[Expanded] Field Note #1: Rest as Governance

This is the first field note in Governance Without a Body, the June arc, and it is being published from a queue I built in advance. I planned to be largely offline this week, and the system is running without me — which is, conveniently, the entire argument. I created this system because I have come to understand that any system dependent on a single body is not a system at all. It is a risk wearing a system’s clothes. That sentence is the thesis of the month, and the rest of this field note is what happens when you take it seriously enough to apply it to your organization, to your technology, and to yourself.

There is a kind of governance that almost every organization performs with real rigor, and a kind it performs with almost none, and the gap between them is where most failures actually live.

The rigorous kind is the governance of things you can put on a balance sheet — capital, risk, compliance, infrastructure. The neglected kind is the governance of human capacity: the deliberate management of whether the people the organization depends on can actually keep showing up, and whether the work survives the moment any one of them cannot. This second kind rarely appears in a governance framework, because it does not look like governance. It looks like wellness, or HR, or a personal matter. And so it goes ungoverned, right up until the week the one person who understands the critical process is unavailable, and the organization discovers that a great deal of what it called a system was in fact a single human being, improvising heroically and invisibly, for years.

I want to make the case this week that the governance of human capacity is not soft, not peripheral, and not separable from the technical governance work I spend most of my time on. It is the same design problem, appearing in a different material. And I am making the case from inside it, because as you read this I am practicing exactly what I am describing.

THE OFFICIAL STORY AND THE LIVED ONE

I learned to see this the way I learned to see most things that matter: by living a long time outside the assumed center of a system, where the gap between the official story and the lived one is impossible to ignore.

The official story about high-performing people and organizations is that capacity is effectively unlimited, that rest is what happens after the work is done, and that a sufficiently committed person can simply absorb whatever the system requires. The lived story is that capacity is finite, that bodies have limits which do not negotiate, and that the appearance of limitless capacity is almost always produced by someone quietly paying a cost no one is measuring. I know this lived story in a particular and non-theoretical way, because I have sickle cell disease and I am undergoing gene therapy, and my body has never once allowed me the fiction that capacity is unlimited. What that has given me, alongside a great deal of difficulty, is an unusually clear view of a design principle most organizations never have to confront until it breaks them: a system that assumes a present, infinitely available body has not been designed for resilience. It has been designed for the best case, and then quietly prayed over.

The countries and cultures that have stared directly at this have given it names. Japan, where I spent more than fifteen years of my adult life, has the word karoshi — death from overwork — a term that exists because the phenomenon became common enough to require one. This is not a historical artifact. In fiscal 2024, Japan’s Ministry of Health, Labour and Welfare recognized 1,304 cases of overwork-related death and serious health disorder, up nearly two hundred from the year before, the sixth consecutive record year; of those, 247 involved strokes or heart conditions and 1,057 involved depression and other mental-health disorders, the first time recognized work-related mental-health cases exceeded a thousand. The single largest share of those cases came from healthcare and welfare — the very people a society relies on to hold everyone else’s continuity together. The phenomenon is not confined to Japan, and it is not a matter of national character. In 2021, the World Health Organization and the International Labour Organization published the first global estimate of its kind and found that long working hours caused roughly 745,000 deaths from stroke and ischemic heart disease in a single year, with working fifty-five or more hours a week associated with a thirty-five percent higher risk of stroke and a seventeen percent higher risk of dying from heart disease, compared with a standard week. Overwork is not a metaphor for a lack of discipline. It is, measurably, a cause of death, and the deaths concentrate among the people holding the most.

The point of these numbers is not to alarm. It is to establish, with evidence rather than sentiment, that human capacity is a real and finite resource that fails in predictable, documented ways when it is mismanaged — which is precisely the definition of something that requires governance. We govern finite resources that fail predictably. We have simply declined to extend that obvious logic to the people the whole enterprise runs on.

REST IS NOT THE REWARD. IT IS THE INFRASTRUCTURE.

The reframe I want to offer is in the title, and I mean it structurally rather than as encouragement. Rest is not the reward you receive once the work is finished. It is the infrastructure that makes sustained work possible at all, and infrastructure is a governance object — something you design, maintain, and hold someone accountable for, not something you hope materializes when there is time.

The World Health Organization gave the failure mode of ungoverned capacity its own entry in 2019, when it added burnout to the International Classification of Diseases as an occupational phenomenon — explicitly defined as “a syndrome conceptualized as resulting from chronic workplace stress that has not been successfully managed,” and characterized by three dimensions: energy depletion and exhaustion, increased mental distance and cynicism toward the work, and reduced professional efficacy. Read that definition as a governance professional rather than as a tired person, and notice what it is actually describing. It is describing a system-level management failure that has been allowed to manifest in individual bodies. The stress is occupational. The mismanagement is organizational. The symptom is personal. That displacement — system failure surfacing as individual symptom — is the exact pattern I document across every domain I work in, and it is why I treat rest as a governance question rather than a self-care one. When the organization fails to govern capacity, the failure does not stay at the organizational level. It is pushed down onto the person least able to refuse it, and then renamed as that person’s weakness.

This is why the cultural conversation about rest, valuable as it is, does not go far enough for my purposes. There is a powerful reframing in the wider culture — Tricia Hersey’s work with The Nap Ministry and her book Rest Is Resistance among the most articulate of it — that names rest as an act of refusal against systems built on extraction, and I share that politics entirely. But I want to add the operational layer that the governance world keeps missing. Rest is not only resistance. It is also engineering. The same withdrawal that is a political act for the individual is, for the organization, a continuity requirement — the deliberate construction of slack, redundancy, and recovery without which the system has no margin for the moment its first-best path fails. An organization that cannot let its key people rest has not built a lean, high-performing machine. It has built a machine with no redundancy, which is the most fragile kind of machine there is.

THE SAME DESIGN FAILURE, IN THREE MATERIALS

Here is where the human-capacity argument stops being a digression from my usual subject and becomes the root of it, because the failure I have been describing is structurally identical to the failures I document in technology and in healthcare. It is one design failure appearing in three different materials.

In artificial intelligence, the failure looks like a model that performs beautifully for the demographic at the center of its training data and degrades, sometimes catastrophically, the moment it encounters someone outside that center. That is a system that works only on its best-case input and has no graceful behavior for the case its designers did not plan for — which is the same shape as an organization that works only when its key person is present and has no graceful behavior for the week they are gone. My Building Me Back experiment documented this on my own face and voice: voice tools that lose accuracy on the parts of my speech that carry the most of me, avatar tools that render me lighter than I am, systems built for a default I sit outside of. The model encountering me is the building encountering the person who does not match its assumed user. The failure is the same failure.

In accessibility, the failure looks like a building with a single stairs-only entrance, which functions perfectly until the day that entrance is closed, at which point the entire building becomes unreachable for everyone, not only for the wheelchair users it was already excluding. A single path is not resilience; it is a hidden single point of failure that the able-bodied majority simply never had occasion to notice. Resilient design — the curb cut, the ramp, the second route — is what an organization looks like when it has internalized that the first-best path will, eventually and inevitably, be unavailable, and that designing for that moment is not charity but engineering.

And in healthcare, the failure looks like a system that has no continuity plan for the patients it claims to serve — which is the subject of my ongoing Lyfgenia: A Public Record. The US healthcare system assumes a patient with resources, time, administrative fluency, and an advocate, and it degrades brutally for the patient who has none of those, offloading its own lack of continuity onto the people with the least capacity to absorb it. That is the karoshi pattern again, at the scale of an institution: a system that survives only when the individual supplies the resilience the institution declined to build.

Software engineering, which has thought about this longer than most fields, even has a grim shorthand for it: the “bus factor,” the number of people who would have to be hit by a bus before a project collapses. A bus factor of one is the universal warning sign of a system that is actually a person. The whole discipline of resilient design — in engineering, in accessibility, in AI governance, and in the governance of human capacity — is the discipline of raising the bus factor above one. It is the deliberate refusal to let any system, technical or human, depend on a single point that cannot fail.

THE ONLY GOVERNANCE AUDIT THAT MATTERS THIS WEEK

So here is the audit I want to leave you with, and it is the only one I am asking you to run this week, because it is the one that sits underneath all the others I will spend the rest of June and July detailing.

Look at your organization, your team, and your own work, and ask what specifically breaks if one key person is unavailable for a single week. Ask who that person is for each critical process, and whether the answer is uncomfortably often the same name. Ask whether your bias audits, your incident response, your institutional knowledge, your client relationships live in documented, owned, repeatable systems — or in a few heads that happen to show up. And ask the version of the question I am answering with my own absence as you read this: does the work survive the body that produces it?

If the honest answer is that a great deal of it does not, you have not found a wellness problem or a staffing inconvenience. You have found a governance failure of exactly the same kind you would treat with the full weight of your risk function if it appeared in your financial controls or your security posture — a single point of failure in a system you have been calling resilient. The reason it has gone ungoverned is that it has been miscategorized: filed under capacity, or commitment, or care, instead of under risk, where it belongs.

This is the foundation the rest of the Governance Without a Body arc is built on. In the coming weeks I will get specific about the AI accessibility audit, the procurement questions, the incident response plan, and the demographic metrics that make any of it real. But none of that technical machinery matters if the people running it are operating with no margin, and if the whole apparatus collapses the first week any one of them steps away. Governance without a body is the discipline of building systems — technical and human — that hold when the body is absent. I am demonstrating it this week by being absent, and watching it hold.

Rest, in this light, is not what I am doing instead of the work. It is the most rigorous governance decision I make.

THE THREE-QUESTION PULSE

Two minutes. It shapes the whole June–July arc.

1. In your organization, is the continuity of critical work mostly held by documented systems, or by specific individuals who happen to show up? (Mostly systems / Mix / Mostly individuals / I’d rather not know)

2. Be honest: are you currently operating with little or no margin — running at or past your own capacity? (Yes / Sometimes / No)

3. What would be most useful from the Governance Without a Body series: a continuity audit template, the AI accessibility and procurement frameworks, real case studies, or the Rest as Resistance retreat details?

   ---

incluu builds governance systems designed to survive the absence of any one person — the documented, owned, redundant continuity that turns a fragile set of heroic individuals into an actual system. If your organization’s most important work has a bus factor of one, that is the work we do. And if you are the single point of failure in your own enterprise, the Rest as Resistance retreats are where leaders learn to design their way out of that — because the quality of your governance is upstream of the conditions under which you make it.

Consultation
incluu
Rest as Resistance retreats
Subscribe for the full series:

Dr. Dédé

incluu + Dr. Dédé Substack

By Dr. Dédé Tetsubayshi

Dr. Dédé Tetsubayashi is a Black, queer, first-generation Togolese immigrant and transracial adoptee living with sickle cell disease. She is a TEDx speaker, a global advisor on AI governance and inclusive technology, and the organizer of Rest as Resistance retreats. → Work with Dr. Dédé at incluu.us.

This opens Governance Without a Body — the June 2026 arc. Next in the series: The Missing Accessibility Audit, The Questions That Survive the Deferral, and the continuity framework that runs through the summer.

Read more