Can an Organisation Forget on Purpose?

AA finance team still runs a manual reconciliation step every month that exists because a specific vendor's invoices, years ago, occasionally double-billed. The vendor relationship ended three years ago. The step is still in the process document, still being performed, and nobody currently on the team was there when it was added — they inherited a rule without the reason, and removing it feels riskier than keeping it, even though keeping it is pure cost with no longer any corresponding benefit. This is the organisational mirror of the half-life question the previous page in this cluster asked. That page was about knowledge an organisation loses by accident when the person holding it leaves. This one is about whether an organisation can get rid of knowledge — or a habit built on it — deliberately, and what happens when it can't.

Voluntary and Involuntary Forgetting FoundationalKnowledge that endures for decades — core principles

PabloPablo Martin de Holan and Nelson Phillips's study of organisational forgetting treats it as a real, nameable phenomenon with more than one shape, not a single failure mode¹. Some forgetting is involuntary: knowledge an organisation once had simply decays away when the people who held it leave and nothing was done to retain it, the mechanism the previous page in this cluster covers in detail. Some forgetting is voluntary — an organisation can deliberately discard a routine, a rule, or an assumption it has decided no longer earns its place, the way a growing company might deliberately retire an approval process built for a ten-person team once it has two hundred people and the process is pure friction rather than a safeguard. The reconciliation step above sits at an uncomfortable point between the two categories: the knowledge of why it existed was lost involuntarily, and the step itself survives not because anyone decided it should, but because nobody with the standing or the information to decide against it ever looked at it directly.

Why Outdated Assumptions Survive FoundationalKnowledge that endures for decades — core principles

ForgettingForgetting on purpose turns out to be harder than remembering on purpose, for a structural reason. Chris Argyris and Donald Schön's account of organisational learning distinguishes an organisation's stated values from its theory-in-use — the assumptions actually embedded in what it does, which are frequently invisible to the people carrying them out precisely because they're baked into a routine rather than held as a belief anyone currently examines². A process document doesn't announce "this step assumes vendor X still double-bills"; it just says "reconcile the invoice," and the assumption that made the step rational survives inside the instruction long after anyone remembers it was ever there to question. An organisation can't deliberately forget an assumption it can no longer see is operating — which is exactly why so much institutional behaviour outlives the conditions that justified it, not from any individual's reluctance to change but from the assumption having become invisible before anyone had reason to challenge it.

Why Organisations Repeatedly Relearn Old Lessons FoundationalKnowledge that endures for decades — core principles

TheThe sharpest version of this problem shows up at field scale, where this site's page on AI's research winters already documents a case in detail: a field can genuinely forget — in the sense that a finding or approach drops out of active attention and funding — and then relearn essentially the same lesson a decade later, often under a new name, because the people who held the original lesson have moved on and nothing durable enough survived their departure to flag the repetition as a repetition. The AI winters page treats this at the scale of a discipline; the same mechanism runs inside a single organisation whenever a reorganisation, a wave of departures, or simply enough time scatters the people who remember why the last attempt at something failed. The second team to try the same approach rarely does so out of arrogance. They usually just don't know there was a first team, because whatever that team learned decayed exactly the way organisational knowledge generally decays once its one reliable retrieval path left the building.

Strongest Objection: "Deliberate Forgetting" Is Easy to Fake After the Fact FoundationalKnowledge that endures for decades — core principles

CallingCalling a loss of institutional memory "strategic forgetting" is a flattering story an organisation can tell itself after the knowledge is already gone, and the two are genuinely hard to tell apart from outside. An organisation that retired a safeguard because it had properly assessed the safeguard as no longer necessary, and one that simply let the safeguard lapse when the person who maintained it left and nobody noticed, can look identical in the record — a rule that used to be followed and now isn't. The only real test is whether a decision was actually made, by someone with the authority and the information to make it, rather than whether a plausible justification for the absence can be constructed afterwards. That test is much harder to apply than it sounds, because the second case's defenders rarely describe it to themselves as an accident while it's happening.

Provisional Conclusion FoundationalKnowledge that endures for decades — core principles

AnAn organisation can forget on purpose, in the narrow sense of deliberately retiring a rule or routine it has properly examined and found wanting — but doing so requires first making an invisible theory-in-use visible enough to examine, which is itself the harder half of the job, not a precondition it can take for granted. Most of what looks like deliberate organisational forgetting, on closer inspection, turns out to be involuntary forgetting wearing a retrospective justification, and the two are worth telling apart before crediting an organisation with a discipline it may simply have gotten lucky with.

Questions for Further Thought

• What would it take for an organisation to regularly audit its own routines for assumptions that are no longer true, rather than discovering them only when a rule's absurdity becomes impossible to ignore?
• Is there a meaningful difference between an organisation "deciding" to forget something and an organisation simply not noticing it still remembers something, once the decision and the noticing happen at different times and by different people?
• Does documenting the reason behind a rule, at the point the rule is created, meaningfully reduce how often this problem happens later — or does the documentation itself just become one more thing nobody reads once the person who wrote it has gone?

Related Topics

• The Half-Life of Organisational Knowledge — the involuntary side of forgetting this page's voluntary/involuntary split is paired against.
• Does a Research Field Sleep? Collective Consolidation and the AI Winters — the same relearning-old-lessons pattern, documented in detail at the scale of a whole research field.
• Is a "Learning Organisation" a Real Claim or a Compliment? — what it would actually take for an organisation to routinely uncover and revise its own theory-in-use, rather than only when forced to.

Further Reading

• de Holan, P. M., & Phillips, N. (2004). Remembrance of things past? The dynamics of organizational forgetting. Management Science, 50(11), 1603–1613.
• Argyris, C., & Schön, D. A. (1978). Organizational Learning: A Theory of Action Perspective. Addison-Wesley.

References