Life, complexity, and evolution

Next to the purely religious arguments -- those that boil down to "it's in the Bible, so I believe it" -- the most common objection I hear to the evolutionary model is that "you can't get order out of chaos."

Or -- which amounts to the same thing -- "you can't get complexity from simplicity."  Usually followed up by the Intelligent Design argument that if you saw the parts from which an airplane is built, and then saw an intact airplane, you would know there had to be a builder who put the parts together.  This is unfortunately often coupled with some argument about how the Second Law of Thermodynamics (one formulation of which is, "in a closed system, the total entropy always increases") prohibits biological evolution, which shows a lack of understanding both of evolution and thermodynamics.  For one thing, the biosphere is very much not a closed system; it has a constant flow of energy through it (mostly from the Sun).  Turn that energy source off, and our entropy would increase post-haste.  Also, the decrease in entropy you see within the system, such as the development of an organism from a single fertilized egg cell, does increase the entropy as a whole.  In fact, the entropy increase from the breakdown of the food molecules required for an organism to grow is greater than the entropy decrease within the developing organism itself.

Just as the Second Law predicts.

So the thermodynamic argument doesn't work.  But the whole question of how you get complexity in the first place is not so easily answered.  On its surface, it seems like a valid objection.  How could we start out with a broth of raw materials -- the "primordial soup" -- and even with a suitable energy source, have them self-organize into complex living cells?

Well, it turns out it's possible.  All it takes -- on the molecular, cellular, or organismal level -- is (1) a rule for replication, and (2) a rule for selection.  For example, with DNA, it can replicate itself, and the replication process is accurate but not flawless; the selection process comes in with the fact that some of those varying DNA configurations are better than others at copying themselves, so those survive and the less successful ones don't.  From those two simple rules, things can get complex fast.

But to take a non-biological example that is also kind of mindblowing, have you heard of British mathematician John Horton Conway's "Game of Life?"

In the 1960s Conway became interested in a mathematical concept called a cellular automaton.  The gist, first proposed by Hungarian mathematician John von Neumann, is to look at arrays of "cells" that then can interact with each other by a discrete set of rules, and see how their behavior evolves.  The set-up can get as fancy as you like, but Conway decided to keep it really simple, and came up with the ground rules for what is now called his "Game of Life."  You start out with a grid of squares, where each square touches (either on a side or a corner) eight neighboring cells.  Each square can be filled ("alive") or empty ("dead").  You then input a starting pattern -- analogous to the raw materials in the primordial soup -- and turn it loose.  After that, four rules determine how the pattern evolves:

1. Any live cell that has fewer than two live neighbors dies.
2. Any live cell that has two or three live neighbors lives to the next round.
3. Any live cell that has four or more live neighbors dies.
4. Any dead cell that has exactly three live neighbors becomes a live cell.

Seems pretty simple, doesn't it?  It turns out that the behavior of patterns in the Game of Life is so wildly complex that it's kept mathematicians busy for decades.  Here's one example, called "Gosper's Glider Gun":

Some start with as few as five live cells, and give rise to amazingly complicated results.  Others have been found that do some awfully strange stuff, like this one, called the "Puffer Breeder":

What's astonishing is not only how complex this gets, but how unpredictable it is.  One of the most curious results that has come from studying the Game of Life is that some starting conditions lead to what appears to be chaos; in other cases, the chaos settles down after hundreds, or thousands, of rounds, eventually falling into a stable pattern (either one that oscillates between two or three states, or produces something regular like the Glider Gun).  Sometimes, however, the chaos seems to be permanent -- although because there's no way to carry the process to infinity, you can't really be certain.  There also appears to be no way to predict from the initial state where it will end up ahead of time; no algorithm exists to take the input and determine what the eventual output will be.  You just have to run the program and see what happens.

In fact, the Game of Life is often used as an example of Turing's Halting Problem -- that in general there is no way to be certain that a given algorithm will arrive at a solution in a finite number of steps.  This theorem arises from such mind-bending weirdness as the Gödel Incompleteness Theorem, which proved rigorously that within mathematics, there are true statements that cannot be proven and false statements that cannot be disproven.  (Yes -- it's a proof of unprovability.)

All of this, from a two-dimensional grid of squares and four rules so simple a fourth-grader could understand them.

Now, this is not meant to imply that biological systems work the same way as an algorithmic mathematical system; just a couple of weeks ago, I did an entire post about the dangers of treating an analogy as reality.  My point here is that there is no truth to the claim that complexity can't arise spontaneously from simplicity.  Given a source of energy, and some rules to govern how the system can evolve, you can end up with astonishing complexity in a relatively short amount of time.

People studying the Game of Life have come up with twists on it to make it even more complicated, because why stick with two dimensions and squares?  There are ones with hexagonal grids (which requires a slightly different set of rules), ones on spheres, and this lovely example of a pattern evolving on a toroidal trefoil knot:

Kind of mesmerizing, isn't it?

The universe is a strange and complex place, and we need to be careful before we make pronouncements like "That couldn't happen."  Often these are just subtle reconfigurations of the Argument from Ignorance -- "I don't understand how that could happen, therefore it must be impossible."  The natural world has a way of taking our understanding and turning it on its head, which is why science will never end.  As astrophysicist Neil deGrasse Tyson explained, "Surrounding the sea of our knowledge is a boundary that I call the Perimeter of Ignorance.  As we push outward, and explain more and more, it doesn't erase the Perimeter of Ignorance; all it does is make it bigger.  In science, every question we answer raises more questions.  As a scientist, you have to become comfortable with not knowing.  We're always 'back at the drawing board.'  If you're not, you're not doing science."

****************************************