The whole "trying to make the thesis more radical than it actually is" is itself evidence of a different kind of conglomerate authorship: universities and research organizations, as well as the system of peer review and common ideas about what makes research valuable, shape the way that theoretical arguments like this one are shaped and communicated. If there is an intensely formulaic (which isn't necessarily to say bad!) genre of writing that is shaped by institutional demands, it'd be literary scholarship.
Also, I'm sure the point has been made before, but I guess BookTok and (to a numerically much lower degree) Substack are taking on the role of the idiosyncratic editor of old.
I was thinking this too -- it seems like a lot of interesting and worthwhile work is marred by this need to be "radical." Collective authorship is an interesting metaphor, but it seems (based on this and other reviews; I haven't read the book itself) like at some point the author forgot it shouldn't be taken literally.
Am going through my comments now. Yes, I think most people who understand this genre of writing recognize the career exigencies that made Sinykin push for the very provocative language in the beginning of the book. It is manipulative, but it also worked--got the book the attention it deserved.
If anything, there's a certain craft involved in being able to make these claims while segregating them from the part of the book that is actually good. In this case, it was executed about as well as it could be.
I don't know if BookTok is replacing the idiosyncratic editor, so much as it's giving that editor the leverage they need. The more weird, off-beat books that become BookTok trends, the more juice editors will have when they try to publish weird, off-beat books. BookTok seems quite resistant to big publisher marketing hype, so in that way it seems to benefit smaller publishers. And one way to make a splash nowadays is by publishing something that genuinely seems different--not because people are looking for something different, but just because the people who _do_ want something different are more likely to talk about it and try to sell it to other people.
(1) Make some outrageous claim: the author is dead; it's the end of history; the medium is the message.
(2) Write a book in which you walk back your outrageous claim: "I just meant we should value the text more than authorial intent;" "I just meant neoliberalism seems very stable;" "I just meant an artist's medium affects what's being communicated."
(3) Spend the rest of your life complaining that people are misreading you because they think you literally meant the outrageous claim.
I would argue that it is good to be able to put an intuition or finding into a big, bold idea, for your own thinking and for your reader's : breaking taboos or formulating whole new concepts forces people to give more attention to the idea, and to open to something new. Even if facts are more conservative, it makes one round them better than by only advancing from the old consensus. Dialectical method, I suppose ?
I think the best thing is to just have a radical hypothesis that you support with compelling evidence, so people in the end agree "Huh, I guess that's true."
But I understand that the timeline of an American academic career often doesn't allow you to develop an idea to completion within a single book.
In this case, there's definitely an element of truth to this collective authorship idea and perhaps some future book can put more work into developing it.
That's close to my idea, actually. I don't think one should defend a bold idea / radical hypothesis if they don't have compelling evidence to support it. But caveats can be found to most radical ideas. I suppose of course that such hypotheses can be developed and made more complex over the course of a body of research, and that it's the right way.
About the collective authorship question, I really would like to refer to Alan Bowness' book that I cited in a comment to another of your posts, but I'm afraid it seems to be terminally out of print in english :( To sum up, it's about the environment, in the modern world, that an artist needs to create something trully innovative, and the different circles of recognition of innovative work.
God you’re good at this. Totally agree about the strength of the book. Sinykin’s insistence on conglomerate authorship has the energy of a college student trying to shoehorn evidence into a flashy thesis he figured his essay needed.
Wow! This is a thoughtful and sensible discussion around a book that I never heard of. The world of conglomerates is certainly very different from the publishing world James Michener described in his novel "The Novel" (1991). As for the possible conglomeratization of literature, I don't think it is a conscious choice by the successful authors, but more of a Darwinism process of selection. Hopefully there will still be readers looking for something more than the dopamine rush.
I just heard of that Michener book! I kind of want to read it. I read so much Michener growing up, and I am surprised, looking back through bestseller lists, by how popular he was.
I really appreciate this nuanced summary as someone who's interested in the subject matter but is unlikely to read the book.
I think the question of whether great books are produced by lone, romantic geniuses or wider social forces is a false dichotomy. Both things are true. Successful books tend to come out of (and speak to) hidden aspects in the culture that are not being voiced AND the people who produce them tend to be cranky, obsessive perfectionists.
I get a bit tired of how the The Death of The Author gets endlessly restated as if it's a new idea. Of course it's true, but the other thing is also true.
That's true. I mean it's funny because I generally am in favor of this kind of systems-thinking, but I myself am such a cranky, obsessive individualist. I myself am impossible to work with and have a deeply personal vision of what I want to put on the page.
Interesting how art has been subsumed so totally into profit-seeking that “success” is now synonymous with “has broad appeal.” I’m not saying that’s *wrong*, but it certainly does have a homogenizing effect.
It was also enlightening to learn about how shifts in the bookstore economy have influenced publishing. I’ve been puzzling over how it can possibly be true that publishers lose money on most of the books they publish — I mean, *how* can that possibly be true? — and it makes a lot more sense now that I’ve learned that, in the past, it didn’t take as many sales to bring a book into the black due to the mosaic bookstore landscape. Thanks for giving me the info to make sense of that mystery
A great post! Well worth rereading actually. These are very important themes you bring up—concerning the literature market and its effects on culture creation (or a lack thereof).
Just a few observations and recommendations:
• Foucault writes in his indispensable essay “What is an Author” that yes, exactly, the entity the author now no longer exists—a caveat: to understand this, one must grasp what he means by the term Author. Obv ppl still write novels and poems, but do they fulfill the functions Thomas Mann or Jane Austen did? The Author is a function!
I link to the essay in my long-form post #2 “The Author is yet to appear” about the privation of authorship in the Middle Ages, not unlike today’s situation.
• Another must-read on this topic is Chapter 4 of Adorno’s Dialectic of Enlightenment, the Culture Industry. He writes in inscrutable fragments, so it must be read in an open-ended contemplative way. Everyone will get something different out of it. The fragments needn’t be read in any particular order. Its a fun read.
Enjoy! I thought I would contribute things from non-anglophone spheres in case you are not familiar.
I think that drawing a direct comparison to film/tv is a mistake on the author's part, right? Even the most collectively-written novel doesn't require anywhere near the same level of coordination as the typical film, just by its nature.
This review does make me think about Harper Lee, though. There's a famous story (I'm paraphrasing from memory here) that she had a wealthy friend who believed in her and decided to fund her life for year, so that she could finally write her novel. At the end of that year, she delivered the manuscript for TO KILL A MOCKINGBIRD. Except, the story isn't entirely true -- the book she wrote that year was GO SET A WATCHMAN. Then, she spent a long time working with her editor, revising and rewriting, until together they created MOCKINGBIRD (and this was a very collaborative process; her editor had a lot to do with the book we all know). The book they produced together is, without a doubt, a work of art, but it's also much more commercially viable and marketable than WATCHMAN. I love it, but when I reread it recently I saw a lot of precursors to trends I dislike in current fiction (unfortunately I cannot elaborate yet, I'm still working through this). WATCHMAN is rough and clearly needed at least one good revision to be ready, but the racial politics and general morality in it are much more interesting and complex than MOCKINGBIRD. MOCKINGBIRD was published in 1960, maybe too early to be affected by these changes, but I wonder what Lee would have produced had she been working under Steinbeck's conditions.
More or less, I may not recall all the details exactly right. I had heard the basic version a lot over the years, with her friends funding her writing for a year. But the story behind Lee and her editor is told in "Furious Hours," a very good book about Lee's attempts to write a true crime nonfiction book (the Mockingbird background may not have fit real cohesively into this, but it worked well enough and was interesting in its own right).
Hm. I confess I'm kinda reflexively opposed to deeply collaborative writing like that, but I'm probably just being romantic. Frankly I get a little itchy even to know how dramatically Gordon Lish changed Carver's work. But now you're making me think: how different would our national literature be if we didn't have that pillar of Mockingbird. I don't suppose it's a life-or-death decision but...it's a flat truth that my joy at the book's existence outshines any pretentious hangups I have about "authorial integrity."
I wouldn't want to do it... I mean, for Mockingbird money, I could get over it.
I actually have done collaborative writing, about 11 years or so ago. Some friends and I did a narrative podcast. The process was (I assume) similar to a TV writer's room, so it was very collaborative. It was a lot of fun, and the collaborative process helped us put together what I still think was a solid product. But I don't know that we would have ever created great art that way.
Some people might take better to being edited than others do. I also think there's a magic that can happen with the right editor and the right writer. I've definitely had editors who knew how to push me to be better--part of that was that they usually convinced me that whatever I wanted was what would ultimately stand.
There is another and I would argue more insidious movement in writing and publishing that of the MFA factory writing dedicated to pushing the novel with speed forward. I have been railing about this type of writing that it sabotages the point of a novel to learn, linger, immerse in the thoughts, language , philosophy. Instead what we get is over blown film treatments. I have a number of essays on this subject conversation in the novel. Please check it out and join the salon conversation.
There should be an entire case study on Harry Potter and it's effect on publishing. I was recently querying agents on a fantasy novel, and I was struck by how so much on manuscript wish lists (romantasy, cozy) was in the long shadow of the boy wizard.
I disagree about the idea of the romantic author. Sinykin DOES challenge the romantic notion of the author because he shows that authors are subject to social forces. The forms their books take are not wholly internal, but made by external social pressure and market forces. Essentially he's proven (even though many people have proved this before him) that creation is not the result of individual interiority. Whether the creative product is good or bad doesn't change that.
Does Sinykin explain the existence and successes or perils of small, specialized publishers like Persephone Books in the U.K. who is re-publishing books from the late 19th & early 20th century written by women ( and therefore, obscured)?
This sounds perfect for my to-be-read pile! I just read Live and Death in the American Novel by Leslie Fielder, and need something new along these lines.
"... authors are themselves influenced by what gets published. They’re influenced both in a direct way (they read the books and imbibe them as influences), and in an indirect way, because authors want to be published, so they write towards publication." -- This sums up the entire argument for me. Times change, tastes change, and an author (or any artist) either responds to those changes or doesn't.
This is why, while I'm not a fan of what a lot of people are trying to do with AI, I think the threat to serious literature posed by it is overblown. (It certainly can't write at a literary level, and there's no evidence it ever will.) Algorithms have been in charge for a long time. When decision makers are just copying notes from people who are also copying, you get a system that, as you correctly said, rewards mediocrity and gaming. If serious literature can survive the disaster that publishing has become, then it can survive AI.
I don't know. This AI is definitely a miracle technology. It's wild what it can do. My perspective is that a world where AI can replace literature will be a world that looks very different from the one where we live. In that kind of world, it'll also be able to do A LOT of other things too, and maybe that will create some new opportunities for people. In the meantime, we have to fill our time however we can.
I'm an AI researcher in addition to a writer. I agree that LLMs can do impressive things, but I don't think they're anywhere close to replacing humans in literary fiction. Language modeling itself seems to be topping out, and if there are advances in AI, they'll probably be in neurosymbolics, and there are tons of interesting applications there for, say, industrial design and drug discovery, but literary fiction isn't on the menu. Neurosymbolics works well when there are objective targets.
Reinforcement learning (RL) could be used to guide machines to create convincing imitations of award-winning literary authors, but (a) it wouldn't be as vital as the real thing, and (b) the costs of training it toward such a target would exceed the benefits. Even if AI can write to a literary level (so far unproven) there is no economic incentive to have it do so, because quality of writing isn't what sells books.
I do think AI will play an increasing role in developing and editing commercial fiction. For Marvel movies and the like, it can write entire scripts. For complex psychological novels, it falls flat. For just one example, AI wants to avoid conflict (because it's RL-trained not to be conflictual with us) and this is just one of the issues that must be overcome to use it for long-form fiction. It would be more work to make AI write a coherent novel than to just write the novel. That said, publishers will definitely use it to skimp on editing, and agents are already using it to read the slush pile (and it will probably do a better job than agents do.) It will have effects on the industry, no doubt.
Of course, if the Singularity is real, all bets are off. Humans might not even be relevant. But we have no idea if that's even possible.
Interesting take, but I think the AI is already doing great short-form literature. I've been getting ChatGPT to write poetic captions for some of my Substack images, and I think the results are often outstanding. It feels like creative collaboration, uncanny.
I read this book, and I found that parts of it were excellent and parts of it left me very confused. Thank you for this explanation and analysis; I think you clarified the confusing and controversial parts very well.
Another excellent post from the Woman of Letters Substack, in this case synthetizing Big Fiction's core arguments and then challenging them in a nuanced manner. I agree that highlighting the role of different actors in the publication process relativizes to a degree (in my view, small) the notion of individual authorship. What it does more effectively is help readers understand how the supply side of an industry they patronize works and how what they see on bookshelves is shaped by these constraints. Like with virtually any other industry subjected to close scrutiny, it puts to rest the nonsensical notion that it operates in a marketplace where actors compete on an even playing field. More interesting (to me) is the broader point that our socioeconomic landscape largely determines the type of fiction that gains currency. Even though this is not Dan Sinykin's argument, by extension it also refutes the widespread (but utterly false) idea that "politics" has no place in literature and that its presence somehow taints or cheapens the otherwise pristine realm of fiction. A lot of the people who claim the latter, I suspect, simply don't like the dominant political orientation of the literary industry but don't want to spell out their own convictions to avoid having to defend them (but I'm straying too far from the argument here.)
I think Sinykin's book, Mark McGurl's The Program Era, and Gail Pool's Faint Praise: The Plight of Book Reviewing in America, are three of the best works to understand the American publishing landscape today. Pool's book hasn't aged as well (it was published in 2007) but this is not its fault since the trends she describes have only become stronger and the decline in quality and number of book review outlets is now more acute. I wish more works like these made a splash beyond academia. The only other one I can think of is MFA vs. NYC but I was never a big fan, as I said in this ancient Goodreads review, the only one I've ever written on that platform: https://www.goodreads.com/review/show/878127810
The whole "trying to make the thesis more radical than it actually is" is itself evidence of a different kind of conglomerate authorship: universities and research organizations, as well as the system of peer review and common ideas about what makes research valuable, shape the way that theoretical arguments like this one are shaped and communicated. If there is an intensely formulaic (which isn't necessarily to say bad!) genre of writing that is shaped by institutional demands, it'd be literary scholarship.
Also, I'm sure the point has been made before, but I guess BookTok and (to a numerically much lower degree) Substack are taking on the role of the idiosyncratic editor of old.
I was thinking this too -- it seems like a lot of interesting and worthwhile work is marred by this need to be "radical." Collective authorship is an interesting metaphor, but it seems (based on this and other reviews; I haven't read the book itself) like at some point the author forgot it shouldn't be taken literally.
Am going through my comments now. Yes, I think most people who understand this genre of writing recognize the career exigencies that made Sinykin push for the very provocative language in the beginning of the book. It is manipulative, but it also worked--got the book the attention it deserved.
If anything, there's a certain craft involved in being able to make these claims while segregating them from the part of the book that is actually good. In this case, it was executed about as well as it could be.
I don't know if BookTok is replacing the idiosyncratic editor, so much as it's giving that editor the leverage they need. The more weird, off-beat books that become BookTok trends, the more juice editors will have when they try to publish weird, off-beat books. BookTok seems quite resistant to big publisher marketing hype, so in that way it seems to benefit smaller publishers. And one way to make a splash nowadays is by publishing something that genuinely seems different--not because people are looking for something different, but just because the people who _do_ want something different are more likely to talk about it and try to sell it to other people.
How to become a famous academic in three steps:
(1) Make some outrageous claim: the author is dead; it's the end of history; the medium is the message.
(2) Write a book in which you walk back your outrageous claim: "I just meant we should value the text more than authorial intent;" "I just meant neoliberalism seems very stable;" "I just meant an artist's medium affects what's being communicated."
(3) Spend the rest of your life complaining that people are misreading you because they think you literally meant the outrageous claim.
I would argue that it is good to be able to put an intuition or finding into a big, bold idea, for your own thinking and for your reader's : breaking taboos or formulating whole new concepts forces people to give more attention to the idea, and to open to something new. Even if facts are more conservative, it makes one round them better than by only advancing from the old consensus. Dialectical method, I suppose ?
I think the best thing is to just have a radical hypothesis that you support with compelling evidence, so people in the end agree "Huh, I guess that's true."
But I understand that the timeline of an American academic career often doesn't allow you to develop an idea to completion within a single book.
In this case, there's definitely an element of truth to this collective authorship idea and perhaps some future book can put more work into developing it.
That's close to my idea, actually. I don't think one should defend a bold idea / radical hypothesis if they don't have compelling evidence to support it. But caveats can be found to most radical ideas. I suppose of course that such hypotheses can be developed and made more complex over the course of a body of research, and that it's the right way.
About the collective authorship question, I really would like to refer to Alan Bowness' book that I cited in a comment to another of your posts, but I'm afraid it seems to be terminally out of print in english :( To sum up, it's about the environment, in the modern world, that an artist needs to create something trully innovative, and the different circles of recognition of innovative work.
Honestly, I appreciate the more optimistic take. Something to think about.
God you’re good at this. Totally agree about the strength of the book. Sinykin’s insistence on conglomerate authorship has the energy of a college student trying to shoehorn evidence into a flashy thesis he figured his essay needed.
Thank you!
Wow! This is a thoughtful and sensible discussion around a book that I never heard of. The world of conglomerates is certainly very different from the publishing world James Michener described in his novel "The Novel" (1991). As for the possible conglomeratization of literature, I don't think it is a conscious choice by the successful authors, but more of a Darwinism process of selection. Hopefully there will still be readers looking for something more than the dopamine rush.
I just heard of that Michener book! I kind of want to read it. I read so much Michener growing up, and I am surprised, looking back through bestseller lists, by how popular he was.
I really appreciate this nuanced summary as someone who's interested in the subject matter but is unlikely to read the book.
I think the question of whether great books are produced by lone, romantic geniuses or wider social forces is a false dichotomy. Both things are true. Successful books tend to come out of (and speak to) hidden aspects in the culture that are not being voiced AND the people who produce them tend to be cranky, obsessive perfectionists.
I get a bit tired of how the The Death of The Author gets endlessly restated as if it's a new idea. Of course it's true, but the other thing is also true.
That's true. I mean it's funny because I generally am in favor of this kind of systems-thinking, but I myself am such a cranky, obsessive individualist. I myself am impossible to work with and have a deeply personal vision of what I want to put on the page.
Interesting how art has been subsumed so totally into profit-seeking that “success” is now synonymous with “has broad appeal.” I’m not saying that’s *wrong*, but it certainly does have a homogenizing effect.
It was also enlightening to learn about how shifts in the bookstore economy have influenced publishing. I’ve been puzzling over how it can possibly be true that publishers lose money on most of the books they publish — I mean, *how* can that possibly be true? — and it makes a lot more sense now that I’ve learned that, in the past, it didn’t take as many sales to bring a book into the black due to the mosaic bookstore landscape. Thanks for giving me the info to make sense of that mystery
Yes I was impressed by that as well.
A great post! Well worth rereading actually. These are very important themes you bring up—concerning the literature market and its effects on culture creation (or a lack thereof).
Just a few observations and recommendations:
• Foucault writes in his indispensable essay “What is an Author” that yes, exactly, the entity the author now no longer exists—a caveat: to understand this, one must grasp what he means by the term Author. Obv ppl still write novels and poems, but do they fulfill the functions Thomas Mann or Jane Austen did? The Author is a function!
I link to the essay in my long-form post #2 “The Author is yet to appear” about the privation of authorship in the Middle Ages, not unlike today’s situation.
• Another must-read on this topic is Chapter 4 of Adorno’s Dialectic of Enlightenment, the Culture Industry. He writes in inscrutable fragments, so it must be read in an open-ended contemplative way. Everyone will get something different out of it. The fragments needn’t be read in any particular order. Its a fun read.
Enjoy! I thought I would contribute things from non-anglophone spheres in case you are not familiar.
Thanks!
I think that drawing a direct comparison to film/tv is a mistake on the author's part, right? Even the most collectively-written novel doesn't require anywhere near the same level of coordination as the typical film, just by its nature.
This review does make me think about Harper Lee, though. There's a famous story (I'm paraphrasing from memory here) that she had a wealthy friend who believed in her and decided to fund her life for year, so that she could finally write her novel. At the end of that year, she delivered the manuscript for TO KILL A MOCKINGBIRD. Except, the story isn't entirely true -- the book she wrote that year was GO SET A WATCHMAN. Then, she spent a long time working with her editor, revising and rewriting, until together they created MOCKINGBIRD (and this was a very collaborative process; her editor had a lot to do with the book we all know). The book they produced together is, without a doubt, a work of art, but it's also much more commercially viable and marketable than WATCHMAN. I love it, but when I reread it recently I saw a lot of precursors to trends I dislike in current fiction (unfortunately I cannot elaborate yet, I'm still working through this). WATCHMAN is rough and clearly needed at least one good revision to be ready, but the racial politics and general morality in it are much more interesting and complex than MOCKINGBIRD. MOCKINGBIRD was published in 1960, maybe too early to be affected by these changes, but I wonder what Lee would have produced had she been working under Steinbeck's conditions.
Is that really what happened? I knew nothing about the story behind Mockingbird. That is so fascinating.
More or less, I may not recall all the details exactly right. I had heard the basic version a lot over the years, with her friends funding her writing for a year. But the story behind Lee and her editor is told in "Furious Hours," a very good book about Lee's attempts to write a true crime nonfiction book (the Mockingbird background may not have fit real cohesively into this, but it worked well enough and was interesting in its own right).
Hm. I confess I'm kinda reflexively opposed to deeply collaborative writing like that, but I'm probably just being romantic. Frankly I get a little itchy even to know how dramatically Gordon Lish changed Carver's work. But now you're making me think: how different would our national literature be if we didn't have that pillar of Mockingbird. I don't suppose it's a life-or-death decision but...it's a flat truth that my joy at the book's existence outshines any pretentious hangups I have about "authorial integrity."
I wouldn't want to do it... I mean, for Mockingbird money, I could get over it.
I actually have done collaborative writing, about 11 years or so ago. Some friends and I did a narrative podcast. The process was (I assume) similar to a TV writer's room, so it was very collaborative. It was a lot of fun, and the collaborative process helped us put together what I still think was a solid product. But I don't know that we would have ever created great art that way.
Some people might take better to being edited than others do. I also think there's a magic that can happen with the right editor and the right writer. I've definitely had editors who knew how to push me to be better--part of that was that they usually convinced me that whatever I wanted was what would ultimately stand.
There is another and I would argue more insidious movement in writing and publishing that of the MFA factory writing dedicated to pushing the novel with speed forward. I have been railing about this type of writing that it sabotages the point of a novel to learn, linger, immerse in the thoughts, language , philosophy. Instead what we get is over blown film treatments. I have a number of essays on this subject conversation in the novel. Please check it out and join the salon conversation.
There should be an entire case study on Harry Potter and it's effect on publishing. I was recently querying agents on a fantasy novel, and I was struck by how so much on manuscript wish lists (romantasy, cozy) was in the long shadow of the boy wizard.
I disagree about the idea of the romantic author. Sinykin DOES challenge the romantic notion of the author because he shows that authors are subject to social forces. The forms their books take are not wholly internal, but made by external social pressure and market forces. Essentially he's proven (even though many people have proved this before him) that creation is not the result of individual interiority. Whether the creative product is good or bad doesn't change that.
I don't think he's proven it. He's definitely stated it though!
Does Sinykin explain the existence and successes or perils of small, specialized publishers like Persephone Books in the U.K. who is re-publishing books from the late 19th & early 20th century written by women ( and therefore, obscured)?
🇨🇦
He doesn't cover the UK, but there is a long section (a third of the book, almost) on US nonprofit presses.
As a sidenote, I love Persephone Press. Extremely good press.
This sounds perfect for my to-be-read pile! I just read Live and Death in the American Novel by Leslie Fielder, and need something new along these lines.
Give it a try. I really enjoyed this one.
Oh I'm definitely going to read it.
"... authors are themselves influenced by what gets published. They’re influenced both in a direct way (they read the books and imbibe them as influences), and in an indirect way, because authors want to be published, so they write towards publication." -- This sums up the entire argument for me. Times change, tastes change, and an author (or any artist) either responds to those changes or doesn't.
This is why, while I'm not a fan of what a lot of people are trying to do with AI, I think the threat to serious literature posed by it is overblown. (It certainly can't write at a literary level, and there's no evidence it ever will.) Algorithms have been in charge for a long time. When decision makers are just copying notes from people who are also copying, you get a system that, as you correctly said, rewards mediocrity and gaming. If serious literature can survive the disaster that publishing has become, then it can survive AI.
I don't know. This AI is definitely a miracle technology. It's wild what it can do. My perspective is that a world where AI can replace literature will be a world that looks very different from the one where we live. In that kind of world, it'll also be able to do A LOT of other things too, and maybe that will create some new opportunities for people. In the meantime, we have to fill our time however we can.
I'm an AI researcher in addition to a writer. I agree that LLMs can do impressive things, but I don't think they're anywhere close to replacing humans in literary fiction. Language modeling itself seems to be topping out, and if there are advances in AI, they'll probably be in neurosymbolics, and there are tons of interesting applications there for, say, industrial design and drug discovery, but literary fiction isn't on the menu. Neurosymbolics works well when there are objective targets.
Reinforcement learning (RL) could be used to guide machines to create convincing imitations of award-winning literary authors, but (a) it wouldn't be as vital as the real thing, and (b) the costs of training it toward such a target would exceed the benefits. Even if AI can write to a literary level (so far unproven) there is no economic incentive to have it do so, because quality of writing isn't what sells books.
I do think AI will play an increasing role in developing and editing commercial fiction. For Marvel movies and the like, it can write entire scripts. For complex psychological novels, it falls flat. For just one example, AI wants to avoid conflict (because it's RL-trained not to be conflictual with us) and this is just one of the issues that must be overcome to use it for long-form fiction. It would be more work to make AI write a coherent novel than to just write the novel. That said, publishers will definitely use it to skimp on editing, and agents are already using it to read the slush pile (and it will probably do a better job than agents do.) It will have effects on the industry, no doubt.
Of course, if the Singularity is real, all bets are off. Humans might not even be relevant. But we have no idea if that's even possible.
Interesting take, but I think the AI is already doing great short-form literature. I've been getting ChatGPT to write poetic captions for some of my Substack images, and I think the results are often outstanding. It feels like creative collaboration, uncanny.
I read this book, and I found that parts of it were excellent and parts of it left me very confused. Thank you for this explanation and analysis; I think you clarified the confusing and controversial parts very well.
Another excellent post from the Woman of Letters Substack, in this case synthetizing Big Fiction's core arguments and then challenging them in a nuanced manner. I agree that highlighting the role of different actors in the publication process relativizes to a degree (in my view, small) the notion of individual authorship. What it does more effectively is help readers understand how the supply side of an industry they patronize works and how what they see on bookshelves is shaped by these constraints. Like with virtually any other industry subjected to close scrutiny, it puts to rest the nonsensical notion that it operates in a marketplace where actors compete on an even playing field. More interesting (to me) is the broader point that our socioeconomic landscape largely determines the type of fiction that gains currency. Even though this is not Dan Sinykin's argument, by extension it also refutes the widespread (but utterly false) idea that "politics" has no place in literature and that its presence somehow taints or cheapens the otherwise pristine realm of fiction. A lot of the people who claim the latter, I suspect, simply don't like the dominant political orientation of the literary industry but don't want to spell out their own convictions to avoid having to defend them (but I'm straying too far from the argument here.)
I think Sinykin's book, Mark McGurl's The Program Era, and Gail Pool's Faint Praise: The Plight of Book Reviewing in America, are three of the best works to understand the American publishing landscape today. Pool's book hasn't aged as well (it was published in 2007) but this is not its fault since the trends she describes have only become stronger and the decline in quality and number of book review outlets is now more acute. I wish more works like these made a splash beyond academia. The only other one I can think of is MFA vs. NYC but I was never a big fan, as I said in this ancient Goodreads review, the only one I've ever written on that platform: https://www.goodreads.com/review/show/878127810