Abstract
The role of scientific advice and the scientific advisor in policy development and execution has been studied and described in many documents, but seemingly always from the point of view of the policy makers. Questions that are discussed in that context are amongst others, how to get unbiased advice, how to know whether an advisor is competent, how to select the subjects for which advice is needed and how to get a complete picture of the problem. In this paper we argue that the scientific advisor needs to consider some other questions too: Am I unbiased; Does the “client” try to influence the scientists’ point of view and if so what to do? How certain is the advice? How to convey uncertainties? How to prevent the user of the advice letting it sound more certain than it is? What to do if the client does not follow the advice but suggests he does? How to prevent the client hiding behind the advisor? If the advisor thinks the problem has more angles than are within his expertise, how to convey that message? How to deal with other views, whether scientific or otherwise? How to deal with confidentiality demands? In the remainder of the paper, we discuss these issues, using actual political and societal examples.