What Science Does
Most generally, scientific theories are useful statements about the world. The world comprises an enormous number of phenomena: things (such as ice crystals or stars), events (storms, radioactive decay), and combinations of the two (planets circling around the Sun, a cat catching a mouse). Science has two main goals: to explain the phenomena we have observed, and to predict the kinds and properties of the phenomena we will yet observe. The power to explain and predict defines what we meant by "useful" in the first sentence.
Science is not the only means of explaining the world. Stories, myths, and religions also serve that goal. However, among all those methods, science is the only one with a significant ability to predict future observations. Even when religion, as then defined, was used for predictions, it was consistently successful only when it was in fact using the scientific method, such as when ancient Egyptian priests predicted floods and eclipses.
How do we know that science has predictive power, and myths and religions do not? What about prophecies? It is important to understand exactly how we can test whether a prediction is successful. The idea of testing predictions is a key concept in science, and a requirement that all predictions be testable distinguishes the scientific method from all other means of explaining the world.
Many people believe that Nostradamus (1503-1566) predicted a slew of historical events, from the French Revolution to the rise of Hitler to the suspicious death of Pope John Paul I to the 9/11 terrorist attacks. However, his quatrains (four-verse poems) are so vague and cryptic that they can only be interpreted after the event they allegedly predict. There is no known case that any of his quatrains was deciphered and used to predict a specific event before it occurred. The Psychology of Belief section of this site explains the human tendency to find coincidences and ascribe special meaning to them.
The Gospel of Matthew claims that the Old Testament prophet Isaiah predicted (prophesied) the coming of Jesus, so many Christians believe that Biblical prophecies had predictive power. But Jews - who also consider the Old Testament their sacred text - deny the validity of Matthew's interpretation of Isaiah's prophecy. In any case, Matthew did not write his account before Jesus (allegedly) fulfilled the prophecy, so even if viewed in the most favorable light, he only explained the events after they happened, rather than predict them in advance.
As those examples illustrate, the prophecies celebrated as true were only recognized by some authority after the event they supposedly predicted. Numerous other prophecies failed to materialize even under most imaginative interpretations. In the 1970 movie "Little Big Man," the Grandfather, after his predicted death failed to come, says "Well, sometimes the magic works, sometimes it does not." This can be said of prophecies in general. That may be satisfactory for some who seek comfort or inspiration in them, but they cannot be considered useful for predicting specific events. Science, on the other hand, demands that its predictions work every time.
Testing Predictions
Consider a very simple example of the scientific method of testing predictions. How many teeth does a horse have? If you are a veterinarian or a game show champion, you might know the answer, but otherwise it is likely that you don't. How would you find out? The obvious way is to find a horse and count its teeth. A statement such as "a horse has 99 teeth" is a scientifically testable prediction, and the act of counting the teeth is its scientific test - an experiment. Performing the experiment will result in a number of teeth (a measured value) that can be compared to the prediction and determine whether the prediction was true or false. In this case, the prediction was false; any theory that gave that prediction would have to be rejected.
But what of a prediction that "a horse has many shiny teeth and is a beautiful animal"? Is this a scientifically testable prediction? No, because there is no way it could be compared to a measured value and found true or false to practically everybody's agreement. Not all statements are scientifically testable. As a rule, a statement is scientific only if it is possible to design a test that could, in case of a particular outcome, clearly demonstrate that the statement is false. Such a statement is technically called "falsifiable," and that is what is meant by the more common term "testable." The price science pays for its rigorous requirements is that it can only consider statements that are experimentally testable. This is why science cannot, and does not try to, answer all the questions people can come up with. A question is scientific only if it can be answered with a statement that is testable (falsifiable).
Science deals with questions such as what things are ("What is water?"), how they work or happen ("How does rain fall?"), how much or how many of something there is ("How much rain falls in Hawaii in an average year?", other questions of measures and quantities ("How fast does a rain drop fall?"),what will happen if we take a specified action ("If I a piece of sodium in water, will it float or sink?"), and how to achieve a desired result ("How can I make a waterproof shoe?"). It does not deal with questions of purpose ("Why does water exist?"), meaning ("What is the meaning of life?"), or intrinsic value ("Is nature more beautiful than art?").
<NEXT>
