Understanding the Zettelkasten as an "Asymmetric" Communication Partner
Theory doesn’t need to be put into practical scenarios for it to be valuable. Theory can, and often does, remain very much impractical until a context arises permitting practical application. Case in point, Luhmann's theory of communication.
Without an IRL context to frame it, Luhmann's theory has for decades been considered controversial, even implausible.1 Its acceptance of non-human interlocutors puzzles those wanting to apply it to everyday conversations, chit-chats, and discourse. Luhmann's essay, "Communicating With Slip Boxes," situates this theory in the context of note-taking, but its effect on zettelkasten users, many of whom don't possess a background in communication theory, has been mixed. Consequently, readers unfamiliar with the nuances in its terminology are forced to take only those bits describing slip box mechanics (section II), leaving behind much of the remaining three-fourths (sections I, III, and IV). Forty five years later, and Luhmann's essay remains only nominally engaged by the audience who could benefit from it most.2
To be fair, "Communicating With Slip Boxes," published in 1981, is one of Luhmann's earliest presentations of his theory of communication, which came into focus in the years following the essay's publication, most succinctly in his essay, "What Is Communication" (1987), as well as in the books Social Systems (1984), Introduction to Systems Theory (from lectures 1991–1992), The Reality of the Mass Media (1995), and Organization and Decision (2000), (among others). While these publications are required reading (they are, after all, primary sources), they too can be difficult reads for the uninitiated. To bridge the gap, readers must often step back and read other scholars' takes on Luhmann's theory, hoping they'll deliver a more conventional and forgiving writing style. Though, as many have come to realize, this isn't always the case.
When reading primary sources and commentary doesn't cut it, I recommend not only stepping back, but stepping to the side, reading how scholars apply a writer's theory to emerging topics.3 To that end, I've found reading how Luhmann's theory of communication applies to communicating with LLMs to be insightful, particularly through Enzo Fenoglio's recent paper, "Asymmetric Communication: Large Language Models and Language Games," and Elena Esposito's book, Artificial Communication.4
From Luhmann to LLMs to Zettelkasten
To get a sense of how this mapping works, there are three Luhmann-derived points one must consider when thinking about communicating with a zettelkasten:5
- In order for there to be a communication event, there must be utterance, information, and understanding.
- Utterance doesn't need to be generated by a human.
- Understanding takes place on the side of the receiver.67
According to both Esposito and Fenoglio, chatting with an LLM qualifies as communication, not because LLM's are sentient or intelligent, but because they produce utterances which can be understood (i.e., distinguished) by you the receiver, resulting in an informational experience (surprise). Fenoglio calls this form of communication "asymmetric," because while LLMs generate utterances, "the norms that make those utterances count as correct or incorrect...remain on the human side." (Contrast this with human–human communication where norms are negotiated between two sentient interlocutors). In other words, asymmetry doesn't negate communication, but it does change its shape.8
Why is this a useful frame for understanding communicating with slip boxes? A zettelkasten is neither sentient nor intelligent. It doesn't participate in assessments of correct and incorrect, nor does it "understand content, meaning, or interpretation." And yet, utterances are produced whenever you surface a note containing a message. You, the receiver, then have an opportunity to understand the utterance,9 which has the potential to result in an informational experience.
The difference between communicating with LLMs and with a zettelkasten is that the messages you interact with in a zettelkasten were created by you.10 Not the "you" you're experiencing now. The past "you." What we typically call the "past self."11 Communicating with slip boxes involves an interaction between present self and messages your past self left behind, which when pulled from the slip box function as utterances. In so doing, the zettelkasten provides opportunities for informational experiences. This is, to use Fenoglio’s term, asymmetric communication. You don't commune with your past self. You don't change its mind, educate or inspire it. You interact with non-human utterances (messages contained in notes), understanding them to whatever degree and in whatever terms are relevant to you.
How the Zettelkasten as Communication Partner Informs Zettelkasten Work
As Luhmann's theory of communication (slowly) begins to filter into conventional wisdom, we have opportunities to translate it into praxis, applying it to previously unavailable situations. Although there are others to be sure (and I'd love to hear from you on this), below are three ways understanding how communication functions in zettelkasten work informs the work itself.
Write for your future self.
Luhmann's theory of communication reinforces the common zettelkasten advice to craft messages (notes) for your future self. Because the context in which you craft a message will shift by the time you reencounter it, make sure your messages are comprehensive enough to be understood later. This means writing in complete sentences, being explicit in what you want to say, and saying why what you say matters.
Information requires surprise, so let the zettelkasten surprise you.
According to classical information theory, a message produces information only when it surprises you. A zettelkasten that regularly reproduces only what you expect is non-informative (see Luhmann on storage and retrieval and it's relation to critical mass). Make sure to establish not only obvious connections, but connections that may not yet be fully realized. Also, consider Luhmann's warning against merely tracing notes point by point. From time to time, widen your search to allow surfacing of topically adjacent messages.
Become an active receiver.
Fenoglio’s claim regarding asymmetric communication (along with Luhmann's claim of how understanding takes place on the receiver's side) puts the onus for interpretation on you. Whether a resurfaced message is right, justified, or relevant is for you to decide. Past-you has no special authority. When an old note resurfaces, and you have a thought about it, respond to it. Agree, qualify, or challenge the message in a new note linked to the old one. Leave contradictions in the box.12
*If you enjoyed this piece, please pick up a copy of my latest book, A System for Writing: How an Unconventional Approach to Note-Making Can Help You Capture Ideas, Think Wildly, and Write Constantly. You may also sign up for my free newsletter HERE. To get my premium newsletter with all the insights, short essays, and zettelkasten scene weekly recaps, click HERE. To read more of my writing, click HERE. To learn more about me, click HERE. For my full website, click HERE.
Esposito, E. (2022). Artificial Communication: How Algorithms Produce Social Intelligence. The MIT Press.↩
You don't need to understand Niklas Luhmann's theory of communication to build, maintain, and use a slip box (ed. Reddit comment). But if you want to understand what he meant when he referred to his zettelkasten as a "communication partner," you do.↩
I call this "paratopical reading," or reading texts adjacent to a topic in order to understand the topic better. Think of reading Marxist takes on architecture to better understand Marxism, or a Buddhist writer's take on gaming to better understand Buddhism.↩
Esposito, ibid. Also, Fenoglio, E. (2026). Asymmetric Communication: Large Language Models and Language Games (arXiv:2607.28137). arXiv. https://doi.org/10.48550/arXiv.2607.28137↩
This isn't meant to be an exposition of Luhmann's theory in full. Only a handful of points needed to contextualize the claim.↩
Esposito, ibid. and Fenoglio, ibid.↩
I'm leaving out Luhmann's theory of information, since it is explained elsewhere in a longer version of this text.↩
Or, as Esposito writes: "[I]n the interaction with machines, we are dealing with a situation in which one communication partner is an algorithm that does not understand content, meaning, or interpretation. It deals only with data.↩
More precisely: distinguish between it and the fact that it was uttered.↩
Setting aside the very real possibility that an LLM may have scraped your own writing, and ends up re-presenting it back to you in some form.↩
By “self” I don’t mean the "Self," or soul, or atman. I’m referring to the self you construct in order to engage in and interpret society based on your experiences, knowledge, and behaviors. Unless you're frozen in time, this self is perpetually reconstituted, with new experiences, new knowledge, and new behaviors informing and causing you to adjust the self you project and with which you inspect. As a result, the self engaging with messages / notes today, your present self, isn't the same self that created them.↩
Doto, B. (2024). A System for Writing. New Old Traditions.↩