About

What this is, and what it is not

The Ontological Atlas is a comparative instrument, not a reference work. It is built to make disagreements between traditions visible. It is not built to tell you what any tradition is, and it should not be cited as though it could.

The short version

Every value on every page is one careful reading of a tradition, made by a person, in order to place it next to other traditions on the same axis. The readings are argued for and they are arguable. Where you see a definite answer in a cell, treat it as the beginning of a claim, not the end of one — and read the evidence and the tensions before you rely on it.

Why a grid at all

Encyclopaedias are organised one topic at a time. That is the right shape for explaining a tradition and the wrong shape for comparing it, because no two entries are written to the same questions. You cannot ask an encyclopaedia which traditions hold that matter is conserved and that the observer genuinely shapes what is observed, and where those two commitments then come apart.

This atlas exists to answer that kind of question. It applies one controlled vocabulary — 39 attributes across six dimensions, plus four meta axes — uniformly to everything it holds: 208 schools of thought, and the thinkers, texts and films read alongside them. Uniformity is the whole point. It is also the whole problem.

The problem uniformity creates

A grid asks every tradition the same questions, including questions the tradition never asked itself. Stoicism has no view about the curvature of space, because the concept postdates it by nineteen centuries. The Milesians have no position on whether information is discrete. When the atlas fills those cells anyway, it is not reporting what the tradition held — it is reporting what the tradition's other commitments most plausibly imply, which is a different and weaker kind of claim.

That move is what makes the comparison possible, and it is also the atlas's principal way of being wrong. Anachronism is not an occasional defect here; it is a structural feature of the method, and you should read every page knowing it is present.

How to read a page properly

The attribute grid is the least trustworthy part of any page, because it is the part that looks most like data. The trustworthy parts are the two that show their working:

  • Per-dimension evidence. Every school, thinker and text argues for its own values in prose, with sources. If the argument does not convince you, the value should not either.
  • Internal tensions. Every classified entry records where it strains against its own classification — the commitment that fits the grid badly, or the objection the tradition has never answered from within. An entry with no tension recorded is an entry nobody has pressed hard enough.

A cell you cannot trace back to a piece of evidence is a cell you should not quote.

Two kinds of thing in here

It helps to know which one you are looking at, because they carry different weight:

  • Classified entities occupy a point in the matrix — schools, thinkers, texts, films. These carry the interpretive burden described above.
  • Probes ask a question about the matrix rather than sitting in it — the dilemmas, thought experiments and historical debates. These are on firmer ground: a thought experiment's framing is a matter of record, and the positions traditions take on it are usually well attested.

Known limits

  • Scale outruns verification. The atlas holds roughly 92,952 individual attribute assignments. No one has audited them all. Errors are certain; the question is only their density.
  • Composite entries flatten real disagreement. Rows such as “Hinduism”, “Islam” and “Christianity” cover traditions whose internal disputes are the substance of their intellectual history. Their tensions notes say so; the grid rows cannot.
  • Fields are not positions. Philosophy of mind, philosophy of science and analytic philosophy are disciplines containing opposed views. Their rows record a mainstream tendency, not a doctrine anyone holds.
  • Some cells should be empty and are not. The vocabulary offers an Unaddressed value on 26 of the 39 attributes, and the schools layer currently uses it in 0 of 8,112 assignments. Every tradition is therefore given a definite position on every axis, including those it never considered. Reducing that is the single largest accuracy improvement available to this project.

Corrections

Disagreement with a value is useful and welcome, and it is most useful when it names the evidence rather than the verdict: which passage, which commentator, which commitment the reading has misweighted. The source and its full revision history are public at github.com/MartinGallagher-code/TimeSpaceMatterObserver.

This atlas is best used the way a map is used: to see how things lie in relation to one another, while remembering that no map is the ground, and that this one was drawn by someone with a view.