In order to be interesting, a nonfiction book must flirt with—but NEVER cross—the Textbook Axis.

UnderstandingComicsCoverI was recently thinking about Scott McCloud’s seminal work on how comics work: Understanding Comics. It’s a comic book that explains all sort of stuff about comics: the tradeoff between a cartoony style and a realistic style; how panels are arranged for effect; the types of panels; the history of comics—everything you never thought about while you read your favorite work of sequential art. It has tons of information and has taught me tons of things that I didn’t know. But I also find it to be incredibly dull.

I have never finished the book. Someday I’ll finish it, but that might be years from now. I’ve been reading it at a rate of 20 pages every three months, usually when I’m on a plane. Twenty pages is about all I can take before I am utterly tired of it.

Because the book is told in comic form, it took me awhile to realize why I found it to be so dull: it’s a textbook.

Textbooks are never interesting. Even when they’re written in an engaging voice, by an accomplished writer, they just don’t operate on that atavistic level that engages a human being’s desire to keep reading or keep watching or keep playing. A textbook can never really get its hooks into you.

Part of this is because textbooks don’t really tell a story, so they don’t engage our narrative sense. Lots of nonfiction books—even non-narrative ones—sort of tell a story. For instance, Taleb’s The Black Swan is the slow unfolding of a thesis about probability. The accretion of evidence and argumentation in some ways functions as a story. Textbooks don’t have that.

But I’ve even enjoyed some nonfiction books that didn’t have a story in them (Bertrand Russell’s History of Western Philosophy, Carnegie’s How To Win Friends And Influence People, David Orr’s Beautiful and Pointless: A Guide To Modern Poetry; etc). I think the key here is information density. Once a work starts putting more than a certain level of information on each page, too much of my attention is devoted to processing and retaining it—I have to stop reading after awhile, because reading the book is mentally exhausting.

However, after thinking about this for awhile, I realized that the nonfiction books which I love the most are the ones that came as close as possible to this level of information-density (a line that I call “The Textbook Axis”)—ones that flirted with becoming textbooks, but didn’t actually get there.

These works provided a mental workout, but the challenge was light enough that it overcoming it was a pleasure. To this end, I have organized, a list of books that I believe fall on either side of The Textbook Axis. I’ve read all the books on the left-hand side and tried to read the books on the right-hand side.

Almost A Textbook


The Feminine Mystique

The Second Sex

A History of Western Philosophy by Bertrand Russell

Every work of actual philosophy

Guns, Germs, and Steel by Jared Diamond

Collapse by Jared Diamond

Robert Caro’s Lyndon Johnson books (and also the Power Broker)

Rise and Fall of the Roman Empire by Gibson

Plutarch’s Lives

Essays by Ralph Waldo Emerson

Lytton Strachey’s Eminent Victorians

Walden by Henry David Thoreau

All of the works on the right-hand side were enjoyable on a paragraph- and sentence-level (this was especially true for Emerson’s Essays. The man was intensely quotable). But the overall effect of the book was quite dull. Because there was too much information on each page!

Of course, people’s Textbook Axes differ in positioning. I’m sure there are people who enjoy reading Walden. And I bet that one’s Textbook Axis changes positions over time. Hmm, maybe I should use Understanding Comics as a litmus test. On the day that I no longer find it to be boring, I’ll know that I’m ready to go out and tackle a whole wide world of heretofore dull books.

2 thoughts on “In order to be interesting, a nonfiction book must flirt with—but NEVER cross—the Textbook Axis.

    1. R. H. Kanakia

      Probably, but like most things, I think there’s a clustering of values, so it is possible to say, in general, where it is in an average, general sort of way (that is, if it wasn’t a very vague and unmeasurable sort of metric).

Comments are closed