic Principle that sound a lot like ID, as you say, but I don't know if those were just written by people with a poor understanding of the concept.

8/18/2006 11:32 AM  
jess said...

*headsmack* E2! I used to be a pretty regular noder there, and used it for everything before Wikipedia got big. Now I don't even think about it unless I'm in the market for a Spenserian sonnet about Ice Cube.

I had gotten the sense from El Intornet that there were Strong and Weak anthropic principles, but there didn't seem to be a strong agreement on how each was defined. I think I'd rather not have these arguments tarred by the same brush!

I agree with you about the probability equations, and I think that's part of the problem. I see the Weak Anthropic Principle (and I'd really rather call it something else!) as boiling down to "we have a necessary and inevitable selection bias." Not in the "doubt all evidence of your senses, science is foolish" sense! Just in a "we can't really draw any conclusions about how everything seems to be perfect for human life to arise, because see above." I think the probability problem may be one facet of this.

Anyway, this was really interesting, no need to apologize. Unrelatedly, you guys should come to an English/physics/math/other mixer at my friend Margaret's on the 9th. It's Grad Pub without the marketing students!

8/18/2006 12:11 PM  
Nick said...

You'll have to tell me your E2 username sometime so that I can look up your nodes.

8/18/2006 4:33 PM  
jess said...

< runs a quick check >

Well, everything that could be embarassing is at least rated pretty high. Please to note I have not noded anything in a year and a half, and most stuff is way older. I should get back on the factual writeup horse, though. Here's my homenode.

8/21/2006 12:22 PM  

Post a Comment

<< Home