In this essay, I discuss some challenges academics face when thinking through the state-of-the art of their field and how Roam can make that easier.
I show how a Zettelkasten alone falls short in tackling these challenges, and how the current capabilities of Roam hint at a powerful solution, something I call the ‘Literature X-ray’.
Before you can stand on the shoulders of giants, you have to climb up there. Before you can contribute, you have to learn what’s already been done.
More than learning, you need to understand what has been done – what claims have been made, what evidence supports them, what gaps remain.
Doing this is challenging, and at least for me, one of the hardest parts of writing an academic paper. I’ve spent the past 10 years trying to find a system that makes that easier, trying every tool that looked promising and even writing custom scripts to hack different pieces of software into working with each other.
None of the solutions I tried felt right, though. There was always something basic missing, and nothing came even close to what I had in mind as the end goal of my system.
But when I encountered Roam, it electrified me. This is the closest any tool has ever gotten to let me do what I want to do and seems on track to become the most powerful tool any academic could wish for to augment and improve their thinking.
Why is that?
When you’re entering a new field you want to contribute to, you’re faced with a number of challenges. You need to understand what’s been done before, the claims that have been made, what evidence supports them and what gaps remain. You also have to deal with different fields of study calling the same concept by different names. And at least in academia, you also have to give precise mention of the origin of any idea you reference that isn’t your own (aka citations). A complex task, to be sure.
Translating Zettelkasten to a digital world
Previous systems have made this complex task easier. Niklas Luhmann is famous for creating in his Zettelkasten a system that allows you to ‘converse’ with yourself about insights and ideas you’ve had in the past – and to easily relate quotes and concepts from different sources to each other. This helps tremendously in building an understanding of the field you’re entering, and in generating new insights to contribute.
It’s unsurprising, then, that many have tried to replicate Luhmann’s system digitally by leveraging the affordances of easier text entry, faster retrieval and portability.
Most of the digital incarnations of the Zettelkasten suffer from serious drawbacks, however, since they translate Luhmann’s ideas too literally onto computers. Daniel Luedecke‘s Zettelkasten app is basically a direct translation, with follow-on notes numbered in the way Luhmann used them.
Other approaches, like the plain-text Zettelkasten systems described in the past by Dan Sheffler or Stian Haklev (and that I’ve used for the last ~6 years) were weakened by their conception of what a note is and how to link them together. All of them are valiant efforts, but I feel they have fewer capabilities than what is now possible.
The promise of Roam
Roam succeeds where these other systems fail. It translates the Zettelkasten successfully into a digital version that doesn’t remain stuck in the mental models of the original, paper-based version. But Roam also goes further, hinting at future capabilities that will make Roam indispensable for any serious academic.
A bold claim, for sure. So let me explain why I think this is the case.
One way to think about how interacting with a Zettelkasten works is what I’d call ‘daydreaming’. You do associative thinking, hopping from one idea to the next, drawing connections as you let your mind wander through past thoughts and ideas. This is fantastic, except in situations where what you want to do is more akin to ‘active recall’ – that is, pulling together all the relevant information for a given concept or project you are thinking about.
And this ‘active recall’ is a large part of what any academic has to do when they sit down to write a literature review for their paper, or figure out what gap in the literature to fill next.
What I want as a researcher is something I’ve come to call a ‘Literature X-ray’: a way to take all the reading I’ve done, all the comments and critiques I’ve written from my reading, and to slice and dice them to spot where things are missing.
For example:
- Give me all the papers that deal with echo chambers in social media, and show me which social networks they study, the methods they use and what I think of them
- Give me all the papers dealing with ‘Autocracy’, and show me how they define the term
- Give me the trail of evidence for the claim that peaceful protest is more successful than violent protest, and show me how credible I think that evidence is
- Let me compare all the definitions of ‘Framing’ – and show me how different clusters of academics research this topic without ever citing each other.
A Zettelkasten will not let you do this, at least not without the considerable effort of manually collecting different pieces every time you come up with a concrete question. But Roam does enable this (kind of) right now and – hopefully – will be able to do this even more effectively in the future.
Creating a Literature X-ray
How can you create a Literature X-ray right now?
Roam lets you do a bare-bones version of this right now through its queries feature. I’ve recorded a video demonstrating this.
Below is a summary of the video. However I recommend watching the full video if you want to understand all the detail.
As long as you tag your blocks the right way, you can write powerful queries that collect the relevant information together.
Let’s use the example of papers dealing with Autocracy and their definition of the term. Creating blocks that mention both [[Autocracy]] and [[Definition]] together makes it easy to collect and compare definitions and my thoughts on the definitions.
I have used what Roam calls ‘aliases’ here to ensure that, when I use the word ‘define’, the accompanying content is included in any queries for ‘Definition’. What you see there as ‘define’ is actually ‘[define]([[Definition]])’.
This is the code for the query.
{{[[query]]:
{and: [[Definition]] [[Autocracy]]}
}}
If I also want to look at what methods these papers use, the query can be easily amended:
{{[[query]]:
{or:
{and: [[Definition]] [[Autocracy]]}
{and: [[Method]] [[Autocracy]]}
}
}}
The code is shown on different lines here to help understanding but needs to be put on a single line in Roam for it to work.
As you can see, this is already vastly more powerful than any pure Zettelkasten approach could be. If you are thorough in how you process the papers you read, you are already able to create a useful approximation of a Literature X-ray. But this approach is barely scratching the surface, given the power that is still untapped inside Roam.
The road ahead
The above example would be enough reason for me to never consider any other existing system of note taking again.
But Roam is only getting started, and I want to put two small thoughts out there on how Roam could be used to create a true Literature X-ray.
1. Right now, queries only work with links. But Roam already has a feature that could super-charge the way we do literature investigations: attributes. Instead of writing a block that links to [[Definition]] and [[Autocracy]], you could also put the text as one block and give this block an attribute along the lines of `definition of:: [[Autocracy]]`. If it was then possible to query for the values of attributes, this would unlock powerful new workflows.
2. Equally interesting would be a feature to create graph views from queries (including ones based on attributes), which would allow investigation of citation and concept clusters.
At the moment, however, the main challenge in building a Literature X-ray lies in building a system for yourself that allows you to get the most out of the existing features. This will require creative approaches to your note-taking and linking between blocks and pages, and thinking through how to leverage your notes through queries.
I’m committed to teaching what I find in my personal explorations around this through my course on using Roam for academic knowledge management, Cite to Write, as well as on Twitter and YouTube.
I’m also interested in how others solve adjacent problems.
Elisabeth Van Nostrand’s approach on extremely careful reading of history books has been a huge inspiration, and I’m very excited to see the #roamcult build new workflows and processes for better thinking.