98: Erik Hoel – Writing, Neuroscience and Genius

Play episode
Hosted by
Will Jarvis

In this episode, we’re joined by writer and neuroscientist Erik Hoel to talk about where all the geniuses went, the secrets of the publishing industry, and the current state of neuroscience. You can find Erik’s work at: https://erikhoel.substack.com/


Will Jarvis 0:34
Awesome. Well, Eric, how are you doing this evening?

Erik Hoel 0:39
Oh, I’m doing well. I’m doing well. Thank you for having me on.

Will Jarvis 0:42
Eric, thanks so much for taking the time to come on the show. Do you mind giving us a brief bio and some of the big ideas you’re interested in?

Erik Hoel 0:50
Yeah, of course, I’m a research professor at Tufts University. But I wear a couple of different hats. The first is as a scientist, I got my PhD working with Giulio Tononi on trying to craft some of the first kind of well formalized mathematical theories of consciousness. And in my research career, I’m interested in consciousness. And I’m interested in emergence and kind of taking subjects that are traditionally philosophical, and making them a lot more hopefully scientific. I’m also an author, I grew up in my mother’s independent bookstore, so I’ve always kind of wanted to be an author and was surrounded by books. So very, very natural to me. So I published a novel last year, the revelations, through Abrams books, and I also run a subset called the intrinsic perspective, where I sort of combined some of my scientific interests with my literary interests.

Will Jarvis 1:47
Gotcha. Do you see kind of a common thread that ties your work together between your writing this work on consciousness and math and, and biology? Like, is there a common thread there? Or is it more just like these are kind of discrete interests, and they don’t really bleed together very much?

Erik Hoel 2:04
Well, the reason why my subject is called the intrinsic perspective, is that I think that we can very broadly view the world as sort of having a scientific image. And that scientific image is what one might call extrinsic, it is based on mechanism, or causation. It is model ball in terms of sort of the physical laws that we know. And then there’s also the intrinsic perspective on the world, which is sort of what everyone is thinking and feeling hidden within the bones of their skull. And that perspective of the worlds is also the world of the writer, right writing is about communication from within your own little well of solipsism to other people who are stuck in their own little wells and solipsism. And you’re trying to sort of put something together, either communicate an idea, or communicate a what it is likeness, communicate what it is like to be a person who lives you know, in a certain way, or at a certain time. And you might do that, for example, by having a character in a novel that’s like that person. And, and so what I’m interested in, is where the extrinsic meets the intrinsic. So it’s sort of like, it’s like, the shelf of the world, right? It’s like, it’s like, it’s like, it’s a place where there are like, Great Titanic, sometimes terrifying, sometimes beautiful forces at work. And that includes, for example, the search for a scientific theory of consciousness, which would be sort of the extrinsic attempt to understand the intrinsic perspective.

Will Jarvis 3:46
SuperCloset is it something like understanding kind of the intersection between quality qualia and reality? Is something like that? Or is that too kind of simplistic?

Erik Hoel 3:57
No, I mean, I think I think it’s a pretty reasonable description. You know, I everything I do is sort of connected via a constellation of, of a couple of different main themes. So for example, I’ve recently been doing some work on dreams, and the relationship between dreams and fictions. And, of course, that’s that that is also in a sense, right? Like in that case, I’m using the intrinsic perspective, which is sort of the phenomenology of dreams, to try to actually come up with sort of like a theory of of art. I mean, that’s really one of the things I’ve been I’ve been working a great deal on is scientifically grounding our notion of an aesthetic spectrum, in something that’s more valid than like, say Steven Pinker’s description of music as auditory cheescake

Will Jarvis 4:52
I love that. I love that. How did you originally come across these ideas? You know, was there a single moment Be like, you know, this, these are the themes I want to explore? Or is it kind of emergent over time, as you read more, you know, as a kid, you said, you grew up in your parents bookshop, which is, which is awesome. Was it just like reading a lot of literature or getting exposed to a lot of different ideas, and it kind of just coalesced over time?

Erik Hoel 5:17
Well, it’s always very hard to, you know, appropriately track your own motives for things, right. But, but in one sense, I mean, there was sort of like a logical decision that was made at a certain point where I was old enough to know sort of what I was capable of, and when I wasn’t capable of, and so I knew, for example, that I would never be capable of being like a string theorists, just, I just don’t have that sort of, like the really, really high level, technical mathematical ability is something that manifests when you’re young, and I just didn’t really have that. And, and so that, that sort of limits your your options. And looking at science, I thought, because of my background in sort of fiction writing, like I was already writing when I was a young teenager, I thought, well, maybe I should go towards something where being creative is, is going to be an advantage, right? Like, rather than being analytical, can I go to something where originality is going to be much more important, right, then, like sheer quantitative ability. And there are a number of like, really open areas in science, where we just frankly, don’t know very much. And we sort of need good, good theories. And so I’ve sort of made my home in those cases in those places. And, and one could view that, almost like how an organism’s adapts to Nietzsche something like that. Right. But I’ve been very lucky along the way. And certainly, I think that there’s a certain serendipity of like, what I work on, and, you know, my progress on it, to the degree that there is or there isn’t.

Will Jarvis 7:08
I really liked that. I really like that. That framing. I am curious, what areas of science do you think, need more original theories need more originality and creativity at this point?

Erik Hoel 7:20
Yeah, so one is in fundamental physics. So physics is physics is not finished. We need, you know, the most brilliant minds to go into physics, and to improve our understanding of the fundamentals of reality and move us past the Standard Model, which, which will one day I can fail, if it hasn’t already. And so clearly, like some very big picture physics stuff, what I was very interested in was consciousness. Particularly because it felt like no one was working on consciousness. I mean, in the time that I became interested in being a scientist, consciousness research has grown significantly, since then it’s still miniscule, it receives less than 1% of 1% of neuroscience funding, I’m making that number up. If there’s some other number, that’s actually true. But but but that’s, that’s, I think, well, within the order of magnitude of how little funding consciousness research actually gets, and there was almost no one working seriously on it. And so that looked to me, like, like an open opportunity to maybe get in there and do some, some fundamental research, you know, unfortunately, at the same time, it’s also perhaps the hardest problem that humanity has ever encountered, I think, in many ways harder than things like new physics, because at least we sort of know what we expect the answer to be with regards to new physics, we sort of, I can’t, I can’t, you know, tell you what those equations will be. But I sort of know very broadly what those equations might look like, right? Or, or let’s say a young hotshot graduate student going into that field might know that. And they all have intuitions about that. With consciousness. It’s, it’s far less clear, right? So we don’t even know what good theories look like. And so there’s a sense in which we’re very, very sort of far away from from understanding it or having a good a good understanding of it. And there’s a couple of other various like, like obvious ones, like neuroscience as a whole is incredibly underdeveloped, although not for not for lack of effort and funding. I think there are some really interesting questions in evolutionary theory and particularly also now in AI. So there’s still like a, you know, a broad number of places where, where the contours of the map are not well drawn in yet.

Will Jarvis 9:57
There’s a lot there’s a lot of whitespace it seems I’m curious, do you think um so I him he has this idea of having like, you have to have an attack on a problem, or else it’s like really difficult to make progress. And it kind of sounds like what you’re talking about with consciousness research, you know, what was your first kind of attack you found to the problem to try and like figure out like what’s going on and try and start trying to form a theory, if that makes sense.

Erik Hoel 10:27
Yeah, so at the time, and I’m thinking back now, when I became sure that I wanted to go to go into consciousness research, and basically to have that be the problem that I spent my my PhD working on. That would be have been around 2008 2009. At the time in America, there was no one working on consciousness in a way that I consider to be serious, except one man whose name is Giulio Tononi, who is probably the first person to propose a well formalized theory of consciousness, that something that certainly is a good candidate for what a theory of consciousness might look like. Right? So it’s really if you think about it as just does this thing even sort of look correct at all? Right? that the answer to that would it is maybe yes. And that seems like an awfully hedge statement. But if you compare it to pretty much everything else, it’s just worlds and leads above it. So I went to go work with with Julio to try to help him develop it as it’s called. His his theory of consciousness based on how information is integrated in complex systems. And that sort of took me down a pretty big, pretty big rabbit hole. But that was my initial interest in reason was just that. Having looked around at sort of the contenders, I saw a lot of things that were basically just metaphors, things like maybe consciousness is a big global workspace. Okay, maybe consciousness is fame in the brain. It’s like, okay, well, let’s, let’s, let’s put some math on that like, like, What do you mean by fame? Right? You mean? The degree of information sharing what what’s in and what’s out? To what degree is something in what degree is something out? How do you get a stream? How do you solve the binding problem? Like, how do you do any of these things? And almost none of these theories had any teeth, as it were, except it it I think, has has some teeth.

Will Jarvis 12:33
I like that. I like that. And I’m curious just for you know, we have a lot of young scientists that Listen, what was your Did you just straight up cold email, you, you look for someone who had a robust theory, you cold emailed them and said, you know, Hey, can I come help? Like, can I come mop the floors? What What was that process? Like? Just mechanically?

Erik Hoel 12:53
Yeah, I believe I actually contacted Julia, after I was accepted to so as I applied to the University of Wisconsin, so he’s, he’s at the University of Wisconsin in Madison. Nice. And it’s a very big research school. But you know, it’s not it’s not Harvard. Right, like, so it’s not, it’s not a sort of place where the Graduate School acceptance rate is 3%. It’s probably, you know, 10%, you know, 12%, something like that. So, if he had been at, you know, Oxford, or MIT or these other places where they’re, they’re basically just rolling dice to see who gets in with like, a 3% chance, right? There’s basically there’s nothing you can sort of do priority that will actually ensure ensure your chances and that, but So luckily, he was he was just at a place that that I was able to get into. I think I only applied to two graduate schools, which is insane. And any, any of like, my career in science has been has been by outside metrics, completely crazy. And I would never recommend almost any of the steps that I took at any point in time. But things like I only applied to graduate schools. One was New York University, and that was to work with with half one lout. We know how come I was at Columbia, though, this was why there was no one at NYU who was doing anything serious with consciousness. But the philosophy department was very good. So they had David Chalmers and some other people. So I figured, okay, I wouldn’t be able to work with any sort of great scientists, but maybe I would be able to sort of still be where a lot of the interesting thought around consciousness was. And that was sort of bleed into my science. And then the other was the University of Wisconsin. And, and I believe I basically just called emailed Julio, after after getting accepted for the initial interview.

Will Jarvis 14:46
Nice. Excellent, excellent. I like that. I want to talk a little bit about your research now. What’s your sense of the biological function of dreams? Like why do we dream like what’s the process for?

Erik Hoel 14:58
Yes, this is a theory that I advanced are sort of put together one might say, but a couple years ago, and it was originally actually born. So so so just to give some people background right, one of the things that Giulio Tononi is very famous for is also his model of why it is that we sleep, which is something called synaptic homeostasis. And the idea is that learning in general, engenders a net potentiation of your synapses, the synaptic strength, this is basically getting too strong as you’re awake, and you need sleep to sort of downscale all the synapses. Actually don’t think that that hypothesis is really going to end up being true. I think it’s a very nice hypothesis. I think it’s like an elegant hypothesis. But there is a sense in which, you know, there, if you’re interested in consciousness, you have to sort of be interested in, in what what happens when consciousness goes away, and why does it? Why does it go away? When it does, and therefore, you know, they’re very good sleep labs. So I have a background in, in sleep research. And I had been writing an essay about the purpose of, of fictions, actually, and why it is that we humans, so love fictional stories. And because this is again, something that’s interested me since since I was young, and I kind of realized that you could you make the analogy that fictions are like artificial dreams. And I don’t think that that analogy is, is that original, but I took it very seriously. So I said, Okay, well, what if we take it incredibly seriously. And therefore we look in therefore, we have two mysteries that look very similar, one of which is why do humans care about what happens at Hogwarts, right and spend a bunch of time talking about all these imaginary lies? And why does the brain make up a bunch of random imaginary hallucinatory BS? Every night, these seams actually possibly quite related to me. And so, so I started developing this theory. And actually, what fell out of it was, basically, I put all the fiction aside to just focus on the theory of dreams that fell out of it, which was this notion that the goal of dreams, the evolved purpose of dreams is to help prevent overfitting, where overfitting is a very common, basically ubiquitous problem that crops up whenever you try to train a complex system that learns like an artificial neural network. And in general, it will often sort of memorize the details too well, and therefore it sort of won’t generalize to new training data. And my thought was, this seems sort of inescapable. And also that mammalian brains can’t ever turn off learning. I mean, we people have really not reckon reckoned with this. But when you’re training artificial neural network, you can sort of stop it at a certain point and say, Okay, I’m done now, right? You’ve been appropriately trained. That never happened, mammals, right? So we’re constantly learning. And this this, the years of plasticity in your brain does not shut off your learning as much when you watch a TV show, as you are when you’re studying for a math class, right, like so the plasticity in your brain never never shuts off. And so there seems like there needs to be some sort of counteracting force. And the proposal is that dreams are basically a combination of techniques that that machine learning theories have already basically found out about. And the brain also stumbled across sort of an intersection of those techniques, and begins to do nightly dreaming in a way that sort of is given your brain augmented data in a way that assists and promotes generalization. And to me, it’s sort of the first theory of dreams that actually gives a good reason why dreams would be so sort of weird and lynchin and dreamlike.

Will Jarvis 19:02
Makes that make a lot of sense. I’m curious, you know, with all your work on consciousness, and you mentioned a lot of Mel terms like overfitting and things like this. Do you think it will be possible for us to make machines conscious? Or is that like, it’s a different thing? It’s different class? That doesn’t make sense.

Erik Hoel 19:21
I, I very much wish I knew the answer to that. Yeah,

Will Jarvis 19:26
those small questions.

Erik Hoel 19:28
Yeah, I mean, so the thing is, is that the way that most people think about consciousness is that they basically pick some sort of pet theory, their pet theory is like, totally wrong. And then they sort of just harp on this pet theory, like for the rest of their lives. This is the normal intellectual cycle of people who talk about consciousness. They just pick something like, maybe consciousness doesn’t exist. I’m going to make up you know, a bunch of reasons why that crazy hypothesis is somehow true, or, or maybe maybe it’s a race just to computer what’s so mysterious, right? And then I’m gonna argue for that. matter, whereas just a big workspace, you know, where is the mystery there and scar you for that, you know, or you have like, there’s a hard problem, it’s never going to be solved. But at least those people are sort of honest, I think about how difficult the problem is. This is actually why despite having a lot of training on consciousness, I’ve really focused on other problems that I found to be like solvable. Like, I think that like, you can have a theory of dreams and that like that, that I’ve advanced, like a pretty, like, I’d like a like a good one like it deserves to be considered. And and that’s because it’s a solvable problem, right? Like, it’s just nowhere near as difficult to come up with, with like, an original hypothesis around dreaming as it is to come up with an actual original hypothesis around consciousness. So with consciousness, I generally wait until I have something really important to say. So I’ve only published like, I don’t know, three or four papers entirely on consciousness at all. And but that leaves me in that epistemic position wherein when you asked me like, are machines going to be conscious? If I was sort of the standard? You know, the sort of sort of standard intellectual talks about consciousness, I would have some sort of ready made answer for you about yes or no, and why, you know, my reasoning as to why but the simple answer is, we have, we have absolutely no idea. And anyone who tells you that they have an idea is absolutely wrong. But I can tell you, yeah, what, but I think we can say some things. And one of those things is that is to judge the similarity or dissimilarity of the artificial neural networks that we have against the brain, and whether in particularly the properties of the brain that are leading theories of consciousness pick out. So as an example, right, one could ask, do these? Do these machines have anything that looks like a global workspace? And the answer, I think, is pretty reasonably No, unless you really stretch the definition of what we mean by global workspace, they generally don’t have that, are they integrated? Well, integrated information theory would say, No, they’re not integrated. So what I sort of prefer to do to answer this question is to say, Listen, almost all the leading theories we have are wrong, but they’re probably somewhat sensitive to the true underlying theory along some axes. And all of them basically look at these feed forward artificial neural networks that don’t have any sort of intrinsic activity, or spontaneous activity. And that, you know, our brain like in the very broad sense of having maybe things like tuning curves, or sort of neural like properties, but they certainly don’t have the structure of the mammalian brain, they don’t have its architecture. They don’t have anything like that. As I said, they don’t even have the inbuilt plasticity, right, you’re trained basically from the outside and manipulated by these outside algorithms, and basically just use Excel spreadsheets. And we can sort of say, you know, the likelihood that they’re conscious is actually probably quite, quite well. So that just described our own situation. It doesn’t really describe the sort of hypothetical that originally asked, but but I am somewhat confident in saying that probably the current techniques, we should not consider as conscious, we shouldn’t sort of take their statements as being based in intentionality, or eyeness. In the way that a human being statements are right when GPT three says, Help, I mean, tortured. You can just say, no, no, you’re not you’re just auto completing the text that I gave you, which was, is there anything wrong? GPT? Three, that would shock me. And then they say, Oh, yes, I’m being tortured.

Will Jarvis 23:41
I got it. I love that. Well, Eric, I really liked that answer, because I think that’s the best answer any human on the planet could give me on that, that that particular question. Yeah, it, I want to take a left hand turn here and talk about aristocratic tutoring. As something you’ve written written a little bit about, first of all, is it your sense that we have less outsides contributors like Beethoven Shakespeare’s etc, than we did? Like in the recent past?

Erik Hoel 24:13
Yes, and just to explain to that, and sort of why you’re, we’re jumping around a bit. So you know, one of the things that I consider to be my job is to sort of, as I said, I don’t I’m not in the top, you know, quintile or whatever, have like analytic ability, but I think maybe in terms of sort of, like originality, or like just just coming up with things. I would probably coordinator at the top I that would be the one cognitive trait that I’m relatively sure of in myself. So what I do is I sort of look around at fields and I look for low hanging fruit that really hasn’t been picked. And one of which was that in education research, and notably, I didn’t write an academic monograph about this because I didn’t think it would get any attention if I did, and I And, and I didn’t want to stumble through, you know the entire academic literature, just to make absolutely sure that no one said anything like this at all. But I do think that it’s it’s relatively original and that certainly I had not heard it before. And so I advanced an idea on my sub stack, which I call it the aristocratic tutoring hypothesis. And that took off recently and sort of went viral, just to give the listeners sort of a background, why we’re jumping into NATO, right. And part of the article, or at least the sort of the opening of it was this argument that if you look at genius, I mean, real world spanning genius, geniuses that everyone would agree upon, particularly people who’ve made a serious, one might say, academic contribution, but who also sort of were publicly relevant, that this sort of genius has, has declined. And it’s also declining along with polymath, the like, even if you look at somebody like Lewis Carroll, who I forget Lewis Carroll’s exact dates, right, but he’s still relatively recent. He’s a mathematician, a children’s book author, right? He’s a logician. He’s doing all sorts of really interesting things. It’s really hard to name a contemporary figure who’s even as sort of Polymathic as what, Lewis Carroll. Right. And so I think, in general, there has been this decline, and then that’s something that is sort of worth exploring, you know, as to why.

Will Jarvis 26:36
Yeah, I want to talk ask about the clients that in here, it seems particularly weird, and the concept of like, the Flynn effect, people’s abstraction ability seems to be getting better over time. But yet we seem to be having less outsized talents. I mean, you are kind of a good example of someone who has been able to at least have interesting insights and a lot of different fields. Do you have any thoughts there? Like, what’s going wrong?

Erik Hoel 27:01
Well, so one thing that I advanced, I mean, I think it’s relatively obvious that there’s there’s something about the public education system, right, that does not do a very good job of making people interested, deeply interested in intellectual subjects from a young age. It the the, you know, one could rant about sort of the public education system, I think, for a very long time. And I would certainly include private education in there just the same, like the structures are almost no difference, right? slightly nicer, slightly better peers, slightly better teachers, but the structure of the entire thing is exactly the same. If you’re at Andover, or if you’re at, you know, big, big inner city high school there, there’s still basically the same structure. So looking historically, though, we see that the the education system has clicked on only very recently. And so the question is, well, what did what did people really do before then? And the answer is that, well, if you were rich, what you generally did was that you hire tutors, and governesses and if you were an aristocrat, especially, and these tutors and governesses would rear the child from a very young age. And a great example of this would be somebody like Bertrand Russell, who was one of the great geniuses of the 20th century, extraordinary, extraordinarily Polymathic he won the Nobel Prize in Literature and also came up with, you know, probably the first, the first presage of girdles Incompleteness Theorem in his paradox around set theory. So this is a very, very smart, intelligent individual, of anyone who’s ever lived, you know, deserves the term genius. And, you know, his education was basically a revolving door of governesses and tutors. And these were often people who were experts in their field. I mean, he was taught by like, Lord Kelvins, graduates to student basically, and like, you know, it’s like Kelvin like that Kelvin. Right? And, yeah, like the Kelvin, right. So, so this is, this is like, if we had, you know, young children being tutored by, you know, excitable charismatic in their 20s. You know, young scientists, like clearly that would have a very strong, very strong effect. But the problem is, is that this this method of, of education is not at all scalable. So, you know, I made clear in the article that I’m not making a moral claim that this was somehow really good, or we should go back to that and aristocrats should sort of separate their their children out. But it is certainly true that I think that that particularly at the very upper echelons of society, there was a huge amount of one on one tutoring going on. And we know that tutoring is is basically the most effective form of education. And that’s been shown in study after study and So we can just look at the decline of tutoring and say, you know, maybe, maybe the simple fact is, is that many of these great geniuses had, you know, what we would now consider almost completely unfair advantages in terms of their education early on. And this is why, for example, the history of intellectual life is often has often been confined to the rich, I mean, you know, we shouldn’t, we shouldn’t mistake that the number of like peasants who made serious intellectual contributions, up until you the industrial era was was so low, I mean, astonishingly low you, like, pick, pick any intellectual from like, 1500 to 1600? And see how many of them were really like, lower class? And the answer is almost none. They’re almost all aristocrats or upper class, and men, and as far as I can tell, many of them had tutors and went through sort of this, this this personalized form of education. And I think that it’s it’s basically the one way we know of to create geniuses, it doesn’t always create geniuses, I certainly would never make that claim. And I wouldn’t make the claim that any child, you know, could sort of be made into a world towering genius. I think it involves a lot of sort of innate talents, although maybe maybe IQ less so than some people would like to believe. I actually think that there’s all sorts of sort of cognitive talents that people can have that aren’t captured by something like IQ. But the but the fundamental fact remains that, you know, all these people are at least seems like this method of education was incredibly effective, and probably maybe one of the few surefire ways of creating geniuses and we even had people who did that people who set out to make their job titles a genius, and in many cases, they ended up succeeding. And there are historical examples of this. John Stuart Mill being probably one of the most famous.

Will Jarvis 31:55
I’m curious, how much of it do you think of how much value do you think was added to be jeans is it there’s two veins, I’m not saying it’s real, but there’s two veins, there’s this idea that you’re not being in curiosity to people, it seems to be like really important. Um, so you’re willing to go out and search because it’s kind of fun. And it’s not just like this rote memorization or whatever else is gone. And then there’s like a mastery learning like Blink two sigma, it’s just much more effective. And you get a lot more information in people’s heads had to wait to do which one do you think is more important? Or are they both just kind of important? And it’s, it’s really difficult to parse out?

Erik Hoel 32:31
Yeah, I actually don’t, I think that there, it’s a bit of a mistake to think that the mastery of the material is what’s making the genius. It could be read that way. Because when you look at the studies of tutoring, you know, they’re limited to what you can measure, like the measurables of education. So of course, they show this big boost in, for example, the amount of crystallized knowledge that a student has, like their memory of things, or something like that. But we all know that that’s only one dimension of what a teacher should impart an example being that you know, what you want in a, in a, someone who’s learning history and what you want in, let’s say, a young historian who was tutoring your child would be to impart for them, like an interest in, for example, ancient Rome, not just memorizing the dates of, you know, all the Emperors of ancient Rome, right, you want to stimulate this sort of intellectual interest. And hopefully, you sort of stimulated enough that it runs their entire life. And maybe when they’re, you know, 75 they crack open a new, you know, book on Marcus Aurelius or something. And who was also had, like 17 tutors growing up. And, you know, that that sort of sparked like, what I would argue would be that probably the the effects of tutoring, extend to all the things that are sort of non measurables in education. So like, not only is one on one tutoring much more effective at, like getting the information into someone’s skull, but clearly all the the extra effects of education, I would argue that it’s probably equally if not more, so effective at those things, like getting the kid excited about the subject, or so on. So that’s my suspicion. But of course, you can’t, you can’t really like measure and prove these things very well. Right,

Will Jarvis 34:26
right. Right. Very, very interesting. I’m curious, I want to talk a little bit about your writing now. That’s okay. You know, have you thought about trying to improve writing lessons over time? Do you have any tips on how to improve one’s writing? Or is it just something where you just, it’s just reps you just put in a lot of work? You talk about interesting things, and that’s how you get better.

Erik Hoel 34:52
Well, I would say there’s a there is a difference between writing fiction and nonfiction. fiction is a bit more perfectible. It’s a bit more like teachable, you can teach someone how to write nonfiction. And you can kind of just copy, copy the greats or the people that you like. You, what you’ll find if you if you write fiction is that if you if you try to just copy the greats or people that you like, you’ll write, like serviceable prose, that it won’t be very good. It won’t be, it’ll be, it’ll just be like soulless, right? It’ll just be soulless. And like fiction is so sensitive to solve, right? Like, is there is there solidness right, like, so when I wrote The revelations, which, you know, to write a novel nowadays, is just like, one of the dumbest things you can do, right? It’s just such a, the amount of time that goes into like, crafting a really good novel is like years and years of work, like you cannot skimp on that. And then sort of, you know, the whole publishing process is incredibly difficult. And getting it out into the world is, you know, insanely difficult. And then it’s just one of like a sea of books that go out at the same time. And unless you’re sort of like a literary industry darling, which you sort of have to have gone to certain schools and sort of write in certain ways or be interested in certain in certain things in order to do that, unless you’re that you really won’t get a huge reaction from it. I still do get some amazing emails and letters and like little little notes that it’s still out in the world, and people find it and they read it, and they say, like, I’ve never read anything like this is absolutely crazy. But to me, I it had to be absolutely necessary. So with fiction, you know, it must be completely necessary, it must be that you just simply had to write this, nothing could have prevented you from writing it. And in that sense, I think that that’s very important, artistically, nonfiction can be forced. Of course, you don’t want to be constantly forcing it, but you can force nonfiction, what I do now is write mostly is write mostly essays on my substack, the intrinsic perspective. And I’ve had some some great reactions to them. And it’s sort of a community has grown up around them, which I really appreciate. But I think a big a big part of that is that the the essay is sort of wide ranging enough that I can use some of the techniques from from fiction. So you know, an attention to language, that sort of thing. And that is just not done as much, particularly in sort of like the world of sub stacks. So I would say, sorry, but just to briefly answer your question about like, you know, writing advice. One, one big thing that I tell everyone is one, you have to have a very definite point of view. So most of the stuff that reads poorly, is because you can’t find the author anywhere, they’re not authoritative. They’re not like making a pronouncement from the mountaintops. This is me, I’m very grounded in myself, here’s my opinion about this, that will lead to good nonfiction writing, if you don’t know who you are. And if you can’t express and by that, I mean, within prose, right? I mean, I don’t mean just throw anyone into an existential crisis. But like, if you don’t know who you are, when you’re writing, that is probably the number one leading cause of of bad prose, it leads people to write like their GPT three, right? It’s like a view from nowhere, right? It’s sort of just like bland nonsense, where you’re just sort of copying what other people have done. And this is genuinely the biggest thing that I find I’ve ever run into. And then I also find that they tried to create too much content, you don’t need that much content, you don’t need to be publishing every day, particularly if you’re doing something like a newsletter. Instead, really focus on putting out really just the occasional really great piece. That’s what people want to read people. There are very few people who you actually want to hear from every day and can give you something original and interesting. Every day. Very, very few people, maybe not.

Will Jarvis 39:15
Exactly, exactly. So so just have a stronger quality bar for publishing is one thing.

Erik Hoel 39:20
Yeah. Yeah, you should sit on stuff for a little while to make sure that it’s, it’s something that that that you would be really proud to show people in written that you’re really excited about. And that’s very difficult because then you need to come up with 100 of those. And you know, if you want a week, right, that’s your two years of running a blog or something like that, then you need to come up with 100 of those. So it’s very, you know, it’s it’s very difficult to do but I’ve been very lucky and and had a really wonderful reaction to sort of my move to writing writing online on this on this substack and I have enjoyed it though. orally. And I also love that it’s, it’s it is an even playing field. I mean, I it’s it’s credibly steep one, but it is an even playing field like most of the ones most of the blogs that I follow that are that do well and put out consistent good content do eventually get subscribers and readers, and people who are interested in a community grows up around them, like most like I’ve almost never seen a blog that’s been running for a while where I look at like how well it’s doing. And I think, oh, it’s it’s totally not where it should be. It’s like, if it runs for long enough, it almost like always fine, sort of where it where it should be

Will Jarvis 40:42
fairly efficient, fairly efficient over time. That’s cool. That’s cool. Can you talk about the publishing industry a little bit and kind of the process of publishing your book and, and kind of what you learned along the way?

Erik Hoel 40:54
I actually just had an essay out on this called Secrets of the publishing industry. And, you know, the publishing a book is an incredibly convoluted and long process. I mean, over years, I think, you know, the book had been finished for maybe four or five years, by the time it came out, although there were some last minute edits, but you’re basically finished. So this is a glacial process. And it’s also a very difficult process, most of the time, people will not get published. I don’t think that there’s a huge significant difference between the manuscripts that get published and those that don’t. I really haven’t observed that. I think probably, there’s, there’s probably more, there’s probably some really great unpublished manuscripts hanging out in the US, I’ll tell you that from from, from what I’ve seen in the process. And it’s very convoluted and you have to do a lot of things yourself from getting like blurbs. Like you know, you your publisher probably is not going to go out and get you a bunch of blurbs from like famous authors, you have to do that, which is crazy. It’s like, oh, how do you how do you find an approach like famous authors, right? So it’s like, for example, like I went to, I went to like Peter, very early on, I went to Peter watts. Peter Watts is one of my favorite sci fi authors. He’s the author of BLINDSIGHT, he won the he won a Hugo award, I think he’s, you know, he’s an excellent writer, given I had briefly corresponded, I basically begged him, he was incredibly kind enough to read the book, and offer a blurb for me, and then okay, now I have a blurb from the Hugo Award winner. Now can I go and talk to somebody talk to somebody else, right, and say, Okay, here’s this blurb, right. And that is at the very beginning, but that shows you like the, the degree of time and intensity of just trying to get the book out there. And then, you know, when the book did came out, I didn’t, I didn’t make a huge amount of money. I didn’t sell it for a huge amount of money. But the publisher did a pretty good print run. So they were sort of like, this was like a, you know, not maybe not their lead book going out. But it was pretty high up there in terms of their their concerns, right. And as an example, this, like Barnes and Noble bought three for every store in America. So I thought, Okay, this is gonna get reviewed, right like this, this will be reviewed. So someone has to notice this, right? I just assumed that reviews happened, right? And that turns out to not be true at all. So people don’t just review a book just because it’s like nationally published and you can walk into Barnes and Noble and find it. Right. So instead, what happens is that there’s all these publicists and the publicist, basically, you know, interact with the book reviewers. And this has to take place years before like, like, like, seriously. So a book reviewer at like a major outlet will not review like a book that they just learned about six months ago, that they have to have known about this book for like 18 months minimum, and asked to have received a copy and like, everything’s planned and like everything. So I didn’t get bad reviews, I actually got a couple, a handful, a small handful of really good reviews. But I just got no reviews, because I just had no idea that the way that this process works is genuinely just through these in house publicists. And you need a really big book deal to have a really good publicist and you need or or a lot more commonly, people will just hire a private publicist and pay a lot of money for it. And that private publicist will will then promote their book, but again, you have to do that yours out, right. You can’t you can do it last minute. And it’s surprising who uses private, you know, private publicists, like you might think you know, Thomas Piketty, you know, famous sort of anti capitalist author doesn’t use a private publicist, but you would be wrong.

Will Jarvis 44:51
You really need someone plugging it and doing it a long time before the book comes out. Have even a shot.

Erik Hoel 44:57
Yes. So you know, my my how Have you witnessed like the total insanity of this pipeline? Yeah. I actually am writing another nonfiction book right now that’s already sold. So I’ve already been through the process a whole second time. So I can’t say like, I’ll never do it again. Like, I literally did do it again. But there’s a very good reason to think that I’ll just be honest, I think substack is going to eat these people’s lunches. Like, these people have no idea what’s going to come. And people have finally sort of cracked the code. People are reading more and more online. And this is where literary life is going to be. It’s going to be online.

Will Jarvis 45:44
could all be online. I’m curious, in this process, have you encountered kind of the MFA mafia? I have this idea that MFA programs

Erik Hoel 45:54
never heard that? I’ve never heard that. But that’s fine. You know,

Will Jarvis 45:59
I wasn’t I was an English major in college. And I always had this sneaking suspicion that, um, you know, it just seemed. George Bernard Shaw has this line, like, you know, it’s just something about, you know, professions, being a conspiracy against the laity. And I just find MFA is it seems like, it always seemed to me like if you wanted to write the great American novel, you go lock yourself in the woods and a cabin and, and start typing, you know, and you do that for a year. You don’t go and get an MFA and learn it and get this like credential. It just seems so antithetical. I don’t know, is that completely off base? And, you know, how has your experience been with the MFA mafia?

Erik Hoel 46:40
You know, now that it’s basically a requirement to get an MFA, there are some serious downsides to having this be the main the mainstream of literary culture. So the first is that, that academisation of literature itself. So the simple fact is, is that these are all people who are coming through the academy. Right? So they are academics, they’re not even really writers, right there. They’re more academics. It’s that closer to their fundamental self image, right? Most, the number of writers who actually make a living from writing novels, is so vanishingly small. Most people are either professors of writing, or they have support from their spouses, and they don’t actually make like a full, you know, they’re not like supporting a huge family from from writing write the number, there are some people who do that, but it’s, it’s vanishingly small, even most of the sort of authors that you’ll find at your local bookstore just aren’t at that level of supporting the writing. So that means that what are they? Well, their professors, they’re academics. And that is a very dangerous thing for for an, for an artistic discipline to be fully fully academic. It leads to solid CISM, it leads to sort of the most successful people being the exact same sort of people who can jump through all the standard academic hoops and score high on all the tests that you need to to get there, right. I mean, publishing a book is much more now like the end step to a process that starts when the person is in third grade. Right? It’s like that’s that that’s publishing a book is sort of it’s like a bunch of hoops from third grade on and then, you know, eventually you jump through enough of them, you sort of get to publish a book, because you’ve got into the MFA program at Columbia University. And you’ve gotten into that, because you did well, as an undergraduate, and you got your undergraduate you went to, you went to Harvard, and you got into Harvard, because you jumped through all the hoops in high school. Right? So it’s sort of like this, this, this, it, you’ve attached literature, to the end tube of this, you know, massive pipeline. And the idea that that doesn’t have any effects, is I think ridiculous. Now does it completely, I wrote actually an essay precisely about this. You’re wondering about the effects of the MFA literature. And, you know, one thing people said was, like, you know, it is absurd to think that the MFA system by itself is sort of completely determinant of contemporary literature. And, and I agree with that, like, contemporary literature is not solely determined by the fact that it has an MFA. But I think the trends in it are, broadly speaking, determined by this and one of the one of the big sort of future pitch that I see is that in through this academisation there’s a slow bleed of public interest. And eventually, what will what happened to poetry will happen to to literary fiction. So essentially, no one will read novels anymore. And no one reads poetry like it just again, you could say, well, poetry survives within the academy, but it’s like it’s a zombie, right? It hasn’t No cultural impact or significance, right? Other than in the rarest of events, right? So it’s completely sort of dead as a discipline. And, and that’s sad, right? You can say that that’s quite sad. But my fear is that the same thing will happen to literary fiction where it’s just, it’s just becomes more and more niche written by more and more of the same people, until eventually, it’s just academics talking to each other. And the whole thing is just a game that, you know, just like now, most poetry that’s put out our books by professors of poetry, that’s the only people who will really seriously publish poetry books. Right, right. Or at least that’s the that’s the majority of it. And, you know, I think that something like that might eventually happens, literary fiction, is sort of continues down this path.

And this is I don’t think it’s super controversial, but strangely, I think, because anything in the literary industry is just, it’s one of the most, you know, like, terrible toxic communities. Online, I don’t mean the entire literary industry, but like, like, like, where you go, like, like literary Twitter is a really terrible, toxic place. And so any little sort of criticism of it is like, you know, you’re criticizing an absolute insane person, right? So any small thing that you say, will be blown up, you know, immensely, or so on. But that just means like, basically, I just don’t really, I don’t really have anything to do with Windows people. And I don’t think that I need to, in order to make by living in way, as a writer, I don’t think I need to really interact with them very much.

Will Jarvis 51:43
Definitely. Well, it also seems like kind of like you said, like, all these hoops you put together and kept to have a master’s degree really limits the perspectives you get in literature, I should say,

Erik Hoel 51:53
there are some contemporary writers that like I really like and enjoy, right, like, but they’re not really who I’m talking about when I talk about literary Twitter, right? I’m not really talking about necessarily the authors even. But, but But yeah, I just want to throw that out. Absolutely.

Will Jarvis 52:12
Not all bad. But yeah, you know, the overall structure can can lead to some negative effects. Are you done for a quick round of overrated or underrated? Sure. Awesome. Awesome. So I’ll throw out a term just give me a sense I’d say overrated or underrated. So the first one is integrated information theory overrated or underrated.

Erik Hoel 52:33
Probably underrated. Underrated.

Will Jarvis 52:38
The free energy principle, Carl, for instance, theory. overrated or underrated?

Erik Hoel 52:43
Overrated? I mean, overrated. Massively overrated. I think no one has given an adequate solution to the Dark Room problem, and there’s no demonstration that the brain currently works by minimizing free energy, just just no actual evidence for that. But yeah, I think it’s extremely freewheeling and that makes me extremely suspicious like when everything is connected and when someone like Karl Friston is the mid author on like, a million different papers and I’m not even sure he read all of you know, it just makes me a bit suspicious of something like that.

Will Jarvis 53:29
It’s good. One last one bookshops overrated or underrated.

Erik Hoel 53:34
Oh, underrated. I love bookshops. booksellers. Like, you know, what’s a town without a bookshop right to go into. I think book shops are some of the most wonderful places on earth. And the people who run them are some of the best people on Earth. And they really are providing like an incredible sort of cultural service to the rest of us. Because it’s not a business where you make a huge amount of money, right? It’s something you get into because you really love it. You can make money but like, you know, you’re not it’s not you’re not. You’re not selling hotcakes. Exactly. Exactly.

Will Jarvis 54:12
Well, Eric, thank you so much for taking the time to come on the show. Where can people find you? Where should we send them?

Erik Hoel 54:21
Yeah, if you if you go to Eric co LL, er ik H O el.substack.com. That’s the best way to keep track of me. Also, you can just type in the intrinsic perspective, or just my name into Google and substack will crop up but but please, if you if you can sign up for it, because it really is the best way to keep track of me and I really do. I really do have a lot of really interesting things planned for it.

Will Jarvis 54:50
Highly recommended. Well, I’ll put the links down in the show notes. Thanks again for coming on.

Erik Hoel 54:55
Thank you so much.

Join the discussion

More from this show