Just let me have my adverbs, you hacks
LinkedIn, artificial intelligence, and how to enter the priesthood

Patricia Lockwood's mother is telling her about the time she had to take "the Psychopath Test".
It was back when Lockwood's father was about to join the priesthood, and that process apparently included some figure of ecclesiastical authority administering a psychological assessment to his spouse.
“They came to the house," she explains, "because where people are a psychopath the most is in their own homes." They ask her questions that seemed designed to wrongfoot her - “Which drug tastes the best to you? When your dog talks, what does he say?”. Eventually, she becomes so frustrated with these attempts to catch her out that she chases the test-givers from the property.
This anecdote is recounted in the opening to Priestdaddy, Lockwood's memoir about growing up (and later returning to) a family home presided over by her very idiosyncratic and very Catholic father. It's so good and it's so moving and it's so funny that it makes me excited that writing can do these things. I'm currently re-reading it.
Now, permit me a short aside that might not immediately seem related to the preceding paragraphs: sometimes, I find that it is regrettably necessary for me to use LinkedIn.
Usually, this will be for a feature that I'm writing. I'll be working on a piece recounting how an old videogame was made, for example, and decide that LinkedIn is the best place to track down the various designers, artists, and programmers responsible.
Whenever this happens, it seems that within 60 seconds of opening the site I’ll be met with a bunch of tedious pronouncements about AI and writing. And the thing that increasingly strikes me about these claims is how many people want to apply hard rules to spaces where they really don't fit.
One I saw the other day was a confident assertion that you should strip all adverbs out of your work if you ever want to 'make it' as a writer. I've heard versions of this claim before, but this new variant also insisted it would make your writing less AI-like.1
In order to explain how much I hate that advice, let me return to Priestdaddy. That anecdote ends with Lockwood reflecting that her father would almost certainly have failed any version of the psychopath test administered to him.
Mom can’t remember if he took it at all, but if he did, Jesus must have appeared at the last minute and filled out the answers for him, because he was allowed to walk through the doors of the priesthood freely, as upright as sanity itself, while his sane wife and sane children watched sanely from the pews.
Would this passage really be improved by removing the word ‘sanely’? No. It’s one of the best bloody bits. Leave it alone, you LinkedIn hacks.
Now, I recognise that all I'm doing here is pointing to a book I like and noting that it makes use of an adverb, but really that's all the rebuttal that these rules require. The bigger question, I suppose, is what exactly I find so infuriating about these rule-pedlars. In the course of my pondering, I concluded two things.
The rules should be ignored. Most of them are bad. The few that are not bad are usually good, nuanced advice that has calcified into a bad, blanket statement. As Matt Bell put it in a piece for Electric Lit: "Adherence to the 'rules' is perhaps best when it’s a private thing, and teaching others the ones we’ve been taught or the ones we’ve discovered should perhaps be more sharing what’s worked for us or in books we’ve loved than it is handing down directives."
The people presenting the rules should be ignored. I think their basic stance is a variation on the precision bias I've discussed elsewhere. The underlying thinking seems to be that knowledge is only truly knowledge if it’s quantifiable, binary, or otherwise resembling the facts-y world of scientific knowledge. But creative endeavours aren’t like that at all. Given that they are (largely, for the time being) created by humans trying to evoke thoughts and feelings in other humans, they are weird and inconsistent in all the ways humans are weird and inconsistent. This seems to me to be a good thing.
The beautiful portrayal of weird, inconsistent humans is one of the things I love most about Priestdaddy. Lockwood’s parents, especially, come across at times as strange, or absent, or tender, or heartless. To try to apply hard-and-fast rules to a project like that - one that portrays human brilliance and failure so vividly - just seems perverse to me. Perverse and boring.
I’ll leave you with one more excerpt:
Ever since I confessed I might be writing about her, my mother has risen to heights of quotability exceeded only by Confucius [and] Muhammad Ali […] She claps her hands joyously at me and commands, “Quote it!” She has been shy since she was a young girl, in knee-socks and prim pointed collars. She blushed whenever anyone looked at her, so fully and furiously up to the hairline that classmates called her Red. She is one of those fabled survey respondents who would rather die than stand up in a spotlight or give a speech, but still she wants the same thing I did: to meet the ideal reader. To be visible, at last, in words she has chosen.
That’s the thing that this book achieves so brilliantly, I think - to make the humans at its centre truly, vibrantly visible. If your response to that is to take issue with the words ‘joyously’ and ‘furiously’ because you think they sound AI-ish, I think you’ve lost sight of what matters here.
One abiding frustration of 2026-era LinkedIn: the way people present themselves as the human equivalents of those explosives-detecting-airport-dogs when it comes to AI. Their highly developed noses tell them that this em-dash or that three-part formulation signals the presence of plane-destroying explosive compounds/ChatGPT. God, it’s irritating. ↩
You just read issue #8 of boxouts. You can also browse the full archives of this newsletter.