Keith Thomas LRB essay on note-taking

I think a lot of you will like this essay, or already like it as it was published in 2010. It’s about 4,200 words, so set aside a little time, should you read it.

Some topics covered: history of note-taking, personal history of developing an approach, the necessity and futility of approaching note atomicity, whether you’re on your way to becoming Lord Acton, being schooled (yet again in my case) by David Hume, contentment with becoming a dinosaur.


This article is fascinating and remarkably relevant, despite not once mentioning any app or even device!

I was deeply struck by him picking out the central ambiguity of “atomic notes” - how do you ever know when you have an “atom”: a single, undivided idea or concept? Every thing only makes sense because of all the other things it is related to.

The other gem is his insistence that notes have to be re-organised and almost re-noted again and again and again for all their purposes (this is Luhmann’s critical insight). This flips most “second brain” stuff on its head - all organisation is temporary and fleeting. Notes are building blocks that you can make many structures from.

There’s lots in there too about what “commonplace books” were really all about.


The omnipresent idea here is something that I noted during writing my own doctoral dissertation - physical notes (whether on scraps of paper, Post-it notes or on index cards) were much more fertile ground for discovery of themes and grounding my writing than digital notebooks were (in SOME not ALL instances). I was not as organized as someone such as Luhmann or currently, Ryan Holiday, yet I was able to get my dissertation and countless journal club papers handled without much suffering.

My point about current indexing themes is this - for all the hysteria surrounding note taking apps and creating perfect PKM systems with cool looking webs of interconnected information - I very rarely read about anything produced of meaningful value using these apps or systems.

This isn’t to say that those things don’t exist (I guarantee someone will come in here with a counter example, because internet) I believe there’s too much emphasis on making the perfect PKM platform placed above just doing the damned hard work that comes with creating something of lasting value.


I’m not convinced that it’s really about whether the notes are physical or digital, but I completely agree with you that “just doing the damned hard work that comes with creating something of lasting value” is the most important thing.

Having used a LOT of note-taking apps over the years I think there’s too much emphasis on creating and storing the note and not anything like enough on whatever the digital equivalent is of moving your pieces of paper or index cards around, making piles of them, even cutting them up into pieces and covering your desk with a creative mess out of which you physically make some sense.

In short, it’s working with the information that’s most important, but strangely digital approaches tend to try to distance you from that. Even those that start with a kinaesthetic approach where you can easily move bullet points and paragraphs (e.g. block-based like Craft or Notion) often end up burying that under layers of styles and prettification. Even mind-mapping programmes do that to some extent with auto positioning and auto-styling.


One of the reasons it’s such a great article is that it comes from a major historian whose books have changed the ways we look at early modern history. And so it’s a brilliant use case for a note-taking life that actually yielded profound work (not something that every YouTubing second-brainer can claim!). I re-read this essay from time to time just to keep reminding me how it can be done!

PS I mean it’s nice to read of systems that are in service of the craft of a writer at the top of their game; rather than hearing (so often) about note-taking systems that exist as their own end.


Very well said. There’s so much attention and hours burned in making PKM setups appear just so and perfect with plugins, themes and complex architectures.

Oddly enough Mind Mapping software was often helpful for me to help me get the structure of a writing project organized only after I’d collected enough readings and notes about a given subject. Sort of like I had a pile up enough bricks before I start planning the building of a house. That’s probably the opposite of how actual home construction goes, but you get my drift.


looks like the article is available to members only?

Sorry about that. Here’s an archive snapshot.


Thanks for sharing. This was a good read.


What a fascinating article, and a wonderful find in the LRB archives. Isn’t it interesting that Keith Thomas can find so many references to note-taking that often sound very like the ideas we now tend to associate with Zettelkasten and yet never mention Luhrman. I wonder when the German sociologist’s work became known in English-speaking discourse.
Despite asking, I never managed to get anyone in my own working life (psychiatry, psychopharmacology) to describe their processes as recounted here.
Lots of thoughtful responses too. I wonder if the promotion of and focus on Obsidian, Notion, etc, etc, on YouTube or podcasts comes from people who are often not academics as such.


My experience is that there was almost zero discussion about the practicalities of research - e.g. how to gather and organise information from multiple sources. There was lots of “research methods” teaching which talked about the integrity of those sources, how to build arguments, how to use statistical analysis etc. but it was just assumed that you would “do something with index cards” and keep a notebook to capture your reading. I did a Masters thesis in the 1980s and heard about Luhmann in the 2010s.

Keith Thomas describes inventing (over a long time) methods that worked for his particular research. That’s how I remember it.

There is a strong place for apps in this whole field. I remember transcribing three very full index card boxes into a very early notes and citation manager (on an Acorn Archmides!) which took hours a day for many weeks, but was absolutely worth it because I could keep everything on a couple of floppy disks, and make a proper back up and because I could search my notes in seconds and produce (dot-matrix) printouts of the research as the basis for writing sections of the thesis and I could make a bibliography in seconds when most people were allowing weeks to do that.


Slow reply as I needed time to read it. Thank you for sharing this, it’s right up my street!

If you fancy an even deeper dive into this subject, this is one of my favourite books to date on the subject (it will probably change, I do like to read about note-taking and organising knowledge): How Romantics and Victorians Organized Information

I don’t really agree with his comment towards the middle of the article that “the sad truth is that much of what it has taken me a lifetime to build up by painful accumulation can now be achieved by a moderately diligent student in the course of a morning.”

And I actually think he contradicts this comment in his closing statement, when he says:

“Historians are like reliable local guides. Ideally, they will know the terrain like the backs of their hands. They recognise all the inhabitants and have a sharp eye for strangers and impostors. They may not have much sense of world geography and probably can’t even draw a map. But if you want to know how to get somewhere, they are the ones to take you.”

Yes, a student might do cmd+f and find their key word across a library of resources. But without actually reading those sources they’ve gained nothing, and as context often matters a brief skim of the paragraph where the key word occurs probably isn’t going to be enough. To continue the local guide analogy, that is the equivalent of a sat nav taking you from point A to point B. Yes you got there and you know a route, but you learnt nothing about the local geography and probably can’t get back from point B to point A without further help. The only point to do that reading was to accumulate ideas and think. A search can narrow your focus, but it can’t do the thinking for you.

There is far too much emphasis on the actual method of note-taking in the “knowledge management as productivity” corner of the internet. I’m glad students might get exposed to these concepts early on, but I wince at the number of comments on Reddit etc., along the lines of “I now have 1000 notes and I don’t know what to do next”.


I took it to mean that as students are now taught generic research skills here in the UK, especially in univeristies — common component of first term first year induction courses, the coping strategies he uses were aquired tediously compared to what a fresher gets to do soon after enrollment.

Absolutely this. The most compelling attribute of Tinderbox, when I started using it some years ago, was the concept of “emergent structure”, which I find essential, whatever tools I’m using.