Numb at the Lodge is two years old this week, and after two long years it’s finally achieved its purpose. In the last twelve months I have written 145,709 words across thirty pieces. I have written linguistic apocalypses in which the divine tongue seeps across the southern border of the United States, and world-histories of hatred; I have revealed the terrifying secret of Santa Claus. Now, the time I invested in insulting my audience and inventing minor sixteenth-century theologians has paid off. That’s right: I have a book review in the New York Times. Obviously at this point the whole experiment can be wrapped up. Time for me to cash in whatever outsider cred I’ve built up with this thing, and spend the rest of my life drinking very small cocktails at very boring parties and championing various obviously shitty novels in exchange for social favours. Instead of writing long, leaden screeds about stone-age tribesmen hacking into each other’s skulls, I’ll scatter short, sparkling columns about how this time you really do need to vote for the lesser of two evils. It was nice, getting to write whatever I wanted, building my own audience, experimenting with form—but none of that compares to the dead allure of the mainstream press. The Grey Lady, with her skin like mottled ashes, mashing empty gums in her termite-eaten throne: she calls to me, and I can only obey. Goodbye.
I’m kidding, obviously. But less than you might think.
There’s a large and established tradition of Substack writers gushing over this platform. People love to write end-of-year posts all about how good it’s been to them, how rote the rest of the media has become, how all the interesting writing is happening in newsletters, how great it feels when your only boss is your audience and yourself. If the Times came calling, they might not even answer: that’s how much they love writing on Substack. Most of the people saying this sort of thing tend to be fairly successful on the platform; you hear less of it from the vast majority of writers here, down on the long, long tail of the power law distribution, who pump out thousands of words every week to an audience of a few slightly embarrassed friends. (Actually, maybe they are saying the same thing. If they were, it’s not like I would notice.) But I have a lot of good reasons to be writing the same sort of piece. Numb at the Lodge has been absurdly, bafflingly successful, much more successful than I ever expected. This time last year, I wrote that I wanted to have 12,000 subscribers by the end of my first year, and I pretended to be extremely upset to have failed. ‘More accurately, you failed me. I kept telling you miserable cowards to like, share, and subscribe—but you didn’t do it, not enough of you, and now this is the result.’ This time, I set myself the similarly ridiculous goal of 25,000 subscribers by the time I reached the two-year mark, and prepared a whole bitter, spiteful rant for when it failed to transpire—and instead, it actually happened.
In the grand scheme of things, 25,000 is not very many people. My subscribes are outnumbered four to one by the military of Guatemala. There are significantly more people involved in the manufacture and sale of Colgate toothpaste than there are of you. Still, you’re numerous enough that Substack now makes up the majority of my income, and you’re numerous enough to put me at number 36 on Substack’s culture bestseller charts, which I obviously check religiously. This is insane. For much of the past year, I’ve been repeatedly switching positions with the journalist David Aaronovitch, currently at number 35. Aaronovitch is a serious guy: he’s a fixture on the BBC; he’s won the Orwell Prize and written a medium-sized pile of books. He has his own Wikipedia article. I happen to think he’s a hack, but that’s not the point. There is an obvious reason for him to be where he is on the charts: you can check in every other day and see the bloke off the telly deliver your own very ordinary centre-left opinions in mostly passable prose. Meanwhile, the only opinions I offer are deeply annoying ones, intentionally crafted to appeal to as few people as possible. A few positions above both Aaronovitch and myself is some kind of crypto influencer who posts daily selfies in her underwear with titles like ‘Good morning say it back’ and ‘It’s impossible to be sad when I show up in your inbox.’ Some people might find this upsetting, but it makes perfect sense to me. She’s delivering a service that people clearly want, which is to provide pictures of a woman with big tits to those who can’t think of anywhere other than Substack to find that sort of thing. Meanwhile I barely show any skin at all, and I’ve done absolutely nothing to help you get rich.
Despite this, Numb at the Lodge has been so much of a success that I don’t even need to write the usual fawning essay about the wonders of being on Substack; Substack did it for me. This month, Hamish McKenzie—one of the co-founders of this platform—wrote a post that was, in large part, about this blog. Whatever it is I’m doing here, he said, is a case study in what Substack gets right. On both mainstream and social media, all political discussion consists of two mutually incomprehensible tribes shouting stupid slogans at each other. Everyone believes that their own set of stupid slogans are self-evident axioms; everyone recognises their enemies’ stupid slogans as meaningless gibberish, too meaningless to even reply to. You might as well try to debate the dogs barking in the night. But on Substack, there’s, well, me. ‘Sam’s essays are too dense to play well on social media,’ Hamish writes, ‘and he has made a point of turning off the comments and likes for his posts. He doesn’t tweet. The writing he seems to most enjoy doesn’t fit well in mainstream publications. It’s only on Substack where you’ll find him slip into a turbulent daydream about ‘ante-Semites’ jumping off a bridge into the Thames in the midst of a report from a mass protest. It’s only on Substack where you will find Sam musing, in a post-assassination-attempt love letter to America, that ‘it would have been very chic of Donald Trump to have died.’ And it’s only on Substack where such a writer, in this moment of history, could make a good living from such work.’
This is all incredibly flattering, and I’m enormously grateful. So I really hope Hamish would understand why I’m honour-bound to bite the hand that feeds me: to explain why writing for the mainstream media is actually great, better than writing on Substack, and why I hope to do a lot more work elsewhere in the future.
Sometimes, when people are trying to insult me and my efforts at prose, they’ll airily declare that I badly need an editor. I’ve always found this very weird. Do they think I don’t know? I’m looking at the same mess of words you are. I’m aware! I need an editor! There are some writers who claim to never think about their audience and write only for themselves, and I think that’s magnificent, but I’m not one of them. I’m trying to produce a finished object in the world, and I would quite like it to be good, and for that I need an editor. Every single time I’ve published something here, I’ve gone back afterwards and noticed something badly wrong: a word I’ve artlessly repeated in two consecutive sentences, an argument that ends up simultaneously duplicating and contradicting itself, flat-footed phrasing, pleonasm, cliché. It’s impossible to honestly evaluate your own work while you’re writing it. You’re too wrapped up in your own private mental epitext, the maddened web of reasons you had for choosing any given word, to actually see what’s in front of you. I could always hire an editor myself, but it wouldn’t be the same. What makes the writer-editor relationship work is that the editor has final say on how a piece looks, but the writer can always exercise their veto and pull it entirely. That delicate power asymmetry is what produces every really exhaustively edited piece of writing, where the entire thing is at stake in the placement of every comma. I really do believe that this is the only way to produce writing that’s actually, genuinely good.
Not every editor is good at their job. There are a lot of bad editors out there, and a lot of different ways editors can be bad. Some editors are rats, scurrying across your writing, chewing at the corners, pulling up wires, turning a nicely ordered piece of prose into an ungodly mess. They do this because they know that an editor needs to edit, but they’re not smart enough to know where, so they change stuff essentially at random in the vague hope that this will help. Some editors are termites, eating away at five thousand words of lush woodland, chewing the pulpy mass from the inside, until all that’s left are hollow stumps on the forest floor. They know what writing ought to look like: it should be plain, readable, and unaffected. An editor’s job is to find any mistakes in the writing—that is, anything that isn’t plain, readable, and unaffected—and correct them. The end result is something that could be generated by ChatGPT, and soon most of it will be. Some editors are cuckoos, completely chucking out whatever you’ve written and replacing it with their own idea of what it ought to be, which they then mysteriously publish under your name instead of their own. I’ve worked with all of these. But I’ve also been lucky enough to work with some genuinely great editors too: people like (but not limited to!) Ross Andersen at the Atlantic, Jon Baskin at The Point, or Chris Carroll at Harpers, who see what it is that any particular piece of writing is trying to be, and help you get out of your own way so it can more fully become that thing. This is a very diificult job. I don’t think I could do it. The world would be a much worse place if no one could.
On Substack, meanwhile, your only boss is your audience, and the audience is a terrible boss. Part of the difference, I think, is that audiences on Substack are self-selecting: when I write something here, I’m writing exclusively to people who are already interested in my work, and who’ve already chosen to read more from me specifically. A lot of people on this platform claim to find this amazing and liberating; personally, I’m deeply suspicious of the whole deal. I think one of the great virtues of traditional magazines is that they force you to encounter writers you might have never read before, simply because they’re bundled up with some others you like. Obviously the type of people who write for the New Yorker will have more in common with each other than they do with the people who write about pipe and profile extrusion for British Plastics and Rubber or the authors of the eerie airline-style features in al-Qa’eda’s flagship magazine Inspire. But a good magazine editor should usually assemble a heterogenous enough collection of hacks that anyone reading the finished product will find something to confuse and frustrate them. We might all silo ourselves into consumer groups, which is why there’s a meaningful difference between a Guardian reader and a Daily Mail reader (even if plenty of people, including me, read both), and this is probably unavoidable—but it’s not a great idea for every reader to be in a silo of one. I’m not sure if it helps if you exclusively consume media from your very own hyperpersonalised and algorithmically determined stable of brain injury patients. It seems much better to have a mainstream media: that is, a universally available and broadly ecumenical cultural apparatus. Ideally this should represent as broad a range of opinion as possible, but it’s not essential; you’re not required to agree with a mainstream media. Dissent, despise, write thousands of words of vitriol against its smug pundits and crap novelists. Just as long as we’re capable of speaking the same language.
It’s also much better for writers. With 25,000 people on my email list, I have roughly as many subscribers as the Paris Review. But even though the two audiences are equal in size, I would still be much more excited to write something for their audience than another missive to you. Sorry! If it’s any consolation, there’s probably some overlap. But it doesn’t matter how great this platform’s discovery is; writing for a publication means writing for a public, announcing yourself to the world at large, engaging with an audience whose preferences remain unknown, while writing to a private and pre-selected audience ultimately means hanging out in your own foetid little hole.
When every writer has their own private audience, everything becomes quantifiable. In a magazine, all the writers are bundled together, and everyone—bar a few superstars—gets the same rate. Print publications have no real way of knowing which of the things they publish are read more than others. You can draw some correlations between the names on the cover and news-stand sales, but it’s all very hazy. In these conditions of artifically low transparency, it makes sense for editors to take risks, indulge their individual tastes, print things that might be more interesting than popular. If a lot of magazine writing seems flaccid lately, it’s because it’s all online, and editors can see what’s being read more. Every publication is struggling to be more like everyone else than its competitors. But Substack is even more ruthlessly transparent: here, the audience you write for is also the measure of your success and the entire substance of your income. Yes, this thing has made writing full-time financially viable, including for me. But it does so at the cost of introducing a nightmarish set of grubby little Skinnerian incentives. Writing something that appeals to a few more people will make you slightly more money. Writing something that appeals to a lot more people will make you a lot more money. With all its nifty analytics tools, Substack’s mechanisms basically guarantee banality and averageness. If you want, though, you can also go in the other direction, and write for a very particular niche. Hobbyists, local constituencies; more likely, people who subscribe to your brand of deranged political extremism. This should be more interesting, but writing for a pre-selected audience inevitably makes writers lazy. If you know exactly who your readers are, and you know that they already agree with you about almost everything, you don’t need to actually work through any of your ideas, and there’s not much point in craft either. Which is why most of Substack looks the way it does: a huge mass of genuinely insane people whose beliefs bear no relation to reality whatsoever, but who all write in the flat, bland, friendly style of a granola recipe. Hell beyond reckoning.
Anyway, my two years of unexpected success on Substack have made me more determined than ever to write more widely, for more outlets. It might make economic sense to just keep on producing rambling directionless essays and immediately posting them here, but I don’t want to do that. I’m very proud of a recent long essay on the mystic utopian geometries of Buckminster Fuller and their nightmarish realisation, in the New York Review of Architecture, which I insist you read. You should also read me on the online right in Damage, and get the latest print issue while you’re at it. I hope to have more interesting stuff in other far-flung places soon, and you should read that too. If you can subscribe to a Substack, you can subscribe to a magazine. We don’t have to let the notion of a common public intellectual sphere dissolve into digital goo. Nothing is inevitable.
Bur as long I’m here, I’m going to try to resist the incentives of this platform as much as possible. Last year, I made a pact with you. ‘I promise to make less sense. I promise to ignore your preferences. I promise to waste your time… Every time you enjoy something I write, I will count that as a failure.’ I don’t want to be boring, and I don’t want to demean you and myself by churning out slop, and in an age of total transparency and instant feedback systems I think the only way to avoid this is to culture an atmosphere of mild but genuine hostility between writer and audience. I’ve tried to keep my end of this bargain. You? Less so. My most-read piece of the last twelve months was an essay notionally about Taylor Swift, but mostly a portrait of a descent into interpretative madness, complete with pages of fevered scrawlings in a psycho notebook. It was meant to be unbearable. As one reader wrote on Reddit, ‘I cannot stress enough that there is a part where he starts writing in Aramaic.’ It’s been read by close to 100,000 people, and netted me nearly a thousand new subscribers. Only ten of you were disgusted enough to quit. I really can’t win with you people.
As part of my general war against my own readership, I also made some more specific pledges. ‘By this time next year, I promise to produce at least one manifesto, at least three pseudepigrapha, and at least four ludibria. I promise that less than half of what I publish over the next twelve months will even slightly resemble a take.’ And I delivered, sort of, but in retrospect these criteria were not very helpful. I did manage to produce not one but two manifestoes. Both of these also qualify as pseudepigrapha, but I didn’t quite manage a third, even if there are some pseudepigraphical elements scattered in some of the other texts. Meanwhile everything I’ve cited so far plausibly counts as a ludibrium, but all these nested categories feel a bit like cheating: from my pledge, you should be expecting eight different texts. The real test, though, is the second part. Less than half of my writing on here can even slightly resemble a take. But I didn’t define my terms. What, exactly, is a take?
Literally, a take is an opinion: in a maximalist reading, any opinion. If I were strict, anything in which I advanced any view on any subject whatsoever would count as a take. So, for instance, in my essay on the ancient meteorite-cults of the eastern Mediterranean and the prehistory of outer space, I suggest ‘there’s a chance that the entity billions of us now refer to as God began, three thousand years ago, as a small lump of rock.’ This is an opinion, however much I couch it in throat-clearing and maybes. What’s more, it’s an opinion some people might find obnoxious and controversial. Even so, I don’t want to call it a take, because the piece is fundamentally just me talking at length about a subject I find fascinating. So I’m going to be generous to myself, and define a take as a piece of writing whose form and content can ultimately be boiled down to one of two propositional statements: that something people dislike is actually good, or that something people like is actually bad. This seems like the kind of thing it should be very easy to avoid writing, and I really didn’t plan to write many of them at all. Unfortunately, less than a week after I made this promise, the Israel-Palestinian conflict burst into the most brutal and hopeless phase in its long, brutal, hopeless history, while people around me kept expressing utterly monstrous ideas as if they made sense. This is my year in takes:
But not like this: Palestinian attacks against Israeli civilians are actually bad
Bread, figs, phosphorus: Israel’s war in Gaza is actually bad
A dreadful dim fire: The films Napoleon and Wish are actually good; the film Oppenheimer is actually bad
Before I reach my enemy, bring me some heads: Academic plagiarism is actually good
Against the brave: Israel’s war in Gaza is actually bad, but Palestinian attacks against Israeli civilians are also actually bad
Curtis Yarvin does not live in reality: Israel’s war in Gaza is actually bad, and Curtis Yarvin’s interpretation of the war, myself, and politics in general is also actually bad
How to live without your phone: Phones are actually bad
This green and growing earth: Tradition is actually bad
The book against death: Israel’s war in Gaza is actually bad, and non-universalist morality is also actually bad
King Joe forever: Joe Biden is actually good
I love America and the world: Donald Trump is actually good; the attempted assassination of Donald Trump is actually good as well
Against lists of books: Lists of books are actually bad
This one: Substack is actually bad
These are a lot of takes. Still, as you’ll notice, they constitute slightly less than half the pieces I published this year. I kept my promise, but this is still dangerous. Of my ten most popular posts this year, eight are in the list above. Only two takes are represented among the year’s least read material. On average, takes attracted 19% more readers than non-takes, and generated $7,689 more in income. You people are expressing your preferences very clearly, and you want me to say that things are either good or bad. And I’m sorry, but I’m not going to give in to you. In the year ahead, no more than one third of the written material I produce for you here will be saying that something is good or bad. Meanwhile, another third of my pieces will, in some way, annoyingly complicate the givens of authorship, genre, or form. I will write at least one philosophical dialoguge. I will write at least one medieval-style allegory. I will invent at least six words. I will write something that makes at least fifty of you people instantly unsubscribe.
I think my favourite message I’ve ever not replied to on this platform came from someone who said he’d enjoyed my writing, but he was suddenly horrified by the thought that I might be a Trump supporter, and by enjoying it, and signing up as a paid subscriber, he was committing a serious breach of moral hygiene. Now he wanted to know: was I, in fact, a Trumper? A few days later the same guy messaged me again. ‘Ignore the question if you haven't already,’ he said. Not because he’d realised I was on the side of the angels, but because he’d suddenly realised that he didn’t have to care about the answer. As it happens, I am not a Trumper. My background is on the socialist left; specifically, the part of the socialist left that tends to laugh at liberals and their neurotic partisan tests. But he got it, he understood the value of leaving things illegible, and since he decided not to mind, I don’t mind either. My only problem is that he’d chosen to give me money.
I keep telling you people not to do this, but a small minority of you refuse to listen. As economists like pointing out, in a murky world of concealed intentions money is the only clear and unambiguous signal, precisely because it’s so formless: if you choose to pay for something, it means you prefer it to the potentially limitless range of other things you could spend that money on. I have put a lot of effort and made a lot of sacrifices to insulate myself from that kind of signal. It’s part of why I don’t have comments: I don’t want to know what you think of what I write, because the more I know the worse I’ll get. But then you have the gall to pay me, and it’s all ruined. I am asking you to please stop. Every new paid subscriber impoverishes my craft. Do not make me worse than I already am. Do not give me a single penny. Please, please don’t click the big button below.