The Art of 'Manufacturing Uncertainty'

By David Michaels, David Michaels, a professor at the George Washington University School of Public Health, served as assistant secretary of Energy between 1998 and 2001.

To many scientists and policymakers in Washington, the revelation this month that Philip Cooney, chief of staff for the White House Council on Environmental Quality, had rewritten a federal report to magnify the level of uncertainty on climate change came as no surprise. Uncertainty is easily manipulated, and Cooney — a former lobbyist with the American Petroleum Institute, one of the nation's leading manufacturers of scientific uncertainty — was highly familiar with its uses.

As an epidemiologist with a special interest in occupational diseases, I share a fundamental problem with the scientists who are studying climate change. Our ability to conduct laboratory experiments is limited; we can't go out and intentionally expose people to carcinogens any more than climatologists can measure future temperatures. Instead, we must harness "natural experiments," collecting data through observation only. We then build models from this data, and use these models to make causal inferences and predictions, and, where possible, to recommend protective measures.