Monday, August 25, 2008

Evolution vs. Thermodynamics

My uni friend Jimmy posted this link online yesterday, and I thought it'd be good to share it with you all. This article, from a math professor in the States, Granville Sewell, states that evolution cannot be possible due to some strange violation of the second law of thermodynamics; specifically that the increase in orderliness contradicts statistical likeliness. While he makes no reference to the validity of intelligent design, there is no doubt that he would find God creating the world in 7 days more probable than evolution.

I did not do so well last semester in thermodynamics, and even then, the course was a introductory course. Nevertheless, though I find some of his theories highly suspect, I have trouble putting on finger on what that is. He makes references to "simple examples" in his book, which we have no access to, and then proceeds to extrapolate these theories to the general case, saying that open systems are just as unlikely as closed systems to tend towards order, even with the greater multitude of tangible and intangible factors.

Arguing that our current situation, and all the steps leading to our current situation, is statistically improbable requires the knowledge that the other situations were probable, but how can we know that if we are currently at the steps we are. It is not as simple as rolling a die 8 times and realizing that our eight 1's in a row was impossible, because that would require knowing the other possibilities. Perhaps all the other universal situations were just as improbable, but equally improbable. Think of it as a six-zillion-sided die, where our current situation is just as crazily ridiculous as any other.

The problem with these counter-arguments is that they have the same flaws as the arguments, in that we somehow have to know the "whole picture of the universe", which is impossible. That leads me to my current dilemma, which is that I know that arguing evolution is impossible from a thermodynamic perspective is in itself impossible... but why?

Really, I'm starting to lose sleep over this stupid question. Gah!


Sloth said...

The solution to your conundrum is that he's misapplying a very specific rule (thermodynamics) to a wholly innapropriate situation (evolution).

Thermodynamics only states that closed systems tend towards entropy. That rule has basically no application to evolution, which deals with massive-ass systems (ecologys and the biology in same) with no easily-definable limits.

All evolution needs to work is gene mutation and replication. That is genes mutate causing species differences, some of which provide a reproductive advantage which is then naturally selected for. The mistake is that it looks like - and is often refered to as - 'adaptation', that is the organisms adapt to thier environment. While this might be the case behaviourally, physiologically all that's happening is random mutations.

While deliberate adaptation looks like it might break thermodynamics (though there's a good argument that it doesn't), random mutation doesn't even come close to violating that law. Indeed; randmoness is an integral part of our understanding of the laws of thermodynamics.

Again, the increasing complexity of organisms, to the extent it exists, is perfectly explicable by references to randomness plus natural selection. There's no need to suggest it somehow breaks the laws of physics, even if it could be shown that those laws applied directly to such a large and varied system.

codespyder said...

Thanks, Sloth. It's much clearer to me now. Originally, my arguments against the article were similar to yours, but far more vague. I made the mistake of assuming that evolution within the entire frame of the universe itself was a closed system, which we don't know to be true. That's what stumped me.