This paper addresses the question of to what extent people's reactions to the Chernobyl 1986 nuclear power plant accident were rational. After a review of studies on Chernobyl published to date, two types of rationality are investi gated : outcome rationality and process rationality. The mechanisms by which people seem to come to subjective probability estimates of nuclear risks are further explored in a summary of an experimental study on the meaning of verbal probability expressions. It is concluded that although the rationality of people's responses depends on the interpreter's focus, these responses are un derstandable and meaningful, and should be taken seriously.