What Science Is (and what it is not)

William Harris came to my university to perpetuate misconceptions last September. I intended to write a summary of the experience, but I could never muster enough anti-venom to deal with his poison. In lieu of a formal treatment of Harris's bullshit, I've decided to (quite tardily) present a short description of what makes a discipline science. This is inspired by my inability get Harris to acknowledge that all scientific disciplines invoke observable causes to the events they attempt to explain. More after the jump.

I will refrain from comparing supernatural and naturalistic explanations. Hopefully, by painting the picture independently of these terms, we can see that intelligent design and creationism are not science without discarding them simply because of their invocation of supernatural causes. And, yes, I realize that this is a debate regarding the philosophy of science, of which I am hopeless underqualified to discuss (see Skipper's explanation of the difference between philosophy and biology). Consider these the musings of an armchair thinker.

One thing I've observed that is common to all scientific disciplines I've encountered is that they are trying to explain something using certain mechanisms. Some researchers look at big questions (in size and scope, not philosophical implications), such as the origin of the modern universe or where all of the life on earth came from. Others are looking at tiny things, like the content of atoms or the insides of cells. Regardless of what they look at, all scientists study cause and effect. They see an effect (life, the universe, or anything that may be in it) and they work to determine what causes it. A physicists wants to know how subatomic particles cause the atom to exist. A chemist looks at how interactions between atoms and molecules cause other molecules to form. A biochemist studies how those molecules cause cellular pathways to function. A developmental biologist works to determine how those cellular pathways cause cells to differentiate. Etc, etc, etc.

What unifies all scientific pursuits is the study of cause and effect. The effect can be anything we have observed, as can the causes. And therein lies the rub. We must have observed the causes to invoke them in explaining the effect. But is it enough to have merely observed the causes? I would argue that a second condition must be met: the causes must have an observable effect (at some level) on what we are studying. If you have never observed a cause or you have never observed a relationship of a cause on the studied effect, it cannot be invoked as an explanation for that effect.

Creationism fails on two counts. Not only have divine (or supernatural, or whatever catch-phrase is currently en vogue) forces never been observed in a controlled setting, they have never been observed to have an effect on the origin or change over time of biodiversity. This is independent of the fact that they are supernatural. For instance, gravity has been observed (if you doubt me, knock a pencil off your table), but it is limited in what effect it has on certain events. I have observed gravity cause my pencil to fall off the table (whether the table exists we'll leave to Sartre), so I can use gravity to explain that event. I have never observed gravity cause my hair to grow or my phone ring, so I cannot invoke it as explanation for those events. It's not enough to have a measurable or observable force; there must also be some evidence of cause and effect between that force and the event we are studying.

No 'fuck you to the creationists' would be complete without explaining why evolutionary biology is science. Let's lay out some of the causes biologists use to explain evolution: sexual and asexual reproduction, genetic mutation, genetic recombination, limited resources, non-random mating, and differential survival and reproduction. We have observed how reproduction leads to the conservative transmission of genetic information from parent to offspring. We have observed that some errors (mutations) arise in that transmission, and that the genetic variation can be shuffled by recombination. We have observed how mutations and recombination lead to variation in a population. We have observed that some individuals survive while others perish when resources are limited and cannot sustain the entire population. We have observed that the individuals that perish may or may not be determined by which variants they possess. We have observed that some individuals leave more offspring than other individuals, and these differences may also be associated with the genetic variation in the population. All of these causes have been observed and they have all been shown to have an effect on variation within a population. They have also been observed to effect biodiversity on larger time scales.