Creation and Evolution Blog

This blog has been superceded, and is only here for archive purposes. The latest blog posts, depending on topic, can be found at one of the blogs at the new location!

Discusses creation and evolution, mostly from a creation perspective.

Thursday, June 23, 2005

Evolution, Chance, and Design (T.O. CB940)

One of the more common criticisms that creationists have to evolutionists is to believe that all of this arose by chance. Evolutionists are always quick to say "natural selection is anything but chance." While I wholly agree with the evolutionists statement, I want to take the time to point out why it doesn't answer the creationist's criticism at all.

This includes a rebuttal of Talk.Origin's response to CB940, though my inspiration for writing this was elsewhere.

So instead of starting out addressing them point-by-point, let me give you a more general discussion.

Is Evolution "Chance"?

In evolutionary theory, there are mechanisms which can be included as evidence of evolution and mechanisms which cannot. For example, if a genome were designed to change itself in a specific in response to a stimuli, or in any specific manner, this would not count as evidence for evolution, because the process for doing this is already there. It is "designed in" for lack of better terminology, whether you believe that design to be evolutive or creative. The evidence could be taken easily either way.

What evolutionists say, however, is that often organisms change in non-specific ways. There are many options for non-specific changes -- nearly an infinitude of them. In each organism, some amount of non-specific changes occur, and those changes which allow the organism to reproduce more are in fact reproduced more often, and those changes which inhibit the organism's reproduction are in fact reproduced less. If no changes were non-specific, and really, if most of the larger changes aren't non-specific, then it is no longer evolution that can be claimed, but a machined designed in a specific way.

So, you see, there are two sides to this -- a generative side and a selective side. The selective side is definitely non-random. HOWEVER, for something to count as evolution and not a "built-in", the generative aspect must be non-specific. When most people talk about randomness they usually aren't talking about randomness in the sense of a mathematical randomness, but simply meaning that it occurs in a non-specific way or direction. As such we can definitely say that while natural selection is surely non-random, the generative mechanism for evolution surely is, otherwise it would have to be progressive creation rather than a true evolution.

To see this in play see Mutations Are Random and Misconception: “Evolution means that life changed ‘by chance.’ ”. In fact, that second link has a great summation of the issue:

Random mutation is the ultimate source of genetic variation, however natural selection, the process by which some variants survive and others do not, is not random.

Can Random Change, When Selected Non-Randomly, Produce ALL the Diversity of Life?

Dawkins has attempted to explain how this is reasonable using computer simulations. The simulations he proposed in The Blind Watchmaker are about generating the phrase "methinks it is like a weasel" by random typing. He proposed typing in random characters, and then letting it "breed" by duplicating itself with minor changes. Whatever phrase is closest to "methinks it is like a weasel" is kept and the rest are discarded, and it starts over from there. However, this does not get at the main problem in two separate places: (1) the intermediate forms make no sense whatsoever. For life to evolve, the intermediate organisms have to make some sort of sense for them to propogate at all (this is another reason why I disbelieve evolution -- the idea that all intermediate forms are stable from molecules to man is just foolish). (2) the model of evolution is actually quite theistic, as there is a model that has to be met.

In Climbing Mount Improbable, Dawkins tries a different, and much better approach. He has models of such things as spider webs. Each spider has a different set of techniques. Each one is slightly modified from generation to generation, and the ones that survive are the ones that catch the most flies. This is a much better model, showing natural selection in a much truer light. However, what Dawkins fails to realize, and others who say that "evolution is not random" also fail to realize, is that Dawkin's model has pre-coded certain types of web-spinning available, and that is what is varied by. By doing this, he only confirms what creationists already believe -- that there is quite a lot of variety in the world, and species can mix it up and produce different offspring. Natural selection only adds the not-quite-as-ingenious-as-they-think compendium that "dead things don't reproduce".

So how can I say that Dawkin's model is good, but still think that evolution is wrong?

Why Dawkins Changes Are Inadequate for Proving Evolution as the Primary Cause of Biodiversity

First of all let me point out that creationists don't believe that all of evolution is wrong. For more information on that point, see another article of mine. What we believe is that new structures are not produced from nothing without the hand of an intelligent agent. Dawkin's simulation used only a shuffling of pre-existing parts, not the creation of new parts. In fact, that's all a program can do -- rearrange existing parts. So what would be required of Dawkin's program to show the kind of evolution that creationists don't believe is possible?

It would require the program to show something new happening. Something not contained within the original pattern of the program. For example, a spider deciding, whether after one or a million gradual generations, "you know what, I think instead of using this web to catch flies, I'm going to do something else entirely." Or likewise, taking the nearest non-spider relative of spiders, and, without precoding the special spider information in the program, using any natural selection method you want, showing how a spider can be produced from something else. Show the generation of the web material itself and the idea of forming a web without any reference to such activities in the program. Also remember that evolution has to explain the existance of so many genetic codal variations (not the sequence, but the base->protein coding itself), how such a change in code would not utterly destroy any existing organism.

Likewise, for the articles I've investigated in Talk.Origin's post, the same problems apply. For example, in the robotics one, they have the program select the best, most mobile arm from a selection process (selected for movement) from a collection of arms randomly assembled from parts. That's all well and good, but evolution claims that the parts themselves were generated by the same process, without any reference to arms or movement in the code for it. Not just a shuffling of existing parts, but the creation of new parts for something not even intended in the code. For example, if the computer selection for movement had discovered a new material when it is not even programmed to look for new materials, and had then selected the best one and even knew the appropriate way to place the new material within the newly created part which it was not programmed to build, and perhaps even found something that performed the task at hand better than the specific movement mechanism it was trying to achieve.

Under those circumstances, I would start taking notice, but not in this simple shuffling of existing parts which both creationists and evolutionists have agreed upon since time immemorial. Remember, breeding was done for thousands of years by creationists, and it was never thought to be inconsistent, because it was not variation, or even beneficial variation, that creationists objected to, but getting something from nothing.

Examples of How This Does and Doesn't Work

If this sort of mechanism really worked, for example, it would make my job as a computer programmer a whole lot easier. To enhance a program, I could simply write another program to make random changes to it, and then test each one to see if it was a better program. I could remove the need to ever be creative! In fact, it could probably make changes I never thought of. Of course, programs don't work like this. The only thing you'll get in this case is a broken program. Now, it is possible to write a program that made changes to itself -- both randomly and in response to stimuli. However, the mechanisms for change and the categories by which it changes would be pre-coded in the program -- the result of design. Perhaps the exact combination would be a chance event, but it would be within the context of a designed mechanism, not outside of it as evolution claims.

So, am I claiming that all randomized mutations will kill the organism? No. In fact, it is the result of good design that our bodies are able to compensate for faults. However, to view the faults as a fundamental part of how the body came about would be foolish.

What About Beneficial Mutations?

So are all random mutations negative? I would say no. Many say, "well, that shows it -- evolution is at least possible, because you could always just imagine a sequence of beneficial mutations to lead you from A to B". However, that greatly exaggerates the character of such changes. First of all, as I mentioned earlier, it would require a sequence of stable intermediates. That such intermediates exist is only theoretical, and I think highly doubtful. However, let's look more precisely about what beneficial mutations look like.

For example, for a temporary beneficial modification, if your backup system was disabled, that would save electrical energy and manpower for other things. It (a) wouldn't be really helpful in the long run, and (b) wouldn't go any further in explaining how the backup system arose, but it could be classified as a beneficial mutation. But no matter how many of those stack up, you still won't be able to convert an image-processing program into a word processor through random changes, no matter what your selection process is (unless you do a non-evolutionary one, like for "methinks it is a weasel").

Information Theory Requires an Intelligent Source for Information

This is always the problem that evolutionists will run into with DNA, because it is a code. As Arthur Wilder-Smith pointed out, the kind of diversity that exists in the world is the result of novel information implanted into life. Life + Time + Energy will not get you the kind of changes required for the types of fundamental changes proposed in evolutionary theory, with any evolutionary mechanism for selection. Time + Energy + Life + Information will, but true information is always imposed by an intelligent party. Dembski calls this the law of conservation of information.

Can code be rearranged, perhaps at random? Certainly, but notice that the rearrangement mechanism rearranges in such a way as makes sense to the organism -- in a designed way, according to designed, pre-built categories. Now, these mechanisms can lead to a vast amount of variation. The variation possible in people from just two parents with no genomic modifications is more than the number of electrons believed to be in the universe, with every member still being fundamentally human. Add in rearrangements, and you get even more. However, all of this variation occurs within the categories and specifications of the code. The genomic modularity hypothesis (which is enjoying more and more evidence all the time) says that the genome can even rearrange itself according to predefined specifications and categories. However, when the code undergoes a random mutation, then the system becomes less stable. Like other codal systems, it cannot withstand too many unplanned changes before becoming unstable.

The Creationist Hypothesis -- Genomic Modularity

I've mentioned genomic modularity before on this site, but I think, since we're talking about genomic change here, it's good to expand upon. Genomic modularity is the creationist hypothesis that genomes are essentially modular in different areas. That some genes are built for changing in specific ways in response to specific events. There is more and more evidence for this, as we are finding hotspots for change in many genomes, and also finding that some organisms can invoke as-yet-not-understood changes in result to environmental stress, like the SOS mechanism of bacteria.

Now, I'm sure that evolutionists will say that evolution can create such changing mechanisms themselves. Hopefully the previous argument will show that this just doesn't make sense. However, granting that, let's figure out if there are any sort of adaptations that evolutionists might even agree that evolution cannot do. I hate to be presumptuous, but I would say that evolution would not be compatible with full-fledged mechanisms for handling an environment that the organism had never experienced. For example, if there was a full-fledged mechanism for handling life in the vacuum of space within an organism, we would find it odd for that to have evolved in population that had never had any contact with space. This would clearly indicate that there was a pre-coded design for doing so, and such a designer.

Some links about genomic modularity from

Does the Nylon Bug Prove that Evolution Works?

This has a lot of relevance to things such as the Nylon bug, which evolutionists have pointed to as an example of evolution at work. The logic is essentially this: (a) nylon is new, (b) bacteria can't possibly have always eaten Nylon, (c) therefore this nylon-eating bacteria must be the result of evolution, and (d) we even know which genes mutated and in what ways, and it even includes a frame-shift.

That is all well and good, but without further experiment we cannot know if this was the result of a designed mechanism or if it is truly evolution at work. For example, it has been pointed out that there were many transposons at work here. Transposons form an integral part of the genomic modularity hypothesis, as essentially "work units" for change. Likewise, the idea of a frame-shift is not unusual for bacteria, as many bacteria read multiple enzymes from a single strand of DNA through frame-shifting. So the creationist hypothesis would be that the bacteria was built to modify its digestive protein based on the abundance of available material. When Nylon abounds but no normal food source exists, the bacteria alters its genome to eat on whatever is available.

Some facts in support of the creationist hypothesis:

  • The change happened quickly -- it is not widely known, but creationists usually propose faster mechanisms for biodiversity than evolutionists -- it's just that, as this essay points out, there are limitations and boundaries to that diversity.

  • The change utilized known transposons, which are one of the keys to genomic modularity.

  • The change affected two genes in concert. Evolutionary theory would have this be highly unlikely in such a short timeframe, but makes perfect sense as a predefined mechanism for change in response to food supply.

  • The new genes were significantly changed, not the gradual kind of change that evolution suggests. Remember, evolution agrees with creationism that completely random generation of genes is ludicrous, therefore such a large change necessitated to produce the enzyme in such a short time without beneficial intermediaries is not indicative of evolution. Evolutionists claim that this was a simple frame-shift mutation, but in fact there were many more changes than that.

Two interesting links on this are:

So how would one test whether the creationist or evolutionist hypothesis was correct? By taking the non-modified form of the bacteria and adding it to nylon in several separated populations. If each population undergoes essentially the same change, then this is obviously a directed mechanism. If only some populations survive, and each that does survives in a completely different way, utilizing completely different adaptations on completely different genes, while it does not completely invalidate the creationist hypothesis, it lends much more credence to the evolutionist one.

Also of interest is one of the concluding paragraphs of the AiG article mentioned above:

P. aeruginosa was first named by Schroeter in 1872.10 It still has the same features that identify it as such. So, in spite of being so ubiquitous, so prolific and so rapidly adaptable, this bacterium has not evolved into a different type of bacterium. Note that the number of bacterial generations possible in over 130 years is huge—equivalent to tens of millions of years of human generations, encompassing the origin of the putative common ancestor of ape and man, according to the evolutionary story, indeed perhaps even all primates. And yet the bacterium shows no evidence of directional change—stasis rules, not progressive evolution. This alone should cast doubt on the evolutionary paradigm. Flavobacterium was first named in 1889 and it likewise still has the same characteristics as originally described.

So, we have a fast-mutating species, and after millions of generations of reproduction, it still retains the basic properties as originally described, and is still identifiable as itself. You may disagree, but I find this quite evident of the idea presented in this essay -- that information cannot arise from nothing, but can recombine in specific, preprogrammed ways for specific purposes, but remains bound to those mechanisms and categories as they were originally designed.

Clearly the mechanisms of genomic change are long overdue for study, as evolutionary assumptions have plagued them (namely, presuming the lack of predetermined adaptation mechanisms). However, for the short time that studying them has been in vogue, the creationist hypothesis seems to be bearing fruit.

This page is powered by Blogger. Isn't yours?