Abstract
If scientists rely on assumptions rather than logic, empirical confirmation, and falsification, they are no longer doing science but ideology – which is, by definition, unethical. As a recent US National Academy of Sciences report put it, “bad science is always unethical.”1 This article discusses several ways in which toxicologists can fall into ideology – bad, therefore unethical, science. In part because of the increasing expense of pollution control, some toxicologists have been reexamining pollution dose-response curves that are non-monotonic, that is, curves in which the direction of some response changes with increasing or decreasing dose.2 Ethanol is a classic example of a non-monotonic dose-response curve because moderate drinking is associated with lower risks of heart disease, whereas heavy drinking is associated with higher risks.3,4 If some low-dose pollutants exhibit adaptive or “beneficial effects,”5 this might suggest re-thinking pollution regulations which presuppose linear no-threshold (LNT) dose-response curves.