To the editor:

Contrary to Todd Wilson’s Oct. 17 letter, science is not in bad shape. The same cannot be said for his understanding of how it works. Repeatability is an essential element of scientific research, but only with respect to the ability of scientists to evaluate theories by reconstructing the observations and logic on which they are based.

Astronomy, geology, paleontology and archaeology investigate unrepeatable phenomena such as the formation of galaxies, the extinction of dinosaurs, the evolution of Homo sapiens and the emergence of ancient agriculture. They use experimentation to test their methods and theories, but NOT to repeat the phenomena they investigate!

Basic criteria for evaluating good scientific theories are: 1. compatibility (do the observed data agree with the theory?); 2. parsimony (does the theory minimize assumptions about unknowns?); 3. completeness (does the theory account for the most observed data?); and 4. predictability (does the theory predict new discoveries?). Evolution is not a religion or a philosophy, but a collection of theories that perform well according to these criteria. Creationist theories, by comparison, consistently fall short.

Science works best when research in different disciplines independently supports a common body of theory. The best scientific theories are compatible with each other, they hold up to ongoing, rigorous testing, and they are constantly improved or replaced by better theories. Science has given us the Internet, cell phones and HIV medication. Just because it is hard to understand does not mean that its waters are “muddied.” The problem may be with the observer.

John W. Hoopes,

Lawrence