Science FAQ

[Durandal]

Mods, I'm going to ask you to make this a sticky in this forum. It's a rundown of the basic scientific method, with definitions of scientific theory and law. I'm tired of dumbshit creationists saying, "It's only a theory," or "Science can only offer theories," without knowing what the fuck they're talking about. So, rather than repeating ourselves, we can just refer them here.

What is science?

Science is a method for analyzing observed phenomena to discover their driving mechanisms. It hinges on making useful, specific predictions and putting those predictions to the test in a controlled environment, or seeing if those predictions match observation.

What is the goal of science?

The goal of science is to describe the universe in terms of equations and predictions. Different sciences seek to describe different parts of the universe. Biology seeks to describe the way living things work, while cosmology seeks to describe the history of the universe and the mechanisms that drove it into being.

What does science not seek to do?

Science is not meant to answer philosophical questions, such as "What is the meaning of life?" or "Is there a god?" Science does not touch on any such questions. As a general rule, science does not ask questions concerning "why" in the sense of an event's ultimate purpose. It only asks, "What caused this to happen?"

How does the scientific method go?

The scientific method begins when a scientist observes an event. The scientist is interested in the driving mechanisms behind that event. So, he studies it and formulates a hypothesis. This hypothesis will include specific predictions which can be tested. The next step depends upon the materials required to test the hypothesis. Either the scientist can perform an experiment in a lab environment, or he can observe the phenomenon further to see if his predictions are inline with what happens in nature. An example of the latter is astronomy. Astronomy deals with energies and masses on such ridiculously large scales that it is impossible for astronomers to run a controlled experiment on the supernova of a star, for example. However, astronomers can look at different supernovae to see if their predictions match what they see.

If the scientist finds that his hypothesis matches what is observed, he submits it for peer review. Before being considered for publication, someone else must perform the experiment and get the same results. In other words, whatever experiment or observation the scientist performs, it must be repeatable and independently verifiable.

If the hypothesis is confirmed by other scientists independently, it is in the running for becoming a theory, depending on what the specific predictions of the hypothesis are.

What must a hypothesis do in order to become a scientific theory? What is a scientific theory?

A scientific theory is a tested hypothesis that has been demonstrated to be inline with observation. A scientific theory must fit all the observed facts dealing with the phenomena, and it must account for all of them in its predictions. It must also, as a part of its predictive capacity, offer a testable explanation for the driving mechanisms behind that phenomenon.

What happens if a theory doesn't fit all the facts and can't explain one or two observations?

The theory is still valid pertaining to what it does explain. For example, Newton's laws of universal gravitation could not explain the location of the perihelion in Mercury's orbital path, but it predicted all the other planets' orbits perfectly. It was only until Einstein came along that he was able to form a theory which could predict Mercury's perihelion along with the orbits of all the other planets. However, Newton's laws are not simply discarded. They can predict motion and gravitation to an acceptable degree of accuracy in most cases. So, we use them in situations where we know they'll make usable predictions.

What is a scientific law? Isn't it a theory which has been proven?

No. Contrary to popular misconception, scientific laws are no more "proven" than theories. There is no way for a theory to be proven absolutely. A proven theory would imply that it be set in stone as a dogma or tenet of science, which would stunt the evolutionary process of science. Since all theories and laws in science are always subject to testing, inquiry and change due to new data, there is no way to declare a theory as proven.

A scientific law is simply a predictive statement that offers no explanations for the mechanisms behind an observation. Newton's second law predicts that F = ma, but there is no explanatory clause in the law. It is purely predictive.

What if two competing theories both make the same predictions, but offer different explanations for the mechanism? How is one judged to be better than the other?

Traditionally, two theories are compared by how complex their explanations are. Whichever theory has the least amount of terms or unknowns in its explanation is the better theory.

For example, take "Intelligent Design Evolution" vs. conventional evolutionary theory. They both make the same predictions and explain the same observations. Which one is better? Conventional evolution. Intelligent design evolution adds an unknown into the mix. It posits that unknown creators must be driving our evolutionary process. This mechanism is purely redundant, compared to conventional evolutionary theory's, which states that evolution is guided by natural selection. This prediction is supported by observation in nature, so there is no reason to assume an unknown (the "intelligent designer") for the mechanism. The intelligent designer explanation offers no extra predictive capacity, nor does it help evolutionary theory in any way. It's a redundant term, so it should be discarded until direct evidence of its existence is observed.


This is incomplete, and I'm sure others will think of more questions to contribute. So, I say we just leave this as a sticky, and people can submit more FAQ entries as time goes on. It'll be an evolutionary FAQ. :)