In this paper, I examine the early history of discussions of participant observation and objectivity in anthropology. The discussions resolve around the question of whether participant observation is a reliable method for obtaining data that may serve as the basis for true accounts of native ways of life. I show how Malinowski in 1922 introduced participant observation as a straightforwardly reliable method and then discuss how – and why – most of the discussants in the 1940s and 1950s maintained that the method is reliable only if the researcher takes a whole number of precautionary measures.