the war atlas · the plates · trade
Commodity flows and the weight assemblages that priced them. back to the book →
nothing is drawn on this plate
10 trade records recorded. None carry coordinates.
6 flows and 4 weight assemblages recorded, none carry coordinates. Every endpoint and every site name was queried against the coordinate stores this atlas actually holds - 22 gazetteer places, plus the three database tables built to carry a place coordinate, which hold 0 rows between them. 11 distinct place names were asked for; 0 came back. No arcs are drawn on the plate. 1 of the 6 flows has no location even in principle: its endpoints are literally "all" to "all", which is a statement about ubiquity rather than a geography that went missing.
0 of 10 trade records are plotted. Region names are not coordinates, and the gap between them is a gazetteer this atlas has never built rather than a rounding error to be estimated across. The zero is a lookup result and not a hardcoded field: the same resolver found Constantinople, Cahokia and Reno on this run, and refused Hedeby and a nonsense string, so it can both succeed and fail.
11 distinct place names were asked for. 0 came back. The stores that were searched, and the number of rows each held when it was searched:
Three of those four stores are the database tables built to carry a place coordinate — holy_site, wonder, innovation_attestation — and all three are schema without data. That is the trap this whole suite keeps finding: the tables designed to hold coordinates are the empty ones, so a search through them returns nothing and looks like an answer. 170 battle coordinates were NOT searched. A battle is an event at a place. Borrowing a battle's coordinate for a trade endpoint or an excavation site would be the invention this exporter exists to refuse.
the control, run on the export that wrote this page
The same resolver was handed Constantinople and the Bosphorus, Cahokia Mounds, Reno and found every one of them, then handed Xyzzy Nowhere Roads, (blank), Hedeby and refused every one. It can succeed and it can fail, so the misses below are a fact about the record rather than a broken matcher reporting a tidy zero. If it ever stops being able to do both, the export aborts and this page does not rebuild.
asked for and not found · Birka · Brazil / Spanish America · Britain · Caribbean · Caribbean / Americas · Carolina · Hedeby / Birka · Iberia · Kaupang · Vastergotland · West Africa
A flow has a topology and a geography. The topology is recorded: 6 flows, each with an origin, a destination, a source tier and a stated falsifier. The geography is not recorded anywhere in this atlas. Drawing these on the world plate would invent the missing half; leaving them out would hide the half that exists. So they are drawn here, between two labels, on paper with no coast behind it.
5 flows name real places this atlas cannot locate. One does not name a place at all, and that is not a gap — its endpoints are recorded as “all” to “all”, which is a claim about ubiquity. The two are counted apart because only one of them can ever be filled in.
cocoa / indigo / cotton[tier B]
Carolina → Britain · Carolina and Chesapeake coasting · Ocracoke Inlet / Beaufort Inlet
Not on the plate: “Carolina” and “Britain” were queried against every coordinate store this atlas holds and came back empty.
would be refuted by · a prize inventory in tier A showing these commodities absent from coastal takings
gold coin (moidores, pieces of eight)[tier A]
Brazil / Spanish America → Iberia · no route recorded
Not on the plate: “Brazil / Spanish America” and “Iberia” were queried against every coordinate store this atlas holds and came back empty.
would be refuted by · a tier A prize inventory showing specie absent from high-value takings
trade goods outbound for purchase[tier B]
Britain → West Africa · Guinea coast slaving run · Cape Lopez / Ouidah roads
Not on the plate: “Britain” and “West Africa” were queried against every coordinate store this atlas holds and came back empty.
would be refuted by · customs series showing Guinea-bound outbound cargoes were not predominantly manufactured goods
Middle Passage[tier B]
West Africa → Caribbean / Americas · Guinea coast slaving run · Cape Lopez / Ouidah roads
Not on the plate: “West Africa” and “Caribbean / Americas” were queried against every coordinate store this atlas holds and came back empty.
would be refuted by · a hull-origin series for period pirate flagships showing slavers as a minority of captured hulls
salt provision, livestock, water, liquor[tier A]
all → all · no route recorded
Not on the plate, and not missing: the endpoints are recorded as “all” to “all”. That is a claim about ubiquity, not a location that went astray, so no coordinate was asked for and none is owed.
would be refuted by · Part 4 provision_share coming in under 10% across a majority of itemised prizes
sugar[tier B]
Caribbean → Britain · Florida Straits to the Azores · Florida Straits / Bahama Channel
Not on the plate: “Caribbean” and “Britain” were queried against every coordinate store this atlas holds and came back empty.
would be refuted by · customs series showing Caribbean sugar exiting by a route not passing a named chokepoint
commodity_flow and route carry source_tier from the A/B/C/D/P/X/S source register, which is a different vocabulary from the T0-T3 confidence tiers. They are not interchangeable and are not merged here. A 2, B 4
Three of these five carry a recorded distance. It is printed as a number, here, because that is what it is. It is deliberately not fed back into the arcs above: every arc there is the same length, so nobody reads a distance off a curve that was never drawn to one.
| route | from → to | direction | distance | chokepoint | tier |
|---|---|---|---|---|---|
| Canaries to the Windward Islands | Canary Islands → Barbados | outbound | 2,700 nm | none named | P |
| Carolina and Chesapeake coasting | Charles Town → Chesapeake | coastal | 450 nm | Ocracoke Inlet / Beaufort Inlet | B |
| Florida Straits to the Azores | Florida Straits → Azores | homebound | 2,900 nm | Florida Straits / Bahama Channel | P |
| Guinea coast slaving run | Gulf of Guinea → Middle Passage | outbound | not recorded | Cape Lopez / Ouidah roads | B |
| Windward Passage transit | Jamaica → Atlantic | homebound | 400 nm | Windward Passage | P |
Hacksilver has no form and no denomination. Its meaning as money is indissolubly dependent on a standardised weight, which means the weights are not accessories to the currency — they are the currency’s arithmetic. That is why a box of cubo-octahedral weights is an assimilation record: whichever standard they answer to is the arithmetic the man carrying them had adopted. 4 assemblages are recorded and 4 of them carry no coordinate, so none of them is a mark on the plate.
coin
the coin as a unit of reckoning · seedcorn calculation
ontology · Antique-Christian concepts
Counted. The only regime where the object's identity as a unit is the point.
ring
the eyrir (ounce) · aurar reckoning — a given number of coins per eyrir
ontology · Odin's eternal and stable gold ring
Value rooted in a THEOLOGICAL image of permanence, converted to coins at a rate. A native standard with a cosmology behind it.
hacksilver
standardised weight · ertog calculation
ontology · none — it 'has no body'
THE REGIME THAT PROVES THE POINT. Hacksilver has no form and no denomination; its meaning as currency is INDISSOLUBLY DEPENDENT on standardised weights. Which means the weights are not accessories to the money — they ARE the money's arithmetic, and that is why the metrology below is the assimilation record.
the weights answer to the local mark / öre / örtug and to nothing else
Vastergotland[T2]
native standard, attested epigraphically
Attested on the 11th-c. Stora Ek runestone: ore around 24.49 g, ertog about 8.19 g. TWO METROLOGIES, SAME HANDS, SAME CENTURY.
source · Stora Ek runestone · no coordinate in any store this atlas holds — not on the plate
the weights answer to an imported standard — the Islamic dirham — inside a Norse purse
Hedeby / Birka[T2]
cubo-octahedral copper-alloy weights
Sits at exactly 5:4 to a dirham on Sperber's Islamic system, and is a multiple of the 0.355 g base unit Steuer identified at Hedeby and Birka.
source · Sperber; Steuer · no coordinate in any store this atlas holds — not on the plate
Birka[T2]
cubo-octahedral weights, lightest specimen
Matches Sperber's Swedish/Islamic system from Birka, where 1.59 g is prominent.
source · Sperber · no coordinate in any store this atlas holds — not on the plate
the record does not settle it, and this column is the honest answer rather than a missing one
Kaupang[T2]
excavated weight assemblage 1998-2003
A POPULATION, NOT AN ANECDOTE. 378 identified weights from the 1998-2003 work on top of 31 from the earlier excavations. Which standard a given assemblage calibrates to is the assimilation index, and it is CONTINUOUS rather than binary.
source · Kaupang excavation reports · no coordinate in any store this atlas holds — not on the plate
best_fit is stored as native / foreign / unresolved - three categories, counted here as foreign 2, native 1, unresolved 1. The Kaupang row argues in its own note that the underlying assimilation index is CONTINUOUS rather than binary, and the record does not hold that continuum. So the assemblages are shown in three bands and never on a scale: a position along an axis would be a reading of a measurement the table does not contain.
The best commodity for assimilation is not the most valuable one. It is the one whose use requires instruction — coffee and tea build rooms, vessels and a ritual around themselves. Furs and ivory have the highest value-to-weight in the record and the lowest knowledge density: they transmit taste and nothing else.
as is
furs, wax, honey, ivory, spices
transmits · taste, and nothing else
Highest value-to-weight, lowest knowledge density. The two are not the same axis and sorting by value gets this exactly backwards.
production technique
silk, paper, porcelain, steel, glass, sugar
transmits · nothing, until the technique leaks
The good travels easily and the technique does not, so transmission events are RARE, LATE and PRECISELY DATEABLE when the technique finally goes. That precision is the analytic value.
use technique
coffee, tea, tobacco, chocolate
transmits · ritual, vessels, and eventually architecture
HIGHEST INSTITUTIONAL YIELD OF ANY CATEGORY, BECAUSE THEY BUILD ROOMS. The best commodity for assimilation is the one whose USE REQUIRES INSTRUCTION — not the most valuable one.
accounting technique
bullion, bulk grain, wool, credit
transmits · numeracy, weight standards, contract forms
THE CATEGORY THAT CHANGES ADMINISTRATION. Ties directly to v49: if numeracy is the substrate of administration, this is the trade category that carries it.
person
captives
transmits · language, naming, story, technique — through the person
AND OFTEN AGAINST THE DIRECTION OF TRADE. The Rus exported slaves, furs, wax and honey to the Abbasids and Volga Bulgars, and imported dirhams. THE EXPORTS TAUGHT THE ABBASIDS ESSENTIALLY NOTHING. THE IMPORT REBUILT SCANDINAVIAN METROLOGY FROM THE BASE UNIT UP. Knowledge ran opposite to value.
knowledge_density is stored as one of "high", "highest", "lowest", "moderate" - an ORDINAL LABEL with no unit, no measured quantity and no defined interval between its steps. The word is what publishes. It is deliberately not a radius, a bar length or a rank number, because every one of those hands the reader a magnitude nobody measured.
Every chip above is the same size, and the rows are in the table’s own order rather than sorted by density. Sorting would be a scale too: it fixes a direction and an interval that four words do not.
