The personal characteristics of the investigator should therefore be irrelevant to the evaluation of the knowledge-claims they make, and the institutions of science should be designed to ensure this (for example, anonymous refereeing of journal articles and research applications)