Learning, entropy and what keeps institutions evolving
In 2024 I wrote here about finding Jane Jacobs’ The Nature of Economies in a second-hand bookshop in Oxford. What I took from her, and have carried ever since, is that development is not a line. It “operates as a web of interdependent co-developments.” Nothing develops in isolation, and the web keeps expanding in its complexity.
Jacobs has stayed with me since 2005. Problem-driven political economy analysis, Problem-Driven Iterative Adaptation, the work on complexity in development that my colleague Harry Jones did when I was at ODI’s RAPID programme: each gave me a stronger sense that the response to development challenges has to work with the context as it is, not as the plan imagined it. Recently I found a new vocabulary that helps me make sense of it. Change in a living system happens in the relationships, not in its elements. And what changes relationships is learning.
This post is part of working out what that means.
In June this year I wrote about entropy and sustainability, after a field visit to a government-led poverty reduction programme in Rwanda. When I wrote it, I didn’t connect it to the post on Jane Jacobs. I do now, because I can see the connection, and it has changed how I think about what I wrote last June.
The second law of thermodynamics says that in a closed system, entropy increases. Order runs down, differences even out, and a hot cup of coffee cools to the temperature of the room. Living systems, and human systems are living systems, hold their order only by staying open, taking in energy from outside, all the time. When the flow stops, the order dissipates and the system winds down.
In June I argued that public delivery systems, as human systems, work the same way. They need a steady flow of energy to fight entropy and avoid sliding into stagnation. I listed what I thought that energy was: funding, technical partners, leadership, data.
I still think the argument holds. What I got wrong was what I meant by energy. I have a better sense of it since the Warm Data course in Finland in Nuuksio, and the time I spent there with Gregory Bateson‘s thinking on systems theory and complexity.
Institutions are made of people. The energy in an institution is its people: their attention, their effort, their willingness to respond. And what keeps that energy from running down is not only more inputs. It is people continuing to notice differences, and the learning that comes from noticing.
What do I mean by that?
Bateson defined information as a difference that makes a difference (Steps to an Ecology of Mind, 1972). Information is not itself energy. The energy to do something with information belongs to whoever notices the difference it points to.
One late afternoon in Nuuksio last August, I went for a walk in the forest. I was following the path. The forest around me was one uniform pattern: trunks, pines, birch trees, moss, stones, green in every direction. Then I noticed something. A change in that pattern. A deer, maybe 200 metres away, standing still and looking at me. It had probably smelled me. I stopped walking and looked at it. I took out my camera and took a photo, then decided not to disturb it and walked slowly backwards. [link to the Photos and Words post, where the photo is] For Bateson, this is where all learning starts. A living system, human or not, can only take in differences, never things in themselves (Mind and Nature, 1979).
The same is true of the data and information we collect in our work on strengthening social systems. We may have a baseline, and some time later a midline. Neither tells us much on its own; the information is in the difference between them. A midline report is full of differences, but until someone notices them and decides to do something, they just remain differences.
So, what holds off entropy in an institution is learning: an ongoing process of people noticing the differences that matter and doing something with them. That is also Jacobs’ web. Nothing develops in isolation, because development happens in that continuous exchange.
Let’s go back to the field visit in Rwanda. The livestock distribution was ready: the financing, the procurement and the logistics were all in place. Then it risked stalling, because ministerial regulations required recipients to own a minimum parcel of land, and most of the households the programme meant to reach did not. A local leader broke the stall. I am simplifying a bit, but he negotiated a local exception for the most vulnerable families as a parallel track for the others.
In June I wrote that he brought new energy into the system. I’d put it differently now. He noticed a difference that made a difference: the regulation and the programme’s purpose didn’t align with the people it was meant for. The energy to act on it was his own. What kept the system from stalling was not an input from outside. It was someone inside who noticed, and responded.
If the energy is in the people, then the real question is what keeps them noticing. I don’t think there is one method or approach. It is a combination of mindsets, methods and spaces. Curiosity and critical thinking when looking at numbers and information. Data that is quantitative, qualitative and experiential: some of it collected, recorded and analysed. And some of it cannot be recorded or documented at all. It exists only in the conversations between people, where different perspectives meet, and it keeps combining in new ways every time people interact. Noticing depends on all of it together, and on spaces where people recognise that everyone sees things differently, and want to learn how others see them.
We tend to treat stability as what happens when nothing is happening. In development, we often define sustainability the same way: the solutions or reforms that continue, as if on autopilot, once a programme ends. I now think it is the opposite. An institution that stops noticing difference is not standing still. It is not sustainable; it is running down, even if nothing looks wrong yet.
The question for anyone working on learning is not only how much information is flowing. It is also whether anyone is noticing.
As always, let me know what you think.