Priority by publication
Make disclosing what you know the self-interested move.
Everything for this step
- Field guide
- The sciences → 15 sections
- Builds
- None yet.
- Reference sheets
- None.
- Sciences it rests on
- Statistics
- Track
- Science and measurement, level 2 of 3 · next: The laboratory as an institution
Why this matters
This is the mechanism the Patterns essay identifies as the ratchet — the reason that after roughly 1700 significant knowledge essentially stops being lost. Before it, technical knowledge was a competitive asset: guilds enforced secrecy, workshops passed knowledge by apprenticeship, alchemists wrote in code. Knowledge was hoarded, so it died with its holders, so each generation re-derived it. The journal inverts the incentive with one rule — credit goes to whoever publishes first, not whoever knows first. Suddenly the way to profit from a discovery is to tell everyone about it immediately. Nothing else on this ladder changes the rate at which everything else on this ladder advances.
The shape of it
- Date and register submissions on arrival. The date is the entire currency. Without a trustworthy record of who was first, the incentive to publish evaporates and secrecy returns.
- Publish on a fixed schedule and keep to it. Irregular publication cannot establish priority reliably, and priority is the product.
- Require enough detail to reproduce. A result described so that only its author could repeat it is an advertisement. This is where the method's recorded conditions become the deliverable.
- Have someone competent read it before it goes out, and keep that person's identity separate from the author's interests. Review is not a guarantee of truth; it is a filter for obvious error and missing method.
- Publish corrections and failures to replicate as prominently as the original. A journal that only publishes confirmations is a literature that cannot self-correct — a defect that survives to the present day.
- Distribute widely and archive permanently. Multiple copies in different places is what makes the record durable; a single archive is a fire away from a lost generation of work.
- Consider a disclosure bargain for practical inventions — a time-limited exclusive right granted in exchange for full public description. Whatever its later pathologies, the original patent bargain existed to end trade secrecy, and it worked.
Doing it, step by step
The summary above says what has to happen. This says how, with the numbers that decide whether it works.
1 · Date-stamp every submission on arrival
The date is the entire currency. Without a trustworthy record of who was first, the incentive to publish evaporates and secrecy returns.
Keep a bound, page-numbered register: every submission gets a number, the date and time of receipt, the author and title, entered in ink the moment it arrives. Give the author a signed receipt with the same number and date. Keep a sealed copy of the submission as received, so the content on that date can be proved.
2 · Publish on a fixed schedule, and keep it
Irregular publication cannot establish priority reliably, and priority is the product. Announce the schedule — monthly or quarterly — and publish every issue on its date, even a thin one.
Print the date of receipt beside every article, so priority is visible in the published record itself, not only in the register.
3 · Require enough detail to reproduce
A result described so that only its author could repeat it is an advertisement. Require every article to state:
- The question and the prediction.
- The method, with conditions — materials, quantities, times, temperatures, instruments — enough for a competent person to repeat it.
- All results, including negative ones, with the measurements.
- The limits — what the work does not show.
This is where the method's recorded conditions become the deliverable.
4 · Review by someone competent with no stake in the outcome
Review is not a guarantee of truth; it is a filter for obvious error and missing method.
Send each submission to at least one reviewer who knows the field and has no interest in the outcome — not kin, not a close colleague, not a direct competitor. The reviewer checks whether the method is described fully, whether the results support the conclusion, and whether the arithmetic holds, and returns written reasons: accept, revise, or reject. Keep reviewers' identities apart from the author's interests.
5 · Publish corrections and failed replications prominently
A journal that only publishes confirmations is a literature that cannot self-correct — a defect that survives to the present day.
- Corrections go at the front of the next issue, numbered, citing the original.
- Failed replications are published as readily as confirmations, with the same standard of method.
- Retractions are clearly marked, with reasons, and the original is marked in every index.
6 · Distribute widely and archive in several places
A single archive makes the accumulated record one accident wide. Send copies of every issue to every settlement and workshop you can reach, and keep complete archive sets in at least three separate places, on archive-grade rag paper — see paper.
Publish an index — by author, subject and date — every year, so the record can actually be found.
7 · Offer a disclosure bargain for practical inventions
A time-limited exclusive right, granted in exchange for full public description. The inventor gets protection; everyone gets the method, and at the end of the term it is free.
Set the terms in published law: a fixed term — England's 1624 Statute of Monopolies allowed 14 years; modern patents run 20 — a description full enough for a skilled person to make the thing, a public register, and no right without the description. Whatever its later pathologies, the original bargain existed to end trade secrecy, and it worked.