Well taken and it goes for podcasts too! My rule of thumb is that no podcast episode contains more than ten minutes of useful information. This leads me to avoid producers like Scott Adams, who always runs an hour and a half because he's so impressed with the sound of his own voice. YMMV.
"I believe that 99 percent of nonfiction authors should not be writing books. They should write essays on Substack" Bravo! I could not agree more.
This thought has been brought to the fore recently by the clutch of recent books about 'Wokeness' (Hanania, Rufo, Henderson et al). I don't thinks there's anything you can usefully say about this subject that cannot be said in a 3000 word essay. People who want to learn more about society works should read good novels....and (somewhat amazingly given the times we live in) these are still being written in quantity.
Love the framing of reading from the outside in. Scott Sumner had a recent post about the background info we need before really understanding and evaluating an economic claim. The best writers, like Arnold, raise these iceberg intuitions into the light. Often they can be suggested effectively in bumper sticker form, despite the complexity around them.
Quoting Sumner here because I love this example. Using the example of price gouging, Sumner writes,
"I can explain to the man on the street why price gouging is actually a good thing, but my explanation won’t sink in. It requires too much background information. Here’s just a portion of what you need to know—and I mean really know in your bones:
1. Supply and demand elasticities are far greater than common sense suggests.
2. Public policy is a repeat game—policies need to be evaluated as a long run regime, not as an ad hoc decision.
3. Retailing is a highly competitive industry, with zero economic profits in the long run.
4. Willingness to pay is far less correlated with wealth than you might assume.
5. Price controls are not an effective way to redistribute income.
6. The economy is not a zero sum game.
All of these ideas (and many more) need to be understood before considering the question of price gouging. And not just “understood” in the sense that someone tells you the words; you need to understand the ideas well enough so that you could persuasively explain the claims to your friends."
It would be nice non-fiction substack writers could be more... dialogical? (I can't think of a better word.) Daniel Kahneman's "Adversarial Collaboration" comes to mind as a decent model... basically, we not only need shorter essays, but essayists who are responding directly to those with alternative viewpoints.
Sometimes yes. I’ve read many a business book that is one good lesson wrapped in self indulgent biography and a selection of pop-psych anecdotes.
But I am also reminded of the Econtalk with Adam Mastroianni on how we learn. So much of schooling is being able to parrot responses without internalizing the lesson. If someone can explain the efficient market hypothesis but still follows stock-picking shows, do they really know it?
Learning is deeper than recitation. Sure, someone might just be able to recite the bumper-sticker lesson of GGS, but it’s possible the bumper sticker is (to use an overused metaphor) the tip of an iceberg of changed thinking about history in a way that makes the mind less susceptible to arguments that contradict GGN’s lessons. Learning requires marinating the brain in an idea, which good books do well.
If non-fiction books are full of padding, signaling, and beside-the-point-ism, what about documentary films? I find them even more padded, and information-sparse, and unnecessarily long.
I think if I read the essay-length version of Thinking Fast and Slow I would not remember a word of it. Maybe information density trades off against effective audience uptake.
One thing I disagree with is information density. The problem isn't that books aren't information
Non-fiction books are made to be written, not read.
I think of them as showing your work. If you have a serious idea, you need to write a book about it to think through and document the implications. Anyone can make an idea sound good in a blog post. It's harder to make it work in a longer form. Doing the work of writing it out is crucial to vetting and creating workable ideas. Thus, books are made to be written.
The reasons to read books are kind of ancillary to this. You want to learn an idea in depth to be able to do more than just recite the basic concepts. You might want to do this for training about the topic, or to better support or criticize the idea, or learn how to write a book of your own on the topic.
Most non-fiction authors do condense their book into one short essay - the introduction. Most of the time you can even find the intro free on Google Books.
"If you have one really important idea, why does it require a whole book?"
Amen to that.
Compared to a book-length treatment, most good ideas would benefit from being refined to a killer essay/article/blog post, with links to supporting evidence, research, and other essays on related ideas. Assembling all of that for a book means having to cut some of the supporting info out (or else the book is too unwieldy).
I did undergraduate studies in both history & economics. At that time, for a history professor to get tenure they needed to publish one or two books. Not true in economics. As a result, the syllabus was crammed with books for a history course, but an economics course syllabus was largely articles. I wonder if the proliferation of books rather essays was due to many disciplines also requiring books for tenure.
Sometimes fiction can be a more interesting and compelling way to present ideas. Labatut’s When We Cease to Understand the World, for example, is a weird yet fascinating work about scientific discovery, unintended consequences and ethics. This is one of the reasons I enjoy fiction as much as nonfiction.
Books are not about reading or writing but signaling culture and that you fit in to this clique or that. The "two or three sentences" is all you're supposed to retain.
"more short essays and fewer books." << yes, this is why YOUR links, and often their links, is where I'm happier spending most of limited time.
Despite thinking that my other favorite blogger, https://www.thenewneo.com/, is more accurate - I'm less interested in most politics now, and more in culture & economics.
Most good writers spend too much time on non-essential parts of the thought, or digressions. I like Freddie's writing style (de Boer's offering to do ghostwriting for others), but he's too voluminous. So are Scott Alexander & the Zvi.
But it's the book that supports the book tour and interviews and more serious discussion of The Idea. Tho Rob Henderson's "Luxury Belief" idea is so powerful, and more accurately explanatory, that he's gotten a good amount of publicity before his book, now available for pre-order "Troubled". Plus his upbringing is different than most upper middle class writers, as was JD Vance, who's fame did follow his book. And biographies, in general, have more human interest than ideas so their added length, when interesting, is fine.
On "Guns, Germs, and Steel", it's at least a few different bumper stickers and far better/ more important than many more recent books, tho missing the genetics of Razib.
A big essay, like what Robert Wrangham wrote before "The Goodness Paradox", is likely 80% of the important thinking, but possibly 120% of what is remembered.
-
Robert Wright's post on the Jobocalypse seems mostly right, but fails to mention some groups like middle managers who collect info from frontline workers and summarize the important points in reports & meetings (& ppt) to the higher VPs. Internal summaries of info used in decision making is what lots of VPs see managers as doing, and mgrBot.ai will soon be able to do that better, faster, more accurately, and more up-to-date.
Another group is gov't employees - the gov't needs programmers to make self-service for citizens better, and to make the laws and the programs more understandable - but it's supposed to be politicians who drive the bigger changes.
I'm looking forward to more e-gov't and less gov't employees.
Will be lots of manual work for humans - for lower pay. Non-STEM college grad "elite overproduction" will become an increasing problem. We need more middle class folk to ... start creating new, small companies. Many won't be so successful, but many of the future upwardly mobile folk will be small business owners, despite Dem policies which make it more difficult to start and run small companies.
What is the goal of the author? If the only goal were to disseminate knowledge, then maybe the Substack essay is best. But if you're a Nobel prize winner, you write the book.
Personally, I'm mostly interested in non-fiction books that seem to run counter to my intuition. And even in that case I'd rather read a review first to make sure the arguments of the book are reasonable and supported by data before spending my time reading it.
If a book makes a case for something I already believe in or intuitively think is right, I'll do the mental exercise of arguing for it myself (citing data, not books or authors!) to check whether I'm actually deceiving myself. Maybe I think I can already argue for that case convincingly when I actually can't.
Well taken and it goes for podcasts too! My rule of thumb is that no podcast episode contains more than ten minutes of useful information. This leads me to avoid producers like Scott Adams, who always runs an hour and a half because he's so impressed with the sound of his own voice. YMMV.
"I believe that 99 percent of nonfiction authors should not be writing books. They should write essays on Substack" Bravo! I could not agree more.
This thought has been brought to the fore recently by the clutch of recent books about 'Wokeness' (Hanania, Rufo, Henderson et al). I don't thinks there's anything you can usefully say about this subject that cannot be said in a 3000 word essay. People who want to learn more about society works should read good novels....and (somewhat amazingly given the times we live in) these are still being written in quantity.
Love the framing of reading from the outside in. Scott Sumner had a recent post about the background info we need before really understanding and evaluating an economic claim. The best writers, like Arnold, raise these iceberg intuitions into the light. Often they can be suggested effectively in bumper sticker form, despite the complexity around them.
Quoting Sumner here because I love this example. Using the example of price gouging, Sumner writes,
"I can explain to the man on the street why price gouging is actually a good thing, but my explanation won’t sink in. It requires too much background information. Here’s just a portion of what you need to know—and I mean really know in your bones:
1. Supply and demand elasticities are far greater than common sense suggests.
2. Public policy is a repeat game—policies need to be evaluated as a long run regime, not as an ad hoc decision.
3. Retailing is a highly competitive industry, with zero economic profits in the long run.
4. Willingness to pay is far less correlated with wealth than you might assume.
5. Price controls are not an effective way to redistribute income.
6. The economy is not a zero sum game.
All of these ideas (and many more) need to be understood before considering the question of price gouging. And not just “understood” in the sense that someone tells you the words; you need to understand the ideas well enough so that you could persuasively explain the claims to your friends."
It would be nice non-fiction substack writers could be more... dialogical? (I can't think of a better word.) Daniel Kahneman's "Adversarial Collaboration" comes to mind as a decent model... basically, we not only need shorter essays, but essayists who are responding directly to those with alternative viewpoints.
Sometimes yes. I’ve read many a business book that is one good lesson wrapped in self indulgent biography and a selection of pop-psych anecdotes.
But I am also reminded of the Econtalk with Adam Mastroianni on how we learn. So much of schooling is being able to parrot responses without internalizing the lesson. If someone can explain the efficient market hypothesis but still follows stock-picking shows, do they really know it?
Learning is deeper than recitation. Sure, someone might just be able to recite the bumper-sticker lesson of GGS, but it’s possible the bumper sticker is (to use an overused metaphor) the tip of an iceberg of changed thinking about history in a way that makes the mind less susceptible to arguments that contradict GGN’s lessons. Learning requires marinating the brain in an idea, which good books do well.
If non-fiction books are full of padding, signaling, and beside-the-point-ism, what about documentary films? I find them even more padded, and information-sparse, and unnecessarily long.
I think if I read the essay-length version of Thinking Fast and Slow I would not remember a word of it. Maybe information density trades off against effective audience uptake.
One thing I disagree with is information density. The problem isn't that books aren't information
Non-fiction books are made to be written, not read.
I think of them as showing your work. If you have a serious idea, you need to write a book about it to think through and document the implications. Anyone can make an idea sound good in a blog post. It's harder to make it work in a longer form. Doing the work of writing it out is crucial to vetting and creating workable ideas. Thus, books are made to be written.
The reasons to read books are kind of ancillary to this. You want to learn an idea in depth to be able to do more than just recite the basic concepts. You might want to do this for training about the topic, or to better support or criticize the idea, or learn how to write a book of your own on the topic.
Most non-fiction authors do condense their book into one short essay - the introduction. Most of the time you can even find the intro free on Google Books.
"If you have one really important idea, why does it require a whole book?"
Amen to that.
Compared to a book-length treatment, most good ideas would benefit from being refined to a killer essay/article/blog post, with links to supporting evidence, research, and other essays on related ideas. Assembling all of that for a book means having to cut some of the supporting info out (or else the book is too unwieldy).
I did undergraduate studies in both history & economics. At that time, for a history professor to get tenure they needed to publish one or two books. Not true in economics. As a result, the syllabus was crammed with books for a history course, but an economics course syllabus was largely articles. I wonder if the proliferation of books rather essays was due to many disciplines also requiring books for tenure.
Sometimes fiction can be a more interesting and compelling way to present ideas. Labatut’s When We Cease to Understand the World, for example, is a weird yet fascinating work about scientific discovery, unintended consequences and ethics. This is one of the reasons I enjoy fiction as much as nonfiction.
Books are not about reading or writing but signaling culture and that you fit in to this clique or that. The "two or three sentences" is all you're supposed to retain.
"more short essays and fewer books." << yes, this is why YOUR links, and often their links, is where I'm happier spending most of limited time.
Despite thinking that my other favorite blogger, https://www.thenewneo.com/, is more accurate - I'm less interested in most politics now, and more in culture & economics.
Most good writers spend too much time on non-essential parts of the thought, or digressions. I like Freddie's writing style (de Boer's offering to do ghostwriting for others), but he's too voluminous. So are Scott Alexander & the Zvi.
But it's the book that supports the book tour and interviews and more serious discussion of The Idea. Tho Rob Henderson's "Luxury Belief" idea is so powerful, and more accurately explanatory, that he's gotten a good amount of publicity before his book, now available for pre-order "Troubled". Plus his upbringing is different than most upper middle class writers, as was JD Vance, who's fame did follow his book. And biographies, in general, have more human interest than ideas so their added length, when interesting, is fine.
On "Guns, Germs, and Steel", it's at least a few different bumper stickers and far better/ more important than many more recent books, tho missing the genetics of Razib.
A big essay, like what Robert Wrangham wrote before "The Goodness Paradox", is likely 80% of the important thinking, but possibly 120% of what is remembered.
-
Robert Wright's post on the Jobocalypse seems mostly right, but fails to mention some groups like middle managers who collect info from frontline workers and summarize the important points in reports & meetings (& ppt) to the higher VPs. Internal summaries of info used in decision making is what lots of VPs see managers as doing, and mgrBot.ai will soon be able to do that better, faster, more accurately, and more up-to-date.
Another group is gov't employees - the gov't needs programmers to make self-service for citizens better, and to make the laws and the programs more understandable - but it's supposed to be politicians who drive the bigger changes.
I'm looking forward to more e-gov't and less gov't employees.
Will be lots of manual work for humans - for lower pay. Non-STEM college grad "elite overproduction" will become an increasing problem. We need more middle class folk to ... start creating new, small companies. Many won't be so successful, but many of the future upwardly mobile folk will be small business owners, despite Dem policies which make it more difficult to start and run small companies.
What is the goal of the author? If the only goal were to disseminate knowledge, then maybe the Substack essay is best. But if you're a Nobel prize winner, you write the book.
Personally, I'm mostly interested in non-fiction books that seem to run counter to my intuition. And even in that case I'd rather read a review first to make sure the arguments of the book are reasonable and supported by data before spending my time reading it.
If a book makes a case for something I already believe in or intuitively think is right, I'll do the mental exercise of arguing for it myself (citing data, not books or authors!) to check whether I'm actually deceiving myself. Maybe I think I can already argue for that case convincingly when I actually can't.