Why Natural Selection is Not Random

Update: Read this article instead.

Every once in awhile (read: all the damn time), a creationist will say evolution is random. Sometimes they say natural selection is random (the words are rather interchangeable among some creationists). But one creationist does us one better and calls both of them random (and the Big Bang, too). So here is an article I wrote quite a few months ago on the topic. The first couple grafs were mainly meant to be topical, so at this point they’re a bit out of date. Deal.

Why Evolution Is Not Random

During a CNN June debate, Republican presidential candidate Mike Huckabee raised his hand when asked whether or not he accepts the theory of evolution. More recently, the Florida Board of Education spent several months deciding if the mere word ‘evolution’ should appear in the curriculum. After many debates, a compromise was met where evolution was referred to as only a theory, not a fact (gravity is also a theory, not a fact). In the Spring 2008 Ben Stein will revive his career on the silver screen. But rather than asking if anyone has seen Bueller, he will be questioning the motives of the scientific community at large. An overwhelming majority of biologists regard the notion of intelligent design – the proposal that life is so complex there must be a creator – as unscientific. Ben Stein sees a conspiracy.

Behind all of these cases is a fundamental underpinning: the desire to bring more people to God. But what is often accepted is the erroneous means to this end. The very public war against the theory of evolution has brought many of these means to light for evolutionary biologists, the crusaders and rottweilers of Charles Darwin’s revolutionary theory.

Perhaps the most vibrant means is the argument against plausibility. To be at all likely, evolution cannot be a random process. Yet this is exactly the case made by many creationists and, indeed, is one of the more popular starting points in a stance opposing the theory of evolution.

One of the reasons creationism could be considered plausible is that it makes complex life likely. If a supreme being exists which can do as he pleases and has the means, then why not create life? This does fail to answer the nature of the origin of a being complex enough to create life (and presumably the Universe), but all things equal, evolution does not address the issue of the origins of life (nor did Charles Darwin ever intend for it to do that). So if one is to parallel the situation, it is well enough to side-step answering the origin of a supreme being for our current scope.

So it follows that if creationism, from at least a certain point, makes complexity likely, then the creationist argument that evolution is random must have a basis in opposing the likelihood of complex life forms. Dr. David Menton of the $27 million Creationist Museum in Kentucky and graduate of Brown University with a Ph.D. in cell biology, puts the creationist standpoint succinctly, saying “Evolutionists feel vulnerable to evolution being pure chance.”

But what of “pure chance”? Evolution consists of many mechanisms, but the two big driving forces are natural selection and random mutation. (To be fair, random mutations should be considered more as just a force rather than a driving force.) So why do some consider these mechanisms to be random? Dr. Menton appeals to the idea that “science is built on a statistical foundation.” Natural selection and random mutations do not result in complex life forms because such occurrences are greatly improbable. Answers in Genesis, the group which runs the Creation Museum, explains further on their website, http://www.answersingenesis.com. “The probability of the chance formation of a hypothetical functional ‘simple’ cell, given all the ingredients, is acknowledged to be worse than 1 in 1057800.” In other words, evolution is about as likely as all the atoms in the Statue of Liberty moving in one direction and then the other, making her appear as though she was waving to all who came to America. It’s possible, but so unlikely that it isn’t worth devoting much thought.

So if evolution is such a stupendously unlikely thing to happen, then why do we give it any credit? Why bother with such odds? If evolution is unlikely, then a mechanism which provides a path to complexity is necessary if the theory is to survive scientific scrutiny – nay, if it is to survive any scrutiny. Natural selection is the answer for most biologists. Ken Miller, a professor of biology at Brown University, perhaps best known for his testimony in the ‘Intelligent Design’ trial (Kitzmiller v. Dover Area School District) in Dover, Pennsylvania (and subsequent appearance on Comedy Central’s The Colbert Report), but also famous for his opposition to creationism, is one such biologist.

“I have no idea why someone would take a term like natural selection and say it is random”, said Miller when reached for an interview.

Miller sees natural selection as one of the essential paths to complex life forms. Such a mechanism gives species the ability to filter out what doesn’t work and leave what does. Professor Miller echoes this notion, saying “[n]atural selection is a distinctly non-random process that acts as a sieve through which genetic changes are filtered.” Just as a sieve filled with various rocks will not end up filtering out its contents randomly, natural selection does not filter organisms randomly.

But how else can it be said natural selection is non-random? In The Origin of Species, Charles Darwin compares it to artificial selection. That is, when humans breed, say, dogs, for particular traits, they are applying a form of selection pressure to a phenotype (a particular dog or particular dogs). This in turn results in the great variety we see among our beloved pets. The key difference here, however, is that this form of selection had a particular goal in mind, i.e. floppy ears, sleek body, fluffy coat, wrinkly skin, etc. Humans were able to apply their foresight and consciousness to the reasoning behind the selection. Nature does not do this.

This notion that natural selection is both a non-random process and an undirected one at the same time can lead to confusion. The concept is essentially that this mechanism lends itself to increasing complexity because it builds in cumulative steps. For a step to be cumulative, it (quite obviously) must be based on the previous step. A random process does not lend itself to cumulative steps because, by definition, it is not based on anything. So in this way natural selection is non-random. But it also does not look to end in the phenotype of a tiger or a bat. It has no conscience, merely results. For this reason, it is undirected.

But the second key ingredient in evolution is random mutation. As Jay Labov of the National Academy of Sciences points out, “[n]atural selection acts on things that are already there.” Without random mutations, there isn’t much there; certainly not enough to account for the great genetic variation seen within species today.

There is dissent, however, from the creationist side. Dr. Menton certainly agrees that natural selection can only act on what it is given (“I believe [it] occurs. I believe in it completely”), but he disagrees that the genetic variation is available for one species to become another. This is because “[r]andom mutations do not provide for the raw material for novel information. It’s like going to Midas and asking for a dozen yellow roses. They just aren’t there,” he says. Without these genetic changes, “[w]e don’t see natural selection producing novel features.” Menton goes further to add that something like a reptile does not have the raw material to produce the features, such as wings, which are seen in birds.

The first issue of whether or not random mutations can add novel information can be answered in day-to-day life. Mutated animals (including humans) are fairly common. A person with an extra finger or a snake with two heads are both examples of organisms which have mutations. These are deleterious (bad) mutations, but they aren’t the most frequent. More commonly, neutral mutations occur. These aren’t particularly acted upon by natural selection because most genes tolerate changes quite well, according to Miller. Sometimes, however, a gene will mutate and it will be beneficial. It may extremely slight, but if it offers any survival advantage at all, it is more likely to survive the sieve of natural selection. For example, a mutation which makes a bacterium immune to antibodies will quickly spread throughout the population.

A second issue is whether or not natural selection can produce novel features. Assuming random mutations do not provide for novel information (they do), natural selection can still produce novel features. Dr. Menton’s example of reptiles and birds works perfectly.

“Reptilian ancestors of birds had wherewithal to produce feathers,” says Miller. When speaking of the more than dozen dinosaur fossils which show feathers, he continues, “One (Shuvuuia deserti) has tested positive for the major protein found in bird feathers.”

What does this mean? Simply, ancestors of modern day reptiles had the information to create novel features. But it is “[e]nvironmental factors [which] may turn genes on and off,” says Labov. Whether or not the genes needed to create the particular feature of feathers show up in a phenotype is determined by need, which is governed by natural selection.

Anne Holden, staff member at the National Center for Science Education, further supports the point of natural selection having great genetic variation with which to work by pointing out that our “DNA can recombine and does recombine during fertilization.” The genome of an offspring is a combination of its parents’ genes, but the way in which recombination can occur is impossible to number.

Holden further cites the adaptive radiation of Darwin’s finches on the Galapagos Islands. As a result of the variation within every organism which is born, the famous finches which where pivotal in Darwin’s formulation of the mechanism of natural selection, had the ability to become distinctly varied throughout the Pacific islands they inhabited. Not only were these finches much different from the familiar European ones Darwin knew, but they were different from island to island. Depending upon the size of the food supply (nuts, primarily), the finches’ beak sizes changed accordingly. A random happenstance of small, medium, and large beaks were not the case on an island where small, hard to get shells persisted. Instead, natural selection non-randomly ‘selected’ for the birds which were best adapted to the task at hand.

It is important to restate the point of this article. Evolution has a strong random element, but natural selection is not a random process. It is this mechanism which gives rise to the great complexity seen in all living organisms today. It does not indicate what the result will be, but it does explain that complexity can be. It builds, in cumulative steps, toward greater adaptability. As a great man once said, there is a grandeur in all this.