Why Evolution Is Not Random.

One of the tricks Creationists like William Dembski use is to tell their victims that evolution would need a hugely unlikely sequence of random chance events to occur to produce any given evolved entity, like a cell, a metabolic process or an organism such as a human being. They then produce some simple maths to seemingly show that the probability of this parody arising by chance is virtually indistinguishable from zero by multiplying all these tiny fractions together to get even smaller ones, and claim this shows evolution is, for all practical purposes, impossible.

Of course, this ignores all the principles of the real theory of evolution by natural selection and presents it as a claim that organisms simply popped into existence fully evolved, pretty much the way Creationists like to imagine it happened, with only the magic missing. An old trick of the charlatan is to redefine the thing being attacked so it becomes an idiotic parody that no sane person could believe, then attack that parody.

They then sit back and rely on the parochial ignorance of their victims to lead them to conclude that therefore it must have been the locally popular god who provided the magic and who did it according to the locally popular creation myth, not realising that they have also now fallen for the false dichotomy fallacy as well as the initial straw man fallacy of attacking an absurd parody instead of the real thing.

I'll try to illustrate how things really happen, and so show why evolution is not a random event, with a simple mind experiment. This analogy has been stripped to its barest essentials so I have had to compromise a little, so please bear with me.

Imagine you have a box of assorted buttons in all shades from blue through green, yellow, orange and red to purple. For this experiment we will initially only be concerned with colour. In reality of course, the buttons will also have other qualities such as size, shape, pattern, number of holes, little stalk with a hole in it or not, metal ring on the back, material (bone, plastic, wood, etc) but for now we are concerned only with the colour.

Now, to make this mind experiment work as an analogy with living things, we need to introduce two unrealistic abilities: the ability to replicate, and the ability to pass on qualities to offspring; in this case the qualities for colour.

The buttons are kept in a box in which there is room for exactly 1000 buttons, so, every time buttons are removed a little room for replication is made and the buttons will replicate to bring the population back up to its maximum.

At the start of the experiment we'll assume there is an equal number of all the different colours. These will have arisen by the random element in this process by small changes in the colour 'quality' information.

Now, you have a member of your family who is keen on knitting and loves knitting blue woollen cardigans. To make two cardigans they remove eight blue buttons. The remaining buttons then replicate to replace the missing eight.

How will the ratio of blue buttons in the box be changed at the end of this process?

Given that each colour started out with an equal number so each colour had exactly the same chance of replicating, removing eight blue ones from the box is bound to mean the chances of a blue button replicating is reduced, so the chance of any of the removed blue buttons being replaced with another blue is reduced and the ratio of blue buttons in the box will probably be reduced. (Note, we can only say probably here because it is possible that all the buttons to replicate are blue, but the probability is high that they won't all be and is in favour of them not being.)

Our knitter then makes another two blue cardigans and the same process is repeated.

Once again the probability is that this results in another reduction on the ratio of blue buttons and of course an increase in the ratio of all the other colours.

It's easy to see how eventually, there will be no more blue buttons left in the box and so no possibility of them being replaced. The button 'gene pool' for colour will have changed and blueness will be extinct, all because of the actions of a 'predatory' cardigan knitter.

So, the factor which pushed this change in gene frequency in an inevitable direction was not a random factor. It was a highly directional. The initially random variation in the information for colour has been given a decidedly non-random direction by selection. The buttons have collectively 'evolved' to make it less likely that they will be selected and the evolution was not only non-random but highly directional. One might forgive an ignorant observer for thinking this looks like someone is intelligently directing things.

Now, to continue with this mind experiment a little more.

Suppose our knitter decides to start knitting multicoloured cardigans and randomly selecting coloured buttons so it is just as likely that, say, red, yellow, purple or green are chosen. Remember, there are now no blues; they've gone extinct.

But, one of the green buttons doesn't replicate quite correctly and produces a rather nasty muddy-brown button, which the knitter never selects because it doesn't go with any of the wool. Quite by chance this button now has resistance to being selected. Note: the 'information' for "don't select me" is no different in quantity to the information for any other colour but the meaning of that information is determined by the knitter not liking muddy-brown. The amount of information has remained unchanged but the meaning of it has changed radically. The button's environment has determined the 'meaning' of the information and that was already present in the form of the knitter's aesthetic preferences.

The longer our knitter continues to select any colour other than muddy-brown, the more muddy-brown buttons will be in the box and the fewer the number of other coloured buttons.

Eventually of course, if the knitter continues to select anything but muddy-brown, the box will contain only muddy-brown buttons, and our knitter will need to adapt or cease knitting. If our knitter is unable to overcome this reluctance to select muddy-brown, the buttons will have won this particular evolutionary arms race, ironically because they were pushed in this direction by the knitter and because the knitter gave meaning to the buttons' changed information. Our ignorant observer might now be forgiven for thinking that whatever was 'directing' events is both intelligent and favours muddy-brown buttons.

So, there we are. The only random event here was the mutation which produced the muddy-brown button in the first place. From then on it was the highly directional, non-random process which ensured the success of the mutation and which caused the evolution - an evolution which was inevitable.

Expand the mind experiment a little more now to include several other factors which determine our buttons, like size, number of holes, material, shape, pattern, etc. Exactly the same principles apply here.

And let's make it more like proper biology by introducing another idea. Most multicellular organisms and many single-celled organisms, include a reproductive stage which involves sharing and redistribution of characteristics so new individuals have two parents and inherit characteristics from both of them. If we stretch credulity for the purpose of this analogy and apply this to our buttons, two buttons would be getting together to produce new buttons. On average, if the population size stays the same, two parents will produce two offspring.

Suppose our knitter only selected large, blue plastic buttons with four holes and shaped like flowers, but at each generation, the 'surviving' buttons shared their characteristics. In this example the numbers of large buttons, blue buttons, flower-shaped buttons, plastic buttons and four-holed buttons would all be reduced and the others would increase accordingly. So we would also have fewer large, plastic, flower-shaped, four-holed red buttons, even though red had not been selected. There would of course be more small, red, round, bone, buttons but fewer small, blue, round, bone buttons, more two-holed buttons, etc.

It should now be easy to see how several different characteristics can evolve in parallel and how selection can quickly change the ratio of particular characteristics in a gene pool, and how the direction and magnitude of the change is not a random process. It should also be clear that the power of selection lies in its ability to quickly amplify the effect of a favourable mutation in successive generations, and that mutations do not have to occur in sequence, as is implied in the Creationist parody described above, but that they can occur in parallel.

It should also be clear that Creationist arguments using information theory and a form of the second law of thermodynamics to purport to show that no new information can arise, are at best based on a fundamental confusion between information and meaning and ignore the fact that meaning is contextual and can change when the environment changes. Like so many Creationist arguments, it ignores observable reality and depends on the credulous ignorance of its target marks; an ignorance which is assiduously maintained by a constant supply of misinformation.

---