In a Comment article on page 454 of this issue, a group of researchers writing about the future direction of the Intergovernmental Platform on Biodiversity and Ecosystem Services (IPBES) note that the organization is “promoting a predominantly science-based understanding of biodiversity”.

This is to be expected, you probably think. Nature, after all, is the international weekly journal of science. Science-based understanding is what Nature does.

Yet, that line from the Comment is not an endorsement of that 'science-based' approach. It is a criticism. In the same article the authors dismiss a reliance on peer-reviewed science as “tunnel vision” and argue that “knowledge that has been standardized and scientifically validated” is insufficient to set useful policy. Above all, they insist, the IPBES must take a different approach from that of the Intergovernmental Panel on Climate Change (IPCC). The desire of the IPCC to produce standardized assessments (it is now working on its fifth) has limited its success, the authors argue, because it has “overshadowed arguably more important tasks: synthesizing wider perspectives about changing climates and spurring action by multiple policy actors”.

This provocative stance raises many questions. First of all, is the criticism of the IPCC fair? The answer is yes and no. That body is set up to draw fire from all sides — the more so because of its clumsy handling of claimed mistakes and the waning of political support for policies to restrict greenhouse-gas emissions. Yet critics of the IPCC too often overlook its heterogeneous nature: goals and practices vary across the three separate working groups, so much so that some have argued convincingly that the IPCC should release not a combined report but three distinct ones. The political clout of speaking with one voice is greater, but a larger target is also easier to hit.

Hostility towards the IPCC's perceived approach perhaps flows in part from how climate change is often presented: as a problem to solve; as something large and external to tackle; and so as something for science to quantify. But it can be viewed instead as a consequence of millions, if not billions, of separate decisions by individuals.

Here, there probably are lessons for the IPBES. If people are seen as major contributors to the problems of climate change and loss of biodiversity, then people's behaviour and attitudes must be a major part of the solutions. That view makes both the problem and the solution more fuzzy and less susceptible to standardization, but it perhaps also makes the problem more realistic and the solution more practicable.

Discussion of values, stakeholders, community partners and engagement — the language of the social sciences — can make some traditional scientists uncomfortable. So what does this approach mean in practical terms for the IPBES? On one level, as the Comment authors suggest, the answer could be as simple as broadening the criteria of what counts as admissible material — learning to value local knowledge and expertise. The challenge of protecting biodiversity from a cumulative death by thousands and thousands of cuts is surely a good place to test this approach. One size cannot fit all. And it does not need to.