The Index Method

Niklas Luhmann, a German sociologist, produced 70 books and close to 400 papers between 1951 and 1997. When he died he left behind two wooden cabinets holding around 90,000 index cards — his Zettelkasten, or slip-box. The cards get described as his archive, the place he stored what he read. They were the instrument he thought with.

Every card carried an address: 1, 1a, 1b, 1a1. A card that continued a thought took a letter. A card that broke off in a new direction took a number. A fresh idea could slot into the network at any point, and nothing already written had to move to make room. Luhmann did not read and then file. He read, wrote a card, and asked what already in the box it belonged beside — then wrote that link by hand. Following those links was how the writing got done. A note from 1952 and a note from 1979 could end up side by side, because the argument wanted them there. By the time enough cards had gathered around a question, the piece was already standing. He copied it out.

In a 1981 essay on his method, Luhmann described the slip-box as a partner in the work — a second memory he was in conversation with, one that kept handing him combinations he had never planned. The index itself was thin: a register of keywords, each pointing to one or two entry cards, and from there you followed the references inward from card to card. His own communication theory held that surprise is the precondition for information — a source that returns only what you expected has told you nothing. The box earned its keep on the days it gave back something he had forgotten writing. He did not consult it to retrieve. He consulted it to collide.

✦   ✦   ✦

The method runs three centuries deeper than Luhmann. In 1685 John Locke published a system for the commonplace book — the personal notebook of quotations and observations educated Europeans kept — built around a two-page index at the front. An entry’s place was fixed by the first letter and first vowel of its keyword: art filed under A-a, education under E-u. Locke had refined the method over twenty-five years of his own note-taking before he let it out. The historian Michael Stolberg, writing in Early Science and Medicine in 2014, traced what it did to the people who used it: the system let a reader break books into fragments and recombine them on the page in patterns their authors had never intended.

The labour in Locke’s method lived in one decision — which single keyword named what a passage was actually about. That decision is a judgment, made in the moment of writing, about what the thing in front of you means. The index entry is a verdict. To keep the index was to read back through your own collection and rule, over and over, on what it had become. The method outlived its author by a century; the London publisher John Bell sold blank books with Locke’s indexing instructions printed into the front.

In July 1945 the engineer Vannevar Bush described the same act for a machine that did not yet exist. Bush had just run the American scientific effort through the war, and he was uneasy about the sheer volume of recorded knowledge outpacing any one mind’s ability to use it. His essay “As We May Think,” in The Atlantic, imagined the memex: a desk that held a person’s whole library on microfilm and let them lay down trails joining any item to any other. His claim was exact. “The process of tying two items together,” he wrote, “is the important thing.” Not the storing. The tying. He went on to imagine a coming profession of “trail blazers,” whose contribution to knowledge would be the paths they cut through the shared record.

So the index is not the plumbing underneath thought. Across Locke, Bush, and Luhmann, the act of deciding what connects to what is itself the thinking. A notebook without an index fails twice over. We cannot find what we wrote, which is the loss everyone notices. We also never stopped to decide what any of it was worth, which is the loss no one notices, because it leaves nothing behind to be missed.

Everyone who keeps notebooks knows the shelf. A row of full ones, each carried for months and written in faithfully, none ever opened again. The reason is not laziness. It is cost. To find one note in an unindexed book is to search the whole book, and the mind refuses that price in the moment it has a question. So we do not look, and the note might as well never have been written.

✦   ✦   ✦

Ryder Carroll reached the same place from need rather than theory. He has ADHD, and every notebook system he tried broke in the same spot: he could not find anything in it. The Bullet Journal he released in 2013 answered with an index at the front and a number on every page. The part he pressed hardest was migration. At each month’s end you move through every unfinished item and decide, one at a time, whether it still earns a place in the month ahead. Copying the list forward would be faster. He made it slow on purpose. The slowness is the deciding.

For a pocket notebook the mechanism asks almost nothing of you. Number the pages. Keep two or three lines at the back, or fold the corner of anything you will want again. Then write the entry: a word, a page number. Writing it down is nothing. The work is the half-second before it, when you weigh whether the thing earns a line at all — which sends you back to read what you wrote and decide what it was. You catch the thought first and judge it later, and the index is where the judging lives. Do that across one notebook and you have been through your own mind once more than you would have otherwise.

Luhmann’s 90,000 cards took forty years, and the cabinet is rightly famous. The act that filled it is not rare. It waits on every blank page, for the price of a number and a word.

Number the pages. Keep the index. What you bother to index is what you have actually thought about.