We Have the Ingredients
We have the ideas: can we do something new with them?
Argyris named the gap between what organizations say they are doing and what they are actually doing — espoused theory versus theory-in-use. Snowden built tools for narrative-grounded sensemaking that respect how human cognition actually works at meso-scale. Ostrom catalogued the design principles of bounded institutions that sustained commons across centuries. Levin is formalizing the mathematics of self-correction across nested scales of biological and cognitive organization. Sapolsky has shown that the same brain placed in a different incentive structure produces different behavior, baboons to humans. Stavros has demonstrated, in hundreds of companies, that changing the ownership structure changes the culture without changing the people. Mondragon proved cooperative enterprises can scale across generations. Sister Mary Scullion has built and sustained Project HOME in Philadelphia for forty years without succumbing to the institutional drift that hollows out most nonprofits. The U.S. Founders built amendment mechanisms into the Constitution because they understood the document would be used by people they would have found alarming, and the structure had to survive that use.
These are facets of one problem: how do we design institutions that detect and correct their own drift from their stated purpose? We are not missing the answer. We are missing the convening.
The silos are worth naming. The crypto and experimental-governance crowd avoids moral and philosophical depth because it is uncomfortable. Mechanism designers avoid embodiment because the math gets messy. Complexity and contemplative practitioners avoid power and scale because they have seen what happens when complexity meets capital. Activists and elected officials avoid system-level experimentation because it is electorally fatal. Each avoidance is locally rational and globally maladaptive — the same dynamical structure the Buddha mapped in dependent origination, the same one Yudkowsky names in coordination failures.
These descriptions are useful only as motivation, not as endpoints. The reason these groups have not converged is not lack of insight in any one of them. It is the absence of a structure in which they meet, articulate what they share, and design an empirical program of testing together.
The hypothesis is concrete enough to test:
Institutions can be intentionally designed to surface and correct the gap between their espoused purpose and their actual behavior, using methods that already exist — Argyris-style language analysis, Snowden-style narrative sensemaking, Levin-style multi-scale modeling of decision authority — and institutions designed this way will sustain alignment with their mission longer and more robustly than institutions designed for conventional metric optimization.
This is testable today, in organizations that already exist. The reason it has not been tested at scale is not theoretical. It is that no one has convened the right people to design the test.
Sister Scullion combined methods that are while individually unremarkable became remarkable put together: long tenure of leadership, embedded residence in the community served, diversified funding to resist capture by any single donor, formerly homeless residents in actual governance roles, faith-based moral substrate without proselytizing, both quantitative metrics and narrative accountability. Each is a known good practice. The integration is what made it work, and the integration is what is missing from most institutional design.
This matters because Project HOME is an existence proof. It is not magic, not super-genius, not lightning in a bottle. It is operational. It can be articulated. It can be transmitted, with modifications, to other settings. The question is whether the principles can be made explicit enough to survive transmission to people who are not Sister Scullion.
That is the same question Chris Argyris and Dave Snowden have worked on for decades. That is the question Levin is now formalizing mathematically. That is what Mondragon answered partially, what Stavros is answering at private-equity scale, what the U.S. Founders attempted at the scale of a nation. The fact that the answers are partial does not mean the work has failed. It means the work is in progress and has been siloed.
We seek to first convene Levin and Scullion because the conversation needs a theoretical axis and an empirical axis before it can usefully expand. Levin can name what self-correction at institutional scale looks like in formal terms — what a cognitive light cone for an organization is, what kind of feedback architecture allows a system to model itself accurately. Scullion can name what has actually worked in forty years of running an institution that did not collapse into the failure modes Argyris and Niebuhr predicted.
If those two axes can be made commensurable — if Levin’s formalism can be tested against Scullion’s lived practice, and Scullion’s practice can be sharpened by Levin’s formalism — then the larger convening becomes worth having. Snowden and Argyris bring the diagnostic tools. Buterin and the experimental-governance world bring capital and the willingness to test. Stavros brings the operational experience of altered ownership. Weyl, Tang, and RadicalxChange bring mechanism design.
Without that initial axis being made explicit first, the larger convening collapses into the same silo problem it was meant to solve. The technologists abstract away the suffering. The activists distrust the technologists. The complexity practitioners refuse to commit to specifics. The mechanism designers retreat to whiteboards. Everyone leaves satisfied that the conversation was important and that nothing concrete should be attempted yet.
The test of whether this is real, and not another aesthetic exercise, is whether it can name the experiments to be run.
A short list, to make the question concrete: an Argyris-gap audit applied to a matched cohort of nonprofits, measuring divergence between mission language and decision patterns over time. A SenseMaker-style narrative collection across staff, residents, and community at Project HOME and three comparable organizations, examining what is structurally visible in one institution and invisible in the others. A Levin-style mapping of the cognitive light cone of decision authority in a value-based health care contract, identifying precisely where the model loses contact with patient and caregiver experience. A small-N comparison of ownership-restructured firms against matched controls along dimensions of mission drift over five years.
None of this is a new civilization. None of it is utopia. None of it requires consensus on contested philosophical questions before action can begin. It is the empirical program that follows from taking seriously what we already know.
The question for Professor Levin and Sister Scullion is whether having this conversation opens a door to helping others that has never been opened before.