One way to begin thinking about science is to think about its aims. (Another is to think about its methodology. Still another might be to consider its subject matter, but it's not clear that there is a distinctive subject matter of science in general, as opposed to specific sciences such as physics or biology.) Here are several aims one might attribute to science.
Some branches of science consist primarily of simple data collection. Some parts of biology may be like this: weíre interested in collecting information about previously unknown species of plants and animals. Similarly, astronomers enjoy finding previously undiscovered planets, stars, and other phenomena in the heavens, and physicists sometimes search for new elements. On the other hand, this surely canít be the entire story. Science is more than just a long list of facts! And long lists of facts are not automatically interesting. Thereís something very odd about the idea of simply going out and hunting for facts, without having any particular question one is attempting to answer, or any particular hypothesis one is seeking to prove or disprove. (Thomas Kuhn suggests that mere fact-hunting is characteristic of the "pre-paradigm" period of a science.)
One way to see that description is not the whole story: suppose we had a complete description of every particular fact about the world, past or present. Would that be a scientific account of the world? (My answer: no.)
One problem with long lists of facts is that they donít automatically tell us anything about the future. We donít just want to know how things are and have been; we also want to know how they will be in the future.
Is that enough? Well, suppose that we add to our description of past and present facts a complete list of future facts. (Maybe it was presented to us by God, or time-traveling aliens; at any rate, suppose that we are entirely confident that it is complete and correct.) It still seems that we would not have a scientific account of the world, that we would want more.
Weíd like, not just to know more about nature, but to be able to harness it to some extent: not just to predict what will happen in the future, but also to be able to make a difference to it. We donít just want to predict the conditions under which someone will become sick; we also want to be able to prevent illness, and to cure it when it occurs. (But is this an essential characteristic of science? Much of science studies phenomena that we have no hope of controlling.)
One reason lists of facts seem unsatisfying is that we are also interested in regularities in the facts, in seeing how they fit together. We don't just want to know facts, we want to know patterns. (Analogy: a list of the R, G, B components of every pixel on a monitor is a complete description of how the monitor appears. But it might be very difficult to use this information to determine what the image on the monitor actually looks like.) We want general facts as well as particular ones.
We're getting closer, but I think there's still a problem here. Suppose I go into a nearly-empty parking lot and look at the five cars there. I notice the following general facts: (1) All the cars are resting firmly on the ground. (None of them is, for instance, floating in the air.) (2) All the cars are made by General Motors. It seems that these are very different kinds of facts. That all the cars are resting firmly on the ground is a result of fundamental physical laws. But the fact that they are all made by GM is just a coincidence. Science is interested in the former kind of fact, but not the latter.
Explanation goes beyond mere description to tell us why the facts are as they are. If we know why an event occurred, then we are in a better position to know when it will occur in the future, so there seems to be a close connection between explanation and prediction. (If the deductive-nomological model of explanation is correct, then in fact explanation and prediction are essentially the same thing.) Similarly, if we know why events occur, we are in a better position to make them occur or prevent them from occurring, so we are more likely to be able to control nature. And explanations typically involve systematic organizations of natural phenomena; for example, if the D-N model is correct, then explanation requires general laws to play a role in the explanans, and general laws provide a very powerful way of organizing phenomena.
Last update: January 12, 2011.
Curtis Brown | Philosophy of Science | Philosophy Department | Trinity University