If systems are brought forth by observers, governing a shared situation becomes harder, because nobody holds the final view. This has been my concern for a long time.
My starting point was to take Maturana’s structural coupling and apply it at scale. We habitually speak of “the environment” as something outside us to be managed. I have argued instead that what we can distinguish as human social life and as the biosphere have always been structurally coupled, in a long history of reciprocal perturbation. Governance that treats nature as an external object to optimise fails, because it assumes we can instruct what we can only perturb. I call the habits behind this (linear causality, prediction, optimisation, the separation of humans from nature) the Anthropocene problematique.¹
I try to be precise about “institutions”. I use the word as most institutional economists do: humanly devised rules and norms, such as property rights, water licences, funding formulas, and professional standards. They may also be informal rules e.g. we have dinner here at 6pm. Institutions are distinct from organisations, thought organisations comprise many institutions. Organisational change is often confused with the need for institutional innovation and change. My work with Phil Wallis on the Murray-Darling Basin shows how institutional complexity, layers of overlapping and often incoherent rules built for an earlier paradigm of stability and control, constrains what adaptive governance can do. The idea of “affordances” helps explain how. A licensing regime or a funding rule makes some collaborative actions easy to imagine and forecloses others, whatever the participants intend.
The reframing I find most useful is to treat institutions as “social technologies”. Drawing on Don Ihde’s idea of a technosphere², in which technologies actively mediate how we meet the world, I argue that a property regime mediates our relationship with the biosphere just as a dam (an artefactual technology) does. That dissolves the line between “hard” artefacts and “soft” rules, and it removes the excuse that institutions are natural facts. They are designed, so they can be redesigned and removed.
Now to goals. If institutions shape what we can do, consider the one that dominates research funding, development aid, and corporate change programmes: the project, with its fixed goals, deliverables, budget, and end date. I call the spread of this form “projectification”³, and it descends from the linear model of innovation, in which knowledge flows from research to development to diffusion to adoption.
The trouble is structural. A project must know in advance what will count as success. In a genuinely complex and uncertain situation, what is worth learning emerges only through the inquiry itself. Checkland’s objection to goal-seeking models returns as an objection to our funding and evaluation machinery.
My alternative is not to abandon planning that seeks purposeful action. It is to “design and enact learning systems”: arrangements of people, institutions, and technologies built to sustain systemic (co-)inquiry, in which participants’ framings, and even the boundary of the situation, stay open to revision. A project asks whether the target was hit. A learning system asks what we have learned together and whether we are better placed to keep learning and effecting change that is improving or transforming. This is also how I understand “social learning as a governance mechanism”, a fourth way of coordinating alongside hierarchy, markets and fixed knowledge production. The difficulty, which Kevin Collins, Phil Wallis and I have documented, is that social learning rarely outlasts a pilot unless it is written into rules, funding, and accountability i.e., into governing practices whether for a program, an organisation or a nation state. Those are precisely the institutions that hierarchy and markets have built.
- Khayame, Houda A., Kevin B. Collins, Ray L. Ison (2026) How may the ‘Anthropocene’ enable and constrain cybersystemic governance? The ‘Anthropocene problématique.’ Systemic Practice and Action Research https://doi.org/10.1007/s11213-026-09764-3
- See his book Technology and the Lifeworld: From Garden to Earth (1990)
- Following the work of Simon Bell and Stephen Morse – see Delivering sustainability therapy in sustainable development projects (2005), Journal of Environmental Management 75, 1: 37-51.
