Science - Historical versus Experimental
Prosthesis has a great post about the difference between experimental science, which deals with predictions and historical science that deals with make intuitions about how something happened in the past, which he calls retrodictions.

From the post

The other type of scientific explanation appeals to historical causes. This type of explanation may or may not appeal to the general regularities or laws that the previous type of explanation appealed to. Instead, this type of explanation is more concerned with what happened in order for some phenomenon to occur. For example, to explain some unique geological feature, we might talk about a certain series of events that occured. Laws, by themselves, can't explain these types of phenomena. Meyer uses the example of the law, "Oxygen is necessary to combustion." This law doesn't explain how a house fire occured. An explanation will talk about a unique event (say, a smoker falling asleep in his chair or a space heater tips over). This type of explanation, rather than making predictions about future events, instead makes retrodictions. Based on some observed phenomena (and perhaps some known laws or regularities), we can retrodict past events or conditions.

I doubt there is much dispute about the accuracy or value of experimental science. When you get into historical science however, you get a lot more assumption and a lot less observation. We have enough trouble being certain of recent events (on crime scenes for instance), that when people start telling you just how something happened thousands of years ago (without eye-witness testimony) you have to take everything they say with a few thousand grains of salt.
- posted by Alan Grey @ Wednesday, May 18, 2005