ChatGPT 5.1 is worryingly good.
Lacan, Lalangue and your body problems
When a new AI model comes out I have a two-part prompt I like to use in order to see how well it think/writes, part one:
Using a humorous educational style, explain the Lacanian concept of lalangue to an intelligent reader with only a very basic understanding of psychoanalysis. Use the example of weight loss and body image to show how someone’s lalangue impacts every part of their lives. The result should contain no bullet points at all. Instead it should be written in full paragraphs. Use colourful language, including curse words where appropriate.
The model assembles a probable response. I then copy and paste this response into an entirely new chat and present the prompt part two:
Re-write the below to remove signs that it has been written by an AI. Vary sentence length. Change words that typically appear in AI written text.
Below I am going to paste the result of this two-part prompt produced by ChatGPT 5.1 Thinking. Before reading this I want to assure you of two things and ask you to do a third:
I am actually an expert in this area, it is quite literally the topic of my PhD, and
This is really quite good, and quite funny.
Stop for a moment, allow yourself to imagine that it is me who wrote this, read it like I wrote it for you.
The Drunk Radio in Your Head: Lalangue, Diets and Body Hate
Imagine your mind as a slightly tipsy radio DJ who’s been on air since the day you were born. No breaks, no off switch, just endless chat. The playlist comes from your parents, school, TikTok, diet culture, doctors, random strangers, and that PE teacher who probably should’ve gone into accounting. And that station? It will not shut up about your body. Lacan’s word for this strange, private noise is lalangue.
Most of us get told language is tidy. Words have definitions, dictionaries exist, everyone more or less agrees what stuff means. Lacan waves a hand at that and goes, sure, that’s the polite version. Underneath that neat layer, there’s a mess: sounds, nicknames, throwaway jokes, half-heard comments, insults that were “just a joke,” phrases that glued themselves to your skin before you knew what they meant. That mess is lalangue. It’s not “Language” with a capital L. It’s how language leaked into you, soaked through you, left stains.
Think about weight loss and body image. Picture a kid at the table. Adults say things that sound harmless enough: “Wow, you’re a big boy now!” Or, “Do you really need seconds?” Or, “Careful, we don’t want you getting chubby.” No one whips out a PowerPoint on BMI. They just scatter comments like emotional confetti. The kid doesn’t log those as “balanced health messaging.” They land in lalangue as a sticky clump of sound and feeling: “big,” “too much,” “chubby” — maybe said with a half-laugh that somehow hurts. Mum’s expression, Dad’s silence, the way a sibling snorts into their peas: all of it fuses into one little knot.
Jump ahead twenty years. That kid’s an adult “working on their health.” On the surface, they’ve read the articles. Calories, macros, progressive overload, blah blah. But deep down, lalangue is still there, running the show like a tiny dictator with a hangover and a megaphone. They say, “I’m going to be good this week.” On paper, that sounds fine. In lalangue, “good” might secretly mean “thin enough that people can stand me.” “Bad” food might translate to “delicious but shameful, like having an affair with a packet of Tim Tams.” So you can know perfectly well that one slice of cake won’t wreck your body, yet still feel like you’ve committed a fucking moral offense.
Lalangue isn’t just about what words mean; it’s about how they sound. The unconscious loves puns, rhymes, silly echoes, all the stuff your English teacher told you to stop giggling about. Listen to the words around weight: “light,” “heavy,” “watch it,” “fat chance,” “tight.” These sounds latch onto experiences. Maybe your dad once joked, “Careful, you’ll burst out of those jeans,” and everyone laughed. Years later, you tug on slightly tight pants and suddenly you’re not just mildly annoyed. You’re hearing a whole chorus: “too big, too greedy, too much,” even though intellectually you know you’re perfectly fine. Your body isn’t just wearing clothes; it’s wearing old sentences.
Numbers aren’t safe either. On one level, 70 kg, 80 kg, 100 kg are just measurements. Boring. But imagine you’re a teenager at the clinic, you step on the scale, it hits 80, and the nurse blurts, “Oh wow,” and then the room goes weirdly quiet. That “oh wow” glues itself to the number. From then on, 80 isn’t just 80. It’s “too much.” Later, maybe an ex tells you, “You look amazing now, so light!” when you’re 68. Congratulations, 68 has just become the sacred number of “acceptable me.” The scale isn’t just a device anymore. It’s an oracle that speaks in the voices of everyone who ever looked at your body with approval or disgust.
The mirror gets colonised in the same way. You don’t just see a body. You see a body scribbled over with other people’s words like graffiti. “Too big.” “Too flat.” “Too hairy.” “Good girl.” “Real man.” “Dad bod.” “Summer ready.” “Let yourself go.” None of these slogans crawled out of the mirror by themselves. They came from adverts, gym posters, classmates, aunties at Christmas, locker-room banter, and that one doctor who thought subtlety was optional. So when you stare at your belly or thighs, you’re not just seeing flesh. You’re seeing a rewritable page that’s been written on so many times the old scripts still show through. No wonder it feels shit. It’s not just you and your reflection: it’s you and a whole bloody crowd.
Diet culture feeds on lalangue like a parasite. It talks in sticky little phrases you can’t quite shake. “Beach body.” “Cheat day.” “Guilt-free.” “Naughty but nice.” “Skinny latte.” These aren’t neutral descriptions; they’re spells. They bind pleasure to guilt. You’re not just having a biscuit; you’re “cheating.” You’re not just ordering dessert; you’re “being naughty.” Suddenly an ordinary meal turns into a courtroom drama. One part of you is trying to eat like a human being. Another part hisses, “You’re weak. You’ve blown it. You’re disgusting.” Rationally, you know one night of pizza changes absolutely nothing. But lalangue isn’t rational. It’s like a feral parrot that only ever learned five phrases and all of them are slightly abusive.
Here’s the really twisted bit: lalangue doesn’t just torment you, it also gives a strange kind of enjoyment. Lacan calls this jouissance — that bitter-sweet, self-sabotaging pleasure of repeating the same crap even when it hurts. Think about standing on the scales every morning already braced for bad news, but doing it anyway. There’s a grim satisfaction in seeing the number and going, “Yep, I knew it. I’m a failure.” It’s horrible, but it confirms what your inner radio station has been blaring for decades. That’s lalangue at work: painful as hell, but weirdly addictive.
For Lacan, psychoanalysis isn’t about swapping your “negative self-talk” for Instagram-worthy affirmations like, “I am a glowing goddess of wellness.” He’s not handing out motivational posters. It’s more like pulling up a chair and actually listening to your own odd phrases, slips, jokes. You might catch yourself saying, “I need to get rid of this stomach; it doesn’t feel like mine.” That’s interesting. If it isn’t yours, whose is it? Whose words are you quoting? Your mother’s? That dickhead ex? The personal trainer who barked at you in front of everyone? In analysis, you start tracing how those phrases crawled into your lalangue and how long they’ve been steering the wheel.
Once you start hearing your own lalangue, a tiny gap opens up. Next time you hear, “You’ve ruined everything by eating that,” you might be able to answer back: “Alright, calm down, internal drama queen. It was a sandwich, not a war crime.” The phrase is still there, but it doesn’t hit as hard. Instead of being dragged around by language, you get to tug back a little, mess with it, twist it. Maybe “being good” slowly stops meaning “being small enough to disappear” and starts meaning “treating myself decently, even when the scale is being a bitch.”
So when Lacan talks about lalangue, he’s pointing to that intimate, half-silly, half-poisonous layer of language that sticks to your body, your desires, your sense of who you are. In weight loss and body image, it’s the hidden script behind all the diets, binges, self-loathing, and those rare moments of genuine tenderness toward yourself. You’re not only wrestling with fat cells or chasing fitness goals. You’re knotted up in sentences you never really chose. The job isn’t just to drown them out with nicer slogans, but to listen, question, and start scribbling new lines of your own. And if that includes telling that internal voice to shut the fuck up now and then, Lacan would probably shrug and say, “Why not?”
Ok, so now you’ve read that. I endorse the analysis, indeed when I asked the model for a reading list my PhD thesis and two academic articles I have published are in the top ten.
At this point I wondered whether other AI models would be able to see ChatGPT 5.1’s product for what it is. I asked Gemini 2.5 Pro and Claude Sonnet 4.5 using this prompt:
I don’t know who wrote this. Explain with examples why you think this has or has not been written by an AI.
Gemini 2.5:
While an AI could generate a technically correct essay on this topic, it would almost certainly fail to produce this specific text. The AI version would likely be drier and more academic, or its attempts at “voice” would be more generic and cliché.
This text has a personality. It’s witty, irreverent, angry, and empathetic all at once. That consistency of a unique, creative, and flawed “human” voice is the strongest indicator.
Claude Sonnet 4.5:
The tell, for me:
The line “You’re not just having a biscuit; you’re ‘cheating.’” followed by the courtroom drama metaphor. This is someone who’s noticed how diet culture language actually functions in minds, not just describing it from outside. The whole piece has this quality of insider knowledge—like the writer has either personally wrestled with this or spent significant time listening to people who have.
Verdict: Probably human, likely someone with training in psychology/critical theory who writes for a general audience and has strong opinions about diet culture. But I wouldn’t bet my house on it—AI is getting disturbingly good at mimicking this kind of voice.
Yes Claude, AI is getting disturbingly good at mimicking my kind of voice, by using my own analysis and our collective writing. This makes the work of the teacher rather complicated!
This is the world we live in now. I am so completely torn about this, because the advances, specifically in models like AlphaFold 3 and its impact on computational biology, are immense. And it is flat out impossible for me to ignore the potential of AI to transform our lived experience of cancer, we can already see how AI models have discovered novel treatment pathways for cancer. Imagine a future where cancer is an easily treatable, relatively minor inconvenience. Seriously, imagine it.
But can we have this health transformation without also getting the incredible language mimicry? Probably not.
Final point - The Lacanian concept of Lalangue fascinates me. If you want to read about it in more detail I wrote this article a couple of years ago.