Last month I introduced a concept from the cyberneticist Stafford Beer that I have not been able to stop thinking about since: the purpose of a system is what it does. Not what it intends, not what it was designed for, not what its participants believe it to be working toward – what it actually, consistently produces. Beer developed this as a tool for organizational analysis, a way of cutting through the gap between stated mission and observable behavior. What I want to explore this month is where that tool becomes hardest to apply and most necessary – at the level of personal identity.
I have finished nearly a dozen Ironman triathlons. I have run ultramarathons through deserts and alpine mountains, swum distances that make people react with a particular expression of confusion and concern. These accomplishments are real; they cost something real, and I am not disowning them. But I want to be honest about something they also produced, which is an identity, and about what that identity cost me in a less visible way.
The identity is layered. There is the general claim: I am an athlete. Underneath it, the specific one: I am an Ironman, an ultra runner, an open water swimmer. And underneath that, something more powerful than either: I am someone who takes care of himself. That last layer is the most defended, because it is not an achievement claim but a self-concept, and self-concepts do not respond to evidence the way achievements do. You can lose a race and still be a runner. It is harder to see an unflattering photograph of yourself and still be, without revision, someone who takes care of himself.
The photograph last summer was not just a confrontation with a body. It was a confrontation between an identity and its output data. And the identity had been working hard, for longer than I realized, to make sure that confrontation never happened.
How Identity Drifts from Reality
The mechanism worth understanding is drift, because that word is more honest than denial and more useful. Each day looks like the day before. The mirror presents the same face it presented yesterday, and the slow accumulation of change is invisible inside the continuity of daily looking. This is not active self-deception. It is a feature of how perception works when the reference point keeps moving alongside you. The person who gains a pound a month does not feel themselves gaining twelve in a year. They feel, each morning, approximately the same as the morning before.
But the identity does more than fail to notice – it actively participates. The mirror is not a neutral instrument; it is interpreted through the story you have already decided is true, and the story has a strong interest in its own continuation. The unflattering angle gets dismissed. The scale, never checked, cannot contradict anything. Yesterday’s training session becomes evidence of a system working, regardless of what the system is actually producing. This is why Beer’s concept is hardest to apply exactly where it is most needed. You can assess a training plan’s output with relative level-headedness, because the training plan is not you. Assessing the output of the system you have built your self-concept around requires letting the evidence supersede the story even when the story is who you believe yourself to be.
The emotional architecture underneath this is worth naming. We are attached to our self-concepts emotionally, in the specific sense that a stable identity produces comfort, the comfort of not having to renegotiate who you are with every piece of incoming data. “I am this” and “I am not that kind of person” are emotional load-bearing structures, and the system maintaining them will filter and discount contradicting evidence not as conscious dishonesty but as an automatic protective function, the way the immune system responds to a foreign body before the brain registers a threat. This is why the person who identifies as healthy can be genuinely surprised by an assessment that says otherwise. The surprise is real. The identity system was doing its job well enough that the gap between belief and reality had become twenty-five pounds before the evidence finally arrived in a form the system could not reframe.
Organizations Have Identities, Too
This same tool operates at institutional scale, on the same mechanism, producing the same blind spots. The organization that identifies as customer-focused but whose output shows unresolved complaints and declining retention is not lying about its values. It genuinely believes the claim. But Beer does not assess what an organization believes about itself. He assesses what it does. The mission statement is the mirror; the customer data is the photograph. The leader who identifies as decisive but whose team has quietly learned to work around delayed and revisited decisions is running the same protective function. The operator who identifies as financially disciplined but has not looked honestly at the margins in longer than they would admit is running a drift of exactly the kind I described in the mirror, each month looking like the last, the accumulation invisible inside the continuity of daily operation.
The fix, in both contexts, is identical. It is not a new mission statement, not a revised set of values, not a more compelling articulation of who you are. It is honest output assessment: the willingness to bring the photograph into the room alongside the mirror and look at both without the story running interference. The operators in this industry who do that regularly, who have built honest feedback mechanisms into how they run their organizations rather than relying on the self-concept to report on itself, are running a fundamentally different kind of system than the ones who are not.
The redesign, at the identity level, is not about abandoning who you are. It is about letting output data into the self-concept regularly enough that the drift cannot accumulate undetected. The person who steps on the scale weekly is not someone who distrusts themselves; they are someone who has decided the story and the evidence should stay in regular conversation, and that when they diverge, the evidence takes precedence. This is the kaizen principle applied at the deepest level. Not small adjustments to behavior, but small, regular doses of honest data that keep the identity calibrated against reality. The adjustment required when you check weekly is always smaller than the one required when a photograph tells you what a year of drift has produced. The system that builds honest assessment into its rhythm never has to be shocked back to reality, because it never travels far enough to require the shock.
There is one more thing to name, because the “I am not that kind of person” defense operates in both directions. It protects the positive identity: I am an athlete, I take care of myself. And it blocks change when that identity is no longer accurate. But it also protects the limiting identity: I am not someone who loses the weight, who changes the system after a year of evidence. Both versions serve the same function and cost the same thing – the gap between who you are and who you are capable of becoming, measured in output data you were not quite willing to look at. The photograph told me something I had not wanted to know. What I did with it was decide the evidence mattered more than the story. That is not a comfortable decision. It is the only one that changes anything.
Challenge for the Month
Apply Beer’s lens to one dimension of your life, and make it concrete. Pick the single identity claim doing the most protective work – I take care of myself, I’m financially disciplined, we’re customer-focused – and name the one output measure you have been avoiding: the scale, the margins report, the retention numbers, the complaint log. Then actually look at it, and put the checking on a schedule you cannot drift past: weekly for the personal metric, monthly for the professional one. The goal is not a verdict. It is to close the distance between the last time you looked and the next time, so the gap never again grows to the size that requires a shock to notice.
