A peek into my zettelkasten
Intellectual Workflow Part 2
In part 1 of my miniseries on intellectual workflow, I explored the idea of the zettelkasten (slip-box) as a note-making and thinking method. In this essay, I will give you a look into my own zettelkasten and discuss some of the particular choices I have made and why. Full disclosure: this is actually my second zettelkasten—or, at least, it is the second iteration of my zettelkasten. I will explain the reasons for this below.
The first and perhaps most fundamental decision I made was to build a fully-physical (as opposed to digital) zettelkasten. There are a number reasons I made this choice. In general, I prefer handwriting and analogue tools to typing and digital ones. I used to default to typing because I found handwriting painful, but since I began using fountain pens, I have experienced virtually no pain when writing by hand, even after extensive writing sessions.
Aside from personal preference, I believe that the physical zettelkasten has several major advantages over its digital counterpart. The limited space of a physical note-card encourages brevity, which supports the practice of limiting each note to a single idea and also often forces deeper understanding—if we can’t paraphrase an idea succinctly, it is often because we don’t actually understand it. The physical zettelkasten requires that we search through it manually rather than using automated search tools, and this manual searching aids recall and facilitates the discovery of new connections as old ideas are revisited. The physical zettelkasten is also highly resistant to obsolescence; it relies upon simple physical tools rather than modern technology, so there is little risk that it will become unusable with time.
There are, of course, some disadvantages to choosing a paper zettelkasten. First and foremost, we miss out on the portability and ease of storage that the digital slip-box offers. Another serious concern is ease of creating a back-up or copy; digital tools make this so simple that it is nearly a non-issue, and while a paper zettelkasten will not fall prey to cyber attacks, it is vulnerable to natural threats such as fire and flooding. Finally, we miss out on a whole suite of digital tools when we opt for a physical zettelkasten, although I would argue that a lot of these digital tools actually automate processes that we ought to be performing manually in order to get the most benefit from the method.
Choosing a physical zettelkasten eliminated the need to select software from the variety of available options, but it did still leave me with some basic practical operating decisions. Specifically, right from the beginning, I had to consider the size and type of notecards I would use, and how I would store my notes. This was where I made my first blunder.
For the initial iteration of my zettelkasten, I chose to use 4x6-inch index cards, blank on both sides. These are easy to acquire and easy to store, and I thought that their size would encourage the brevity that I wanted to practice in my notemaking process.
Unfortunately, it became obvious to me pretty quickly that the 4x6 cards were a bit too small for my preferences. Instead of being brief and succinct, my notes were often underdeveloped and lacked the clarity required for long-term use. As I added links, I frequently ran out of space and had to use a second card to contain the overflow—which isn’t inherently problematic, but was a bit of an annoyance. It was in large part for this reason that some two months into my zettelkasten use, I made the decision to start over.
I ordered some 5x8-inch cards (lined on one side) and steeled myself for the work to come. Once I had my cards in hand, I extracted information from the notes in my first slip-box, developing ideas further where needed. All told, it took about a month to transfer to the new zettelkasten, but I think it was worth the effort. My notes now tend to contain better-developed ideas and have more space for links. The lined cards also make for neater writing, and I can always flip them over if I want to create a diagram or other note that benefits from being on unlined paper.
I also purchased some coloured cards for specialised use. As of right now, I only use yellow cards in addition to my normal white cards, and these are used for direct quotations (there is debate over the appropriateness of including quotations in the zettelkasten, but I personally prefer to include them in moderation). I anticipate using other colours in the future to indicate special note-types such as structure notes (notes which contain information about the structure of a topic as found within the zettelkasten), questions to explore, and other types of notes as needed.
As for storage, I currently keep my notes in a basic storage box that I happened to have on-hand. There are more elegant solutions out there, and I do not plan to use this particular box forever, but I was certainly glad that I hadn’t spent any money on storage when I realised that the 4x6-inch cards were not working for me. The day will likely come when I seek out nicer storage, but first I need to be certain of what my precise requirements are—which means working with and living with my zettelkasten for longer than just a few months.
The other big change I made when beginning the second iteration of my zettelkasten concerned my approach to numbering (assigning each note its permanent identifier) and titling (although I had already modified my personal titling guidelines partway into my first zettelkasten).
In my first zettelkasten, each card began with a number, starting at 1, and had a title indicating the topic contained in the note. This meant that the second card in any given line of thought would be labelled something like 1a or 5a, resulting in a lot of use of lettering. In the second iteration of my zettelkasten, the first card was instead assigned the identifier 1.1, adopting the approach described here by Bob Doto. I have found this to be a much more flexible approach and, perhaps somewhat oddly, one that has offered more freedom from a sense of imposed hierarchy—it no longer feels to me like my related notes are organised hierarchically by virtue of their permanent identifier.
When I first began titling my notes, I would often simply write a few basic words about their contents—something like “Woman’s vocation” or “The goal of education”, or simply “God”. I later read the advice (unfortunately, I can’t recall where) that the most effective titles are ones that make specific claims or statements, and I have tried to observe this practice ever since. A new title, then, might be something like “God is the Prime Mover” or “Education is formation”. This has made my notes much more usable, because the title gives a better overall sense of the content, and I have also found that it has made my notes much more memorable, helping me to actually retain the information in my mind as well as in my zettelkasten. Just seeing the title is usually enough for me to recall the basic content of the note, which was not the case when my titles were vague and made no such truth claims.
One thing that has largely remained the same throughout my zettelkasten usage has been my approach to linking—both to other notes, and to references. When I cite a source, which I generally do even when the abstracted idea does not, strictly speaking, require a citation, I typically format my reference as “(AuthorLastName Year, page #)” and I have a corresponding note card in my references section that carries a permanent identifier of “AuthorLastName Year” rather than the numeric-alpha identifiers of the main cards. When I link to other notes within my zettelkasten, I try to write a few words about the card that I am linking to, so I know where the link leads without having to actually follow it, and then I include the card’s permanent identifier in square brackets. I will then either create a reciprocal link on the other card if the link is bi-directionally relevant, or I will record a backlink—literally—on the card by writing the identifier of the card with the link on the back of the linked-to card, if the link is uni-directional.
So far I have focused mainly on the notes at the heart of my zettelkasten, but I think it is worth also briefly discussing reference section, which I already touched upon, and the index. Without these, the zettelkasten simply wouldn’t function, and while I had no difficult establishing a reference system, I found creating an effective index to be much more of a challenge.
A quick look first at my references, then, since they are straightforward. As already mentioned, most of my source references are stored on individual notecards with a permanent identifier consisting of “AuthorLastName Year”. If necessary, letters can be appended to the year to allow for multiple publications by the same author in the same year. I also make exceptions for particular ancient sources, such as St. Thomas’ Summa Theologica, trying to stick closer to the academic conventions established for such sources. The reference cards are then organised alphabetically.
In the main content of the reference note, I begin by citing the source in an approximated Chicago-style Name-Year format. While I aim to format my citations correctly, I don’t worry about it too much; the primary goal for me is to be able to identify the work, and not simply to copy the citation into a written paper in its exact form. So are all of my citations formally correct? Almost certainly not, but I’m okay with that.
Under the bibliographic information, I list any notebooks where I have recorded notes from the source, including any commonplace books that contain quotations from the source. And below the notebooks list, I list the notes in my zettelkasten that originated from and link to the source in question. In theory, on the back of the source note, I also include a brief description of the source, but in practice, I have been pretty lazy about actually doing this, despite believing that it is a good idea.
This leaves us with the index. I considered a variety of options for this, including one index card per letter of the alphabet, one index card per keyword, and so on, but I was unhappy with all my potential solutions. And then, of all people, (dead) philosopher John Locke stepped in to save the day.
In his short work, A New Method of Making Common-Place-Books, Locke describes how he indexes—you guessed it—his commonplace books. You can read his method for yourself, but I chose to use a slight variation of this method for indexing my zettelkasten. Basically, each notecard in the index is identified by two letters: the first letter of the word, and the primary/first vowel in the word (if a word begins with a vowel and contains no other vowel—for instance, if I had a keen interest in ants—its single vowel serves as its first letter and primary vowel—”ants” would be filed under A/a).
I find this method of indexing strikes a nice balance between excessively crowded cards, which I think would result if each letter of the alphabet had only a single card (whose entries, by nature of the successive addition of keywords engendered by the zettelkasten method would not themselves be alphabetised), and the excessive proliferation of index cards that would result from assigning an individual card to each keyword.
I link to a suitable entry-point for keywords as I deem appropriate (it’s not an exact science) and record the back-link on the notecard that is that entry-point. The back-link is probably unnecessary, but I like to have it there for record-keeping purposes.
And that’s all there is to it (at least so far). I hope that this look into my personal zettelkasten was helpful. I am still fairly new to the practice, and by no means its best example, but I know that seeing other people’s real-life examples has helped me to understand the method and to avoid many potential pitfalls. Hopefully I can pay forward this assistance!
If this essay was interesting or helpful to you, I invite you to share it with someone else who might enjoy it, and to subscribe to my substack if you haven’t already done so.
Next time I’ll be wrapping up my miniseries on intellectual workflow with a look at my entire working process. In the meantime, I’d love to hear your thoughts about intellectual workflow in the comments below. Is the zettelkasten method something that could work for you? Are there particular areas of workflow that you struggle with? Let’s get the conversation going!
Thank you for reading, and I’ll see you next time!