An action gap, not an evidence gap
The day was organized around a single admission, delivered early and never seriously contested: the world is not short of scientific knowledge about sustainable development. It is short of the institutions, incentives and timing needed to use it.
Robbert Dijkgraaf, who takes over as President of the International Science Council in October, made the same diagnosis from the science side and added a point about timing. Policy windows are mostly frozen, he argued, and briefly fluid; the intervention has to arrive during the thaw.
Astra Bonini of UN DESA located the blockage in institutional design. The science of the SDGs is integrated and full of interlinkages; the machinery meant to act on it is not.
From the floor, a delegate from Saudi Arabia’s Ministry of Education supplied the sharpest formulation of the problem, arguing that the binding constraint is a government’s capacity to absorb evidence at all: you can pour science into a state with no absorption plumbing, and nothing moves.
Where the science is produced
Hanif’s second warning was that the imbalance in who produces sustainability research is becoming an imbalance in who develops. Joshua Amponsem put a number on it that went unchallenged for the rest of the day.
The corollary was voiced by Rani Nashu of the Pacific Academy of Sciences, who reframed capacity as a question of authority rather than knowledge. Pacific communities, she said, do not lack knowledge, leadership or will; what they lack is decision-making authority and resources. Marie-Hélène Zérah of IRD extended the point to method, arguing against one-size-fits-all prescriptions in favour of policies that are flexible, context sensitive and grounded in local realities.
The missing data is impact data
The most concrete measurement argument of the day came from Amir AghaKouchak of UC Irvine, and it is one SDGCounting will keep watching. Science can already deliver hazard monitoring and prediction, largely for free. What it cannot supply is what happens next.
Forecasting a drought is a solved problem. Knowing what a drought does to hydropower output, crop yields or school attendance is not, because almost no one collects and shares the impact data systematically. His proposal was institutional rather than scientific: that the UN standardize impact reporting the way agricultural and water statistics were standardized before it. Without that, the models improve and the decisions do not.
Beyond 2030
A closing conversation among Marianne Beisheim (SWP), Norichika Kanie (Keio University) and Edward Carr (SEI) turned to what a post-2030 framework should keep, drop or invent. Beisheim and Kanie both sit in the lineage of the Global Sustainable Development Report, written every four years by an independent group of fifteen scientists, with the next edition due in 2027. Kanie’s contribution was to argue that the interface is misnamed: it is a science, policy and society interface, and the third term is the one usually missing.
Brendan Case of Harvard pressed a more uncomfortable question about the goals themselves, noting that the SDGs concentrate on a narrow band of human outcomes and that optimizing them can impose real trade-offs elsewhere. Carr closed on the structural problem nobody solved: the group knew what needed doing, and the incentives to do it, and to stick with it, were what was missing. The two hardest levers, reforming academic and political incentives and confronting organized anti-science disinformation, drew no concrete answer from anyone in the room.
Why it matters for the SDGs
Science Day sits directly on SDG 17 (means of implementation, data and the Technology Facilitation Mechanism) and SDG 9 (research and innovation capacity), with the day’s substance ranging across climate, cities, water and health. The session ran the same morning as the small island developing states review, and the two rhymed: both described a world where the analytical tools exist and the institutions that would act on them do not. For a project concerned with counting, AghaKouchak’s point is the durable one. The SDG indicator framework measures states of the world reasonably well and measures the consequences of shocks hardly at all.
Watch & read
- UN Web TV, recording of Science Day 2026 (9 July 2026).
- Global Sustainable Development Report 2027, the next edition from the independent group of fifteen scientists.
- International Science Council, the lead convener.
- IDSSD 2026, UNESCO’s science-for-sustainable-development conference · Full HLPF 2026 coverage.
Quotations are lightly edited from an automated (Otter.ai) transcript of the UN Web TV recording and should be read as close paraphrase. Speaker names and titles were reconciled against public records; the transcript garbled most of them. Several audience contributors could not be identified reliably and are cited by role or institution.