The Sept. 22, 2011 issue of the Gauntlet ran an article entitled Limits of the Scientific Method by Louis Joubert. I have found this article to be most erroneous in its criticism of science. Part of this stems from the failure to clearly define science. The author proposes to "stick to the definition that seems to usually be in people's minds." This I find to be an unhelpful definition, and one that allows the author to totally confuse the scientific method with scientific theories, and thus much of his article is criticism of theories and not method. I will mainly consider the article's criticism of scientific theories, but any discussion of science must touch on the scientific method itself.
In his article Joubert that current scientific theories are doomed to be destroyed. It is with this wording (and what it implies) that I must disagree. It implies that what we know about the natural world is unreliable, and that at any moment our entire understanding could be shattered by a new paradigm.
Such a claim shows a lack of understanding about the nature of scientific theories. Isaac Asimov put it best when he wrote "theories are not so much wrong as incomplete." The process of science gradually refines and improves theories-- it is like building on past foundations instead of tearing them down and starting again.
We can illustrate this with an example: at first the ancient Greeks (and other ancient civilizations) theorized that the sun and all other planets revolved around the Earth. Then, Copernicus introduced a model where the Earth revolved around the sun.
The first theory was indeed wrong, but not totally wrong. It got the revolving part right, just not the order of planets. Later theories would improve upon Copernicus, and show that the planets orbit in ellipses, not perfect circles. Copernicus was wrong, but not totally wrong. He got the orbiting part right, just not the paths of orbits. These replacement theories build upon previous theories, keeping what works and replacing the incorrect aspects with more correct ones.
Another criticism that Joubert raises is the changeability of science, a point with which I strongly disagree. Change is not bad for scientific theories-- indeed we should be glad to have it, lest our theories become dogmatic and stale. Or worse than that, they could become useless. Theories that fail to accurately describe the world would prohibit technological progress. Imagine for a moment that our theory of gravity was incorrect. There would have been no moon landing, no space travel, and no artificial satellites orbiting the Earth.
The very fact that scientific theories are changing is a virtue. It means that the process of science is still underway, that we are constantly checking up on our theories, and that we are actively trying to falsify them. Indeed, it is the very process of science that makes scientific theories reliable.