The Limp - How Patterns Form In Organisations

Around 5,600 people in the UK walk, hobble or get carried into A&E with a sprained ankle every day. It accounts for three to five per cent of everything A&E sees, and it is a fraction of the real number, because most of us never go. We sit on the sofa with a bag of frozen peas and watch the swelling come up over a few days in the colours of a Monet, purple through green into an unhappy yellow.

The ankle is the most commonly injured joint in the body because it is the joint absorbing the shock every time the ground turns out not to be where you thought it was.

The body is built to cope with that. Well, sort of.

Within a second of going over, before the pain has finished arriving, the body is already changing how you walk. Weight off the bad side, shorter stride, a small roll through the opposite hip. Nobody taught you this and you could not have done it deliberately if you tried. It is a system adjusting to an incident, the way ants pour towards a breach in the nest wall.

Six weeks later the ligament has healed. Here is the funny thing: roughly four in ten people never go back to moving the way they did before. Repairing a ligament and unlearning a walk are separate jobs, and only the first one happens on its own. A 2020 meta-analysis of people with lasting ankle instability found the work redistributed up the leg, with load taken off the ankle and handed to the knee and the hip. The hip takes on a job it was not built for, the spine adjusts to the hip, the neck adjusts to the spine. Two years on, something gives. Usually the knee. The physio treats the knee, and for a fortnight it is better. The ankle is a distant memory, and anyway, the knee is the problem. Right?

No stage of this is a mistake. At every point the body solved the problem directly in front of it, using what it had, and each solution became the ground the next one stood on. At no point did anything go back to check whether the ankle was still causing pain. It is motorway maintenance setting up a diversion, fixing the road, and never taking the cones away.

The same thing, in a building.

Say to a room of senior people that their organisation is run by patterns nobody chose and they agree. They reach for phrases like the walls have ears, or the building has a memory. Ask how those patterns got there and the conversation stops.

The answer is in the limp. Something costs somebody something: a launch that failed in public, a redundancy round handled badly, a founder who would not let go, a client who sued. Somebody adjusts to carry the load, and because the adjustment works, it stays.

Then the injury heals or the people involved leave, and the adjustment carries on. Ten years later it is being taught, unconsciously, to new starters by people who were at school when it happened.

The cost, the pain, shows up somewhere else entirely. In the meeting that can no longer change anything. In the objection that has nowhere to land. In the customer who has been absent from the room so long that nobody notices the absence. Load comes off the ankle and goes up the chain, and the place it hurts is not the place to treat.

Why nobody can tell you why.

In the mid-1970s Richard Nisbett and Timothy Wilson set up a table in a bargain store and asked shoppers to pick the best of four pairs of nylon stockings hanging side by side. The pairs were identical. People chose the one on the right roughly four times as often as the one on the left, and when asked why, they talked about the knit, the sheerness, the weave, the workmanship. Eighty reasons between them, and not one mentioned position. Told directly that position might have swayed them, almost everyone denied it, some of them looking at the interviewer as though he had lost his mind.

The reasons people gave were genuine answers to the question, produced on demand, to fill a gap, and they arrive feeling exactly like a memory. Which makes asking people why they do certain things a poor way of finding out their motivation.

Chris Argyris taught at Harvard for most of his career and spent a good part of it on the corporate version of the same problem. His best-known study looked at young management consultants, brilliant people who had never really failed at anything, and found they were the worst learners in his sample. The moment anyone suggested they had got something wrong, they blamed the client, the brief, the market. His hypothesis was that their sense of who they were depended on being right, and that organisations get built around protecting exactly that.

What makes it stick.

There is a run of fox tracks through my mother's garden. The foxes change year on year as new ones are born, some stay and some leave to find their own territory. The run does not change. It comes in at the same corner, crosses on the same line, and takes a wide bend around a spot where a shed stood until it collapsed and was cleared away. The bend is still there, taken by foxes that never saw the shed. Cutting the corner would be quicker. Following the line somebody else wore is easier.

Dirk Helbing was a theoretical physicist at Stuttgart when he and his colleagues published this in Nature in 1997, with a companion paper that year in Physical Review E extending it to animal trails. A used route becomes easier to use, ease attracts use, and the traffic bundles onto fewer and deeper lines until the alternatives stop being real options. They tested the model against the campus of the University of Brasília, where students had worn their own path system across a piece of high modernist planning. The simulation, run on a rule as simple as trodden ground is easier to walk on, produced the same shapes.

The same selection was demonstrated on people in 1971, by Karl Weick, then a psychology professor at Minnesota and later the man who gave organisational studies the idea of sensemaking. Weick and Gilfillan gave small groups an arbitrary procedure for a task, then replaced the members one at a time across eleven generations. Awkward procedures were dropped by about the fourth generation. Easy ones survived the whole run, with equally easy alternatives available throughout. What lasted was whatever was cheapest to keep doing, and cheapness keeps no record of where it came from.

Which is the trouble with values on a wall. A sign saying keep off the grass addresses a choice, and that choice was cashed in by the first animal across.

Building over it.

Mfuwe Lodge in Zambia went up in 1998 across a route elephants had been walking for far longer, to a wild mango tree that now stands in its courtyard. Every year from late October the herd walks through the open reception to reach it, up the steps, across the tiles and out the other side. At least three generations of the same family have made the trip. The lodge ended up designed around them, lobby left open, because closing it was never going to work.

Reorgs run into this constantly. The boxes move, the reporting lines are redrawn, and the traffic keeps going down the old line because the old line is still the cheapest way through. Six months later somebody reports that the new structure has not bedded in.

After everyone has gone.

Donald Campbell spent his career at Northwestern teaching social science how to tell a real effect from an artefact. In 1961, with Robert Jacobs, he turned that on culture itself. They sat confederates in a group judging how far a pinpoint of light appeared to move in a dark room, an illusion with no correct answer, and had them insist on a wildly inflated figure. Then they removed the confederates one at a time, replacing each with a newcomer. Traces of the invented norm carried on for four or five generations after the last confederate left.

Muzafer Sherif had invented that method as a young researcher in the 1930s, having arrived in America from Turkey, and went on to run the Robbers Cave study, taking two groups of eleven and twelve year old boys to a summer camp to manufacture a war between them and then take it apart again. By 1976, in his seventies, he was still on the question, and he and Mark MacNeil repeated the 1961 design with the planted norm made less extreme. Still invented, still arbitrary, but within the range of something a reasonable person might say. Every original member was gone by the fourth generation. At the eighth, more than a quarter of what the newest arrivals said still fell inside a range made up by strangers they had never met. The wildly extreme version collapsed within four.

Arbitrary norms survive on plausibility. Anything obviously daft gets corrected by the next intake, and anything that sounds like judgement gets inherited and defended.

Nobody says we keep the customer out of the room. People do say we're careful about who we expose to clients, and mean it, long after the last person who knew why has retired.

Treating the knee.

The standard response to a pattern is to work where it hurts. Trust scores are down, so there is a customer-centricity programme. Nobody speaks up, so there is a psychological safety workshop. Middle management looks like the blocker, so middle management gets the budget.

All of it lands on the knee, which is already carrying load that was never its to carry and now has a development plan as well. The spend is real, the evaluation scores well, and the same conversation is waiting a year later under another name.

Where it opens.

You cannot talk somebody out of a limp. They do not experience it as a limp. They experience it as walking, and they will defend it honestly.

Recognition is what moves it, and recognition only happens under certain conditions. It has to happen while the pattern is running rather than while it is being discussed. The load has to become visible for long enough that the person can feel what they are carrying and see whose it was to begin with. And it closes again, fairly quickly, as the system reorganises around the gap.

The shed came down years ago and the foxes still go round it.

Notes on the evidence.

Ankle sprain incidence. Approximately 5,600 acute ankle sprains present to emergency departments in the UK each day, accounting for three to five per cent of all ED attendances (figures repeated across the UK clinical literature, including the SALI cohort study protocol, BMC Musculoskeletal Disorders, 2021, and the West Midlands population-based epidemiology study). Community incidence in European population studies runs at roughly five to seven per thousand person-years for sprains that present for treatment. Honest label: the daily figure is an estimate carried forward across several papers rather than a single national count, and unreported sprains are not captured in any of it.

Compensation after injury. DeJong, A. F., Koldenhoven, R. M., & Hertel, J. (2020). Proximal adaptations in chronic ankle instability: systematic review and meta-analysis. Medicine & Science in Sports & Exercise, 52(7), 1563–1575. doi:10.1249/MSS.0000000000002282. The estimate that roughly 40% of lateral ankle sprains progress to chronic ankle instability is the figure commonly reported in this literature (Doherty et al., 2016; Hertel & Corbett, 2019). Honest label: cross-sectional, laboratory-based, typically small samples, and a 2026 umbrella review rated confidence in most of the underlying reviews as low or very low. The kinetic-chain finding is consistent and repeatedly observed. It is used here as an analogy rather than a causal claim.

Explaining behaviour we cannot see. Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: verbal reports on mental processes. Psychological Review, 84(3), 231–259. doi:10.1037/0033-295X.84.3.231. The stockings study is reported in that paper: four identical pairs on racks in a bargain store, chosen left to right by 12, 17, 31 and 40 per cent of shoppers, around eighty reasons given and none referring to position. Honest label: a review paper rather than a single experiment, and among the most cited in the field. Argyris, C. (1991). Teaching smart people how to learn. Harvard Business Review, 69(3), 99–109, and Overcoming Organizational Defenses (1990).

Routes that deepen with use. Helbing, D., Keltsch, J., & Molnár, P. (1997). Modelling the evolution of human trail systems. Nature, 388(6637), 47–50. doi:10.1038/40353. Helbing, D., Schweitzer, F., Keltsch, J., & Molnár, P. (1997). Active walker model for the formation of human and animal trail systems. Physical Review E, 56(3), 2527–2539. doi:10.1103/PhysRevE.56.2527. The Brasília comparison appears in the second paper, using material documented by Klaus Humpert. Honest label: these are simulations. The claim is that a simple reinforcement rule reproduces trail structures observed in real green spaces and animal trail systems.

The lodge. Mfuwe Lodge, South Luangwa National Park, Zambia, built in 1998 on an existing elephant route to a wild mango (Cordyla africana) in what became its courtyard. The herd passes through the open reception between late October and mid-December, and at least three generations of one family have returned annually. Documented by the operator, the Zambian tourist board and international press. Honest label: an observation, not a study.

Traditions that survive replacement. Weick, K. E., & Gilfillan, D. P. (1971). Fate of arbitrary traditions in a laboratory microculture. Journal of Personality and Social Psychology, 17(2), 179–191. doi:10.1037/h0030461. Jacobs, R. C., & Campbell, D. T. (1961). The perpetuation of an arbitrary tradition through several generations of a laboratory microculture. Journal of Abnormal and Social Psychology, 62, 649–658. doi:10.1037/h0044182. MacNeil, M. K., & Sherif, M. (1976). Norm change over subject generations as a function of arbitrariness of prescribed norms. Journal of Personality and Social Psychology, 34, 762–773.

In MacNeil and Sherif the natural norm was around four inches. The moderately arbitrary planted norm ran from nine to fifteen inches around a mode of twelve. All three confederates were out by the fourth generation, and at the eighth transmission generation over a quarter of judgments still fell inside the planted range. The most arbitrary condition, fifteen to twenty-one inches, was down to around five per cent by the fourth transmission generation. Honest label: small laboratory studies with few replications per condition, run on male high-school students judging an optical illusion. They model cultural transmission rather than measure it. The convergent finding across all three, and the cultural evolution research that followed, is that transmission strength varies inversely with how obviously arbitrary a norm looks to the people receiving it.