Uncertainty complexity and dissent make climate change hard to tackle with normal scientific procedures. In a post-normal perspective the normal science task of "getting the facts right" is still regarded as necessary but no longer as fully feasible nor as sufficient to interface science and policy. It needs to be complemented with a task of exploring the relevance of deep uncertainty and ignorance that limit our ability to establish objective, reliable, and valid facts. This article explores the implications of this notion for the climate science policy interface. According to its political configuration the Intergovernmental Panel on Climate Change (IPCC) adopted a "speaking consensus to power" approach that sees uncertainty and dissent as a problematic lack of unequivocalness (multiple contradictory truths that need to be mediated into a consensus). This approach can be distinguished from two other interface strategies: the "speaking truth to power approach," seeing uncertainties as a temporary lack of perfection in the knowledge (truth with error bars) and the "working deliberatively within imperfections" approach, accepting uncertainty and scientific dissent as facts of life (irreducible ignorance) of which the policy relevance needs be explored explicitly. The article recommends more openness for dissent and explicit reflection on ignorance in IPCC process and reporting.