We find Stephan Lewandowsky and Dorothy Bishop's framing of science governance to be overly simplistic and in need of a firmer evidence base ('Don't let transparency damage science' Nature 529, 459–461; 2016).

The authors' analysis is biased by its reliance on testimonials from the narrow range of invited experts at last year's Royal Society meeting (see go.nature.com/zptirs). Complex issues associated with openness and transparency also need to be taken into account (see S. Jasanoff Law Contemp. Probl. 69, 21–45; 2006).

The authors present important topics such as expertise, disciplinary boundaries and communication as simple dichotomies. These divisions overlook extensive nuanced evidence from the social-science literature about who counts as an expert and under which conditions (see, for example, go.nature.com/xdfzrn).

In our view, governance issues around openness and transparency should not be framed only by the research community. The debate must also include representatives from across the broad range of public viewpoints.