From the editorial with the same title in the Dec 2002 issue of the Scientific American:
"...Inevitably, scientists will sometimes be just plain wrong - they make mistakes. Interpretation of evidence leaves room for error. Moreover, scientists aren't saints. They can be swayed by careerism, by money, by ego. Biases and prejudices can blind them. As individuals, they are no more or less flawed than those from any other walk of life. Over time, however, science rises above narrow interests and corrects itself more reliably than any other institution through such practices as the open publication of results and methods."...
"All scientific knowledge is provisional. Everything that science "knows," even the most mundane facts and long-established theories, is subject to reexamination as new information comes in. The latest ideas and data are the most provisional of all. Some recantations will be unavoidable. This is not a weakness of science; this is its glory. No endeavor rivals science in its incremental progress toward a more complete understanding of the observable world...
How should the public weigh the recommendations of scientists? The greatest mistake is to wait for 100 percent scientific certainty or agreement, because it will never materialize. Conclusions vetted by the professional community might turn out to be wrong, but they generally represent the best-supported views currently available. People are free to disregard those views, but they shouldn't delude themselves that they are being more reasonable by doing so. Perfect certainty belongs only to the gods. The rest of us have to make do with science, imperfections and all."
Read the article at: www.sciam.com