Nobody finishes reading my books

Or anyone else's either

Several years ago, University of Pennsylvania law professor Amy Wax wrote a critical review of my book Just Babies.¹ One of the things that bothered me was that Wax plainly hadn’t read the whole book. She got to the chapter on sex and stopped, with two chapters left to go.

But then I checked and realized that the positive reviewers also didn’t seem to make it to the end. Their reviews focused on the first chapters; at best, they skimmed the rest. Actually, by making it to chapter 5 and reading it closely, Wax was unusually persistent.

It’s not just me. The philosopher Martha Nussbaum once complained that reviewers of her book Upheavals of Thought: The Intelligence of Emotions focused on the preliminary outline of her theory in the first chapter and ignored all the nuances and qualifications in the many chapters that followed. They apparently didn’t read most of her book.

This bothered me. I’m a fan of Nussbaum and feel she deserves better. Actually, I own this book of hers and read it with pleasure. Well, not the whole book. Just the first chapter. Have you seen how big it is?

How often do people make it to the end of books? The mathematician Jordan Ellenberg did some number crunching, analyzing passages marked by Amazon Kindle readers and estimating the percentage who finished. This percentage is what he calls the Hawking Index, named after Stephen Hawking’s notoriously unread book A Brief History of Time. Here is the Hawking Index for some popular books:

This is a crude measure. It might underestimate the proportion of people who finish books—maybe they keep reading but stop highlighting. Or maybe it’s an overestimate; readers who mark up a book might be unusually persistent.

But I believe in the general finding—people don’t tend to finish books, particularly when it comes to certain sorts of non-fiction books. Certainly, I don’t tend to finish books. The picture on top of this post is from my office at work. I’ve opened just about every one of these books and read the first page. I haven’t finished more than a dozen.

Why not? Well, I’m not going to finish any of these books in one sitting. I will start to read, and then I will put the book down. Finishing it requires that I pick it up, over and over again. And the odds of me picking it up after I put it down are not 100%.

From there, it’s just math. Suppose there is a 1% chance that at the end of any page, I will put a book down forever. This means that once I read the first page, there is a 99% chance I’ll start page 2. To get to page 3, the odds are 99% times 99% = 98%. The odds I will get to page 100 are 36%. The odds of getting to page 300 are 5%—and most of these books are over 300 pages.

Maybe this is a strange way of looking at it. After all, often the odds are much better than 99%. Some books are so gripping that when I put them down, I can’t wait to get started again. (Maybe I’m reading The Goldfinch, with its 98.5% Hawking Index.) I once stayed at a vacation house with some friends, and after everyone went to sleep, I picked up Scott Turow’s Presumed Innocent, just checking it out to see if I wanted to take it to the beach the next day, and then I stayed up all night reading it. If you look closely, you’ll notice a novel on my shelf—Destiny by Tim Parks. I finished it with pleasure. There is also an academic book published by a university press—Evil Men by James Dawes—that I wouldn’t have dreamt of giving up on.

But for most of the other books, I read the beginnings, maybe leafed through the rest, and never picked them up again.

---

Some qualifications: As I mentioned, there are plenty of books that people do finish reading. Good novels, obviously. Even some bad novels—I’m a fan of Stephen King, and I’ve never given up on one of his books, not even his stinkers. I’ve read biographies of Derek Parfit, Elon Musk, and Sam Bankman-Fried and finished them all. Good biographies have a story-like flow that carries you to the end.

My focus here is more specific—on academic books and non-fiction books for the general audience—such as Capital in the Twenty-first Century, A Brief History of Time, Thinking Fast and Slow (from the Hawking list above), and the (substantially less well-known) books that I write myself.

Now, of course, some people do read such books from cover to cover, and I’ve even heard from people who’ve finished my own books. (The title of this post is meant as hyperbole—comical exaggeration. If you tell me in the comments that you have, in fact, finished reading at least one book, this will show that some people don’t even read past the beginning of Substack posts.) I don’t trust the Hawking index, and perhaps I shouldn’t extrapolate too much from my own experience. I tend to buy more non-fiction books than most people, in part because, as a professor, I pretty much read for a living. Maybe I give up on more books too.

There are people who finish more books than I do, such as Rob Henderson—a superb writer who worked in my Psychology lab at Yale many years ago. In a post called How I Read, he writes

In a given year, I read about 40-50 books cover to cover, read excerpts and chapters of perhaps another 100 or so, and skim many more.

40-50 books a year is a lot of books to finish! (And I highly recommend his advice on how to establish a routine to get all that reading done.) But then there’s the other 100 or so other books he mentions. Even Henderson, then, finishes only about 1/3 of the books that he picks up.

Tyler Cowen, a brutally selective reader, finishes a smaller proportion.

The important thing is to be ruthless with the books that are not good. Just stop reading, put them down, usually throw them away, don’t give them away – if you give them away you could be doing harm to people …

Sometimes readers just go on and on with blather, or with personal detail that has no relevance to the argument. Or there are just pages of terminology and it’s like, well, you might still give the book a chance, but you start turning the pages more rapidly. And you’re just waiting for some bit of meat, you’re like out there desperate, giving the author still a chance, and then at some point you’re like “No, sorry. ” Zap – throw it in the trash, on to the next one.

---

What’s the real problem here?

I’m not complaining that people aren’t reading my books. (If I ever were to say that, feel free to quote Livia Soprano at me.) I’m actually really pleased with how many people read my books (or at least the beginnings of them).

I’m not complaining that too few people are reading long books of the type that I’m interested in. Given how good other sources of information and entertainment are, it’s impressive that people read these books at all.

I’m not complaining about a Cowen-like selectivity. There’s no need to keep reading books that aren’t working for you.

I’m not complaining about skipping or skimming parts of books. I like biographies, but I just skip past all the boring parts about grandparents that biographers feel compelled to put in. Even some of my favorite books have parts I’ve quickly leafed through.

Here’s what really bothers me. Books such as Upheavals of Thought are meant to be read from cover to cover. You can skip or skim a bit, but the argument is developed throughout the book, and if you give up after a chapter or two, you won’t fully appreciate the ideas that the author is trying to convey. There is a serious mismatch between what the author wants and what actually happens.

Consider a couple of analogies:

#1: Omakase dining refers to a multi-course Japanese meal prepared by a chef. It takes a long time to eat; it’s usually expensive; and it can be mind-blowing. If you’re in Toronto, I’ve heard nice things about Shizuku, which offers a 22-course meal for $270 Canadian dollars.

How many people stay until the end?

I bet you never thought to ask that question. Barring emergencies, everyone stays until the end, because they paid for the whole thing and the whole thing is pretty damn good. It would be strange to walk out halfway through just because you’re not in the mood for more Japanese food or have somewhere else to go. I bet the chef would be pissed. The experience is meant to be completed.

#2: A while ago, I saw Killers of the Flower Moon, directed by the great Martin Scorsese. It is 3 hours and 26 minutes long. It took about 200 million dollars to make and has an amazing cast. I thought it was terrific.

The movie theatre was less than half full. This isn’t a movie for everyone. And that’s fine—if Scorsese complained, he’d deserve the Livia Soprano response: Poor you! If his goal is to fill theatres, he should make movies set in the Marvel Cinematic Universe.

But now imagine that once the movie started, people began to trickle out. An hour in, one out of every ten seats is occupied. When the movie ends, there are just a few people left. If this happened in every theatre, then, no matter how much money it made, Killers of the Flower Moon would be an artistic failure. Like Omakase dining, this Scorsese movie (like almost all movies except for pornographic ones) is meant to be experienced from beginning to end.

It’s fine if you don’t want to read Upheavals of Thought (or experience Omakase dining, or go to Killers of the Flower Moon). But if you do, you should expect to stay to the end.

If you don’t share my worry about the mismatch of purpose (what the creator intends) and use (how the book is read), think of it in terms of waste. Nussbaum spent years of her life (I’m guessing) writing the whole book, and she likely put equal time into the ending as she did into the beginning. Editors, copyeditors, proofreaders, and someone from the legal department carefully reviewed the entire book. Readers paid for the whole book, supporting all of this labor. If most people just read the first chapter, so much of that time, effort, and money was for nothing.

---

Whose fault is it? I don’t blame Nussbaum; she’s a strong writer, much better than most. I don’t blame the readers either. There is no moral obligation to finish a book that you’re tired of reading.

I blame … the system. Authors are expected to write non-fiction books that are about 70,000 to 100,000 words long. Maybe this was a reasonable length in the past, but now there are too many other distractions in the world, and few of us have the Sitzfleisch anymore for that kind of long book.

You might think the market would correct this. Suppose Omakase dinners were 61 courses. Suppose most movies were over five hours long. People would lose interest, and in response, restaurants and movie studios would ratchet things down, working to give people what they want.

It’s a mystery to me why this sort of correction hasn’t happened with books. One concern is that many academic books are purchased mostly by libraries, which don’t prioritize readability. An even gloomier thought is that many people don’t buy books to read; they buy them to own, and they like to own hefty books.

---

Forget about why this is the case; what’s an author to do?

Don’t worry about it

So many of the rewards of writing books have nothing to do with whether anyone reads them. There are advances and royalties. There are opportunities to go on podcasts, be interviewed by fancy people, and give lectures, sometimes for money. In some fields of academia, you need to publish a book to get tenure. Your parents might be proud of you; you will always have a cheap gift to give to family and friends. When you’re feeling blue, you can look at the book on your shelf and say I wrote that.

Write for those who stay

Suppose, though, that you have ideas and arguments that you want to convey, and this only works when your book is read from cover to cover. So be honest with yourself; acknowledge that there is a small audience that you’re targeting—small, but, despite the hyperbolic title of his post, it’s probably not zero—and write for them.

Focus on beginnings

You can compromise: Put all your energy into a good first chapter, ensuring that someone who gets this far will understand what you’re on about. For the rest of the book, just pad and recycle old material.

Write books where the order doesn’t matter

My most recent book Psych was an introduction to psychology, and it contains 15 chapters, each on a different topic. People can read the chapters in any order they want, and since some of the most interesting ones were the final ones—Chapter 14 on mental illness and Chapter 15 on happiness. This is the rare book where more people probably read the end than the beginning.

Write books for captive audiences

I assigned Psych for my Intro Psych course. (I gave them the book for free, by the way—the morality of professors profiting from students is a topic for another post.) If students didn’t read all the chapters, they wouldn’t do well in the course, so this motivated them to read the whole thing. Unless you are Chairman Mao, though, there aren’t many other contexts outside of a classroom where you can pull this off.

Write really short books.

Like On Bullshit, by Harry Frankfurt. 80 pages and a New York Times bestseller!

Give up on writing a certain sort of books

If you’re writing novels—I can’t wait for The Goldfinch II!—self-help books, biographies, books that can be read out of order, books that people are forced to read, and really short books, keep at it. But if you’re writing the sorts of long books that people don’t tend to finish, and you’re not doing it to get tenure or impress your parents, maybe you should stop. Communicate your ideas in other ways.

I wince to write this. I like writing long books, and I like buying them and starting them, even if I usually don’t finish. I do not want to find myself in the same company as Sam Bankman-Fried, who once said this in an interview:

I’m very skeptical of books. I don’t want to say no book is ever worth reading, but I actually do believe something pretty close to that. I think, if you wrote a book, you fucked up, and it should have been a six-paragraph blog post.

Thomas Chatterton Williams just wrote a wonderful article called The People Who Don’t Read Books, and he picks out this quote of SBF’s for particular scorn. And then Williams makes a moving case for the power and value of the sort of books I am talking about here.

when a book succeeds, even partially, it represents a level of concentration and refinement—a mastery of subject and style strengthened through patience and clarified in revision—that cannot be equaled. Writing a book is an extraordinarily disproportionate act: What can be consumed in a matter of hours takes years to bring to fruition. That is its virtue. And the rare patience a book still demands of a reader—those precious slow hours of deep focus—is also a virtue.

Yes. I love this. Williams is writing here for true readers, capturing our sense of struggling with a long book—“those precious slow hours of deep focus”—and I’d like to see myself as part of this community, someone who not only appreciates these virtues but who embodies them.

But this wouldn’t be honest. I don’t really have the patience he praises, and I’m not sure how many of Williams’ other readers have it either. My worry is that most of us aren’t much better than SBF. We don’t read books either. We just read the beginnings.

1. This was originally written right after I started Small Potatoes, and not many people read it. So I’m reposting it with some minor edits and corrections.