First, two conjectures:

1. Basically nobody has read Knausgaard’s My Struggle all the way through. My evidence: (i) it’s really long; (ii) there’s about 1000 pages in the middle that most people, including the author, think is substantially weaker than the rest; (iii) most commentary, even from people who talk as if they’ve read the whole thing, focuses on the first book or two; (iv) there is some absolutely mind-blowing, wonderful, hair-raising stuff in the sixth book. More concretely: I’ve heard a bunch of people describe My Struggle as provocative, but the evidence they give tends to come from the first 100-500 pages of the series, even though book VI gives, e.g., an argument that the ethos of modern Western day cares comes, psychoculturally, right out of Nazi values. And people claim that the series is structurally innovative, but again, the evidence for this usually comes from the beginning of the series. I’ve never heard someone mention that there’s a c. 100-page analysis of a Paul Celan poem right in the middle of Book VI. The best explanation of all this is that people simply never got to (or even near) Book VI.
2. Basically nobody has read The Mathematics of Poker all the way through. My evidence: (i) it’s long; (ii) much of it is difficult; (iii) most commentary, even from people who talk as if they’ve read the whole thing, focuses on the overall approach or the beginning. So, for example, the poker world is full enough of people who like to poke fun at and/or find amusing details in poker books, but I’ve never heard anyone other than me mention that near the end of Mathematics of Poker there’s a page or two where every lowercase s is italicized (so that you read things like “this issue”). And here’s a blog post from 2014 in which a respectable person claims that the whole risk of ruin analysis is completely flawed. Whether or not the claim is right or wrong, the issue is (I think!) pretty serious, at least as a matter of clear communication. Again, the best explanation of why all this happens only in a few comment threads from 8 years after the book’s publication date is that basically nobody ever made it to, much less seriously worked through, that part of the book.

This all reminds me of:

First, Warren Buffett’s famous claim that you should simply read a lot of relevant material because “knowledge builds up, like compound interest.”

Second, a claim from this wonderful tweetstorm about gaps in results from number theory:

Apparently it’s frequently the most famous people who discover these gaps. Roommate said it was surprising the degree to which a lot of what famous people do is actually what everyone is “supposed to do,” e.g. deeply understanding techniques, checking things for yourself…

And third, Byrne Hobart noting that that claim “generalizes across other fields.” I can affirm it’s true in philosophy, poker, and software. People often ask what they can do to generate original contributions or comparative advantages. Usually they vastly overestimate how common it is to have gone through the basic intellectual background in a field. If you’ve actually read the book (actually checked the proof, actually implemented the algorithm), you’re probably way ahead of the field. Much expertise is simply doing this over and over.

Addendum 2022-10-05: This note from Donald Knuth suggests that he thinks that, for many problems in The Art of Computer Programming, “nobody has been sufficiently motivated to check the finer points out carefully as yet.” In context, “check out the finer points” seems clearly to mean something like “actually work through properly.”

Addendum 2023-08-12: ChatGPT is, in my experience, very bad at summarizing the plot of, retrieving factual information about, and even knowing how many books are in War and Peace. I’m sure my prompt engineering is lacking, but there’s also a lesson here about how much training data was generated by people who had actually read War and Peace.