Safari - Rob Colson: The Science Book

A logical system for the scientific process was first put forward by theEnglish philosopher Francis Bacon in the early 17th century. Building on thework of the Arab scientist Alhazen 600 years earlier, and soon to be reinforcedby the French philosopher Ren Descartes, Bacons scientific method requiresscientists to make observations, form a theory to explain what is going on, andthen conduct an experiment to see whether the theory works. If it seems to betrue, then the results may be sent out for peer review, in which people workingin the same or a similar field are invited to pick holes in the argument, and sofalsify the theory, or to repeat the experiment to make sure that the resultsare correct.

Making a testable hypothesis or a prediction is always useful. Englishastronomer Edmond Halley, observing the comet of 1682, realized that it wassimilar to comets reported in 1531 and 1607, and suggested that all three werethe same object, in orbit around the Sun. He predicted that it would return in1758, and he was right, though only justit was spotted on December 25. Today,the comet is known as Halleys Comet. Since astronomers are rarely able toperform experiments, evidence can come only from observation.

Experiments may test a theory, or be purely speculative. When the NewZealand-born physicist Ernest Rutherford watched his students fire alphaparticles at gold leaf in a search for small deflections, he suggested puttingthe detector beside the source, and to their astonishment some of the alphaparticles bounced back off the paper-thin foil. Rutherford said it was as thoughan artillery shell had bounced back off tissue paperand this led him to a newidea about the structure of the atom.

An experiment is all the more compelling if the scientist, while proposing anew mechanism or theory, can make a prediction about the outcome. If theexperiment produces the predicted result, the scientist then has supportingevidence for the theory. Even so, science can never prove that a theory iscorrect; as the 20th-century philosopher of science Karl Popper pointed out, itcan only disprove things. Every experiment that gives predicted answers issupporting evidence, but one experiment that fails may bring an entire theorycrashing down.

Over the centuries, long-held concepts such as a geocentric universe, the fourbodily humors, the fire-element phlogiston, and a mysterious medium called etherhave all been disproved and replaced with new theories. These in turn are onlytheories, and may yet be disproved, although in many cases this is unlikelygiven the evidence in their support.

"All truths are easy to understand once they are discovered; the point isto discover them."

Galileo Galilei