Do Notes Preserve Knowledge, or the Moment Thought Happened?

Geoffrey Chen

How a Person Builds a World · Season One: “Reading, Memory and Judgment” · Article Three

A beautifully complete set of reading notes may contain dozens of quotations, chapter headings, keywords and carefully designed tags. When the writer opens it years later, however, there is no indication of why any sentence was copied. Every line came from a book, but none reveals the question that was alive at the time, the point that produced hesitation, or the judgement that began to change.

Another note contains only a few lines: a page number, an unfinished question and a conclusion later crossed out. It preserves little information, yet it allows the writer to see the movement of thought again—what seemed true before the passage, where resistance appeared, and why the author’s conclusion was not accepted.

Which note preserved more? If knowledge is retrievable information, the first is plainly richer. If knowledge also includes the path from material to judgement, the second may have retained something more important.

Notes perform two different kinds of work

Research on note-taking has long distinguished two functions. The first is encoding: selecting, compressing and reorganising material requires processing that may itself support learning. The second is external storage: paper or a file retains material outside biological memory for later use. Early experiments and later reviews have treated these as separable effects; the task, the manner of recording and whether the notes are reviewed all matter. An integrative review of the cognitive costs and benefits of note-taking surveys these distinctions and the inconsistent findings that make simplistic prescriptions unwise.

This is more fundamental than the familiar argument about handwriting versus laptops. Handwriting can be mechanical transcription, while typed notes can involve demanding synthesis. A visually impressive page may represent little interpretation. A single question in the reader’s own language may require deep processing. The medium affects speed, searchability and physical action, but it cannot decide by itself whether thought occurred.

From a wider perspective, a note is a form of cognitive offloading. We alter the environment so that a task need not be carried wholly by internal memory. Research on cognitive offloading considers reminders, lists, physical rearrangement and other external supports together, asking why people transfer cognitive requirements into the world and what follows. Risko and Gilbert’s review of cognitive offloading provides a metacognitive framework for the field.

Offloading is neither cheating nor evidence of intellectual decline. Complex knowledge practices could not exist without writing, indexes, calendars, archives and diagrams. The serious question is what we offload and which capacities remain available after the transfer.

A quotation preserves the author’s words, not necessarily the reader’s position

Quotation has irreplaceable value. It retains exact language, supports verification and prevents a reader from passing off a paraphrase as the author’s wording. When edition, page and source are clear, a quotation is a route back to the text.

Large collections of quotations can nevertheless create an illusion: that preserving sentences preserves understanding. A quotation answers, “What did the author write here?” It does not automatically answer, “Why did I stop here?”, “Which problem made this relevant?”, or “Which part did I accept and which part did I resist?”

Notes containing sources but no relations resemble pages moved into another warehouse. They can be found again, but their adjacency is unexplained. Tags group material without replacing the judgement that formed the group. Labelling a passage “freedom” does not tell a future reader whether it concerns legal status, psychological experience, political power or moral responsibility.

A living reading note should therefore distinguish at least three voices:

  1. Source: what the author actually said.
  2. Restatement: how I currently understand the passage.
  3. Response: which judgement it changed, supported or resisted.

When these voices merge, a note may grow increasingly fluent while becoming harder to verify. When they remain distinct, the note can respect the text and preserve the reader’s formation at the same time.

The most valuable trace may be where a judgement changed

People prefer to record conclusions because conclusions look finished. Years later, however, the traces before a change often explain more.

Consider a note that says: “I assumed clearer rules were necessarily fairer. This is the first point at which I see that greater clarity can make it harder for somebody outside the classification to explain their situation.” It is not a summary of the whole book. It preserves a change in conceptual structure. A future reader can see where an old judgement met a counterexample and how another became possible.

Notes of this kind should not erase hesitation, error and revision in the name of neatness. Crossed-out sentences, earlier versions and later objections show that thought was not written once. The Library of Congress guidance on keeping personal digital records specifically notes that drafts and earlier versions may preserve important information absent from final documents. The archival principle also applies to intellectual notes: a final statement preserves the result, while revision can preserve formation.

Not every hesitation deserves permanent retention. A knowledge life is not a total log of mental activity. Selection is itself a judgement about which transitions could help a future person understand the problem and which were merely noise.

A perfect system can begin to manage thought

Note-taking tools easily draw a person into endless preparation: building a taxonomy, refining templates, migrating platforms and reorganising old files. The system becomes increasingly complete while the number of genuine questions declines.

Manageable objects are more reassuring than unfinished thought. Quotations can be counted, links multiplied and folders optimised. A question that remains difficult to formulate exposes a gap in understanding. A support for thought can therefore become a structure for avoiding thought.

The quality of a note system should not first be measured by its capacity. A more useful test is whether it supports four actions: returning to a source, reconstructing context, discovering relationships and correcting an earlier judgement. A system that provides rapid retrieval but cannot distinguish quotation from inference is an information database. A system that preserves only immediate response without provenance cannot support serious research.

A good structure need not be elaborate. An important note that can answer the following questions is often sufficient:

  • What is the material, and where exactly can it be found?
  • Why am I saving it now?
  • How do I state its meaning in my own words?
  • Which earlier problem or note does it meet?
  • What is my present judgement, and under what conditions should it be reconsidered?

These fields are not intended to force thought into a form. They preserve a route that can later be travelled again.

A note is not a second brain but a meeting across time

“Second brain” is an attractive metaphor, but it implies that an external system can replace internal understanding. Records certainly supplement memory and can preserve more material than any person can hold consciously at once. A file, however, does not decide what matters or know why an old conclusion has ceased to be adequate. Search produces a match, not a renewed understanding.

It is more accurate to call notes a meeting place between a present and a future self. The present writer knows the context but cannot know what the future reader will ask. The future reader has new experience but may have forgotten why the note was made. A good note leaves enough evidence between them to recover a question, not merely retrieve a sentence.

This is also a form of knowledge without a single knower. A complete thought is not wholly present in current consciousness or wholly contained in a file. It is distributed among the book, page reference, quotation, personal paraphrase, revision history and later use. Knowledge becomes a living judgement again only when these parts are reconnected around a new problem.

Notes can preserve both knowledge and the process by which thought happened, but the two do not automatically arrive together. Sources, quotations and versions make faithful return possible. Questions, responses and revisions reveal how a person changed. If one must choose, the preservation of more information is not necessarily more valuable than the preservation of a genuine turn.

A note worth keeping does not need to prove that the writer had already understood. It should allow a future self to discover where thought began, why it changed direction, and where it can be continued.

Primary sources and further reading

Continue reading: Explore the How a Person Builds a World series.

Discover more from Geoffrey Chen

Subscribe to get the latest posts sent to your email.