This paper focuses on the concept of post-normal science, originally proposed by Silvio Funtowicz and Jerome Ravetz, as an advanced method of knowledge generation for policy, and reflects on the ethical motivations for both its theoretical meaning and its practical realisation. In order to put the reflection on the why and how of post-normal science in a broader contemporary and future-oriented context, I will first elaborate on what I call ‘the politics of hypothesis’ and argue that the fundamental challenge for science that aims to advice policy today is not the problem of strategic manipulation of scientific advice by politics, civil society or the market, but rather the problem of dealing with the lack of evidence in situations where politics, civil society or the market ‘need’ that evidence to (urgently) inform, criticise or justify specific actions or practices. Confronted with the need to deal with incomplete and speculative knowledge, in many cases, scientific hypotheses have become the ‘end products’ of science themselves, and society has no other choice than to deal with them in a responsible way. The challenge of science in these cases is therefore not any longer the production of convincing proofs, it is the construction of credible hypotheses. Against this backdrop, a second part will recall how the normative motivation for post-normal science was originally worked out by Funtowicz and Ravetz. I will re-emphasize why and how the argumentation of Funtowicz and Ravetz in favour of the democratisation of science and the opening up of the dialogue to include opinions, beliefs and lay knowledge of ‘non-experts’ is ethical. Consequently, the third part proposes some paths for further ethical reflection with regard to the value and meaning of post-normal science in the ‘post-normal age’. I will briefly elaborate on the concept of transdisciplinarity, the idea of ‘co-creation’ of complexity, the concept of holism and the idea of the unavoidable moral authority of the present generation in intergenerational ethics. The reason is that each of these explorations is at the same time and in its own way an invitation to reflect on who we are as humanity, and on how we can possibly make sense of things for the better. I hope to make clear not only how these concepts and ideas may inspire the ethics of post-normal science, but also that they should become topic of concern in post-normal science dialogues themselves. Finally, in conclusion, I will situate the original ethical motivation for post-normal science in a broader perspective on responsible knowledge generation ‘in face of’ the complexity of complex problems and argue that the overall ethical motivation for postnormal science is to enable an emancipatory and (respectfully) confrontational dialogue and not to come to a full understanding of the complexity of a complex problem or to ‘proof’ specific hypotheses. I will consequently suggest that, responding to the ‘ethical appeal’ of complexity, ‘reflexivity’ with respect to the own position, interests, strategies, hopes, hypotheses, believes and concerns would be the fundamental ethical attitude for anyone concerned. Reflexivity as a ‘quality’ of the science-policy interface does not only require science to reform in order to enable it to generate more reflexive policy advice, but also politics to reform into inclusive and deliberative forms of decision making prepared to become inspired and instructed by that reflexive policy advice. Post-normal science can in that sense be understood as a ‘method’ that can generate reflexivity at the science-policy interface and even as a deliberate political ‘act’ of knowledge generation involving science, political decision makers and civil society.