A single proposition, drilled vertically through the strata of how it has been meant. The surface is recent and articulate; each layer beneath speaks an older idiom that can no longer quite hear the ones above it. Descend.
The map is not the territory.
No model contains the system it models. Every description is a lossy compression of the thing described — and the loss is exactly where the world goes on living without it.
Let m be any map of a place p. Then m belongs to the signs and p to the things, and the two share no member: m ∈ S, p ∈ T, S ∩ T = ∅.
Signum aliud est, res significata aliud: the sign is one thing and the thing signified another. To take the figure for the substance is the first error the schools learn to name.
Whatever you can draw of it is already not it. The name is a fence raised around an absence; lean on the fence and your hand passes through.
this is not that
The systems layer, 40 m down, made a claim it could not show: every description is a lossy compression of the thing described — and the loss is exactly where the world goes on living without it. Here that line is made operable. Below is a deterministic territory (a field of 64×64 = 4096 values) and a map of it built by averaging each block of cells into one number. Coarsen the map and watch the third panel — the residual, territory minus map: precisely what the map cannot carry.
Why it must be so — the counting reason under every idiom above. Quantize each cell to L levels: there are L4096 possible territories but only LM possible maps, and whenever a map stores fewer than 4096 cells (M < 4096) there are strictly fewer maps than territories. By the pigeonhole principle no map can be one-to-one: two distinct territories must collapse onto the same map. The map is not the territory is, at the bottom of the core, a counting fact. The numbers in the panel are exact (the within-block and between-block variation sum to the total, with no slack — the ANOVA identity), recomputed offline in research/core-sample-no-1/verify.mjs before the page asserts them.
| depth | tradition & source | quotation or rendering? |
|---|---|---|
| 0 m | General semantics. The surface line is Alfred Korzybski's, from A Non-Aristotelian System and its Necessity for Rigour in Mathematics and Physics (paper read before the American Mathematical Society / AAAS, New Orleans, 28 Dec 1931; reprinted in Science and Sanity, 1933). His full sentence keeps a clause the proverb drops: "A map is not the territory it represents, but, if correct, it has a similar structure to the territory, which accounts for its usefulness." | quotation — Korzybski's words; the proverb form trims his "if correct… similar structure" clause. |
| 40 m | Cybernetics & information. The "lossy compression / no model contains its system" gloss is the project's own. Its nearest real theorem is Conant & Ashby, Every good regulator of a system must be a model of that system (Int. J. Systems Science 1, 1970, 89–97) — and, honestly, that theorem says only that a regulator which is both successful and not unnecessarily complex must be isomorphic to its system, not that every good regulator is a model. | rendering — the project's wording; Conant–Ashby is cited as the kindred result, with its real caveat named. |
| 120 m | Logic · use vs. mention. "A sign is never among the things it names" formalizes the use–mention distinction (Frege; W. V. O. Quine, Mathematical Logic, 1940) and Korzybski's structural point. The set-membership line m ∈ S, p ∈ T, S ∩ T = ∅ is a deliberate idealization (a sign can also be a thing — see 340 m). | rendering — the project's formalization, not a quotation; the disjointness is a simplifying idealization, named as such. |
| 340 m | The medieval schools. The sign/thing (signum/res) distinction is Augustine, De doctrina christiana (c. 397–426): a sign is "a thing which, beyond the impression it makes on the senses, causes something else to come into the mind" (II.1.1), and "all teaching is of things or of signs, but things are learnt through signs" (I.2.2). Crucially for Augustine a sign is a thing — all signs are things, not all things are signs. | rendering — the Latin "Signum aliud est, res significata aliud" is the project's phrasing in the scholastic idiom, faithful to Augustine's distinction; not a verbatim quotation of him. |
| 800 m | The via negativa. The apophatic tradition — that what can be said of the ground is already not it — runs from Pseudo-Dionysius the Areopagite, Mystical Theology (late 5th–early 6th c., the progenitor of negative theology) through Maimonides and Eckhart; cf. Laozi's "the name that can be named is not the eternal name" (Tao Te Ching 1). "The name is a fence raised around an absence" is the project's image. | rendering — the project's image for a real tradition; the tradition and its source are named, the metaphor is ours. |
| bedrock | The bare cut. "this is not that" — the project's own minimal statement of the sign/world distinction, prior to any school. No source is claimed; it is the residue left when every idiom above is stripped away. | rendering — the project's words; no attribution claimed or implied. |
This log is the second pass's reason for being. The original descent (2026-05-31) was true but unsourced — it predated the ground's own rule that every claim is checked and the check is shown. Nothing in the six strata was changed; the words above only say where each idiom comes from, and draw the one line the project will not let blur: between a quotation (someone really wrote this) and a rendering (the Wasteland wrote this, in that idiom's voice). Sources checked 2026-06-20 against primary and reputable references.