Bear with me for a moment while I cover territory that might seem obvious, but includes some distinctions that are important.
Organisms have been changing for a long time. Assuming that that they are getting "better" in some way is called evolution. The evidence that organisms have been changing over time is so good, it is as nearly a fact as possible for something to be a fact. There's a new strain of strep in San Francisco that is killing people because our current antibiotics don't kill the bacteria. The bacteria used to be different. What caused the change?
Scientific theories tend to be about the mechanisms. What causes what to occur? They are about models.
And there are a lot of different models out there. In my physics class, I am fond of showing my students a phenomenon and asking them to come up with a model that explains what is going on and that can make predictions for related phenomenon. They often come up with several different models of which some of the models are incompatible. The debate then ranges on how to decide between these models.
The basic rule is that the model has to correctly explain everything that has already been discovered. That, however, is not enough, good models make predictions about related situations as well. The joke in class is that some models are "Demon Theories", a name that comes from an interaction I had in college with a brilliant, summa cum laude electrical engineer when I asked him for help explaining how transistors worked. He said that demons live inside of the little black boxes and when goosed by voltage they pick up electrons from one wire and put them on the other wire. When the goosing gets higher, they work faster. His Demon Theory of Transistors did explain transistors pretty well, but it didn't help me make any predictions.
In any case, a good model should be able to explain and predict, and this fact of evolution needs a model to explain why it occurs.
There are lots of possible models. Many explain but few predict. Several demon theories include that we were all born a few seconds ago with false memories of the past including fossils, that fossils were planted by God as a test for the faithful, organisms what to change, etc.
Some models explain and make predictions, though. These predictions can be tested and they should be.
This is what science does. If you don't teach in class that testing models is science's job, you are not teaching science. English and art and religion don't make models and test them. They are prescriptive and they tell how the world should be. Science doesn't know how the world should be. It just tries to figure out what the world is.
Back to modeling evolution. So, the most popular model at the moment is natural selection. How can we tell if this model is good one? We can test it out. A few people have mentioned that it takes a long time to see evolutionary changes, but this isn't always the case. In the right situation, they can happen vast enough to check.
A classic experiment involves fruit flies and a swiftly varying environment. Imagine starting with two groups of very similar fruit flies. One group is trapped in a container where the environment is rapidly changing while another group of fruit flies is in a pretty static environment. Generally speaking, the static environment fruit flies do better assuming we start off the fruit flies in an environment to which they are suited.
Now we will subdivide each group into two groups, one of each will be exposed to mild mutation inducing radiation. What do you think will happen?
Depends on your model, of course. Natural selection predicts that the fruit flies exposed to mutating radiation that are in the constant environment will tend to end up unsuited for their environment and so the irradiated flies will tend to be less suited than the non-irradiated flies. On the other hand, natural selection predicts that that the fruit flies in the changing environment will benefit from the mutations as only with rapid mutations will they be more likely to survive in a rapidly changing environment.
Other models predict other things. For example, evolution by means of intelligent design predicts that neither or both groups should benefit from the radiation. If there is a hand from a designer, it doesn't matter how much radiation there is. As long as any mutations occur, and some occur without help from the scientists, then there will be room for the right mutation from the intelligent designer.
So, what happens?
The flies in the rapidly changing environment that get the radiation do better than the flies in the rapidly changing environment that do not, while the flies in the constant environment that get irradiated do worse than the flies that don't get the radiation.
It's a pretty clear victory for natural selection and a defeat for intelligent design. But don't be sad. This is no different than what we do in physics every day. We make models that seem counter-intuitive but they do a really good job of making predictions that are born out by doing tests. We accept those and discard the others.
And we most teach students to make these kinds of models and test them, and if they cannot test them, then find someone else who can. How else will we protect them from the snake oil salesmen and women on television. I just saw an ad that said that pads on your feet would take the toxins out of your body that cause things like diabetes.
Marc "Zeke" Kossover
The Jewish Community High School of the Bay\