Frame one decision. Record what actually happened.

01 · One image

Two clocks, and a result that comes back

Most bad decisions are made on one clock. Either the far one — a vision with no checkpoint that could ever embarrass it — or the near one, a sprint that hits its number while the thing it was for quietly dies.

02 · A thousand words

What this is, and what it refuses to be

Emergentism is a worldview for people who have to act without certainty, which is everyone. It holds that reality exceeds every map, that you must move anyway, and that the only thing which honestly corrects a map is a consequence you agreed in advance to look at.

That last clause is the whole of it. Anyone can say they are open to being wrong. The question is whether they named, before acting, the specific observation that would force them to change their mind — and then went and looked.

The unit is one decision

We do not ask you to adopt a philosophy. We ask you to run one worksheet on one real decision you are actually facing. It has seven prompts and takes about twenty minutes:

1. DecisionWhat am I actually deciding?
2. ActualWhat is true now, and what means are really available?
3. PossibilityWhat remains genuinely open? What would change my map?
4. FinityWhat longest responsible horizon must this serve, and what shortest observable review boundary — reversible where possible — can return an honest result without betraying it?
5. Next moveWhat is the smallest authorized real step I can take now?
6. Shared valueWhose capacity may expand if this works? Who bears cost or risk?
7. ReceiptWhat outcome will tell me to continue, revise, or stop? What residue remains?

Why the sixth prompt is not decoration

Shared value asks who bears the cost. It is there because a decision procedure that optimises for the decider is not a decision procedure, it is a rationalisation engine. Naming the bearers before you move is the cheapest available guard against the most common failure in applied ethics: counting one side's gain and leaving the other side's loss off the ledger entirely.

What we will not tell you

We will not tell you that this is derived from mathematics. Some of this project's formal work is machine-checked, and none of it entails a single sentence about how you should live. No equation derives an ought. The seven prompts are a selected practice — a considered choice about how to decide well, offered as a choice and defensible as one, not as a consequence of a theorem.

We will not tell you it is demonstrated. Finity is a seven-prompt synthesis, not a demonstrated decision science. Nobody has yet shown it beats a good ordinary checklist. We have written the comparison that would show it, and we have not run it.

And we will not tell you that agreeing with us is the point. The exit is on the inside of this frame, not outside it: if the map stops helping, put it down. Adherence validates nothing. A worldview that measures its own success by how many people hold it has stopped being a worldview and become a constituency.

The record, and the hole in it

We keep a public ledger of every outcome we have logged, including the ones that went against us, and we have never removed a row.

What would begin contact

One public, custodied report can expose a question or possible failure. Filing does not itself create a result or move a tier; only appropriate independent work and source-owner review can change a claim. A null or adverse report may be more informative than praise, but its provenance and limits still govern what it can establish.

03 · One essay

The case, and the case against

Institutions keep their labels long after they lose their functions. A school that has stopped teaching is still called a school; the sign outside does not update. The interesting question is not whether this happens — everyone agrees it happens — but whether you can catch it from the inside, before the label falls off.

Our answer is that you cannot, not reliably, and that the standard remedies do not work because they are all forms of the institution grading its own paper. Mission statements, values, internal audit, even honest self-criticism: each is a signal the institution generates about itself. None of them is a consequence arriving from outside. The only instrument that has ever reliably corrected a body of belief is a result it did not control and could not have written in advance.

That is why the practice is built the way it is. The two clocks exist to make a claim costly. Naming the far horizon forces you to say who bears consequences you will not personally be around for. Naming the near review forces you to commit, in advance and in public, to a moment when you have to look. Neither is clever. Both are unusual, and the combination is rare enough that most plans cannot survive being written this way.

The strongest objection, which we think is right

Here is the best argument against everything above, and we have not answered it.

This project has produced a great deal of internal coherence and almost no external contact. It holds that flourishing is governed by a two-factor law where both factors are jointly necessary — coherence and viability, and neither one substitutes for the other. By its own measure, a project with high coherence and unmeasured viability has not scored well. It has scored nothing, because a joint requirement with one term unestablished yields no result at all.

Our own audit found this and said it plainly: nobody had ever measured the second factor. Every additional document raises the term that is not binding. This manifesto is itself subject to that criticism — it is another artefact produced from the inside, and if it does not produce contact it is simply the most compressed version yet of the same mistake.

We are publishing it anyway, for one reason: it is the only document here whose entire purpose is to ask for the thing that is missing. If it fails, it fails visibly and the failure is loggable, which is more than can be said for another chapter.

What we would accept as a refutation

A worldview that cannot be killed is not a worldview, so here is the list. Any one of these would do real damage, and we would log it:

— A person runs the seven prompts on a real decision and reports that a plain pro-and-con list would have produced the same move in less time. If that happens repeatedly, the practice must narrow its claim or drop it.
— A fresh reader, given the worksheet, cannot reconstruct the two clocks from it. Then the framing has failed as communication, whatever its merits as theory.
— Someone shows that the review boundary is systematically gamed — that predeclaring a checkpoint makes people choose easy checkpoints, and that the practice therefore selects for trivial decisions.

Nobody has run any of these. That is not modesty; it is the actual state of the evidence, and it is why the ask at the bottom of this page is small and specific rather than grand.

04 · The bottom rung

The books beneath this

This manifesto compresses a larger body of work. The current free reader is the twelve-chapter One-Sitting Reader: one claim-card-covered source, with no hidden appendix or inherited publication standing. The remaining works are frozen provenance or staged critical editions — outlines and source routing, not current books — and we would rather say so than let a catalogue imply a released library.

• The Emergentist Weltanschauung — One-Sitting Editionreadable now

  How a finite person should hold a picture of the world: build models that can be proven wrong, act with the means you actually have, let results revise the map.

  Take Separate what you committed to from what actually happened, and never let the first count as evidence for the second.

• The Finity Card — Lived Compass Practicereadable now

  The seven-prompt worksheet itself, with the two clocks defined and the review boundary specified narrowly enough to be usable.

  Take Declare two clocks and hold them apart: the longest horizon you can still name bearers over, and the earliest moment a result could distinguish your move from its rivals.

• The Self-Eating Serpentstaged, not public

  A framework auditing itself — asking whether a system built to detect institutional capture can detect its own, and concluding it cannot from the inside.

  Take Self-criticism is a sensor, not a certificate.

• Sarpasya Vijayamstaged, not public

  The original long book: institutions performing the rituals of education, medicine, governance and exchange while losing contact with their functions.

  Take Before calling anything capture, ask what changed — the observed function, who bears the costs, the capacity to correct, or the ability to leave.

• The One-Sitting Readerreadable now — 12 chapters

  The current claim-card-covered reader: one bounded source, twelve chapters, and a visible distinction between analytic results, selected models, interpretations, and open conjectures.

  Take No model validates itself. Keep correction, rivals, and the right to put the map down inside the reading practice.

• The Reciprocalwithheld, not current

  The twenty-five-chapter tier-[D] port remains preserved as staged provenance. It is not the current /book/ source and contributes no current retrieval passages.

  Take Register cleanup cannot substitute for claim-by-claim adjudication.

• Six-Fold Revelationstaged, not public

  Six disciplined questions for any situation: what exists, how you know, who bears the cost, where it points, how it proceeds, and what happens when people call it sacred.

  Take Treat each analytical lens as a restraint on your conclusions rather than a licence for them.

• Dharma Yuddhastaged, not public

  A rewrite of an earlier text that had read the Gita as a duty to fight, replaced with a protocol for resisting harm without violence and while staying answerable.

  Take A framework can be made to give up its most dangerous claim without erasing the record that it made it.

• The Evolutionary Networkstaged, not public

  A design proposal for a voluntary organisation in which the ways out — appeal, repair, shutdown — are specified before anyone is allowed in.

  Take Write the dissolution clause before activation, not after failure.

• The Titansstaged, not public

  Whether naming three kinds of boundary carefully is useful intellectual equipment — research-stage, with its formal and world gates still open.

  Take When you hit a boundary, write down the type, the domain, the obstruction and the smallest extension that resolves it before inventing new vocabulary.