Jun 09 2010

Is Evolution Science?

Evolution is, undeniably, a science. It is not only science, it is a robust and highly successful line of research and a powerful explanatory model.

But there remains confusion in the public as to exactly what it means to be scientific, on both sides of the evolution/creation debate. Even, at times, among proponents of evolution. The following comment to a recent blog post expresses some of this confusion.

Using the scientific method as the criterion, neither creation nor evolution is an established scientific theory. Evolution is stuck on the 3rd step of the 7-step process, establishing a testable hypothesis. After 150 years, the evolution hypothesis is still being “tweaked” and the 4th step, testing the hypothesis has not yet occurred. As of this date, no peer review publications have been presented that go beyond more observations or modifications to the hypothesis. So, why is evolution being pushed as a “fact”? The scientific method also requires that a proposed principle be “falsifiable”, that is, there is a method to prove the hypothesis false. Since creation is based on the existence of God, and God cannot be proven by science, it is a matter of philosophy or faith, not science. Actually, evolution suffers from that same deficit and is, until the falsifiable requirement is met, more of a philosophical speculation than a scientific theory. An excellent resource regarding the creation-evolution debate can be found at http://sechumanism.blogspot.com/p/secular-humanism.html

The author of the comment is Patrick Vosse, and the “excellent resource” he links to is a blog with a single blog post promoting his book, Secular Humanism. Perhaps he was referring to the book itself. Either way, given that he gets this one point so hopelessly wrong it is difficult to consider his writing even minimally informative let alone a reliable resource regarding evolution.

He claims that evolution is not science because it has not fulfilled the seven steps for “the” scientific method. While that article “the” is often used in haste when referring to scientific methodology (I am sure I have done so occasionally) it is misleading to speak of “the” scientific method as if it is one thing. In fact, science encompasses a range of methods.

The seven steps Vosse is referring to go something like this:

But this is a grade school cartoon of the scientific method, useful for giving students an overview to help them with their science fair project, but not as a starting point for serious philosophical discussions about empiricism.

Often a scientific investigation will begin with an observation, or even an idea. Most relevant to this discussion, the “test with an experiment” bubble is very narrow in its concept of how science progresses. The bubble should really read, “test the hypothesis.” Experimentation is only one way to test a hypothesis. Another way is to see if it is compatible with existing knowledge, especially established laws of nature. If a hypothesis violates the laws of thermodynamics, you can probably chuck it.

Further, you can test hypotheses by making further observations. Observational studies are a critical part of medical science and all historical sciences. We cannot build suns in a laboratory, but we can observe how they behave.

Evolution is largely an historical science, and so much of the hypothesis testing has been observational. But there is also a great deal of experimental data that supports evolutionary theory

Vosse wrote:

“As of this date, no peer review publications have been presented that go beyond more observations or modifications to the hypothesis.”

This is a rather bold, and entirely false, assertion. Vosse clearly does not know the first thing about evolutionary science. There are countless published observational and experimental studies that support various aspects of evolutionary theory – common descent, natural selection acting on variation brought about through mutations and recombination, and the specific relationships among organisms. Much of this evidence is summarized at talkorigins.org. But just to give one example, the long term E. coli experiments of Richard Lenski are a dramatic counter example.

Vosse next brings up the related issue of falsifiability, and claims (again, falsely) that evolutionary theory, as currently formulated, cannot be falsified. This claim is just astounding, and again demonstrates that Vosse should not write about topics that he knows functionally nothing about.

Since Darwin there have been thousands of opportunities to falsify evolution – to deal the theory such a major blow that it would have to be abandoned or at least significantly modified. Evolutionary theory has survived every such challenge. For example, Darwin introduced his theory of natural selection before genetics were discovered. The lack of a known mechanism of inheritance was a serious problem for Darwin. It is most significant that Darwinian evolution requires the persistence of favored traits – not the unlimited dilution of such traits in the larger population. Genes – discrete units of inheritance that are not diluted with reproduction but can be copied intact – provided a mechanism that allowed for Darwinian evolution.

The whole of genetics also supports common descent – the pattern of base pairs and the amino acids they code for are found in an exquisitely evolutionary pattern. Common descent could have been destroyed by looking at the genes of organisms, but instead genetic analysis has provided its strongest support.

The geographic and temporal pattern of fossils also supports evolutionary theory. As J.B.S. Haldane famously quipped – all it would take to falsify evolution is a single rabbit skeleton in a Precambrian layer. We have never found the equivalent of a Precambrian rabbit.

Vosse’s comment that the evolution “hypothesis” is still being “tweaked” is such a non sequitur it’s difficult to know what to even make of it. Scientific theories tend to be modified and tweaked over time – that does not make them less of a scientific theory.


The meme that evolution is not testable or falsifiable is a persistent one, despite the fact that it has been knocked down over and over again by those who actually know what they are talking about. I doubt my humble blog post will end this meme – fighting pseudoscience is often like an endless game of whack-a-mole. But at least I can do my part in giving this one mole a good whack.

398 responses so far