Creationists often state that they are not anti-science – it’s just that evolution isn’t science. They then go on to state that science is defined by what is observable, testable and repeatable and since we cannot observe, test or repeat evolution it does not fit the bill. The problem being that their use of observable is restricted to direct observations of phenomena. Their use of testable excludes predictive tests that are commonly used in science and their use of repeatable seems to be limited to repeating the event of creation itself (or sometimes simply the mere act of abiogenesis).
In science, the act of observation is rarely direct. Talkorigins.com, a great site for information about evolution, points out that the most important scientific advances arose from indirect observations. The existence of electrons and other subatomic particles being a popular example. I have raised this point a number of times to creationists and their common rebuttal is that there is a distinct difference. Electrons are still there and we can interact with them at that moment. In contrast, evolutionary events occurred in the past and cannot be observed directly. There are a number of problems with this position though. It is connected to their notions of repeatability, which I will get to in a moment. The creationist position also over looks the countless examples of evolution that have observed in the lab or indirectly in nature. The italian wall lizards of croatia are a wonderful example of this – macroevolution, or changes at the species level, occurred and we observed it by examining the species before and after an interval of time. Finally, the creationist position completely overlooks the fact that events in the past leave a sort of footprint that can be observed and used to deduce information about the original event. In a similar way, these historical tracks provide a means to test the theory.
We can all remember our highschool science classes. The teacher would give you a hypothesis and turn you lose with some laboratory recipe designed to test it. This is our first exposure to the scientific method – you come up with a hypothesis and then you test it with available evidence or you concoct an experiment to put the hypothesis to the test in a controlled manner. Testing evolution is a bit tricker than your average highschool science lab but it can still be done. One method is to generate evolutionary change in a laboratory. This is often done with a species that reproduces rapidly so that many generations can be spawned in short periods of time. (This is why fruit flies are of such interest biologists, a point that is lost on Governor Palin.) Another method of testing the theory is to look at its implications against the known physical evidence, such as the fossil record. To me, on of the strongest pieces of evidence for evolution comes in this form; the distribution of life on this planet.
Evolution is a branching process – when changes occur it breaks the species into two distinct groups differing by a single trait [1] with each of the new groups inheriting the traits of the previous subgroup. There is a necessary, and well studied, mathematical consequence of this type of process called a nested hierarchy. Each subset of species must be contained within the larger subset at the level above. For example, near the top of the hierarchy, two types of cells have emerged; eukaryotes and prokaryotes. Every organism evolving from the eukaryotic division will have those types of cells and can become multicellular while every prokaryotic organism will be single celled and will reproduce by binary fission. These traits are inherited from the parent and it forms the seed for an existing hierarchy that cannot be broken within the evolutionary framework. If you find a single example that violates this, say a multicellular prokaryotic organism, you will have falsified evolution. You can therefore test the theory by forming this nested hierarchy and then going to see if it exists in nature. If it does not, then evolution cannot be true but, if it does you have evidence to support the theory. Regardless of the result, this is a test of the evolutionary theory that does not require direct observation of the process itself but rather observation of its result.
This brings us to repeatability. There are some events that are extremely rare and are extremely difficult to reproduce in a laboratory setting. So, if we limit our notion of repeatability to the rare events then yes, we are going to have a hard time repeating them. But this is not to say that it is impossible. We have built a multi-billion dollar facility at CERN in order to reproduce the conditions just after the big bang. This is certainly an achievement that would have been thought impossible in the early days of quantum mechanics. However, this is a side issue since these one time events are not what the scientific community is referring to when they require repeatability. When talking about repeatability, we actually mean that whatever tests you develop, or experiments you perform, must be independently repeatable by another worker. This is important because everything in science undergoes peer review. If you do an experiment that no one else can repeat then there is no independent way to confirm your results. Obviously the tests of evolutionary theory can be checked independently by workers around the world and therefore meets the standards of repeatability.
Despite what creationists claim, evolution is a valid scientific theory and it rests on extremely firm empirical and experimental evidence. I often tell my friends that I view evolution as being one of the most successful scientific theories to date, second only to quantum mechanics.
[1] I don’t know if examples of more than one difference has been found. If anyone knows of example of this please let me know.
