Dismembering Evolution 8: Avida

In this world of technology no theory would be complete without computer models being used to demonstrate the feasibility of the theory. In other words, program in the parameters of the theory and see what you get. If the simulation supports the theory you are good.

There is one problem though with any computer simulation. It simulates the parameters that are plugged in whether they happened in real life or not. When these parameters are based on massive amounts of assumptions that are not supported by laboratory experiments the program is running entirely off guesses. Even complete fabrications can be programmed into a simulation as any video game proves.

One example of just such a flawed computer program was one Richard Dawkins ran to show that random mutation in the early world could have resulted in the swift changes that are observed in the Cambrian. He selected a verse from Shakespeare, programmed in 27 variables, the alphabet plus a space, and had the simulation randomly change each slot. By randomly I mean random slots and random result. In just forty five generations the selected verse was completed to 100% accuracy. Subsequent runs showed an 80% likelihood of achieving the same result within forty five generations. Pretty spectacular huh?

There is just one teeny-tiny, huge, massive, glaring, fatal flaw to this program.

He programmed it to lock in the letters when the right one fell into place.

That’s right. Once a desired mutation was achieved he removed an essential variable, the possibility of losing the mutation. As both scientists and lay persons here have stated often, evolution is not a one way process. All is capable of being changed in such a way as to advance, or to regress. But don’t take my word for it, just the comments and links left in the discussion of previous articles in this series.

According to evolutionary theory there is no such thing as a locked in change. What is beneficial one generation may be fatal the next due to environmental changes. A mutation may or may not be passed on to future generations. A mutation may or not be repaired by the DNA as it happens, or the mutated cell may simply be attacked and destroyed by the organism’s immune system. What’s more, is the mutation even possible to pass on? Most are not transferable to future generations because the germ cells are not the cells that are mutated. This, of course, does not apply to binary fission such as in bacteria, or to viruses. So Richard Dawkins programmed a fatally flawed simulation to prove a point he could not make honestly. Bad scientists! Bad! No treat for you!

A much more important, and also much newer computer simulation is called Avida. This computer program is truly fascinating.

It has gone through several revisions until it is now the most exciting bit of research simulation for studying evolution there is. Initially, with the parameters being exactly what evolutionary theory states it had the problem of actually wiping out biological diversity. Then I was altered to reward any change whatsoever, and that solved the problem of destroying diversity, resulting in a rich environment of simulated organisms, but also not achieving this result in the time allowed by geologic history.

This problem demonstrates a weakness in evolutionary theory every time someone says that evolution is not random, but sequential. In sequential evolution the number of generation the simulation needed to produce a diversity resembling the Cambrian too until past the modern day. True to Darwinist fashion, rather than looking at evolution for the problem, the programmer looked at the program’s parameters. After much tweaking he decided to bombard the earliest simulated life forms with incredible doses of simulated radiation. This produced a random mutation rate sufficient to produce the appropriate diversity of life in the appropriate time.

Amazing huh?

But there are a few problems.

First, such a high radiation level is shown to prevent abiogenesis by the destruction of amino acids and proteins. Second, it is shown not to mutate cells, but to kill them. Third, eve if something were to evade the death trap of light it would be shielded form the mutating effects of the radiation. Third, assuming something survived completely unshielded and was randomly mutated, and successive generations continued the trend for a hundred million years or so, this destroys the notion that evolution is not random. It actually reduces the soft definition of evolution that is becoming fashionable, “The change in the frequency of alleles” to bull-spit because evolution absolutely required massive random mutation to do anything to begin with. It takes early evolution completely out of the soft definitions, and even takes it out of the idea of natural selection in the classical sense because extreme mutation is not compatible with that theory.

Scientists studying evolution are going nuts, absolutely raving about Avida. Never mind that the parameters it requires to operate according to evolutionary theory are actually in conflict with evolutionary theory. Never mind that the source of the conflict is the generally acknowledged problem of UV radiation of the kind required in this simulation is simply lethal to life rather than mutating it into many forms. It is the first computer program to successfully produce a model of evolution based entirely on random changes supported by a reward (a simulated environmental factor, actually, it is more food).

What about it’s failures? What about its failure to punish the occasional change with a reduction in food, reduced viability, or even death? What about the lethality of the early parameters? What about the way it actually decimates the concept of evolution NOT being random? What about the conditions of early evolution being completely destructive to abiogenist as well? None of this matters because it provides a computer model that works.

More importantly to the skeptic, what about the parameters of the program in the first place?

This program follows a lot of assumptions, just like evolutionary theory itself. And just like evolutionary theory, this program has its serious flaws when compared to reality. To my way of thinking, the vital change that was needed to the early environment, which is generally considered to be a reasonably accurate simulation of the early Earth, and is also a set of conditions that would wipe out all life and pre-life, actually provides AGAINST evolution and abiogenesis. Without abiogenesis evolution cannot happen. If early life is destroyed by radiation the evolution cannot happen. Current laboratory experiments show that the radiation levels that are simulated absolutely destroy all life and pre-life. Lesser radiation does not provide the required mutation levels to form the diversity of life in the required geologic time. These basic incompatibilities only seem to bother skeptics, like myself, and not the highly educated scientists who should be able to look at this and see the exact same issues.

So, what do I think about computer simulations? I think they are proven to be a powerful tool for demonstrating what will happen within a given set of parameters. I think that if the parameters are demonstrably good then the tool is good. I think that if the set of parameters is questionable then the tool is not much good at all, and only shows an unrealistic simulation of what the programmer wishes were true.

And, no, there is no computer simulation to date that provides a working model of evolution under realistic circumstances.