La Posición The Position
Governance is a design problem long before it is a political one.
Every organization past a certain size develops the same quiet failure: the parts are sound and the whole is not. The weld passes inspection. The calculation is right. The schedule is defensible on its own terms. The project still lands three months late, and the post-mortem finds no one who was negligent. The failure happened in the seams — between departments, between disciplines, between the people who knew and the people who decided — where no job description reaches and no one owns the outcome.
Complex system governance takes those seams as its subject. Its central claim is unglamorous and testable: a complex system stays viable only when a specific set of functions is actually being performed somewhere, by someone, well enough and connected well enough to matter. Identity and direction. Operational control. Coordination between units that must not collide. Learning. Watching the environment for the thing that will invalidate the current plan. When one of those functions is missing, duplicated across rivals, or starved of the information it needs, the resulting dysfunction is not a mystery and not a morale issue. It has a name, a mechanism, and a predictable shape.
We read that literature closely — primarily the complex system governance work of Keating and colleagues, standing on management cybernetics and general systems theory before it — and we test it where the consequences are physical: capital projects, control systems, plants that must not be shut down to accommodate a reorganization. Then we write down what happened.
Los Dominios Where We Work
Three domains, one question: what is actually governing this?
Management
Most management systems measure what is easy to count and route it upward, where it arrives too late to act on. We treat metrics as a control loop with a gain, a delay, and a failure mode — not as a performance verdict. The question is never are people trying. It is does this organization find out in time.
Team-building
Teams rarely fail at the center. They fail at the seams, where one group's output becomes another group's input and nobody is accountable for the handoff. Interventions aimed at trust and morale leave the coupling untouched. We go after the coupling first and let the morale follow it, which it generally does.
Systems engineering
Requirements, interface definitions, and verification are governance instruments that happen to be written in engineering language. Done honestly, they are the only durable record of what the organization agreed to. Done as paperwork, they document a consensus that never existed.
Research and implementation are the same activity here, separated only by who pays for it. Fieldwork feeds the writing; the writing disciplines the fieldwork.
Lo Que No Somos What This Is Not
A word on the politics, since the field has acquired some.
Systems thinking picked up a political reputation it does not deserve. The vocabulary is partly to blame — whole systems, collective behavior, central coordination — which reads to a working American manager like an argument for a larger central authority and a smaller sphere of personal judgment. It is a fair suspicion. It is also backwards.
Read the axioms and the opposite falls out. A viable system pushes decision authority down to the lowest level that holds the relevant information, because any other arrangement forces decisions to be made by people who cannot see the problem. Coordination exists to keep autonomous units from colliding, not to replace their judgment. And a metasystem that absorbs the work of its parts has not governed them — it has eaten them, and it dies of the indigestion: overloaded at the center, blind at the edge, slower every quarter. Centralization is not a political preference in this literature. It is a pathology with a name and a predictable failure curve.
So there is no manifesto here, no economic program, and no opinion about who ought to run anything. The claim is narrower and more useful than that. Every organization is governed whether or not anyone designed the governance. Designing it deliberately beats inheriting it by accident. That is the whole argument.
Apuntes de Campo Field Notes
What we are working on.
-
Governance Is Not Government
Why the most centralizing reading of systems theory is also the one its own axioms rule out — and what that means for anyone trying to fix an organization from the middle of it.
-
Control Is Information, Not Authority
A capital portfolio where every gate landed on schedule and nobody could explain why. The answer was not discipline. It was observability.
-
Teams Fail at the Seams
Coordination and control are different functions with different failure modes. Confusing them is why so much team-building money buys so little.