The view that writing down your ideas tends to clarify and improve your thinking is nowadays widely expressed. Academia in particular cares a lot about good writing, and so both researchers and teachers have taken it upon themselves to argue that good writing improves the writer as much as the reader. But the concept seems to have escaped into the wild as well. All kinds of bloggers, thought leaders, and businesspeople agree that writing forces you to think more clearly.

I suspect that the idea has gotten more attention recently because of the advent of Large Language Models (LLMs) like ChatGPT and company. As the name suggests, LLMs are really really good at generating large quantities of polished, grammatically correct text. The pushback from writers against LLMs has been varied, and “writing is thinking” is one of the common arguments; if writing is thinking and you stop writing, clearly you will be worse at thinking.

For this post, though, I’m less interested in such weighty concerns; instead, I’m curious about the history of the idea. Investigating the history of ideas is interesting (to me, at least)—many ideas are so old that they likely predate any records of them, while others are surprisingly new. For a rough upper limit of age, because “writing is thinking” presupposes writing, it can’t be older than the invention of writing. To narrow down my guess, I would suggest that “writing is thinking” corresponds with the widespread teaching of writing—as noted earlier, “improving your thinking” is good motivation for learning to write.

Let’s check my guess. William Zinsser published a relatively well-known guide to writing in 1976 (On Writing Well); he noted in the preface that “Writing is thinking on paper.” On the more academic side of things, we could look at a 1967 paper on teaching English as a second language: “The process of learning to write is largely a process of learning to think more clearly.” When we lean on a linguistics paper to find more references, we find that the concept is traced back to some of Marshall McLuhan’s work in the early 1960s, though McLuhan was making a far more general argument than we’re considering. How to Read a Book (1940) claims that being able to restate a book’s thesis in your own words is a suitable test to check if you truly understood the material.

So we can reasonably say that “writing is thinking” gained widespread acceptance in the middle of the 20th century, but probably existed a few decades before than, given that it is treated as uncontroversial at the time. All well and good and roughly in line with our original hypothesis. But then I came across this line in Descartes’s Discourse on the Method (translated in long-winded 19th century style by John Veitch):

[Writing down my results] commended itself to me, as well because I thus afforded myself more ample inducement to examine them thoroughly, for doubtless that is always more narrowly scrutinized which we believe will be read by many, than that which is written merely for our private use (and frequently what has seemed to me true when I first conceived it, has appeared false when I have set about committing it to writing)…

Alas, I do not have as complete of command of pre-modern philosophical works as one might hope, so I can’t say if there are any earlier expressions of the idea. But “writing is thinking” is, at the very youngest, just under 400 years old.

References

Academic support for “writing is thinking” is often expressed informally through writing centers, workshops, and editorials: Berkeley, Purdue, MSU Denver, and UT Austin,

Bloggers and other amateur writers justify their activities by claiming that writing generally improves their thinking: Psychology Today, marcusdtaylor.me, writingisthinking.com, blog.stephsmith.io, Paul Graham (2022), and Paul Graham (2024)

Both amateur and professional writers have argued that people still need to write “by hand” instead of using LLMs to maintain critical thinking ability: opencontent.org, a Nature editorial, companies that create writing software, academic studies, and a Psychology Today editorial.