As I’ve explained in a number of earlier sideline pieces, midway through the first draft of my Darwin book (which is still a work in progress) I became aware of the Zettelkasten system of note-making, and quickly became besotted with both it and the associated Obsidian app (other apps or analogue alternatives are available). I ended up taking a few months retrospectively converting all my existing notes to the system, then filling in some of the more glaring gaps in my notes.

Earlier this week, Chris Aldrich put out a call for model examples of Zettelkasten output processes. What he’s after is best summarised in his response to one commenter:

Now that you’ve got [your notes] and they’re linked, how do you actively revisit and reuse them? What does that portion of your process look like? Do you actively use them to write papers, articles, blogposts, other? How is that done?

In my own case, the answer is yes, I do actively use my notes to write both long-form pieces (the more recent chapters of my book), and some of my longer-form ‘sideline’ blog posts (including this one). It’s still early days, but, as I wrote before, there’s no going back.

My research and writing is a convoluted process, but, in response to Chris’s call, I thought it would be an interesting challenge to try to describe how I went about drafting a recent chapter of my book, which was fairly typical under my new system—and not entirely unlike how I worked in my pre-Zettelkasten, pre-Obsidian days. I shall, as Chris suggested in the comment quoted above, begin from where I already had a set of interlinked notes in place…

Chapter background

The completed chapter I shall be discussing is currently entitled Trying to Explain Too Much—although, knowing me, this might change in future drafts. Its original working title was Use It or Lose It. The change in title was due to a change in scope of the chapter, which, as we shall see, occurred as a direct result of my interacting with my notes inside the Obsidian app and elsewhere.

My original intention for the chapter was to explore how Charles Darwin explained the development of useful new biological traits, and the diminishing of old ones that, due to changes in circumstances, are no longer as useful as they used to be. Unsurprisingly, Darwin often invoked natural selection to give what we would now consider the correct ‘Darwinian’ explanation for such changes. But in some cases he gave the wrong ‘Lamarckian’ explanation of Use and Disuse.

Like many people, I’d always been baffled by the occasional, undeniably ‘Lamarckian’ passages in On the Origin of Species, bearing in mind Darwin is generally credited with having discredited such thinking.

Further developing my notes

My original note about ‘Use and Disuse’ merely summarised the basic idea, pointing out it was bogus, although Darwin sometimes referred to it uncritically. One thing I wanted to get my head round was why did Darwin think it was a genuine phenomenon? It turned out he thought so because pretty much every other biologist of the time thought so. Darwin took ‘Use and Disuse’ effectively as a given. As I read more around the topic, I began to add new notes to Obsidian with titles such as:

Lamarckism

Darwin and Lamarck compared and contrasted

What we think of as ‘Lamarckism’ had little to do with Lamarck

Darwin switched between natural selection and ‘use and disuse’ to explain different traits

Darwin thought injuries could sometimes be inherited

Variation, ch.24: Laws of Variation – Use and Disuse, etc.

This last note refers to a chapter from Darwin’s book The Variation of Animals and Plants under Domestication, in which he expanded on some of the themes only touched on in On the Origin of Species. As I explored and made notes about this and a few subsequent chapters, it slowly dawned on me that I was going to have to go off on another one of my tangents and begin researching one of Darwin’s other ideas: his hypothesis of Pangenesis.

Pangenesis was Darwin’s attempt, in the absence of any understanding of modern genetics, to explain a mechanism for biological inheritance. It’s a subject usually glossed over in polite Darwinian circles these days, as Darwin’s proposed mechanism was hopelessly wrong. I’ll spare you the details. But, having politely avoided looking into the subject for many years, I ended up doing a deep-dive, amending or splitting many existing notes, and adding some new ones. I didn’t appreciate it at the time, but the key new note was one entitled:

Phenomena Darwin tried to explain with Pangenesis

One of the supposed phenomena Darwin tried to explain with his hypothesis of Pangenesis was our old ‘Lamarckian’ friend, Use and Disuse.

At around this point, as is my habit when trying to work out where I’ve got to, and to devise a basic outline, I took out my trusty Leuchtturm1917 notebook and scrawled out a rough mind-map of my potential chapter:

Chapter mind-map

Note the side-comment half-way down on the left:

D liked his theories to explain multiple accepted phenomena (cf. Nat Seln)

Thanks to a similar passing observation in my Writing Journal three weeks later, this brief side-comment was to lead to the aforementioned changes in scope and title of my chapter.

Writing Journal

As I’ve written before, the single best piece of writing advice I ever received was to keep a writing journal. I’ve been doing so now—on and off, but mostly on—for over a decade. My Journal saw me through my first book, and is now seeing me through my second. It has always been an electronic Journal in simple ‘Markdown’ (plain text) format. So, when Obsidian, which is also in Markdown format, came along, my existing Journal was easily incorporated into my new system.

Among other things, I have traditionally used my Journal to think out loud to myself about my work in hand: the progress I’m making, the problems I’m encountering, and so on. Many of my best ideas have arisen by writing to myself like this.

In March 2022, I had been struggling to work out how the existing draft chapters of my book were all going to fit together. So, as I explained in a subsequent sideline piece, I broke my golden rule and went back to re-read my existing chapters before the first draft of my entire book was finished. Half-way through this re-reading, I recorded the following in my Journal (the text in bold indicates links to notes in my system that I can’t link to from here) :

Today, continued my reading in among chores:

As I’d already realised, there is clearly going to be some overlap between [my completed chapter] As Monstrous as a Whale and my planned next chapter on use and disuse, etc. […]

Also, [my existing chapter] Small Change Writ Large mentions Darwin trying to shoehorn the parallel roads of Glen Roy into his applications of Charles Lyell’s theories of uplift and subsidence (geological). When all you have is a hammer… This would tie in nicely with the use and disuse chapter, where Darwin made a similar mistake. He recognised Glen Roy as a blunder, but never Use and Disuse.

Developing my notes, re-reading my work, and thinking out loud in my Journal had led me to an interesting observation: Charles Darwin did indeed like his theories to explain multiple accepted phenomena, and I now had three examples. In chronological order, they were:

Darwin applied his friend Charles Lyell’s thinking on uplift and subsidence to explain a number of apparently unrelated, recognised geological phenomena. (In one case, the parallel roads of Glen Roy, he turned out to be very wrong.)

Darwin applied his own theory of evolution by means of natural selection to explain a wide range of apparently unrelated, recognised biological phenomena, from species classification to embryology, and from vestigial organs to the geographical distribution of species.

Darwin applied his hypothesis of Pangenesis to explain a wide range of other recognised biological phenomena, from how wounds heal to how plants reproduce though budding. (But at least one of these recognised phenomena, Use and Disuse, was bogus.)

Then came my lightbulb moment: Darwin had identified a number of phenomena that he wanted to try to explain with his proposed mechanism of inheritance, but at least one of those phenomena was wrong… I immediately created a new note:

Darwin never stood a chance with Pangenesis

The scope of my chapter immediately expanded, and its title needed to change: I decided I would now write about Darwin’s habit of applying general ideas to a wide range of phenomena in order to convince other scientists of the power of those ideas. But, as the new chapter of my title made clear, in the case of Pangenesis, Darwin bit off more than he needed to chew: he was Trying to Explain Too Much.

Obsidian graph of my Trying to Explain Too Much note

Outlining my chapter

I now had an interesting new take and scope that required me to re-outline my chapter. There was no way I would be able to mind-map this one on a single page in my trusty Leuchtturm1917, so, as always, I created a detailed outline in the electronic note dedicated to the chapter.

I make a habit of outlining chapters in Obsidian as it allows me to structure them with indented bullet points, and to link individual bullet points to supporting notes, including notes on original sources. I also make the bullet points into checkboxes, so I can check them off as I make my way through the outline as I’m drafting the actual chapter.

I can’t include an image of my chapter outline here—it’s far too long—but here’s a small section to give you a feel for how it looks (the light-purple text is linked to notes in my system):

Section of chapter outline

Writing the chapter

All that remained was the small matter of actually writing the chapter. I don’t do this in Obsidian: I think it would be asking for trouble to mix notes and their end-products in the same place. For the time being, my writing app of choice is Ulysses, but plenty of others are available—even, heaven help you, Micros✽ft W✽rd. [Postscript: Shortly after publishing this article, I changed to iA Writer as my writing app of choice—primarily because it uses more standard Markdown than Ulysses.]

Having links to original sources in my outline makes the compilation of references for the chapter far easier than it used to be.

In summary

The above is an attempt to describe how I went about writing one chapter of my book. I use the same basic approach for all my chapters, namely:

make lots of linked notes about stuff I happen to find interesting;

continue to develop those notes, splitting them into smaller notes when they become too wide-ranging;

write Journal entries and draw mind-maps to explore what I’ve discovered;

keep playing with my notes;

await a lightbulb moment, when two or more notes suddenly make an unexpected new connection in my brain, and I think, “Oh, that’s interesting!”

create a detailed bullet-point outline of my chapter, complete with links to supporting notes and references;

write the chapter;

compile the chapter references with the help of the chapter outline links;

repeat until the first draft of the book is finished;

then comes the fun part.

It’s far more complicated than that, obviously. Different parts of this process are going on all the time. While working on one chapter, I’m also capturing and working on unrelated—for the time being at least—notes on other topics that interest me, including stuff that might well end up in future books.

Apologies for such a long, nerdy piece. I do hope I wasn’t, like my hero, trying to explain too much.