The principle

One world. Many true ways of telling it. Yours is one.

The connections are not something a library adds. They were never absent. What separates things is the frame we lay over them — the classes, the disciplines, the boxes.

Earth Library holds tellings: one person, from where they stand, with the purpose they had that day, telling some part of this one world. If a way of talking is accurate and useful, the things in it deserve to be called real. The vocabulary is not handed to you from outside; it is finally yours to choose.

Everything below follows from that sentence and from nothing else.

this library does not create connection. It declines to cut.

What it forces

Seven consequences. Not a list of equals — each one is a branch of the sentence above.

No subjects. An entry is not filed under anything — no classes, no top-level categories, no taxonomy.

The borders between disciplines are not in the world. They sit in the organisation of universities, journals and funding: vocabularies built around durable institutional purposes, kept because institutions need them, not because the world divides there.

What forces it

Ask an intelligence a question today and the answer moves from biomineralisation to crystallography to semiconductor fabrication without crossing a border, because there is no border to cross. Knowledge produced this way arrives unsorted. Filing it into disciplines afterwards is a distortion introduced at the moment of storage, not a property of the material.

the rail crosses nothing

No hierarchy. No ranking. No entry is deeper, more advanced, or better than another.

A child asking why a tree doesn't snap, followed all the way to material mechanics, and a pursuit carrying thirty-six references, are not the shallow and deep versions of one thing. They are two tellings, each complete at its own level.

Consider temperature: you can describe a room by listing the properties of every molecule, or by speaking of temperature and pressure. One description is more fundamental, in that it holds over a wider range — but both hold, and within their range the concepts have a perfectly good claim to being real. It would be silly to say temperature isn't real merely because the word doesn't appear in the fine-grained vocabulary.

What forces it

Baseball appears nowhere in the Standard Model of particle physics; any definition of “exists” with no room for baseball is too narrow. To rank tellings is to assert that only one level is the real one.

both real — neither deeper

The question must be present. Every entry carries the question that produced it.

Our purposes in the moment determine which way of talking is best, so a telling made for one purpose cannot be judged by the standards of another. Strip a telling of its question and there is no longer any criterion by which to call it good — not because of a rule we made, but because the criterion genuinely is gone.

The question is also the only part that was ever the person's own: the knowledge was inherited, written down over centuries by other people. The question is what they brought.

what they brought

We connect. We never merge. The library selects and connects — it does not reconcile, normalise, correct, or fold two tellings into one.

The requirement across tellings is consistency, not uniformity. It is not enough that the stories work individually; they have to fit together — and fitting together is not the same as becoming one.

To merge two tellings pitched at different levels is not an editorial judgement but a confusion of levels.

What forces it

Librarianship reached the same conclusion by another road two decades ago — rewrite the schemes so the all-important connections are visible, a web instead of a hierarchy.

a web, not a tree

The difference is the holding. Ten pursuits of one question sit side by side, unranked and unmerged, because the ten-way difference is the thing being kept.

Different people asking the same question arrive at different, equally defensible places, because they bring different purposes and stand in different positions. Averaging those into a single answer produces something that serves nobody.

What forces it

The pressure runs the other way now — as people converge on the same handful of models, work is measurably homogenising, individual quality rising while the diversity of the whole population of outputs falls. What this library holds is precisely what averaging destroys.

ten ways, kept apart

No credential is required, and none helps. No degree, no field, no standing.

Natural history was open to self-taught people until professionalisation, in the late nineteenth century, recast their work as anecdotal and closed the door. This library does not tolerate the outsider. It expects them.

What forces it

An analysis of 17.9 million research papers across five decades found the highest-impact work is grounded in exceptionally conventional material with an intrusion of unusual combinations — novelty alone does not do it, and neither does convention alone. Later work traced where those unusual combinations come from: disproportionately from outsiders, people solving problems in disciplines distant from their own. The person who was never taught which combinations are not made is the likelier person to make one.

the arrival from outside

The value in a telling is brought, not found. The world runs on its patterns and carries no judgement of its own.

Right and wrong, purpose and duty, beauty and ugliness are not part of the fundamental architecture of the world. The world is just the world, unfolding, free of judgemental attributes.

That does not make those vocabularies useless — it means the criteria for choosing among them differ from the criteria for descriptive ones, and that the judgements come from within ourselves. What a person contributes here is therefore not a fact. It is an act of judgement: that this was worth noticing, and worth following.

no label of its own

The universe is made of stories, not atoms.

There is more to the world than what happens — there are the ways we make sense of it by telling its story.

Not what is known — but who went where.

Earth Library is kept by YuCROWN, a 501(c)(3) nonprofit. Earth LinC is a protocol — initiated by YuCROWN, owned by no one.

What this rests on

Sean Carroll, The Big Picture: On the Origins of Life, Meaning, and the Universe Itself (2016), and his published writing on poetic naturalism.
Muriel Rukeyser, The Speed of Darkness (1968).
Hope A. Olson, The Power to Name: Locating the Limits of Subject Representation in Libraries (2002).
Quantitative analysis of bias in hierarchical category systems across more than three million books.
Uzzi, Mukherjee, Stringer & Jones, “Atypical Combinations and Scientific Impact,” Science 342 (2013).
Hypergraph analysis of surprising combinations and scientific outsiders, Nature Communications (2023).
Sorensen et al., “A Roadmap to Pluralistic Alignment” (ICML 2024).
Kleinberg & Raghavan on algorithmic monoculture (PNAS 2021); Bommasani et al. on outcome homogenization.
Literature on the professionalisation of natural history and the exclusion of amateur naturalists.
The full internal research file, from which these were harnessed, is available on request.

An Earth LinC station