A surface and a floor
Every study has a face a non-specialist can read and act on, and a layer directly beneath it with the numbers, sources, and assumptions a specialist needs to trust it. Neither is decoration for the other.
A working roundtable · the metacrisis, made legible
Chaosfragmentation
Authoritarianismrigid control
The Third Attractordistributed coordination
Two of these basins are traps. The third is the way through.
The metacrisis is the way today’s crises — ecological, economic, technological, institutional — interlock and accelerate one another, outrunning any fix aimed at just one. Left to run, it settles toward one of two failure modes: chaotic collapse or rigid control. The Roundtable maps the narrow third path — distributed, antifragile, alive.
The bench
Eight seats, one table
Tap a seat to hear its lens.
at the table
A study
Eight lenses converge on one subject
How a study is built
Legible on the surface. Defensible all the way down.
The method is borrowed from the people who mapped this terrain — Daniel Schmachtenberger's three generator functions (sensemaking, rivalrous dynamics, exponential technology) and Nate Hagens' biophysical lens — and held to one discipline above all: verify, never rubber-stamp.
01 / TWO LAYERS
02 / ONE FACET
The whole, one piece at a time
We don't try to explain the metacrisis in a sitting. We take one tractable facet — an energy shock's household incidence, a commons dilemma, a sensemaking failure — and make that piece genuinely understood.
03 / NO RUBBER STAMPS
04 / THE SAME LENS, TURNED INWARD
Reproduce before you integrate
External analysis is reproduced from seed and raw data before it enters a study. If a number won't reconcile, the work halts and says so. Correlation is not promoted to cause, and an annual rate is never mistaken for a monthly one.
We publish our errors
A standard you only apply to others isn't a standard. Find a discrepancy in a study and we check it against the data, not against our preference — and if it holds, we amend the study and log the change in the open.
Flag a discrepancy →The library
Studies and infographics
Each study publishes as a lay-facing surface backed by a reproducible analytical package. The first is the reference implementation — every convention the table runs on was forged building it.
Reference study · Economic incidence
The Uneven Month
A 2026 Strait-of-Hormuz oil shock, traced through U.S. households. The national bill concentrates in dollars at the top of the income distribution and in share of income at the bottom — the poorest bracket loses about 8.9% of after-tax income, the richest about 1.6%. Regressive in 100% of 100,000 bootstrap replicates. An exposure-and-incidence study, not a forecast — and it began by catching an annual inflation rate that had been applied as a monthly one, overstating the hit some fifteenfold.
Read the study →
In the queue
Next from the table
A commons dilemma, a supply-chain fragility, a sensemaking failure — convened study by study. The bench rotates to fit the question.
Avenues, not verdicts
Every path forward carries its evidence on its sleeve.
A study that only diagnoses is half-finished, and a study that promises a fix it can't back is worse than none. So every avenue we surface is tagged by how much evidence stands behind it. The reader always knows which is which.
Established
Well-supported by replicated evidence or settled mechanism. Acted on with confidence; the burden has been met.
Contested
Real support, real dispute. Promising and worth pursuing, but the disagreements are named rather than buried.
Exploratory
Plausible and worth testing, not yet earned. Offered as a direction to investigate — never dressed up as a conclusion.