I once heard that thick books generate a higher profit margin because people are willing to pay more for thick books. A change in thinking is needed: thinner books should be more expensive than they are today because buying them also saves time that would be lost reading thicker versions of the same books.
I have courses on Udemy. 265000 students. 197000 have never started. In the last 12 mths 96% of my students did not look at even one lecture. It’s not just books…
I think readers are making a choice -- and are thus taking some kind of risk -- each time they decide how much of a book to read. If it has a good beginning but gradually turns out to be a crappy one that wastes your time, it would be a loss to spend hours reading it from cover to cover. But if it turns out to have fruitful content throughout the entire book, it would be a loss -- or an incomplete gain at best -- to just read the first chapter and call it a day.
My personal thought is that whatever book it is, it's best to finish the whole thing. Even if it turns out to be crappy, there must be a reason why you think it's crappy, and by exploring that reason you still take away some kind of new knowledge or rediscovery of personal philosophy. This encourages me to finish whatever book I read each time, even if sometimes it takes more time than I expect.
I love these reflections, Paul. As a nonfiction author, I think there's a communicative obligation to have the important qualifications up front alongside the main idea. (This is one of my complaints about Against Empathy.)
I also tend to conclude my introductory chapters by inviting readers to skip ahead if they get bored or bogged down, and to read the chapters in any order they like, though my guess is that few actually take this advice. It makes sense to write chapters fairly modularly anyway, since few people will remember the details of Chapter 2 by the time they're on Chapter 7.
I know this seems defensive, but I’m surprised by your complaint about AE. God knows there are a lot of reasonable ways to disagree with my views, but I begin the book in an unusual way—much of Chapter 1 responds to criticisms of earlier versions of my view, and so I am (perhaps boringly) careful to be clear about what I mean by empathy, what the scope of my argument is (it applies to morality, not other facets of life), and I even discuss cases where I agree that my argument doesn’t work. Whatever the flaws of AE, I do think that I meet up with the obligation you’re talking about.
It has been a long time, but my experience as a reader was that the book started out seeming very bold but accumulated qualifications along the way, resulting by the end in a more moderate position than I thought was suggested by the beginning.
I seem to remember writing a blog post about the author's obligation to guard carefully against obvious misreadings of their work and using Against Empathy as an example in which those misreadings weren't adequately prevented -- and I seem to recall your replying to that post. Oddly, I can't seem to find that post anymore. Maybe you convinced me I was being unfair and I took it down? The vicissitudes of memory....
Thanks for this post, which I enjoyed and identified with. I found your Substack because I started reading 'Against Empathy' (10% down so far, and it's a good read as well as interesting). I don't know yet whether I'll get to the end: like you, I begin far more non-fiction books than I finish. But there's the difference. I normally read a whole novel because, as you say about films, it's an artistic experience that I value.
Non fiction is different. Then I'm in a conversation with writers. Reading non-fiction nourishes my own thinking – that's why I do it. I can get distracted because an author has given me an idea that I want to go away and play with: that's what you've already done with 'Against Empathy'. I'll read on, but I might get distracted by another thread of thinking. Yesterday I finished a book by Kay Redfield Jamison I began two years ago because the moment to engage with her thinking had returned.
My point here is that there is a fundamental difference between reading as an intellectual practice, developing one's own thinking by engaging with any of the available resources – long books, short books, blog posts, podcasts, magazine articles, academic papers etc. – and reading for pleasure. Both are creative, but only the first requires the reader to engage with and understand the architecture of the whole. Doing that with non-fiction would mean limiting oneself to the tiny proportion of books that even an avid reader like myself can get through in a lifetime. Having studied literature as an undergraduate, it took me years to accept that I didn't have to finish a book just because I have started it. Now, I read in a much more creative way.
PS I found the part of Pierre Bayard's 'How to speak about books you haven't read' very interesting in this respect. It's notable that the French title 'Comment parler des livres que l'on a pas lus?' has a question mark at the end because the book is a philosophical enquiry; the absence of that punctuation mark in the English translation suggests that it was being marketed as a self-help book.
Agreed. One of the most common reactions I have to a non-fiction book I've started is "I'm reading the wrong book!" That is, the book has stirred up an idea or question in me that needs to be investigated first before this book can be profitably engaged with (if I decide to return to it). And that's OK! Or, at least, that's what I think, and since I'm in charge of what I read, what I say goes!
I’m surprised at how many book *reviewers* you suggest don’t read the whole book. If you are a reviewer, you have one job, which is to read the whole book, so you can let other people know whether or not they have to, and if they don’t have to, you should let them know which parts they should read, which requires you to have read all the parts.
It would be worth comparing this to academic papers. Most people never get past the abstract, and the ones that do usually don’t finish. There’s a good number of papers where I’ve just dipped into relevant sections without reading from the beginning.
We need to think about how people actually use papers and books, and write them to be used effectively in that way, rather than writing them assuming that most users read from beginning to end.
I finish 90% of the books I start. Including Hawkings, and his twice. But I am from a previous age, admittedly.
Of course there are readers and there are readers. I do not include my college freshmen/sophomore students, many of whom are at this point in time essentially functionally illiterate, and do not read.
And there are books and there are books. I do not include those books that are like unto an overripe cheese: you don't have to eat the entire thing to know it is bad.
Motivational compromising: perhaps there are too many books and too few that are good. Money, Samuel Johnson to the contrary, is an insufficient reason alone to write books that matter. And books that matter, it seems to me, require a habit of reading and a sustained habit of thinkng.
And time, which is admittedly short for all of us today.
But Sam Altman's satement? I wonder. What we mistake for intelligence in some people is little more than a local inflamation of the brain, bestowng upon the owner a narrow (sometimes single) range of savant abilities.
I'm still reeling from the numbers you quote here. Food for thought. Many thanks.
As a full-time academic who doesn't have much of a social life because he lives in a wheelchair now, I decided to do an experiment and actually record how many books I read (cover to cover) in a single year. I'm going from December 2025 to December 2026, in part because I had a productive Christmas books-wise. This is because I'm sceptical that few people get to read more than I do, given my situation, and yet a little while ago I saw some stat claiming that 4% of Americans read over 50 books a year.
50 is a lot!
1/4 through the experiment, and although definitive results will have to wait until the end of the year...50 is a lot!
I read your post, but not to the end... (I did finish Psych though - and A Brief History of Time - and wished both had been longer).
My impression is that the books I read to the end seem to be the ones where I already knew most of the material. This means either I DO already know most of it (Brief History), or it's written in such a way that I feel I do (Psych probably).
Do people actually stop reading, or just put to one side for finishing later? This might simply mean a shiny new book has arrived, or you want to let that topic soak for a while and read something else.
Anyway, professional authors excepted, most books are written cos the author wants to write them. Readers merely provide downstream validation.
Shitty when it comes to book reviewers or books queried to publishers. I guess the reader/consumer has their right but still quite a bit disappointing. G.I. Gurdjieff's Beelzebub's Tales to His Grandson was once the most abandoned book found on trains or in cafés in Europe after it had first come out. I've tried to read it several times and will probably go back again at some point. Although Gurdjieff wrote in a way to purposefully make it a struggle. I couldn't get through Cormac McCarthy's early work and still don't think I can. From Blood Meridian on I'm all in. Just read the Passenger and am saving Stella Maris for two or three books down the road. Had enough of David Foster Wallace's Infinite Jest by the halfway point so quit that with no plans to go back. Although with fiction, I try to force myself through it even if just to know why it didn't work for me. I can see not getting through whole works of non-fiction in one go depending on the subject and especially if it is in your own library and you plan to revisit it at some point. Oh well. Whatta ya do? Tough world
Ha! "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." Delightful And, of course, a spur to read all the way to the end in case there was another trap
I have recently been stressing out about this exact issue because I think the last few chapters of my book are actually the best...but who will get there?
If you haven’t sent the final draft off to the publisher, then include a note in the first paragraph telling readers to skip ahead to the last chapter!
I once heard that thick books generate a higher profit margin because people are willing to pay more for thick books. A change in thinking is needed: thinner books should be more expensive than they are today because buying them also saves time that would be lost reading thicker versions of the same books.
As a time-starved person, I think most non-fiction books are 3-4x longer than they should be.
I think you are right. Lots of non-fiction books should be (or started as) an article.
I have courses on Udemy. 265000 students. 197000 have never started. In the last 12 mths 96% of my students did not look at even one lecture. It’s not just books…
I think readers are making a choice -- and are thus taking some kind of risk -- each time they decide how much of a book to read. If it has a good beginning but gradually turns out to be a crappy one that wastes your time, it would be a loss to spend hours reading it from cover to cover. But if it turns out to have fruitful content throughout the entire book, it would be a loss -- or an incomplete gain at best -- to just read the first chapter and call it a day.
My personal thought is that whatever book it is, it's best to finish the whole thing. Even if it turns out to be crappy, there must be a reason why you think it's crappy, and by exploring that reason you still take away some kind of new knowledge or rediscovery of personal philosophy. This encourages me to finish whatever book I read each time, even if sometimes it takes more time than I expect.
One minor point, “On Bullshit” was turned into an 80 page book, but it’s just a 20 page paper that was reprinted on small pages with big font!
I love these reflections, Paul. As a nonfiction author, I think there's a communicative obligation to have the important qualifications up front alongside the main idea. (This is one of my complaints about Against Empathy.)
I also tend to conclude my introductory chapters by inviting readers to skip ahead if they get bored or bogged down, and to read the chapters in any order they like, though my guess is that few actually take this advice. It makes sense to write chapters fairly modularly anyway, since few people will remember the details of Chapter 2 by the time they're on Chapter 7.
I know this seems defensive, but I’m surprised by your complaint about AE. God knows there are a lot of reasonable ways to disagree with my views, but I begin the book in an unusual way—much of Chapter 1 responds to criticisms of earlier versions of my view, and so I am (perhaps boringly) careful to be clear about what I mean by empathy, what the scope of my argument is (it applies to morality, not other facets of life), and I even discuss cases where I agree that my argument doesn’t work. Whatever the flaws of AE, I do think that I meet up with the obligation you’re talking about.
It has been a long time, but my experience as a reader was that the book started out seeming very bold but accumulated qualifications along the way, resulting by the end in a more moderate position than I thought was suggested by the beginning.
I seem to remember writing a blog post about the author's obligation to guard carefully against obvious misreadings of their work and using Against Empathy as an example in which those misreadings weren't adequately prevented -- and I seem to recall your replying to that post. Oddly, I can't seem to find that post anymore. Maybe you convinced me I was being unfair and I took it down? The vicissitudes of memory....
Thanks for this post, which I enjoyed and identified with. I found your Substack because I started reading 'Against Empathy' (10% down so far, and it's a good read as well as interesting). I don't know yet whether I'll get to the end: like you, I begin far more non-fiction books than I finish. But there's the difference. I normally read a whole novel because, as you say about films, it's an artistic experience that I value.
Non fiction is different. Then I'm in a conversation with writers. Reading non-fiction nourishes my own thinking – that's why I do it. I can get distracted because an author has given me an idea that I want to go away and play with: that's what you've already done with 'Against Empathy'. I'll read on, but I might get distracted by another thread of thinking. Yesterday I finished a book by Kay Redfield Jamison I began two years ago because the moment to engage with her thinking had returned.
My point here is that there is a fundamental difference between reading as an intellectual practice, developing one's own thinking by engaging with any of the available resources – long books, short books, blog posts, podcasts, magazine articles, academic papers etc. – and reading for pleasure. Both are creative, but only the first requires the reader to engage with and understand the architecture of the whole. Doing that with non-fiction would mean limiting oneself to the tiny proportion of books that even an avid reader like myself can get through in a lifetime. Having studied literature as an undergraduate, it took me years to accept that I didn't have to finish a book just because I have started it. Now, I read in a much more creative way.
PS I found the part of Pierre Bayard's 'How to speak about books you haven't read' very interesting in this respect. It's notable that the French title 'Comment parler des livres que l'on a pas lus?' has a question mark at the end because the book is a philosophical enquiry; the absence of that punctuation mark in the English translation suggests that it was being marketed as a self-help book.
Agreed. One of the most common reactions I have to a non-fiction book I've started is "I'm reading the wrong book!" That is, the book has stirred up an idea or question in me that needs to be investigated first before this book can be profitably engaged with (if I decide to return to it). And that's OK! Or, at least, that's what I think, and since I'm in charge of what I read, what I say goes!
I’m surprised at how many book *reviewers* you suggest don’t read the whole book. If you are a reviewer, you have one job, which is to read the whole book, so you can let other people know whether or not they have to, and if they don’t have to, you should let them know which parts they should read, which requires you to have read all the parts.
It would be worth comparing this to academic papers. Most people never get past the abstract, and the ones that do usually don’t finish. There’s a good number of papers where I’ve just dipped into relevant sections without reading from the beginning.
We need to think about how people actually use papers and books, and write them to be used effectively in that way, rather than writing them assuming that most users read from beginning to end.
Hear, hear.
This is bewildering to me.
I finish 90% of the books I start. Including Hawkings, and his twice. But I am from a previous age, admittedly.
Of course there are readers and there are readers. I do not include my college freshmen/sophomore students, many of whom are at this point in time essentially functionally illiterate, and do not read.
And there are books and there are books. I do not include those books that are like unto an overripe cheese: you don't have to eat the entire thing to know it is bad.
Motivational compromising: perhaps there are too many books and too few that are good. Money, Samuel Johnson to the contrary, is an insufficient reason alone to write books that matter. And books that matter, it seems to me, require a habit of reading and a sustained habit of thinkng.
And time, which is admittedly short for all of us today.
But Sam Altman's satement? I wonder. What we mistake for intelligence in some people is little more than a local inflamation of the brain, bestowng upon the owner a narrow (sometimes single) range of savant abilities.
I'm still reeling from the numbers you quote here. Food for thought. Many thanks.
As a full-time academic who doesn't have much of a social life because he lives in a wheelchair now, I decided to do an experiment and actually record how many books I read (cover to cover) in a single year. I'm going from December 2025 to December 2026, in part because I had a productive Christmas books-wise. This is because I'm sceptical that few people get to read more than I do, given my situation, and yet a little while ago I saw some stat claiming that 4% of Americans read over 50 books a year.
50 is a lot!
1/4 through the experiment, and although definitive results will have to wait until the end of the year...50 is a lot!
Jordan Ellenberg is so underrated. I loved *Shape* and *How Not to be Wrong*. (Both of which I finished!)
I read your post, but not to the end... (I did finish Psych though - and A Brief History of Time - and wished both had been longer).
My impression is that the books I read to the end seem to be the ones where I already knew most of the material. This means either I DO already know most of it (Brief History), or it's written in such a way that I feel I do (Psych probably).
Do people actually stop reading, or just put to one side for finishing later? This might simply mean a shiny new book has arrived, or you want to let that topic soak for a while and read something else.
Anyway, professional authors excepted, most books are written cos the author wants to write them. Readers merely provide downstream validation.
Shitty when it comes to book reviewers or books queried to publishers. I guess the reader/consumer has their right but still quite a bit disappointing. G.I. Gurdjieff's Beelzebub's Tales to His Grandson was once the most abandoned book found on trains or in cafés in Europe after it had first come out. I've tried to read it several times and will probably go back again at some point. Although Gurdjieff wrote in a way to purposefully make it a struggle. I couldn't get through Cormac McCarthy's early work and still don't think I can. From Blood Meridian on I'm all in. Just read the Passenger and am saving Stella Maris for two or three books down the road. Had enough of David Foster Wallace's Infinite Jest by the halfway point so quit that with no plans to go back. Although with fiction, I try to force myself through it even if just to know why it didn't work for me. I can see not getting through whole works of non-fiction in one go depending on the subject and especially if it is in your own library and you plan to revisit it at some point. Oh well. Whatta ya do? Tough world
Ha! "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." Delightful And, of course, a spur to read all the way to the end in case there was another trap
I have recently been stressing out about this exact issue because I think the last few chapters of my book are actually the best...but who will get there?
If you haven’t sent the final draft off to the publisher, then include a note in the first paragraph telling readers to skip ahead to the last chapter!
Too late!
Excellent post. I almost never finish non-fiction books either (including, alas, Jordan Ellenberg's...), but I did finish Descartes' Baby!!