The Permanent Impermanence: Can the NHS Learn to Dance with Change?

Part 2: Nature Does Not Have a Filing System

In my last piece here, I asked what a health system designed for constant iteration would look like. This week I want to go one level deeper, because I think the answer starts with a question we rarely ask: are we even looking at the right kind of problem?

Peter Checkland addressed this in his 1981 classic, Systems Thinking, Systems Practice. His argument was deceptively simple. The world is not made up of systems. Systems are intellectual constructs, tools we invented to help us inquire into a complex world. Nature does not consist of physics, chemistry, or biology. These are man-made divisions, convenient ways of carving up the task of investigation. They served us well for a long time. But when we started applying the same logic to human problems, we ran into trouble.

Checkland drew a line between two kinds of thinking. Hard systems thinking treats a problem as an engineering task: define the goal, design the solution, implement it. It works when the problem is clear. Soft systems thinking starts from a different place. It recognises that in complex human situations, the problem itself is often not well defined. The task is not to engineer a solution but to create a structured process of inquiry and learning.

The NHS has spent most part of its recent life being restructured. New structure, new plan, new legislation. Each reorganisation assumes that if we get the architecture right, the system will perform. But the challenges the NHS faces, an ageing population, heterogeneous communities, the blurring of health and social need, do not sit still long enough to be engineered. They are soft systems problems dressed up in hard systems clothing.

We have also replicated the physics-chemistry-biology perspective inside the health system itself. A person does not experience their health in separate departments. Neither does a community. Neighbourhood health has some potential here, but I’m not sure we’ve properly defined it yet.

The reflexive feedback loop I wrote about last week is, in Checkland’s terms, a soft systems idea. It does not assume a fixed goal or a designed endpoint. It assumes that learning is the output, and that the system’s job is to keep the inquiry going.

The Modernisation Bill and the 10-year plan are written in the language of hard systems thinking. That is not necessarily a flaw. Structure has to exist somewhere. But if the structures we build do not contain within them a genuine capacity for soft inquiry, a way of asking whether we are even solving the right problem, then the next reorganisation is already being written.