How to run a Zettelkasten in an outliner

Niklas Luhmann was a German sociologist who published around seventy books and four hundred articles. Asked how, he pointed at a wooden cabinet: ninety thousand index cards, each holding one idea, each filed next to the card it belonged beside rather than under a subject heading. He called it his Zettelkasten — his slip-box — and said he did not think alone. He thought with it.

The method survived him because it is not really about cards. It is about three rules that make a pile of notes turn into an argument.

One note, one idea

A note that holds three thoughts can only ever be filed under one of them. So each note gets exactly one idea, and it is titled as a claim you could disagree with rather than a subject you could file under.

"Attention" is a folder. "Attention is a filter, not a spotlight" is a note. The second one can be argued with, extended, contradicted by the next note you write. The first one just sits there collecting things.

This is the rule people abandon first, and abandoning it is what turns a slip-box into a junk drawer.

In your own words, or not at all

Luhmann never copied passages. He read, closed the book, and wrote what he thought it meant. If he could not, he had not understood it — and a note he did not understand would be a note that lied to him in six months' time.

This is slower than highlighting, and the slowness is the feature. The friction is where the understanding happens; a highlight defers the thinking to a version of you who will never do it.

Filed next to its neighbour, not under a heading

A new card went behind the card it most closely related to, wherever in the box that was. Over time this built chains of argument that nobody planned — which is the part that made the box worth having. A category tree only ever gives you back the structure you already had in your head when you built it.

In an outline, this is nesting. A new note goes under an existing note, not under a new heading you invented to hold it. Zoom into any note and it becomes a page of its own, with its neighbours beneath it. When you cannot remember where something went, ⌘K / Ctrl+K finds it by name — search is the index you never have to maintain.

What is in the template

Four branches, and one of them is deliberately almost empty:

• Inbox — anything you have not turned into a note yet. Empty it, do not curate it.
• Notes — the box itself. One idea each, titled as a claim, filed under the note it belongs with.
• Sources — what you read and the handful of lines you actually took. Author, year, page, so a claim can be traced back.
• Index — three or four ways in. Not a taxonomy. The entry points you actually start from.

The Index is small on purpose. Luhmann's own was strikingly modest for a box that size, because the box is navigated by following links between neighbours rather than by browsing a hierarchy — the index only has to get you in.

• ZettelkastenLuhmann's slip-box. One idea per note, written in your own words, filed beside the note it belongs with.
  • InboxAnything you have not turned into a note yet. Empty it, do not curate it.
    • That thing you thought of in the shower
    • A line from the book on your desk — page number, so you can find it again
  • NotesOne idea each. Title it as a sentence you could disagree with, not as a subject.
    • A note titled as a claim is a note you can argue with later"Attention" is a folder. "Attention is a filter, not a spotlight" is a note.
      • Why I think this is true
      • What it contradicts
      • Where it came from — author, year, page
    • Write it in your own words or do not write itCopying is not understanding. If you cannot say it differently, you have not got it yet.
  • SourcesWhat you read, and the handful of lines you actually took from it.
  • IndexThree or four ways in. Not a category tree — the entry points you actually start from.
    • Attention
    • Writing
  • How to run it

Folded bullets open with a click; every line is yours to rename or delete. Open it →

How to run it

Empty the inbox daily. Anything you cannot restate in your own words goes in the bin, not into Notes — a slip-box is not a to-read list. When you write a note, find the note it argues with and put it underneath. Once a week, read a branch at random and see whether the note below still deserves to be there.

The box becomes worth having when it surprises you: when two notes filed months apart turn out to be the same idea, or opposite ones. That takes months, and it is the whole point. Nothing about the first fortnight feels productive, which is why most people quit before the thing they came for arrives.

What to skip

Skip the numbering. Luhmann's 21/3a7 addresses existed because a physical card cannot be in two places, and nothing in a digital outline needs them — they are the one part of the method that was a workaround rather than an idea.

Skip the elaborate tag taxonomies too. If you find yourself designing the filing system rather than writing notes, you have replaced the method with housekeeping about the method.