The most deliberately engineered large-scale collaboration infrastructure in existence — designed from first principles to produce cross-boundary knowledge. What it achieved and where its ceiling is are both instructive for WhatIfWe.
The European Union's Framework Programmes for Research and Innovation — of which Horizon Europe is the ninth generation — represent the most sustained attempt in history to engineer a large-scale collaboration infrastructure from first principles. Since 1984, successive programme generations have refined the mechanisms for forming transnational research consortia, distributing funding to incentivise cross-border partnership, measuring the outputs of collaborative research, and iterating the design based on what each generation revealed.
Horizon Europe requires that funded projects include partners from at least three different EU member states, and actively incentivises broader transnational and transdisciplinary consortia. Its Missions — on climate adaptation, cancer, healthy oceans, smart cities, and soil health — are explicitly designed as cross-disciplinary challenges that cannot be addressed by any single scientific discipline. Recent programme generations have also begun integrating social sciences, humanities, and — in emerging frameworks — indigenous and traditional knowledge into research design.
What nine generations of design iteration have produced is not just a funding mechanism, but an accumulated understanding of what makes cross-boundary collaboration work — and what consistently prevents it. That accumulated understanding is one of the most practically useful bodies of knowledge for anyone designing a new collaboration infrastructure.
The programme has also been honest, in its evaluations, about what it has not achieved. Cross-disciplinary collaboration — genuinely integrating different scientific disciplines, not just assembling them in the same project — remains difficult to produce and difficult to measure. The friction between disciplinary cultures, between academic and applied knowledge, and between formal science and local or traditional knowledge, has not been eliminated by funding architecture alone.
Deliberately engineered collaboration infrastructure — designed, iterated, and refined over decades — can produce cross-boundary knowledge at scale. And nine generations of design reveal that funding architecture alone is not sufficient: what produces genuine integration is something that happens in the people, not just in the incentive structures.
Horizon operates within a geographic boundary (primarily Europe), a sectoral boundary (primarily science and technology), and a cultural boundary (primarily academic and institutional). The deeper ways of knowing that lie outside those boundaries — contemplative, indigenous, spiritual, artistic — are structurally marginal in the programme's design, even as recent iterations have begun to gesture toward them.
More fundamentally: Horizon is an infrastructure for funding and incentivising collaboration, not for developing the human capacity to collaborate across genuine difference. The evidence from its evaluations suggests that this is the binding constraint — that the obstacle to genuine cross-boundary knowledge generation is not funding or incentive structures, but the interior capacity of the people involved to hold radical difference without collapsing it.
The most important thing nine generations of Horizon design reveal for WhatIfWe is what they have not been able to solve: that genuine cross-boundary knowledge generation is constrained not primarily by the incentive architecture, but by the interior capacity of participants to genuinely engage with ways of knowing that are structurally different from their own. This is the finding that points most directly toward what WhatIfWe is attempting.
WhatIfWe is not a funding programme and does not need to be. What it is attempting to design for — the development of the interior capacity to hold radical difference, the cultivation of the quality of attention that genuine synthesis requires — is precisely what Horizon's infrastructure cannot produce, and what its evaluations have repeatedly identified as the missing ingredient. The two projects are not in competition. They are working on adjacent problems, at different levels of the same challenge.