Field Note #7 ∷ On the Need for Updating One's Priors

"Keep moving, and get out of the way."
— Merlin Mann & John Roderick, "Roderick on the Line"

Sometime around 2008, I found myself in a bunker beneath SHAPE — Supreme Headquarters Allied Powers Europe in Mons, Belgium — sitting in a meeting scheduled for ninety minutes that was now entering its third hour.

The fluorescent lights hummed softly above the table.

The agenda had long since been exhausted.

What remained was a kind of institutional performance: the careful choreography of who spoke when, who deferred to whom, and which objections were entered into the record without any real expectation that they would change anything.

I was the youngest person in the room by a considerable margin.

I remember thinking: how does this happen to an organization?

Not as a complaint. As a genuine puzzle.

That evening I went home and ordered a copy of Parkinson's Law off eBay for ten dollars. First edition, 1958, John Murray. It arrived with a Cambridge University library stamp inside the cover — Board of Extra-Mural Studies, Stuart House.

It has been sitting on my shelf ever since.

In 1955 C. Northcote Parkinson — a British naval historian in his day job — published in The Economist what he described as a scientific law: work expands to fill the time available for its completion.

Everyone knows the phrase. Almost no one has actually read the essays.

His book itself is savage — and, in the way of the best satire, entirely serious. Parkinson was not making a joke about procrastination. He was describing a structural mechanism by which bureaucracies grow independent of — and eventually inverse to — the work they were originally created to perform.

He even had the numbers.

Between 1914 and 1928, the British Admiralty's capital ships in commission fell by 67%. Officers and enlisted men fell by 31%. Yet Admiralty officials increased by 78%. The bureaucracy administering a shrinking navy was itself expanding at nearly six percent per year.

The mechanism he identified was not laziness or malice. It was structural: the incentive to multiply subordinates rather than rivals; the work those subordinates generate for one another; the resulting need for coordination; the coordinators who then require coordinators of their own.

Seven officials now doing what one once did — not because anyone consciously chose that outcome, but because each individual decision along the way was locally rational.

Parkinson called the terminal stage the Age of Obstruction.

The official who has passed through Qualification, Discretion, Promotion, Responsibility, Authority, and Achievement arrives finally at a stage where their primary institutional function is to prevent others from doing what they themselves once did.

They are not villains.

They are simply done updating their priors.

The essay goes considerably further. Parkinson works out, in what he presents as rigorous algebra, the precise career stages through which any ambitious person must pass — from Qualification through Discretion, Promotion, Responsibility, Authority, Achievement, Distinction, Dignity, and Wisdom — before arriving finally at Obstruction. He even calculates, given a starting age Q (22 years), exactly when the crisis occurs between an incumbent X (62 years) and their destined successor Y (47 years). His arithmetic is devastating — and entirely too recognizable to anyone who has been in the workforce more than a few years.

Parkinson also traces the alternative path — the one invariably taken by Y, if the incumbent X stays too long. Having arrived at the Achievement state but thwarted, the alternative sequence ends in Frustration, Jealousy, Resignation, and Oblivion. As he puts it, "It will be apparent, however, from the other symptoms described, that the man still in a subordinate position at 47 (or equivalent) will never be fit for anything else."

Parkinson was writing about bureaucracies. Four years later, Thomas Kuhn arrived at the same diagnosis from inside science itself.

In The Structure of Scientific Revolutions, Kuhn argued that paradigm shifts rarely occur until long after the evidence becomes overwhelming. The anomalies accumulate for years — sometimes decades, even millennia. The defenders of the dominant framework become increasingly elaborate in their explanations of why the anomalies do not really matter.

And then, gradually, the generation that built its identity around the old framework retires.

The next generation — which never grounded its intellectual life on the old paradigm to begin with — simply moves on.

Max Planck put it more plainly: "Science advances one funeral at a time."

This is not a cynical observation. It is a structural one.

A footnote worth pondering: Parkinson published his essays in The Economist in 1955, and as a book in 1957 — four years before Kuhn began circulating the manuscript that would become The Structure of Scientific Revolutions. I couldn't find evidence that Kuhn read Parkinson. The Economist had broad transatlantic reach among exactly the kind of educated generalist Kuhn was, and the book was an immediate sensation.

It is nonetheless tempting to imagine the two in convivial conversation.

What seems more likely, and in some ways more interesting, is that they arrived independently at the same structural observation — one from naval history, one from the history of science — which is itself a kind of evidence that the mechanism they were describing is real.

A prior is not merely a belief held in the mind. Over time it becomes a career, a reputation, a community of practice organized around that belief.

Updating the prior is therefore not merely an intellectual act. It can feel like a small form of self-dissolution.

And most people — quite understandably — resist dissolving themselves.

The tragedy is that institutions cannot wait indefinitely for individuals to resolve this tension.

So they wait for the funeral — and call it adaptation.

What has changed, perhaps, is the environment in which this dynamic now unfolds.

For much of the twentieth century, our institutions could afford to wait. The surrounding world moved slowly enough that generational turnover, however inefficient, was still a viable mechanism for adaptation.

That assumption is becoming less reliable.

We now operate inside what is increasingly described as a polycrisis: overlapping systems under stress, interacting in ways that are difficult to model and harder to manage. Infrastructure, climate, geopolitics, and information systems no longer evolve independently; they are tightly coupled.

In such an environment, the cost of waiting for funerals is no longer theoretical.

Over the past few decades of working life I have watched some version of this dynamic play out in organizations I care deeply about: standards bodies, security communities, and the quiet governance institutions that help keep the Internet functioning.

These spaces are filled with brilliant people who did genuinely important work — people who built things that matter and earned their authority through real contributions.

But some of them — not all, but some, and often those who arrived earliest and stayed longest — eventually drift into Parkinson's Age of Obstruction without ever realizing it.

The signals are remarkably consistent across institutions.

New proposals are evaluated first by their provenance rather than their substance. Dissenting voices are carefully managed rather than seriously engaged. Institutional memory — which is genuinely valuable — begins to function less as a resource and more as a veto.

The founding cohort's priors, formed under earlier conditions with different tools and constraints, gradually become the implicit standard against which all new thinking is measured.

And usually found wanting.

The irony, which Parkinson understood perfectly in 1958, is that this phenomenon tends to be most advanced inside the most successful organizations.

Success reifies the priors that produced it.

Frameworks that worked become encoded into process, hiring, incentives, and the unspoken grammar of what counts as a serious contribution.

By the time the founding generation reaches the Age of Obstruction, the institution itself has become a machine for enforcing the very assumptions that once made it successful.

Reform from within does not usually fail because reformers are weak.

It fails because the institutional immune system is functioning exactly as designed.

This is the point where I have to be honest — with you, and with myself.

I am forty-eight years old. I have spent more than a decade thinking about variations of the same problem space. Over that time I have developed frameworks, vocabulary, and intellectual and social commitments that I find genuinely compelling. I have also staked a considerable amount of professional credibility on very specific claims about how systemic infrastructure risk works.

In other words, by now I have significant priors of my own.

The question Parkinson forces me to ask — uncomfortably, but fairly — is this:

Where am I in this sequence?

Have I passed through Responsibility and Authority into something that is beginning to resemble Obstruction disguised as conviction? Are there anomalies I am failing to see — or perhaps choosing not to see — because acknowledging them would require me to revise ideas I have already built part of my identity around?

I do not know.

Incompleteness — as an earlier Field Note suggested — is not a flaw in complex systems. It is a necessary condition of their existence.

The same may be true of the person writing this.

The willingness to keep asking the question may, in fact, be the entire practice.

One useful method: stay close to the youngest generation in your workplace. That is where the paradigm tensions surface most clearly, and most honestly — before they have learned to manage their dissent into cynicism.

The Age of Obstruction rarely arrives as a dramatic event. More often it forms gradually, in the dark, through a thousand small decisions not to update. Each decision locally reasonable. Each one a tiny calcification.

Until one day you find yourself sitting in a bunker, wondering how this happened to your organization — and realizing, with a quiet shock, that you have become part of the problem.

Anyone sitting in traffic, hating being stuck, must at some point admit they are also traffic.

Learning requires something institutions quietly discourage: the willingness to look foolish, to fail publicly, and to admit that one does not yet know what one is doing.

In that sense, the most difficult habit to sustain as we age may be the simplest: treating our own priors as working hypotheses rather than settled conclusions — and cultivating, against the grain of hard-won expertise, a genuine appetite for being wrong.

The copy of Parkinson's Law is still on my shelf. Sixty-eight years old now. The Cambridge library stamp inside the cover remains perfectly legible.

The last line of the book's final chapter — a discussion of retirement, and the various ways institutions eventually persuade people to step aside — reads like this:

"When it comes to forcing our own retirement, our successors must find some method of their own."

Parkinson wrote this as a joke.

These days I read it more as a quiet intergenerational covenant.

The work of remaining updatable is not something anyone else can do for us.

Our institutions age the same way people do: through the quiet accumulation of assumptions that once made perfect sense. They age at a human rate. The world no longer does.

Our successors are watching.

The real question is whether we will make them wait for our funerals — or whether we might help them climb a little sooner, the way we once wished someone might have done for us.

The systems we build will outlive us. Our task is to ensure they will still be able to learn.

If you received this via email, the canonical archive version lives here: propertools.be/fieldwork/field-note-7-on-the-need-for-updating-ones-priors/