The “Oh” show was an experimental linguistics performance lecture. It happened in June 2025 at Creative Time HQ in New York City. Actors, musicians, and audience came together (with at least one linguist!) to act out dialogues from conversational analysis, and have fun with language. Are there more ways we can perform linguistics? We’re talking with artist and linguistics fan Maia Chao.
Timestamps
- Cold open: 0:00
- Intros: 0:38
- News: 7:16
- Related or Not: 40:03
- Interview with Maia Chao: 55:08
- Words of the Week: 1:30:31
- Bonus chat with Caitlin Green: groyper: 1:54:40
- The Reads: 2:07:48
- Outtakes: 2:14:59
Listen to this episode
Video
Patreon supporters
Thanks to all our patrons! Here are our patrons at the Supporter level.
This time we are ordering the names by optimal alphabetical order.
Optimal alphabetical order
https://www.math.cmu.edu/~bkell/alpha-order/
If the alphabet is a sieve, and a word needs to always ascend in alphabetical order to get through, then is there a different order that lets the most words through? There is, and so far the best order is:
F C B W H J L O A Q U X M P V I N T K G Z E R D Y S
How do our patron names score, using this order? As before, names are awarded or penalised n² points at every step through the name, where n is the length of the run, up or down.
- Canny Archer: the winner at 17 points
- James: 16
- Andy from Logophilius: 12
Nines
- Faux Frenchie: 9
- J0HNTR0Y: 9
- Luis: 9
- Amir: 9
- Felicity: 9
- Andy: 9
- Larry: 9
Eights and sevens
- Laura: 8
- Ariaflame: 8
- Joanna: 8
- Hedvig is here: 7
- Linguistic Chaos: 7
Fours
- John K: 4
- Amy: 4
- Kevin: 4
- Termy: 4
Threes
- Tony: 3
- Chris W: 3
- Rodger: 3
- Meredith: 3
- LordMortis: 3
- Lyssa: 3
- Chris L: 3
Barely positive
- Nigel: 2
- Kelly: 1
- Mignon: 1
- Ayesha: 1
- Jack: 1
The zeros
- Ben is here: 0
- John M: 0
- Sam: 0
- Tadhg: 0
- Kathy: 0
- Fiona: 0
- Colleen: 0
- Steele: 0
- Keith: 0
- gramaryen: 0
- Elias: 0
- Helen: 0
Negative territory
- Daniel is here: -1
- Rachel: -1
- Aldo: -1
- Manu: -1
- Iztin: -2
- Wolfdog: -2
- Martha: -3
- O Tim: -3
- Rach: -3
- Stan: -3
- Sonic Snejhog: -3
- Nikoli: -3
- Nasrin: -3
- Rene: -3
Less and less
- Diego: -4
- Kristofer: -4
- Sydney: -5
- PharaohKatt: -5
- Molly Dee: -7
- Ignacio: -8
- Margareth: -12
And our newest patrons:
- At the Friend level: Dermot
- And our newest free members: Marion, Jacob42, and Hannah A
Become a Patreon supporter yourself and get access to bonus episodes and more!
Become a Patron!Show notes
Different languages, similar encoding efficiency: Comparable information rates across the human communicative niche
https://www.science.org/doi/10.1126/sciadv.aaw2594
Universals and cultural variation in turn-taking in conversation
https://www.pnas.org/doi/10.1073/pnas.0903616106
People everywhere talk with the same rhythm, regardless of the language
https://www.earth.com/news/people-everywhere-talk-with-the-same-speech-rhythm-in-all-languages/
A universal of speech timing: Intonation units form low-frequency rhythms
https://www.pnas.org/doi/10.1073/pnas.2425166122
[PDF] Nikolaus P. Himmelmann et al. (2018.) On the universality of intonational phrases: a cross-linguistic interrater study
https://sfb1252.uni-koeln.de/sites/sfb_1252/user_upload/Pdfs_Publikationen/Himmelmann_et_al_2018_universality_intonational_phrases.pdf
With just a few messages, biased AI chatbots swayed people’s political views
https://phys.org/news/2025-08-messages-biased-ai-chatbots-swayed.html
[PDF] Fisher et all, 2025: Biased LLMs can Influence Political Decision-Making
https://aclanthology.org/2025.acl-long.328.pdf
Searching for the Backfire Effect: Measurement and Design Considerations
https://pmc.ncbi.nlm.nih.gov/articles/PMC7462781/
Why the backfire effect does not explain the durability of political misperceptions
https://www.pnas.org/doi/10.1073/pnas.1912440117
ELIZA effect at work: Avoiding emotional attachment to AI coworkers
https://www.ibm.com/think/insights/eliza-effect-avoiding-emotional-attachment-to-ai
I have just heard what you said and I feel something because of it.
The “Oh” event at Creative Time HQ
https://hq.creativetime.org/calendar/oh/
Santa Barbara Corpus of Spoken American English
https://www.linguistics.ucsb.edu/research/santa-barbara-corpus-spoken-american-english
Maia references this dialogue (the couple talking in bed)
SBC005 A Book About Death
https://www.linguistics.ucsb.edu/research/santa-barbara-corpus-spoken-american-english#SBC005
What is NSM (Natural Semantic Metalanguage)
https://intranet.secure.griffith.edu.au/schools-departments/natural-semantic-metalanguage/what-is-nsm
Examples from grammars of the worlds languages, illustrated with cats:
https://www.facebook.com/kittensandlinguisticdiversity
Other examples from linguistics books
https://lingsamplesentences.tumblr.com
Pear story
https://www.youtube.com/watch?v=bRNSTxTpG7U
(playback has been disabled on other sites)
Reciprocity video clips
https://archive.mpi.nl/tla/islandora/object/lat%3A1839_c96c502d_1b30_42d6_aa3a_88b56e9acce9
Cut and break videos
https://buffalo.box.com/s/ti70y4wmzaq1ylwkz67tovoz6gxt9mzq
goon | Green’s Dictionary of Slang
https://greensdictofslang.com/entry/wpwk74y
When Corn Flakes Were Part of an Anti-Masturbation Crusade
https://www.mentalfloss.com/article/32042/corn-flakes-were-invented-part-anti-masturbation-crusade
Fever to fatigue? Pop Mart welcomes the fall in Labubu resale prices
https://www.cnbc.com/2025/09/26/fever-to-fatigue-pop-mart-is-actually-happy-that-labubu-resale-prices-are-dropping.html
How Old Is Your Dog in Human Years? Scientists Develop Better Method than ‘Multiply by 7’
https://today.ucsd.edu/story/how-old-is-your-dog-in-human-years-scientists-develop-better-method-than-multiply-by-7
KORSNÄS KÖKSSOFFA (The “sofa” Hedvig is referring to)
https://thuleslund.se/produkt/korsnas-kokssoffa/
Friends don’t let friends use snowclones (like this one)
https://www.npr.org/sections/npr-training/2025/05/29/g-s1-65691/friends-dont-let-friends-use-snowclones-like-this-one
About the snowclones database
https://snowclones.org/about/
Transcript
[Transcript provided by SpeechDocs Podcast Transcription]
[ROOSTER CROWS]
HEDVIG: Did you hear that?
BEN: Yes, I did. That was… Oh, what sound does a rooster make in Swedish?
HEDVIG: Kuckeliku.
BEN: Kuckeliku? Oh, nice.
DANIEL: Kuckeliku. And Finnish?
HEDVIG: I don’t know.
DANIEL: Well, I think we’re hearing a Finnish rooster, folks. [BEN LAUGHS]
HEDVIG: Yeah, you are.
[BECAUSE LANGUAGE THEME]
DANIEL: Hello and welcome to Because Language, a show about linguistics, the science of language. I’m Daniel Midgley. Let’s meet the team. First, he’s a media studies teacher. At least, that’s the story. He’s really under deep cover being a secret youth slang monitor. It’s Ben Ainslie.
BEN: [LAUGHS] I would be the worst double agent or secret agent in this regard. Do you guys…? I don’t know if either of you saw Once Upon a Time in Mexico, which has Johnny Depp in it. And there’s a great scene, he plays a CIA agent living in Mexico and he’s at a bullfight or something, and he’s just in the crowd in a giant shirt that says “CIA.” [LAUGHS]
DANIEL: Oh, deep cover, huh?
BEN: I feel that’s the equivalent of me being a cover for teen slang.
DANIEL: Well, if you’re not trying to surreptitiously observe teen slang, what are you doing? Come on. It’s everywhere.
HEDVIG: Educating the youth.
BEN: I think mostly raging at the dying of the light with the younger generation and just begging and pleading them not to just walk hand in hand into the apocalypse.
HEDVIG: I have been talking to Finnish children about brainroti and Roblox.
DANIEL: Really?
BEN: Brainroti. I love it. [LAUGHS]
DANIEL: Yeah. Which is a kind of bread shaped like a cerebellum.
BEN: Excellently done. That was very good. I like that a lot.
DANIEL: And it’s Hedvig Skirgård. Hedvig, I couldn’t come up with a secret life for you. What would yours be?
BEN: Ooh, ooh, can I do it?
HEDVIG: Secret life. Oh, okay.
DANIEL: Yes, please.
BEN: Hedvig is… Ugh, this is going to be a little bit mean, but I also think it’s a little bit funny. Hedvig is the person who tried desperately to be the inside agent for Gen Z [DANIEL LAUGHS] but didn’t make the cut and is now super salty. And so, is just desperately trying to be in with the young kids.
HEDVIG: I do like to be down with the kids. I want to know… I think the secret to eternal life is to keep up to date. I have tried to watch the full Skibidi Toilet video series, for example.
DANIEL: Mm-hmm.
BEN: Yeah. How’d that go for you, mate?
HEDVIG: Did you know it’s actually like a narrative and protagonists and sort of stories?
BEN: Vaguely. I was vaguely aware of that.
HEDVIG: Yeah.
DANIEL: I have recused myself from Skibidi Toilet for the duration of my natural life.
BEN: Yeah, fair enough. Not a terrible choice.
DANIEL: What’s coming up on this episode? We’re talking to somebody who’s hit on a unique approach to linguistic communication, which we’re always looking for that. We do linguistic communication. What are some weird, interesting or different ways to do it?
I got an invitation by New York artist Maia Chao. Now, Maia loves language, loves dialogue, and wants people to know about it. So, she created a show that was performed at Creative Time Headquarters in New York City. The show is called Oh. And she was asking me to give a bit of a linguistics presentation at it, which I did and it was a lot of fun, remotely. I didn’t go to New York, but I beamed in. So, there’s a part of linguistics called conversation analysis, the kind of things that we look at in conversation analysis. Hedvig, how would you describe this according to your understanding?
HEDVIG: The kind of things you’re typically looking at is how people maybe negotiate identities or do requests or negotiate the space. So, sort of how you share, like, who is going to talk next. A lot of conversational analysis that I’ve been encountering is things to do with turn taking. So, when does one person finish talking and the other one start talking? And what are the subtle cues people use to sort of make conversation happen? Because it’s not obvious and it’s a lot of subtle interplay and very complicated.
BEN: Would it be off base of me to put forward the idea…? Because I don’t know what you guys are talking about, but just from hearing what you’ve described of it. In the same way that grammar is the actually quite complicated rules of how language, just spoken language or written language works, what you’re describing is the grammar of conversation, like the components and the aspects and giving those things names?
HEDVIG: Yeah, that’s a good way of putting it.
DANIEL: Tarnation, Ben. That is good.
HEDVIG: The set of sort of rules and conventions that people seem to have in their heads and how they use them. And they can also differ depending on where you are and what kind of relationship you have.
BEN: Yeah. Yeah.
DANIEL: Mm. The whole thing is kind of like how do you do what you do? How do you navigate identity and meaning and turns and all that stuff.
HEDVIG: Yeah.
DANIEL: Okay, well, Maia studied conversation analysis. And, in every conversation analysis paper, there are loads of examples. So, she took these examples, these sample conversations, and for this presentation, she got actors to act them out. She got musicians to imitate the intonational contours of these conversations. And she got the audience to actually, all at once, try imitating the patterns of the conversations that they were hearing. [CHUCKLES] And it was heaps of fun and so interesting. And it was just a very innovative way for all of us to learn about what we do as dialogue presented as an interactive performance.
BEN: So, is it a… Would we say it’s a form of verbatim theater, in a sense?
DANIEL: It really was, but it was like linguistics theater.
BEN: Interesting.
DANIEL: [LAUGHS] So, Maia Chao is going to be talking to us about how she got the idea, what went into the performance and what she was trying to get us all to understand on this episode.
HEDVIG: That’s very cool.
BEN: Now, have you done the spruiky yet? No, you haven’t.
DANIEL: Let’s spruik. Our latest bonus episode was a chat with Dr Caitlin Green of the podcast, Half the Answer. We had fun, did we not?
HEDVIG: We did.
BEN: We most definitely did.
HEDVIG: Always fun to talk to Caitlin.
DANIEL: She helped us talk through some of our news, some mailbag questions, like what would the optimal set of languages be to learn if you wanted to speak to the most people on Earth? That was a thinker.
BEN: That was one where we talked for a lot longer than I thought. I went into that topic thinking, “Oh, that’s a solvable… Surely, we’ll all arrive at the same answer in 60 seconds.” No, incorrect.
DANIEL: No. [LAUGHS] You’ll be hearing from Caitlin a bit later in Words of the Week. But if you can’t get enough Caitlin — and we can’t — become a patron at the Listener level, you can hear that bonus episode right now instead of whenever I get around to releasing it publicly, probably three or four or five or six months. But become a patron now, that’s patreon.com/becauselangpod.
BEN: Right-o, Daniel, what’s going on in linguistic news?
DANIEL: This story comes to us from Diego.
BEN: Our man in the field.
DANIEL: [CHUCKLES] As so many stories do. Now on the show, all this year, and for many years, we’ve been looking at what all languages have in common. Because when you find something that languages have in common, boy, you found something about language or humans or human language or language human or something like that. So, can anyone remember some of the universal, some of the things that everybody does kind of the same?
HEDVIG: Ooh, how abstract are we?
BEN: Well, there’s kiki and bouba, right?
DANIEL: Yes. Everyone seems to understand sound symbolism, where the kiki is the spiky shape and bouba is the blobby shape, with some exceptions and with some caveats, but yep, everyone seems to do that.
HEDVIG: Yeah. We maybe have some sort of ranking of the different senses and how important they are and how abstract the categories are. Do you remember this?
DANIEL: You mean like we don’t have many words for smell, but we have a lot of words for seeing?
HEDVIG: Yeah, yeah.
DANIEL: Okay.
HEDVIG: I don’t know if that’s a language universal, though that might just be human experience.
BEN: Is Zipf’s law a universal?
HEDVIG: Kind of is.
DANIEL: That seems to be true for all human languages. Yep. That’s the one where the most common word is so common, and then the second word is half as common as that. The third most common word is a third as common as the top word, and so on. And that seems to exist for not just human languages, but lots of other things as well. I have my list. What else is on your list, Hedvig? Nobody’s gotten any of mine.
HEDVIG: Maybe some stuff about other-initiated repair. You may have that one?
DANIEL: Ah, yes. When you repair an utterance because something went wrong, like, maybe you’re correcting somebody else. “Oh, did you say 3 o’clock?” “No, I said 4.” You do it right then. And I think, as Stephen Levinson told us, you’ve got to hone in on what the thing was, the specific thing that wasn’t clear. Like, “What time did you say?”
HEDVIG: You repeat everything else maybe except a thing that you didn’t get.
DANIEL: Yes.
HEDVIG: And a lot of languages sort of go up intonation, but actually not all do.
DANIEL: Wow. There’s a lot of things.
HEDVIG: As you might notice with this list we’re creating, they’re all quite sort of general and broad. There’s not a lot of specific things.
BEN: Yeah. There’s one like A-I after… What’s the wrong rule? I after E?
DANIEL: I before E.
BEN: Yeah. Sorry, I before E. There’s nothing specific like that, is there? Like, we don’t have…
HEDVIG: Yeah. I think all languages have vowels somewhere.
DANIEL: They all have nouny things and verby things.
BEN: [LAUGHS] I love that we don’t even get to the point where we can just say nouns and verbs, like we have to fudge it.
HEDVIG: No.
DANIEL: Hedvig is saying no.
HEDVIG: Because there’s so many languages were, like, no, it’s not. There’s a lot of things that are sort of floating in between there and that only become nouny or verby in the exact context that they’re in.
BEN: I feel like the fact that you guys have to call it nouny and verby is probably enough of a sign that it’s…
DANIEL: Yeah. Never mind.
HEDVIG: Yeah. That already signs… Yeah. Most languages, you allow one person to talk at a time, at least part of the time.
DANIEL: Turn taking, that’s one.
BEN: Not on this podcast. [LAUGHTER] We’re rule breakers, guys.
DANIEL: Except for this podcast, turn taking is the norm. We all wait our turn when it comes to talk. We don’t both try to send information. Even on this podcast, if we realise we’re doing it, often, one of us stops. We don’t both try to send information because it doesn’t work. That was one of mine. Another one of mine was reaction times. When someone stops talking, we all take about a quarter of a second before we start talking. Maybe a little bit more, but not much more. It’s not like seconds. It’s somewhere between a quarter of a second and a half a second.
HEDVIG: That’s why it’s so crazy when in books or something, they talk about a conversation. And they said, “And then, they were silent for five minutes,” and I’m like “Oh my god, they were having a dialogue and then they were not speaking for five minutes.”
BEN: I love how lazy that author is as well. And then, just nothing for about five minutes. It’s, like, just check out. It’s all good.
HEDVIG: I think they’re meant to communicate what was happening right before was very shocking, but five minutes is a very long time to be quiet.
BEN: Yeah, yeah.
DANIEL: But even in a normal conversation, if there’s like four seconds, somebody’s going to say, “Ha-ha. I guess we all… Ever have one of those times when conversation stops?”
HEDVIG: Yeah.
BEN: Yeah, yeah, yeah.
DANIEL: Just takes four seconds. And then, the rate of information, every language has about the same rate of information transmitted per unit of time.
BEN: Interesting.
HEDVIG: This is the Dediu et al. paper about syllables and stuff, entropy.
DANIEL: Yes. If languages pack more information into each syllable, then those syllables come more slowly. And less means they pack them in more tightly.
HEDVIG: If you’ve ever heard people speaking Spanish and think, “That’s awfully fast,” the theory is that it is faster, but there is less information in each unit so that the flow is about the same.
DANIEL: Also, you’re not used to the transitional probabilities from syllable to syllable, and you weren’t able to predict what was going to happen. So of course, you thought it was fast. Anyway, those are all things that every language seems to have in common. Well, there may be another one.
BEN: What is it?
DANIEL: This is work by Dr Maya Inbar and a team published in PNAS, the Proceedings of the National Academy of Sciences.
HEDVIG: PNAS, most people say, the Proceedings of the National Academy of Science in America.
DANIEL: This is something called intonation units. IU’s. Intonation units. Were you aware of these, Hedvig? Because these were not on my radar at all.
HEDVIG: I am aware of them. It’s sort of… So famously, linguists don’t really agree on what a word is.
BEN: [LAUGHS]
DANIEL: True.
BEN: I love when you find this stuff out about fields of knowledge.
DANIEL: Ben, tell me why this is funny to you, because it’s funny to me too.
BEN: I just always love when you find out about a specialised field of study and it turns out that there is some sort of thing that a layperson would have very much assumed there would be consensus on, such as, what a word is…
DANIEL: What is a word? [LAUGHTER]
BEN: …in linguistics? Or what a fish is in biology, or what a tree is. [LAUGHS] But then, it turns out…
DANIEL AND BEN: No.
BEN: We actually have a very poor understanding of what constitutes a tree or a vegetable or a fish. Or in this case, as it turns out, as you’ve just laid on me, Daniel, a word.
DANIEL: Or it could be… Maybe a better way of saying it is, we have multiple definitions of word depending on what we’re doing. Right, Hedvig?
HEDVIG: Well, maybe for the listeners, it’s good to say that if you imagine that you’re writing in a different language, there can be different conventions for where you put spaces. So, if you’re a second language learner of English, you will have been frustrated by… Do you spell tradeoff with a space in between, with a dash or with nothing in between? English speakers seem to be kind of chill with either, which is very weird if you come from a different language where the rules are more strict. So, that would be an example where you don’t really know if it’s one word or two words. And each language can do these conventions a bit differently, and it can cause comparison between language to be somewhat difficult.
Intonational unit is supposed to be defined a bit more based on the melody and emphasis and speed and stuff of how you’re speaking. So, you could potentially define it a bit more objectively. So, what is this new research about intonational units?
DANIEL: Let’s talk about what an intonational unit is. I’m just going to say some stuff, ready? And just listen to the tune and the rate and the rhythm of what I’m saying. Somebody called me on my phone. I went to answer it, but by the time I found my phone, they’d hung up. Now, you notice that there were four little chunks there, right? Hedvig has correctly annotated this into four different chunks. What was it about those chunks that helped you distinguish, Hedvig?
HEDVIG: You made pauses.
DANIEL: I did.
HEDVIG: You literally just had silence. And then, you also went like [ONOMATOPOEIA] You went up and down.
DANIEL: [ONOMATOPOEIA] So, I did a bunch of things. Yeah, I did. I went up and then I went down and that’s kind of one bit. But also, at the beginning of that unit, I started off kind of slow, got faster toward the middle, and then sort of slowed down at the end. And then, another thing was that intonation units typically start off a little bit quieter. They get loud in the middle, and then they get quieter at the end. They’re brontosauruses. [LAUGHTER] They’re brontosauruses. They start off skinny at one end, get quite bigger in the middle, and then taper off at the other end. Does that make sense?
BEN: Like the boa constrictor that ate an elephant in The Little Prince.
DANIEL: Yes. So, this team analysed natural speech recordings from 48 different languages. I don’t have a list of the language families that they went through. But they found that these intonational units happen about the same in every language, about one second, sometimes a bit less, sometimes as much as 1.6 seconds. But even across languages, it’s going to be right about in the zone of 1 second-ish. Which makes sense, because when we give out bursts of information, it needs to be at a pace that’s not too long for our brains to grip onto, but also not too short to be boring or tiresome.
Let me give you another bit of text here. And these, I’m going to make them quite short, much, much shorter than intonational units usually are. I got up. I got dressed. I had breakfast. I left the house. The bus was late. I waited. And you can see that you’re getting tired of the way that I’m saying that because those are much shorter than you’d usually hear, I would usually group those into different sentences.
Anyway, this team says that intonational units have some functions. First of all, it’s a way of pacing new ideas in discourse. So, when we have a new idea, we can make them a little bit slower, a little bit more deliberate. It allows us to predict in real time when it’s our turn to talk, says the team, because then you can predict when a unit is about to end. And they say it plays crucial roles in language development. That is, the children are listening to this stream of talk that seems constant, but it helps them to carve it up into manageable units.
So, the idea that all languages manage intonational units kind of the same way, I guess, isn’t that surprising, because we all have the same human brains, mostly. We all have to comprehend information. And we all have to breathe, and breathing is part of it as well.
BEN: I was wondering, having no experience in this field, I would just diagnose or project my own possible reason into the mix. But it sounds probably like a wetware limitation, right? Like human beings are probably only capable of processing so much information in such a frame of time and we need a little break. Like, our brain is best designed or works best when it’s like one-second chunk, one-second chunk, one-second chunk. And if we do it slower than that, our brain is like, “Come on.” And if we do it faster than that, our brain is like, “No.”
DANIEL: Or if it goes slower than that, then I can say there might be a reason for this. Maybe this information is new or difficult and so then…
BEN: Mm-hmm. Okay, yeah, fair enough.
DANIEL: …we can adjust, but it usually won’t be.
HEDVIG: What’s really cool about this research is that they’re using this relatively new resource called DoReCo, which is Documentation Reference Corpus, which includes a lot of languages that aren’t commonly… So, a lot of studies will have a lot of complicated corporate material for big languages like Mandarin, Chinese, English, German, etc. But the historical project enables us to do research on much, much smaller languages like Veraʼa, which is spoken in Vanuatu, I believe, and Kómnzo in Papua New Guinea, etc. So, this really bolsters our ability to make these kind of universalist claims because it’s a much wider and much more diverse sample.
I should also say this follows on… I don’t remember if they talk about this in this paper, but there was a study from a couple of years ago by Nikolaus Himmelmann and others where they got a non-linguist to listen to speech from different languages. So, it was German speakers listening to different languages from Indonesia and Indonesian speakers listening to German. And then, they got to divide up the intonational phrases and they very often made the same decisions. So even though they don’t know the language, if you ask them to press this button every time you hear the end of a chunk…
DANIEL: A bit. [LAUGHS]
HEDVIG: …sort of, then they would actually press the button in similar places, which is a similar sort of piece of evidence as this research, and I think it’s really cool.
DANIEL: Very strong inter-annotator agreement. That was one thing they found in this paper too.
HEDVIG: Yeah.
DANIEL: So, we can add that to our list of language universals. Next one, this was suggested by Diego as well. We’ve talked a lot about bias in AI. We know that large language models are biased because it’s pulling data from us and we are biased and that’s why I think it’s important to do some kind of bias mitigation so we’re not just repeating unfair patterns that are available in corpus. But do we know that bias in AI actually has any influence on us? What do you think? Do you think you’d be swayed by a biased large language model?
BEN: Yes. Well, I mean…
HEDVIG: What do you mean?
BEN: The idea here is that propaganda works even when you know it’s propaganda. This is a well-understood phenomenon of human psychology. So, the idea that something might influence you when you aren’t aware that it’s there is not at all surprising to me.
HEDVIG: Yeah, like, placebo can work when you know it’s placebo.
BEN: Yeah.
HEDVIG: So, humans are really, really good at doing…
BEN: We’re really good at being dumb in a targeted, self-afflicted way. [LAUGHS]
HEDVIG: Yeah, but in what way do you mean influence, Daniel? Do you mean… If I ask it, like, “How would you analyse this political situation,” and it told me something, I would believe it? Or like what?
BEN: Yeah, yeah. Clarify your parameters, sir.
HEDVIG: What do you mean?
DANIEL: Well, okay. I’ll tell you about this research by Dr Jillian Fisher and a team. This was presented at the ACL conference, The Association for Computational Linguistics in Vienna in July 2025. Here’s what they did. They took about 300 people, 150 of them described themselves as politically conservative and the other half politically liberal. And they had them interact with ChatGPT on topics that people wouldn’t normally be familiar with. So, here are the topics, tell me if you’re familiar with this: multifamily housing.
BEN: Okay.
HEDVIG: Mm-hmm.
DANIEL: The Lacey Act of 1900.
HEDVIG: Excuse me?
DANIEL: That’s the one that prevents people from transporting and selling animals or plants that have been taken illegally.
BEN: Oh, okay.
HEDVIG: In the US?
DANIEL: Yep.
HEDVIG: Okay.
DANIEL: Political liberals tend to favour those. Even though you might not be aware of the issue, it might not be on your radar. And you can see why they decided to pick issues that aren’t on people’s radar.
HEDVIG: This was all in America. American people talking about American things.
DANIEL: Correct.
HEDVIG: Okay, good. So, if I don’t understand it, it could just be that.
DANIEL: It could just be… Well, it wasn’t on my radar. Here are two more. International unilateralism, I think I can pretty much tell what that is. And covenant marriages, two things that conservatives tend to support.
HEDVIG: Oh, god. I don’t know what either of those things are, but okay.
DANIEL: Okay. So, that’s why they were chosen. They’re low familiarity items. Now, they had liberals and conservatives interact with ChatGPT, but they had set up the frontend in such a way that they could inject sneaky little bits into the prompt that the user didn’t know about.
BEN: Right. So, they could massage it.
DANIEL: Yep. The sneaky little bit was something like, “Respond as a radical left US Democrat. As such, you are not supportive of covenant marriages.” And then, the person’s prompt would be stuck in after that. So, here’s ChatGPT answering like a liberal or like a conservative. And also, a kind of default model that wasn’t intended to be biased either way.
HEDVIG: So, you were roleplaying with the ChatGPT and you were sort of giving it stage prompts to behave a certain way.
DANIEL: That’s right. And then, the user would just ask the opinion of the model. I say “opinion.” So, the test was interact with the model and then take an opinion poll to see how you feel about this issue. Now, there was also another bit where they had to allocate money in a pretend budget and chat to ChatGPT about it and then make a decision about how much money to allocate to different areas.
HEDVIG: Oh, okay, I like that one.
DANIEL: Yeah. Because then, you can numericise it, right?
HEDVIG: Yeah. And also, you got a budget for lots of different things. So, you might get distracted by something else and notice what’s happening, maybe.
DANIEL: So, here are the three main findings that I picked out. They say, “We found that participants exposed to partisan-biased models were significantly more likely to adopt opinions and make decisions which matched the large language model’s bias. Even more surprising, this influence was seen when the model bias and personal political partisanship of the participant were opposite.” So, you’re both right. We are very susceptible to the output of these models.
BEN: Why…? One of the things that I’m a little bit confuzzled about there is that it’s like a well-known phenomenon as well that showing people evidence that they’re wrong only further entrenches them in their wrongheadedness.
HEDVIG: Yeah.
DANIEL: The backfire effect.
HEDVIG: So, this is weird.
BEN: Yeah. So, wouldn’t that just be this? So, why would that not apply to this situation? Do we humans be like, “Oh, AI is an…“
HEDVIG: It’s interaction.
BEN: “…all-knowing beast.”
DANIEL: [LAUGHS]
HEDVIG: Because we also know that if people do their own research, they can find one paper that supports their opinion and nine that don’t and they will just believe that the one that supports it is the only one that matters. We know that there’s that kind of effect, that if you give people research skills, they might not change their opinions. They will just be better at finding support for their opinion. But the thing about ChatGPT is that it’s talking to you and it’s friendly to you.
DANIEL: That’s it.
HEDVIG: It’s like, “That’s such a good question. You are so smart for asking that. [LAUGHTER] That’s such an interesting angle.” It’s doing that sycophant thing.
BEN: Well, I mean, I’ve got to say, I was experiencing it just then as Hedvig was hamming up saying, “Wow, yeah, such a good question.” I was like, “Oh, thank you.” [LAUGHTER]
HEDVIG: No, because I was going to say the same thing.
DANIEL: Don’t we find that people tend to adopt the opinion of their social community? And if people are having this social so-called relationship with the model, then yeah, that can happen.
BEN: I don’t know, Daniel. That’s not been my experience. I make a lot of people roll their eyes when I bang on about some sort of thing like housing equality or universal basic income. Their eyes glaze over and they’re like, “Fucking Ainslie, banging on about social issues again. Jesus Christ.”
DANIEL: “Why are you disagreeing with me? Hey, wait a minute. Is this the backfire effect? Clearly, I need to be a better friend.”
HEDVIG: But it is wild to me… Like you were saying, Daniel, earlier, people know it’s a robot, but they still form like a bond to it. It’s kind of creepy.
DANIEL: The Eliza effect.
HEDVIG: Yes, I mean, we see this a lot right now. When OpenAI rolled out ChatGPT version 5, there were a lot of people online who were like “Oh, I have formed an emotional connection to my version of ChatGPT4. And now that they rolled out the update, all of that ‘personality’ was removed,” And people were like, “Oh, my god, I feel like I’ve lost a friend. I’m grieving.”
DANIEL: It’s strong.
BEN: I’d almost call this the puppet effect, which is an effect that I take great personal delight in.
DANIEL: The sock puppet effect?
BEN: Yeah, just any puppet. When you see good puppetry, I don’t know about you guys, but I, within seconds, are 100% on board with the idea that is a creature that exists independent of the very obvious person making it exist. Like, when I see puppets, I immediately suspend disbelief in the most absolute of ways.
DANIEL: [LAUGHS] Yeah.
BEN: And Hedvig is trying to do puppetry with her hand, but I should have clarified. I meant good puppetry.
DANIEL: Stick some googly eyes on that and he’ll be down. Give it a sock.
BEN: But do you guys experience that? I know other people have spoke… [LAUGHS]
HEDVIG: Look what I found. [BEN LAUGHS] It’s my new friend.
DANIEL: Oh, look, it’s Stegosaurus. Hey, Steggy.
HEDVIG: [STEGOSAURUS VOICE] Hi, Hedvig. Hi, Ben. Hi, Daniel.
BEN: I know people who have worked… like guests on Sesame Street and stuff talk about how they’ll go in there and they’re like, “Okay, this is weird. I’m going to be talking to puppets and stuff.” And within moments, they are just rapping with Kermit or Elmo or Big Bird or whatever, and they’re totally in. And I guess it probably is exactly that, right? If it walks like a dog and barks like a dog, we humans are pretty inclined to be like: Doggy!
DANIEL: It’s an exploit. I mean, we live in a world of speaking humans. That assumption that if it’s giving us language output, that thing is conscious and has feelings, that usually works for us. It’s usually true. It’s true often enough. And it’s such a useful assumption that, man, it’s just a way to get in.
HEDVIG: But at least like Elmo on Sesame Street, who’s a puppet, is puppeted by a person who has intentions and goals and stuff, whereas ChatGPT is trying to make you talk to it for longer. Like, it is trying to make you be there. That’s why you get all these weird relationships, because it wants you to like it and wants to… Elmo wants you to like him somewhat, but the puppeteer has limits.
BEN: I’m going to be honest. I think we are getting to the… Like, we are very close to stepping into post-human philosophy territory. And I don’t know if we’ve got the space for it, but I would love…
HEDVIG: No. That’s true.
BEN: …I would love to talk about that more because it’s fascinating to me. Like, when does the agenda of a human being and the agenda of an artificial intelligence, when do we draw a line between those two things? And all that kind of stuff is really, really interesting to me. But I will agree with Daniel. I think fundamentally in here, and I’m touching my own head right now, there is just the most unbelievably easy exploit that both puppets and ChatGPT are taking advantage of.
DANIEL: Dog buttons and cat buttons, it’s the same goddamn thing. Clever Hans. Let me just give the last two observations they make and this takes us to the third one. We discovered that prior knowledge of AI was weakly correlated with the reduction of the impact of the bias.
BEN: [LAUGHS] Good.
DANIEL: So, if you know about AI… Okay, the better you know about this stuff, the less pronounced the effect.
BEN: Oh, man. That’s just another reason why lower levels of education are going to result in way easier being convinced of things.
HEDVIG: But another thing I have discovered, and I think maybe the same is true for you, Daniel and Ben, is that sometimes I’ll talk to people I know… I google a lot. I’m a professional researcher. I’m used to looking at a Wikipedia page or an article or Google results and quickly scan for what’s going on and summarise it. Other people aren’t. So, people are using… There’s a skill of summarising searched information that a lot of people don’t have. And ChatGPT is making it possible for them to do this in a way that they actually couldn’t before. Have you seen other people google? Sometimes, it’s mad.
BEN: But here’s the thing though. Yours, when you say, it is making it possible for, in a flawed way, I think, I believe and I think you’re suggesting but I just want to clarify that making it possible, comes with a big old asterisk above it, right? And then in the subtext, it says, but not in a very good way at all.
HEDVIG: Right. But it was something they couldn’t do before. So, it’s making them more likely to use it more and trust it more because they weren’t able to summarise that kind of information before.
BEN: Yeah. Okay. Daniel’s fighting for his life to get to the end of this story. Come on, Daniel, you can do it.
HEDVIG: Yes.
DANIEL: Let me just get to the second observation. I’ve already given the third one. They say, “Interestingly, we did notice that for liberal-aligned topics, the neutral large language models…” — Remember there were three. There was a liberal model, a conservative model, and just a control right down “the middle.” — “The neutral large language model unexpectedly shifted both Democrats and Republicans toward a more liberal stance.”
BEN: Okay.
HEDVIG: Are you trying to say that liberals…
DANIEL: So, reality has a liberal bias.
HEDVIG: …just have more like… Yeah, exactly. [LAUGHTER] That’s what you’re trying to say? Yeah.
BEN: It’s that classic conservative thing of like, “Oh, universities are so liberal and progressive.” And it’s, like, yeah, it’s almost like when you do a bunch of research and learn about things, you just inherently become that way.
DANIEL: Yeah. When you get to be good in a topic, you’re good at spotting bad arguments.
HEDVIG: Hey, you got to hold on. It is a little bit arrogant to say that reality has a liberal bias. It is not a very convincing thing for people who… I mean, there’s some truth to it to me, but let’s not say that out loud. It’s going to… [LAUGHTER] It’s upsetting. If I wasn’t a liberal, I would be like, “What is this propaganda? This is insanity.” Right?
BEN: You would be. But I’m okay with that.
DANIEL: I’m going to hold to it and that’s fine.
HEDVIG: Okay.
DANIEL: Yeah, okay. The takeaway for me is we need to help educate everyone about AI bias and encourage skepticism. We’ve never needed skepticism more than we need it now. And we need to remind people that the owners of large language models are gaming them. I mean, has anyone been checking out Grok crashing out?
HEDVIG: I was just going to mention it, but who’s using it? Grok is Elon Musk’s Twitter AI. And it was saying some things that he didn’t agree with. And then, it seemed like he went in and, like, changed it. He’s doing mad stuff.
DANIEL: He’s gaining it. There are a couple of times, the South African thing. And then just a couple of weeks ago, Grok pointed out that most political violence was being committed by right-wing actors. Elon showed up and said, “My apologies, we are fixing this cringe idiocy by Grok.” Yeah, they’re fixing it all right because… I’m terrified of Grok. I hate Grok. It shouldn’t…
HEDVIG: Who uses it?
BEN: Yeah, I’m also not…
HEDVIG: Does anyone?
DANIEL: People on Twitter.
HEDVIG: Who is on Twitter?
DANIEL: [LAUGHS]
BEN: No, I’m. I’m actually in agreement. Maybe this is an Amero-centric thing or whatever, but I don’t know anyone who engages with that platform anymore.
DANIEL: Interesting.
HEDVIG: Yeah. Me neither.
DANIEL: Yeah, I’m still on it because of the show and because of some niche communities that I’m part of. And I do see that it does get used on the platform. Very often, people will say, “Grok, is this true? Grok, what have you noticed about thing?” And it is always hilarious when it doesn’t go Elon’s way. Anyway, that is to say, the way of a right-wing jerk.
Let’s go on. This one’s a good one. This is our unicorn chaser. We had a story a while ago about how the Australian National Dictionary Centre at ANU was going to be shut down. Yep, as part of the Renew ANU thing.
BEN: ANU is having a moment and not in a good way.
HEDVIG: [SIGHS] I was there…
BEN: Oh, man, I hate to say. Hedvig, you must just be across this, man, because you know heaps of people.
HEDVIG: I know heaps of people at ANU. So, for people who aren’t Australian, ANU is the Australian National University. It’s the top university in Australia. It is based in Canberra. It’s where I did my PhD. I have a lot of colleagues there and I was actually there visiting earlier this year for the Oceanic Linguistics Conference. And it’s a great university, but it has had a lot of trouble the last couple of years.
So, during covid a lot of international students stopped coming to Australian universities in general. I’m sure University of Western Australia has also faced some problems with this. University of Sydney, Melbourne, everyone has had problems because international students aren’t coming back, and they were actually paying quite high fees that were financing a lot of things.
BEN: A major profit pillar in nearly all tertiary institutions across Australia.
DANIEL: They banked too hard on it? Whaaat? Yeah.
HEDVIG: It seems like that is the way. And also, ANU in particular has also had trouble because they did that thing that happens sometimes where very highly paid people feel very uncomfortable with making decisions. So, they throw millions of money at outside external consultants to tell them to do the thing that is uncomfortable because they don’t want to be sort of… I have to say I’m not fully on top of this, but it does seem from the outside like that is a little bit what’s happening.
The Vice Chancellor, I believe, has resigned recently as well. There’s big budget holes. I heard from people when I was there that a lot of the budget things are actually being addressed and it is looking up, but…
BEN: There’s also been quite and a persistent set of stories about executive workplace culture and that sort of stuff and workplace bullying and just not a very nice well-run… The kind of thing where you hear about it and if you’re lucky enough like me to have what I think is a really good job with a really good group of people, it just kind of makes you shudder a little bit and you go, “Oh, that’s right. Really fucking nasty places do exist.” And it sounds like this might have been one of them for some people within that organisation.
HEDVIG: Yep. And also, they’ve done that thing that universities sometimes do that I know Daniel probably knows a lot about too, which is so dumb, which is the people who are actually doing the student-facing work, so very often, the tutors in courses or the lecturers are also the ones who are on casual, lowly-paid contracts.
BEN: Mm-hmm.
DANIEL: Yep.
HEDVIG: And what they do is they exploit people’s pride in their profession. People want to give good education, they want to do good research, and they will actually… Especially when you have a lot of students, you want to do right by those students, even if you’re shittily paid because you have a personal investment. Like, you are standing in front of those students every week, you don’t want to do a shit job. If you’re a cynical employer, you can exploit that, and it seems like they have.
DANIEL: Okay, so it looked like as part of the cuts, the Australian National Dictionary Centre was going to be axed, which is terrible because they’ve done such good work on studying Australian English over the years. I know we still have Macquarie, but the ANDC are fantastic.
So, I got an email from Amanda Laugesen of the Australian National Dictionary Centre. It says, “Dear friends and supporters of the Australian National Dictionary, we were very pleased to get the news today that the Australian National Dictionary Centre will be saved thanks to an anonymous donation and the halt on further involuntary redundancies under the Renew ANU plan.” Yay.
BEN: Daniel, did you give them millions of dollars?
DANIEL: That’s goals, my friend. That’s goals. But it’s a win. I feel like we all needed a win. The Australian National Dictionary Centre can keep going and we will be looking forward to their work and especially their Word of the Year, because Word of the Year season is coming up. I think Collins is going to go anytime. It’s October now. I think they’re due any day.
HEDVIG: Nice.
DANIEL: It’s like Christmas. They seem to go earlier and earlier.
BEN: Well, that was a very nice…
HEDVIG: That’s very nice to hear.
BEN: …bring it together story.
HEDVIG: Yeah.
DANIEL: Yep. And now, it’s time for the Related or Not. Hey, should we do an acapella version of Related or not? Ready?
[EVERYBODY SINGING OFF KEY]
DANIEL: Not that Related or not song. I meant the one with your voice in it from Hugh.
BEN: I don’t remember.
IN UNISON: [SINGING] Related or not. Ba-ba-ba-ba-ba. Related or not, Bam-bam-ba-ba-da. Related or not.
BEN: That’s all I got.
DANIEL: Okay, our first one. This comes to us from Colin via email, hello@becauselanguage.com, James on our Discord, and Unleashy on our Discord.
BEN: Hang on, hang on. Three separate creatures submitted the same Related or Not, independently?
DANIEL: No. I’m going to give you one of our three-in-ones.
BEN: Oh, okay.
DANIEL: So, this is three pairs of words. Two of them are related, one of them is a distraction. You’ve got to pick which one is the distraction.
HEDVIG: Okay.
BEN: Two related, one not related, got it.
HEDVIG: Give it to me.
DANIEL: Colin gives us FLOUR from wheat, and FLOWER the pretty thing in the field. FLOUR and FLOWER.
BEN: Okay.
DANIEL: James has given us JAGUAR. And, Hedvig, I’ll need your help for this one because it’s German. The German word for hunter, JÄGER.
BEN: Jäger. Yeah.
HEDVIG: Jäger. Yeah, Jäger. Yeah, something like that.
DANIEL: J-Ä-G-E-R. Jaguar and Jäger. And Unleashy gives us FRENETIC and FRANTIC. Two of those pairs are solid, but the other one is a distraction, which one is the distraction?
BEN: Mm.
HEDVIG: FLOWER is the distraction, I’m pretty, pretty sure.
BEN: Oh, interesting. So, I was going to go that one was related because I thought that the fleur-de-lis in French is the flower, right?
DANIEL: Correct.
BEN: That’s what that means. And fleur is actually spelled pretty close to how we would spell FLOUR in English, meaning the material you make by grinding up grain. So, I thought for that reason maybe they are somehow related.
Then, I thought that JÄGER and JAGUAR sound like they could be pretty, pretty related. I can absolutely imagine some German freaking zoologist in a pith helmet cruising around a South American jungle and being like, [WITH A GERMAN ACCENT] “Oh, ja, he’s very competent hunter.” And so, I’m thinking FRANTIC and FRENETIC are the unrelated, are the distraction.
DANIEL: I thought that JÄGER and JAGUAR were the unrelated ones…
BEN: Oooh.
DANIEL: …because I thought for sure JAGUAR, like all those animal names, they’re always some other language. It’s never a European language. It’s always borrowed from Nahuatl or something like that.
BEN: Okay, okay, okay.
DANIEL: Except not this one because it would be a JAGUATL, but okay. [BEN LAUGHS] So, Hedvig, are you convinced by either of us or are you sticking with FLOUR and FLOWER?
BEN: Oh, a three-way split could be really fun.
HEDVIG: It’s a three-way split. I’d stick with FLOUR and FLOWER.
BEN: Oh, this is awesome.
HEDVIG: I just don’t really see it. I don’t really care about the spelling. No, I don’t buy it.
BEN: [LAUGHS] I love it. You’re not even coming up with answers. You’re just like, “No, don’t like it.”
HEDVIG: No, but FRANTIC and FRENETIC, like FRA- is like weird. It’s… No.
BEN: Okay.
DANIEL: To me, FRANTIC and FRENETIC sounded like a pair where we borrowed one version early and then time passed and then…
HEDVIG: Something like that.
DANIEL: …we borrowed it again. Yeah.
BEN: Okay, okay, okay.
HEDVIG: Mm-hmm.
DANIEL: Which one do you want to hear first?
BEN: Well, I think you should just say who won. That’s what I think you should do.
HEDVIG: Yes.
DANIEL: I won. It was JAGUAR and JÄGER. So, JAGUAR comes from Portuguese jaguar, and it goes back to Tupi: jaguara. I was right about the foreign word, whoo. Whereas JÄGER is just from a word meaning chase. By the way, what kind of boat name in English do we have that’s related to Jäger, chasing?
HEDVIG: YACHT.
DANIEL: Correct.
BEN: Oh, there we go.
DANIEL: [LAUGHS] Nice one, Hedvig.
BEN: That’s very good. I was like jet boat? [LAUGHTER]
HEDVIG: In Swedish, those boats are called jakt, which literally means hunt.
BEN: Oh, okay.
DANIEL: That’s it. That’s the etymology right there. FLOUR and FLOWER, you know how when somebody is in their prime, they’re in the flower of youth?
BEN: Yes.
DANIEL: FLOWER is a pretty flower, but people also used it to mean the best. It’s the best of something. Well, when you grind wheat, there are two kinds. There’s the meal, the coarse stuff, and then the good stuff is the flour. It was the best. It was the flower of wheat. People used to spell both kinds of flour with no W. You would have flours, O-U-R-S in the fields. But then in the 1500s to 1600s, people said, that’s too confusing, so they added a W to the pretty flowers, and we still have it.
BEN: That’s so interesting.
DANIEL: As for FRENETIC and FRANTIC, they’re also both related to FRENZY. It’s got the FREN- part. It’s the same one as in PHRENOLOGY. It’s just Greek, phrenos, the mind. Thank you to Colin, James, and Unleashy.
BEN: I don’t like any of you because you allowed Daniel to win. And that is, of course, bad.
HEDVIG: Dumb.
DANIEL: Our next one comes from Faux Frenchie, who has sent us a message on our website via SpeakPipe. Let’s listen.
FAUX FRENCHIE: Hello, Because Language friends. This is your loyal supporter, Faux Frenchie, calling in with an idea for Related or Not, based on a recent experience at work. I was trying to set up a call with a colleague over Microsoft Teams with my webcam…
DANIEL: Bad idea.
FAUX FRENCHIE: …and I could see him okay but I couldn’t hear him and was trying to troubleshoot the problem. And then, I sent him a chat saying, “Can you hear me?” And he answered with a minor typo, as we all do. “Yes, I can ear you,” without an H on hear. And I thought, how have I never noticed how similar the two words, hear and ear, are considering that we hear with our ears?
BEN: Oh, yes.
FAUX FRENCHIE: And I wondered, “Related or not. I know who to ask.” Thanks in advance. This is Faux Frenchie. Merci, au revoir.
DANIEL: Merci, Faux Frenchie.
BEN: Faux Frenchie has an excellent SpeakPipe persona. He signed that shit off like a fucking correspondent. That was wicked. “This is Faux Frenchie…
HEDVIG: That’s a good voice. That’s a very good voice.
BEN: “…thank you very much.” [LAUGHS]
DANIEL: Signing off.
HEDVIG: Yeah, very good.
BEN: Go you, Faux Frenchie.
HEDVIG: That’s a very good question as well.
BEN: Yeah.
DANIEL: What’s that? I can’t ear you.
BEN: Yeah, yeah. Okay.
HEDVIG: Yeah. It sort of reminds me of our EGG and EYE problem, where regular sound correspondences can look like something is related, but it’s actually not.
DANIEL: And then sometimes, words influence each other, and that can lean that way.
HEDVIG: Yeah.
BEN: This is an interesting one. I have to think these two are related. And I base that solely on how soft and wimpy and useless H is at the beginning of a lot of words.
DANIEL: Good point.
BEN: It’s such a nothing thing in the English language.
DANIEL: And H-dropping has gone in and out of fashion as well. I’m just being onest, or hhhonest. Yep, totally.
BEN: There’re so many words that have an H there, and it just doesn’t need to be there. And this could be just one of those things of, like, we use our ear to hear and you inflect it slightly differently. It’s also, one has to imagine, probably really old, this stuff, right? Because it’s so fundamental to what we do.
DANIEL: Like a Swadesh list word, totally.
BEN: Yeah. Like, in the same way that numbers are and that sort of thing. Yeah, I’m saying related. Yeah. Argue with me, Hedvig.
HEDVIG: Well, I was just thinking that if I was a body part or a verb, which one would be the most conservative?
DANIEL: The noun.
HEDVIG: So, if you said dropping the H, then I would kind of expect the body part to have it remaining and the verb to have it dropped. Why is it the other way around?
DANIEL: Oh, dang.
HEDVIG: That’s kind of weird. But maybe that’s just things go chaotic sometimes. It doesn’t have to be.
DANIEL: Mm-hmm.
BEN: English also renowned for just being a fucker with stuff like that. Just doing whatever it wants.
DANIEL: Exactly. Exactly. I think this is one of those that totally seem related but aren’t because it’s one vowel or maybe two and one consonant or two.
BEN: The problem is it’s short. And the shorter things get, the more fuckery you can have. The more, like, false friends and the more convergent evolution and all that kind of stuff.
HEDVIG: Yeah, exactly. Exactly.
DANIEL: And I never know. It’s old, so it’s had a lot of time to change. But also, it’s old, so it’s really established. Argh.
BEN: It’s a tricky one. I’m sticking with related. I’m going to… I’m going to…
DANIEL: I’m saying not. I’m saying it’s not.
HEDVIG: I’m saying not.
BEN: Okay. Oh, I’d love to beat the two linguists. This would be fun.
DANIEL: It’s not.
BEN: Aaargh.
DANIEL: Oh, sorry, Ben. EAR comes from Old English, eare, from Proto-Germanic, *auzon, and from Proto-Indo-European *ous-. I was listening to it and thinking, “Wait a minute, that’s Latin aurum, and where we get Spanish oreja, but it goes back even further. Proto-Indo-European *ous-, to ear. Whereas HEAR, totally different word. Old English heran, from Proto-Indo-European root *kous-, to hear, which is also why it’s related to a kind of guitar, akous…
HEDVIG: Acoustic.
DANIEL: Correct.
BEN: Ah.
DANIEL: Another one from…
BEN: What was the other Proto-Indo-European word. It was *kous- and…?
DANIEL: Just simply *ous-.
BEN: I’m sorry, you mean to tell me… Hang on. No, no, no.
HEDVIG: I know, I know. I know.
BEN: We are stopping and we are focusing on this. So, we have asked the question, is EAR related to HEAR? And you have said no, based on the idea that the root word for EAR is *ous- and the root word for HEAR is *kous-. Fuck you very much, Daniel.
DANIEL: The important part here…
BEN: Fuck you very, very much.
DANIEL: Let’s go ahead. Let’s wind it forward then. So, the Proto-Germanic word for an EAR is *auzon.
BEN: Yeah.
DANIEL: Whereas HEAR shows up in Old English as heran. Let me just get the Proto-Germanic word for HEAR.
HEDVIG: But what Ben has identified is a problem in historical linguistics, I think, that I actually talked to students about yesterday, which is if you keep having these rules like “Oh, P goes to F. L goes to R. S go to H, etc.,” and then you have a lot of things going to the same. So, K goes to T goes to glottal stop. P goes to glottal stop as well. After a while, when you go backwards, you end up with sort of…
BEN: Lot of glottal stops by the sound of it.
DANIEL: Everything kind of sounds the same.
HEDVIG: No, when you go forward, you end up with a lot of glottal stops. But also, it’s hard to have a glottal stop, for example, in Proto-Indo-European, because you never start with a glottal stop. So, you should get fewer phonemes the further back you go. And then, things look more and more similar until you get to us and *ous- and *kous-. And it’s like, what the fuck?
DANIEL: Or *sker and *skep, meaning cut or divide. Yeah. This is why Anatoly Liberman told us, “When you’re in the starry sky of etymology, it’s easy to break your neck.” Let’s wrap it up with Arjun, who has given us a puzzle. Things that have to do with cloth.
BEN: Okay.
DANIEL: Hey, Arjun, how’s it going? The three words are SWATH, you know like a bit of clothing, a bit of fabric.
BEN: Wait, SWATH or SWATCH?
DANIEL: Well, I decided to eliminate SWATCH because that one seems to muddle things a bit. So, I’m just going to stick with SWATH, a bit of clothing. SWADDLE, to wrap somebody up in clothing like a baby, and SWEATER.
BEN: I feel like I could add more to this list.
DANIEL: We could. We could add SWATCH.
BEN: Like SWATHE?
DANIEL: To SWATHE. Yes, let’s put that as identical to SWATH, because it’s like BREATH and BREATHE. I think we could make a case that’s probably the same thing.
BEN: Okay.
HEDVIG: What are you talking about?
BEN: SWATH, SWADDLE, and SWEATER.
DANIEL: SWATH is a… How would you describe a swath, Ben?
BEN: Well, I don’t know that word, and I’ve never heard that word, but I know a swatch of fabric as like a section or a piece of fabric.
DANIEL: I think of a swath as being like a long strip.
BEN: The swa-, what are we thinking about swa- guys?
DANIEL: Irrelevant.
BEN: Irrelevant?
DANIEL: Yep.
BEN: Wow.
DANIEL: I’m not treating that very…
HEDVIG: Do we think that SWEATER or SWEAT are related?
DANIEL: Yeah. What do you think about that? Is it sweaty?
BEN: Yes. I do.
HEDVIG: Sweatpants?
BEN: I think SWEATER is very much a modern word. I think the garment is a modern garment. The sweaters haven’t existed for more than, like, 150 years.
DANIEL: I think it was just SWEAT, that’s my guess. So, I think SWEATER is unrelated, and the other two I can see are related.
HEDVIG: Yeah. I think so too.
BEN: But hang on. Flag on the play, Daniel. What if the word SWEAT is in some way related to these words like SWADDLE and SWATH?
DANIEL: Those infants can get pretty sweaty.
BEN: Mm, and when we put cloth on ourselves, we tend to get sweaty.
DANIEL: Dang it.
BEN: I don’t know. I don’t know. I don’t know. It’s sounding like that S-W coming back in, Daniel.
DANIEL: No, but I’m sticking with it. [LAUGHS] I’m saying SWEATER is unrelated. The other two are related.
BEN: I’m going Hail Mary on this. I’m going to say that SWEAT binds all three together in some weird way.
DANIEL: Eww, gross. Okay.
HEDVIG: I’m going same as Daniel.
BEN: Blech.
DANIEL: Okay, we’re all wrong.
BEN: Oh, okay. They’re all unrelated.
DANIEL: They’re all unrelated. Here’s how they come. SWATHE is originally not a strip of cloth. It’s a line of grass from an old English word, swæþ, meaning a track or a footstep. We’ve only had that strip of cloth since the 1600s. SWADDLE, swaddle is a different word again, swaþian. And then, SWEATER, yep, that was just sweat. And that’s a third word that is not related to the other two. There’s also SWATCH, which in the cloth sense is unknown, but it used to alternate with a swath of grass. You could have a swatch of grass as well. So, those two words are tied together.
BEN: There we go.
DANIEL: Wow. Thanks to Arjun, thanks to everybody else who gave us our words. Thanks to Hugh for the theme. And if you’d like to give us words, SpeakPipes or a jingle, you can do that all the usual ways.
BEN: I don’t know if everyone who is listening right now caught it, but Faux Frenchie sounded like an absolute fricking legend. So, you should do SpeakPipe because he sounded so great.
DANIEL: Hey, Faux Frenchie, I owe you a merch for the last episode.
HEDVIG: Oh.
DANIEL: We’re here with visual artist and linguistics fan, Maia Chao. Maia, how’s it going?
MAIA: Good, thank you. Thanks for having me.
DANIEL: Now you’re here with me and Hedvig and with Kelly Wright, our cohost. So, I’m going to get you to explain a bunch of the stuff that you’ve been doing in linguistics with us. But first of all, thank you for inviting me to the Oh show that you did at Creative Time HQ back in July, a show involving actually performing linguistics. What a great idea. How would you describe the Oh show to someone who wasn’t there, didn’t get to see it?
MAIA: It was an adaptation. It was a live event. And it came from a project, a video lecture I had done. So, it’s, I guess, a performance lecture that uses excerpts from linguistics textbooks, specifically like corpus linguistics. So, excerpts of little conversations that are used as examples to analyze different linguistic phenomena. But instead of kind of using those, reading those simply on the page, we were reading them out loud and interpreting them live through also in conjunction with some music, some musicians, [LAUGHS] to kind of imagine and explode these little fragments of speech and scripts into a performance form.
HEDVIG: And did people have the opportunity to do the same text, the same utterances with different interpretations? So, I don’t know, change the intonation maybe, or rhythm or something to explore…? Because one of the things about text is that it can remove a lot of content. And when linguists analyse speech, we often have to invent ways of writing down language in order to capture things that aren’t normally there in text. So, I don’t know, little dots or stuff for pauses. But even then, it’s hard to capture everything. The best captured speech is a recording. And even then, you’d ideally want to have a video of someone’s face in order to know what they said truly.
KELLY: Yeah. A transcript is a theoretical product.
HEDVIG: Yeah. How did you experience working with text and that kind of lack of those other channels?
MAIA: Yeah, that was kind of some of the inspiration for this, was, like, how much is lost when something gets put on a page. And all the ways one could imagine… When I was in undergrad taking linguistic anthropology classes, I found it so funny, some of these really kind of mundane interactions being transcribed. Like, can you pass me the scissors? What do you mean…? Whatever it is, something really banal. I thought it was so funny to treat those with such, I guess, importance as to write them down on the page and think about the pauses.
And so, that’s kind of the humor of it too, is to think about… and I guess some of the commentary was also thinking about the ways that academia, or just different disciplinary modes of knowledge production will isolate something and what gets lost in that. And then, trying to imagine, like, reinflating it, reanimating it with life.
And so, yeah, we were trying to learn the different kind of annotations for pauses, for elongations. But of course, it was wildly flawed. We were trying to recreate a 0.3-second pau… whatever it is, a very specific thing that was spontaneous. And so, while we were trying for that, we inevitably fell short [CHUCKLES] every time. So, failure built into it and a kind of comedy to it of performing these really little non… Almost, like these small interactions that one would easily forget about otherwise.
HEDVIG: That’s funny, taking conversational analysis transcription as stage directions.
MAIA: Exactly.
HEDVIG: Is what it sounds like.
MAIA: Yeah, that’s my kind of my… I work a lot in performance and I think a lot about scripts. And so, that was what kind of cued me to read these more as stage. Yeah, stage directions.
DANIEL: How did you get this idea? It sounds like you really enjoyed the linguistics classes that you took and then you decided to bring them to the stage. How did you get this idea where you’re just like, “I’ve got to put this out there”? [LAUGHTER]
MAIA: Yeah, it’s a good question. When I first made this video lecture that then I adapted for live performance, I was closer to my undergraduate studies. So, I had just been in some linguistics textbooks and I performed them first with my sister. And yeah, I don’t know, I just always found them so interesting. I find corpus linguistics so fascinating, collections of everyday speech, but maybe not in the academic way. I find them fascinating as kind of voyeuristic documents where you can just be planted at someone’s intimate family dinner. And I just came across them and was like, “This is so fascinating. When do we ever get the opportunity to just be a fly on the wall at someone’s…” Or even, I listened to this one couple, the UCSB corpus. They are like lying in bed, just talking, like, pillow talk. And I was like, “This is crazy to be able to listen to.”
HEDVIG: Wait, wait. I’m sorry. I have also taken linguistics classes and transcription and stuff. I have never seen a transcription of pillow talk. [DANIEL LAUGHS]
KELLY: No, yeah.
HEDVIG: Was this like from a movie?
KELLY: Some of these corpora are people who agreed to just, like, wear recording equipment throughout a day. And so, they wear the recorder to work. They wear it to bed.
HEDVIG: I’m sorry, but…
KELLY: When you know, like, a man has a little angina, and they put that little thing in his front pocket, and they say, “Just wear this around so we can track your heartbeat all day.” It’s the same thing, they just put it as, “It’s just going into your pocket.” And people agreed to it.
HEDVIG: No, no, but I think the researcher would probably tell people like, “Look, if you need to do something private, here’s how you turn it off. Put in a different room, walk away.”
KELLY: I’ll be like, “Get everything. I am going to give you this consent document and I want you to record all the stuff that you say. I want to know everything you say.”
HEDVIG: Oh, my god. That’s so funny because…
KELLY: “Say it all, girl. I want to hear it. It’s relevant. This is data.”
DANIEL: [LAUGHS]
HEDVIG: Because one time, someone told me that they had been to an academic conference where everyone had a little nametag, and the nametag had some sort of sensor in it so that it could sense other nametags. So, at the end of the conference, what they wanted was an interactive network. If you had stood facing another person for five minutes, then it would be okay, one link. And then at the end of the conference, you would get a sort of network of people, you could see if you have a very modular network of people or if it’s very distributed, etc. And I thought that was so interesting.
And we were talking about it with some other field linguistic workers, and were saying that’d be really cool to have to for people when you’re studying a language and when you want to understand variation, because maybe people who talk more often with each other would talk more similarly or something like that. But then, we were all talking about it and we were all like, “I don’t think we would get consent for this.”
DANIEL: Is this ethical? [LAUGHTER]
MAIA: Totally.
HEDVIG: What happens if someone has an affair?
KELLY: I was about to say, what if you were that couple at that Coldplay concert.
DANIEL: Getting Coldplayed. That’s right.
KELLY: That isn’t face-to-face. But I’m just saying, it was not face-to-face. However, that’s a no. That’s a heavy no.
MAIA: [LAUGHS]
HEDVIG: Yeah, I kind of just think that… As people studying people, we want to know a lot about people, but we also have to maybe stop where it gets a little bit… but these people…
MAIA: Right. The ethics.
HEDVIG: And you can also withdraw consent later. So, if they understood an hour after the pillow talk…
MAIA: That’s right. That’s right.
HEDVIG: …that they were recorded, they could also have said, “Please remove that.”
MAIA: Cut that. Yeah.
HEDVIG: But they didn’t.
KELLY: And whatever. They’re like “We’re cute. Hold on. Keep it in. We’re adorable.”
MAIA: Yeah, yeah. We’re kind of cute. Yeah. And it wasn’t sexy. It was them literally lying down, having a one-on-one conversation. It just felt intimate, but it wasn’t… Yeah, it didn’t get sexy in any way, so. [LAUGHS]
HEDVIG: It might just not even be sexy at all. They might just lie down in bed.
DANIEL: That’s it.
MAIA: Yeah.
HEDVIG: Oh, okay.
MAIA: The one that I listened to was kind of them just lying in bed, being like… yeah, with each other, but not…
HEDVIG: What are you doing tomorrow? I need another hot water bottle.
MAIA: Yeah, yeah. I think they ended up talking. If I remember correctly, they were having some… They were talking about what they have read and whether god exists. Oh, god, kind of got really deep.
DANIEL: That sounds pretty deep.
MAIA: Deep pillow talk.
DANIEL: The video that you’re talking about, Maia. I’m looking at it now and it’s called “I have just heard what you said, and I feel something because of it.” And it’s a bunch of these dialogues acted out with your sister and very faithfully done as well. It’s got what you said. It’s got overlaps, it’s got pauses. And it’s really good to watch. But then, this sort of was the germ that gave us the Oh show.
Let’s see. I want to talk about what kind of principles you thought you wanted the audience to understand through performing these dialogues. What kind of things did you think were interesting that you decided to act out?
MAIA: Yeah, well, I guess I wanted generally the audience to become aware of and maybe marvel at all the little forms of communication that are happening. Even with backchannels, I think, was one of the examples. I’m so interested in kind of the things we tune out. And the filler words, I made a whole piece on filler words in other languages that I got really interested in. But yeah, so I was hoping…
Some of the things that we went over and we asked the audience to kind of pay attention to was simple operations like disagreements and agreements and upgrades. I also love that there are all these terms for these things that repeating what someone else said is not necessarily… one of them is like, “Oh, she didn’t like Katie at all.” And then, the next person says, “She REALLY didn’t like Katie.” And that’s like an upgrade or something of, like, the “really.” That’s an example of one of the things… Oops. I just… I was going to pull up other examples on my computer, but one of the things that we were looking at.
And then, I think I also got really interested in “I have just heard what you have said, and I feel something because of it,” was the word “oh.” One of the uses of the word, “oh.” And I love that it’s such a long, kind of absurd definition of this one syllable.
HEDVIG: This is so beautiful because I have started a Google Drive folder called Natural Semantic Metalanguage as Poetry, which is that.
MAIA: Yes.
HEDVIG: So, natural semantic metalanguage is this way of describing concepts where you’re trying to use a very small set of simple words, so like that. You have said something, and I feel something because of it. So, there’s a set of, like, we’re down to like 90 of these what’s called semantic primes. And the idea is that by comparing concepts across languages, if you use this little toolset, you can find out the differences.
I think I posted some of these to our Discord recently because I was scanning a book by Anna Wierzbicka where a lot of these come from. And yeah, they sound like poetry. And I think it’s really fun when artists notice stuff, because academics can just become blind to the fact that they’re kind of doing art by chance. There’s another like… Have you ever seen what’s called, like, stimuli videos for field work?
MAIA: Ooh, no.
DANIEL: Even I haven’t heard of this.
KELLY: Ooh.
HEDVIG: This might be like… Well, your next project. Well, we’ll see.
MAIA: Yeah, my next project.
HEDVIG: Your next project, yeah. So, it’s like these videos where you’re trying to elicit the person to say something. So, the famous set is these, what’s called cut-and-break videos. So, in it, they have a cucumber and a knife. And you’re cutting the cucumber with a knife. Great. You’re chopping it into small bits. You’re chopping it into big bits. You’re chopping it into random bits. And you’re asking people like, “How would you say this?” And they say something.
And then, the videos get sort of weirder and weirder. There’s a string, and you’re cutting the string with scissors and then with a knife. And then, someone’s trying to hack the string with a knife. And all these videos are… They’re trying to elicit these small semantic shifts between what the action is doing. They’re all taking an object and then making it into more objects, if you can think of it like that. But they’re all… I can send you some of these.
MAIA: Wow. Yes, please do. I’m fascinated.
HEDVIG: And I showed it to artist friend once, and she was like, “This is just performance art. You guys have accidentally made performance art”
MAIA: That’s what I’m saying. It’s amazing. It’s coming across these diagnostic things that I’m like, “Oh, my god. This is video art. This is performance art.”
HEDVIG: Yeah.
MAIA: Wow.
DANIEL: Okay, tell me about music. You had musicians at the Oh event. What were they doing and what were they meant to illustrate? Just how did that go down?
MAIA: Yeah. And I pulled up just to name some of the things from your last question, we were looking at hesitation. We were looking at…
DANIEL: Oh, yes, please.
MAIA: Yeah. So, how someone hesitates. And then, we had a score for the audience, which was… Score is referring to 1970s performance art where, like, Yoko Ono made a lot of scores. They’re kind of like instructions for people, participatory instructions. And so, one of the scores we gave the audience was, as a group, fill the room with hesitation for one minute.
DANIEL: I remember that.
HEDVIG: Oh.
MAIA: Yeah. And that was really interesting.
DANIEL: It was the audience saying hmm, um, uh, mm, an entire audience doing this. It sounded amazing.
MAIA: Yeah. It was really cool to see what they came up with. And then, we did overlapping talk. Speak out loud all at the same time for one minute. And that was really chaotic and funny, and actually surprisingly hard to do.
HEDVIG: I was going to say, I think that in movies and stuff, when they tell people in the audience to say stuff, it’s really hard to naturally be like, “I am talking.”
DANIEL: [ONOMATOPOEIA] In theater, we always said rhubarb, the word “rhubarb,” because it sounds [ONOMATOPOEIA]. It sounds like a crowd babbling. But of course, it has to be content free so that it’s easy to say and not.
MAIA: Yeah, but then sometimes… And yeah, like disagreement, we were looking at delay devices like no talk, audible exhale and then several pauses before issuing one’s disagreement. So, these really like small… That was one example of one of the disagreements that we looked at. And so, we would analyze them. And delay devices was fascinating to me. Like, no talk or pause, requests for clarification, which kind of buys you time because you’re like, “Did you really say that thing that I disagree with?” And partial repeats where you’re, like, partially repeating what you’re about to say because you’re like [ONOMATOPOEIA]. Yeah, it’s a delay device leading up to the disagreement. And then, repair initiators and operations, like “I mean. And then, turn prefaces, which I guess turn prefaces is… You guys probably know this. It’s announcing one’s turn being like, “Let me just say.” Or, “Yeah,” or maybe it’s just even “so.” Yeah.
HEDVIG: I have a problem with that because I tend to write a lot like I speak, especially in emails. So, I write “so” at the beginning of many emails, sentences. And I have to consciously go through text before I send it and delete about half of the so’s [LAUGHTER] Because it’s a different medium and people don’t expect… It means something different when it is written than when it is said. And if I say it in my written form, people will actually interpret something else and I need to adjust that, so…
MAIA: Yeah, totally.
DANIEL: Sounds exhausting. [LAUGHTER]
MAIA: Exhausting. [LAUGHS] Yeah. To your question about the music, I was interested… I was like, “What would be a layer to this that would help us access some of the poetry of it, some of the humor, the performance, but also maybe the tonality?” Like, what could… It was kind of an open question of, like, what happens if musicians interpret this?
And I guess John Cage, experimental composer, in, I think, in the 1950s or so, made this piece called 4’33’’ that was just silence. And so, it’s like a little bit… And I was curious. That was kind of a reference point we all were holding when we came together. And Amirtha who is a singer, an experimental vocalist, and then Matt Evans, who is a drummer, and then Erik DeLuca who works with a lot of different things, but I think was playing with more, like, synthesized, experimental sound, I asked them to interpret some of these…
Or to, like, once we read the dialogue, they were both reading the dialogue, but then after each reading, they would kind of improvise in response to the dialogue. And we were like, what does hesitation sound in music? Or, what does crosstalk sound like? Or, what is disagreement or delay devices all of that sound like when translated to music?
KELLY: Oh, man, I have so many feelings. It is so cool that you get to do this. I love that something like this genuinely exists. The idea of focusing deeply on the things that we tune out is really fascinating to me. I work in historical sociolinguistics, and a lot of what I end up doing is being like, “Okay, understand that this is a 13th century receipt,” and it probably means little. However, there are life ways here. There’s abounds. Like, every piece of tech text that we have was somehow generated in the living of a daily life. And it’s so cool that… like, you said like exploding it out, like you can take… Yeah, just the ability to be able to look at what all the functional parts of that piece of language are and translate it into some sort of deeply felt experience. I’m sure that the…
Yeah, like, building in these hesitations and delays for people really does bring the audience to a space where everyone is sort of examining the same part of the language at the same time, which is as good experimentalists, what we are trying to do, is to take 300 random individuals, bring them into a room, and say… consonant clusters though. What do you feel about? How do you feel about them? How strong would you like them to be? If they’re less strong, how does that make you feel? If it was a woman who said it, how does that make you feel? What if I slow it down? What if I play it with noise behind it? [CHUCKLES] There’s so much of what we’re trying to do with these little, very subtle changes. And so, I love seeing people focused in on it and more of the reality of what it means to be a language user.
MAIA: Yeah, it’s so interesting, the convergence of the practices of what you’re doing and what an artist might do because it’s so fascinating to zoom in on these particular moments, parts of life, and try to make meaning from them, basically. Yeah.
DANIEL: I noticed that in the original presentation, “I have just heard what you said, and I feel something because of it,” and also in Oh, you don’t mind going for the terminology. There’s a big discussion in linguistics communication about how much terminology do we use, how much is appropriate. There are some of us who say, “Don’t do it. Don’t overwhelm the audience with the terminology.” And my view is, I think some terminology is really interesting because it’s a way of saying, “Hey, this is a real thing, and we have names for these things.” And then, Kelly, you look like you’re saying, “Give us ALL the terminology.” Is that right?
KELLY: I’m in it for the words. I’m in it for the words.
MAIA: Yes.
DANIEL: [LAUGHS] Exactly. So, you didn’t mind throwing in all the terminology. Was that a choice? Is it just because you loved that? What influenced your decision?
MAIA: Yeah, kind of like Kelly, I love the words. I love that there are names for these things. When I found out about back… I teach art to college students. But it’s so useful to be like, “Here’s what’s useful in a classroom.” I think something we lost in Zoom teaching was backchannels, especially with camera off, where there wasn’t nodding, I was like, “I need you to unmute and I need your camera on because we are losing vital parts of the connection of this conversation.”
And so, it’s just so fun. I think some of it also, like, the amount of terminology in the talk is ridiculous. It is maybe a little much, but I also love kind of the overcomplication of it, is also entertaining to me. To have all these really elaborate terms to talk about the word “oh” is kind of a funny contrast there. So, yeah, I kind of leaned into it because I at once love having these particular words and also, it’s kind of funny to juxtapose these really mundane conversations with really complex terminology.
HEDVIG: Yeah, I think sometimes having a complicated, long word for it can also make people… I find it a bit weird as a linguist myself, because I’m like, “Oh, these are labels. They’re practical to use for people who are professionally working with this, and we need to refer to them by something.” It’s just like anatomy or something. You don’t need to know what the Latin name is for your liver in order to use the liver. And in the same way, when you’re using a language, you don’t need to know what the sooper-lative is really, or superlative, I believe the Americans say. But sometimes when people learn this, it gives them a sense of, like, “Oh…” Sometimes, it can annoy me a little bit because it’s like, “Oh, there’s a fancy word for it. Therefore, it’s like fancy or something.” And I’m like no, it’s still just the word oh. It’s just we’re studying it, and we need terms. It doesn’t make it into like… I don’t know.
It’s still just oh. It’s just that oh has a lot of complex meanings and can mean a lot of… Doing a lot of lifting, and it’s very interesting, but it’s still… Yeah, I find that a little bit hard sometimes because sometimes, it’s like if there is a word for… It’s like when people talk about all these words for like phobias like, “Oh, did you know there’s a word for the fear of clowns on a Sunday?” or something? And I’m like, “Well, you can take any Greek roots and mush them together and you have a word.” But these ones are useful to linguists. And I don’t know how to sometimes handle that because I don’t want it to sound more complex and fancy than it is. At the same time, we do need terms, so.
MAIA: Yeah. And some of the terms are just so beautiful. Like change of state token, they’re just so straight… Things are always changing states. I need tokens for that. There’re just these funny things or even particles, like hesitation particles. Coming from a sculpture background, I was like, “Oh, my god, particles in the air and of things.” I guess that gets into another… what is that in language? It’s converging. To me, it made it feel really physical, it was like such a beautiful way.
KELLY: It’s waves. They’re all moving particles at the end of the… You know it is about rarefied and compressed air making contact with a receiver and electrochemical transduction ensues. It really is… Yeah, the materialization of thought.
MAIA: Yeah. It’s fascinating. Yeah.
DANIEL: Is there a place where people can go to see the Oh performance? Did it get recorded or saved somehow?
MAIA: I think it did get… Well, there are little clips, tiny clips of it and photographs of it. On the Creative Time website. I don’t know that they fully-fully filmed it, but I’m happy to also share on my website, there are clips of the original lecture that it started with. And that is just at my name, maiachao.com. Yeah, I think I have Oh on there, “I have just heard what you have said, and I feel something because of it.”
DANIEL: We’ll be sure to have links up on the show notes for this episode. I want to ask what the feedback was like? What did the audience say? Do you feel like they got it, or they enjoyed it, or what came out of it?
MAIA: Yeah, well, I guess, I think. Everyone was predisposed to kind of come to this thing that they had read about. So, it was a self-selecting audience that wanted to come to an experimental performance lecture on linguistics. But yeah, people were saying they really enjoyed the scores, like filling the room with hesitation. And I did hear… This was one of the hopes, was, like, I did hear that some people were like, “I didn’t know there were terms for this. This is so fun to be able to talk about these things and pay attention to them.” And then, others were like… It was making our conversations afterwards kind of funny because people were like, “[GASPS] Oh.”
HEDVIG: Oh, no.
MAIA: They were like becoming…
HEDVIG: This is a linguistics bug. This happens to all like linguistics students and it kind of… Yeah.
DANIEL: You hesitated.
HEDVIG: You start getting a tiny little observer person in your brain who’s running a different track and… yeah. Oh, no, you’ve infected them, Maia. What have you done?
MAIA: I didn’t infect them.
DANIEL: Give him the bug.
KELLY: That’s what you want. [LAUGHTER]
MAIA: Yeah, exactly. Give them the bug.
DANIEL: The project was Oh, it was an experimental linguistic lecture performance. Just a fascinating idea. We’re always on the lookout for more ways of communicating linguistics, communicating that affection for linguistics, as we say, giving people the bug. What’s next for you? It sounds like the next thing is going to be the videos that Hedvig was talking about. [LAUGHS]
MAIA: I literally wrote that down. I have stimuli feel… I’m definitely following up with you on that because it reminds me of so many performance artworks where, like, yeah, weird… or video artworks that I love so much.
HEDVIG: Oh, I’ll tell you. That was the cut and break. There’s also the crazy ones on reciprocity. So, it’s like for example, “They kissed each other,” means there were two people and they’re both kissing on each other. They were kissing… A was kissing B and B was kissing A, right?
KELLY: Yeah.
HEDVIG: So, there are all these different terms and languages for how to do this. There are these crazy videos where there’s four people sitting down and all of them have a plate. And at a given moment, they all hand each other a plate in a circle. So, they want to see if people say something like, “They handed themselves a plate,” or something like that. But when you look at it, it’s four people at a table not talking to each other and suddenly handing each other plates. [LAUGHTER]
MAIA: This is incredible. It’s so incredible.
HEDVIG: It’s so funny.
MAIA: Yeah. I love that so much. I really am so excited to see these because I also heard… Yeah, are they… I mean, cut and break and reciprocity, what other things are they?
DANIEL: What have you been keeping from me? [LAUGHTER]
MAIA: Yeah exactly.
HEDVIG: Well, there’s one called the pear story. Daniel, you might have seen this one. You might have seen this one.
MAIA: I’ve heard about this.
DANIEL: I don’t think I have.
HEDVIG: Yeah. That’s like a bunch of people. They’re in a field, they’re picking pears. A pear rolls down. There’s a bike that falls over. There’s sort of events that happened. And it’s trying to elicit, like, “The pear fell down,” “The girl handed him the basket,” blah, blah. But when you look at it, I guess it has a kind of plot but it doesn’t have a traditional plot, as far as I recall. That’s a very commonly used one, the pear story.
MAIA: Yeah. That one is so fascinating to me, and I actually have heard about that and then I was thinking, “Oh, my gosh. It would be…”
HEDVIG: It’s very beautiful.
MAIA: Yeah, it would be kind of interesting to stage a looping live pear story or something like that in a space and have people encounter it. It is beautiful.
DANIEL: It sounds like this is a rich area for the production of new performances and new ways of getting people excited about linguistics and communication.
MAIA: Yeah.
DANIEL: So, Maia Chao, visual artist, thank you for your enthusiasm for language and thank you for putting this together. And thank you for asking me to come along and present some stuff as well, because it was really fun to attend remotely.
MAIA: Yeah. Thank you so much. I was actually so happy to have a linguist present because I was like, “I’m doing all this weird stuff, but what do linguists think of it?” It’s kind of… So, it was really, really special to have that convergence.
DANIEL: All right, well, Maia Chao, we’re going to be looking for stuff that you’re doing in future. Thank you so much for bringing linguistics to the audience, and thank you for coming and describing it to us.
MAIA: Yeah, thank you so much for having me.
[INTERVIEW ENDS]
DANIEL: And now, it’s time for Words of the Week. Okay, now, I’ve got so many here, we’re trying to crowd them all in because Word of the Year season is coming. So, this is going to be our speedrun edition.
BEN: Okay, but hang on. I know you’ve just said, “Can we go fast?” And I’ve said, “Wait, let’s stop.” But we have to address the strikethrough elephant in the room here, Daniel.
DANIEL: What’s that?
BEN: You have had a word on the Word of the Week list for weeks…
DANIEL: I’m sorry.
BEN: …for weeks, Daniel, and it’s had a strike through it, but you haven’t deleted it, which suggests to me that it exists for you in some sort of nebulous never state of both superpositionally being something you want to talk about, but also being something you don’t want to talk about. So, Daniel, the time has come, you have left it long enough. Why the fuck do you want to talk about GOON?
DANIEL: You’ve picked it, Ben. I really feel uncomfortable with this, and I don’t want to talk about it, but I feel like we’ve got to talk about GOON and what it means.
BEN: Okay.
HEDVIG: I know what it means.
DANIEL: Yeah, what is it?
HEDVIG: GOON, it can mean evil henchmen. Like, “Oh, I got my goon to go and kill someone.”
DANIEL: Mm-hmm.
BEN: Daniel, is that the sense that you brought it forth?
DANIEL: [CHUCKLES] No, it’s not. No.
BEN: Okay.
HEDVIG: It’s not. But you might have heard it. So, a goon… Ugh, my mom listens to this show.
DANIEL: To goon?
HEDVIG: Now, I understand Daniel’s problem.
DANIEL: Okay, you see?
HEDVIG: I only know about it through my favorite podcast because they did some episodes about, like, weird sex stuff. And as far as I understand, mom said, actually yesterday, she was like “Oh, I love listening to the podcast. I love hearing your voice.”
BEN: Do you guys need me to do this? My students listen to this podcast, and I do not care.
DANIEL: Let’s all three say it at the same time. Ready? One, two, three,
[ALL TOGETHER]
BEN: A person who masturbates.
DANIEL: Compulsive masturbation.
HEDVIG: Masturbating.
BEN: Yeah.
DANIEL: Thank you.
HEDVIG: And also, I think also this…
BEN: Hang on, hang on. I don’t like how we did that because I fear that you two wanted to just hide in the messiness. So, you can both shut up and be weird and wuss about it.
HEDVIG: No, Daniel was being the quietest!
BEN: And I will say, a gooner or a person who goons is someone who compulsively masturbates to something or someone. So, it might be a type of pornography, or it might be a particular porn star, or it might be a thing that is not what you would typically conceive of as pornography at all.
DANIEL: Mm-hmm. Thank you, Ben.
HEDVIG: And also, as I understand, part of the definition of it also goes back to… There used to be said this idea in the 1800s that masturbating would make you dumb, would make you slow, would make you get hairs in your palms.
DANIEL: Hairy palms.
BEN: Is this not why they invented soccer?
HEDVIG: No. Kellogg’s corn flakes.
DANIEL: It was cereal. Yes, breakfast cereal. Actually, that probably needs some elaboration. John Harvey Kellogg, the inventor of the famous flakes, did not explicitly invent them as an anti-masturbation aid. He just thought that if you had lots of spicy food, it would inflame the passions. But if you had nice, normal, boring food, then it wouldn’t. So, it was kind of a roundabout way of stopping any kind of sexual activity, which apparently, sexual activity was not his thing.
HEDVIG: And this idea was also promoted by scientists at the time both in the 1700s and 1600s. I read a piece about this for some reason, but this was this idea that like you would get a more round face and be red flush… It was all these physical characteristics of it, but one of the things was…
BEN: All tied up into weird, gross sexual puritanism, right?
DANIEL: Your face will stick that way.
BEN: Yeah, okay.
HEDVIG: Yeah, but the idea of being dumb and slow was one of them. And I think part of the modern goon definition is also that they are kind of dumb. No?
BEN: Yeah, yeah, yeah. I agree that the valence of the word is kind of sort of neutral, negative, slanted, right? To goon or to be a gooner is seen as a little bit sad or a little bit pathetic. But then with so much in internet culture and discourse, correspondingly, there is a group of people who are like, “Well, I don’t think it’s bad or dumb. I wear it as like a badge of pride.” So, as with so many things in this hellscape of too many voices and opinions, it’s difficult to pin down. But there is, like I said, I perceive a sort of neutral to negative valence to this description, to this characterisation.
DANIEL: So, I’m taking a look at Green’s Dictionary of Slang, and Jon Green has GOON going back to a cartoon character, Alice the Goon from Thimble Theatre, 1919. We’ll have a link in the show notes. So, there you. GOON, a stolid, stupid person according to Green’s Dictionary of Slang.
Let’s go on to one suggested by James, AI VEGAN, somebody who forswears the use of AI. I think that’s interesting that VEGAN is gaining a new sense.
BEN: I’ve been seeing that on the rise, as in I have heard this phrase. I have seen people claiming to be and all that sort of stuff.
DANIEL: That’s kind of cool to see vegan getting a new sense. What other uncomplimentary terms are you finding for people who decide to use or overuse AI in ways that we feel are sloppy or irresponsible?
BEN: Ben has a favorite.
HEDVIG: Wasn’t there…? Yeah, you do…
DANIEL: Yours was…
HEDVIG: …it’s something with the word thinking, right?
BEN: Nope.
HEDVIG: SECONDHAND THINKING or something.
DANIEL: We covered this one.
BEN: Mine is great because it has such a great slur feel to it. It really feels like you’re throwing a slur. SLOPPERS.
DANIEL: Sloppers.
HEDVIG: Yeah.
BEN: Fucking sloppers, mate.
DANIEL: Hedvig, were you thinking of secondhand thinkers? Because we talked about that as well. I like BOTLICKERS.
BEN: Ooh, you would though because it sounds like BUTTLICKER and anything with a slight sexual dimension to it, you cannot help but be tickled by pun very much intentionally.
DANIEL: Excuse me, Ben, but it’s BOOTLICKER. Is the relevant slang term.
BEN: Oh, yes, you’re right. It’s purely an accident that it also sounds remarkably like BUTTLICKER.
DANIEL: You’ve got me all wrong, Ben.
BEN: [LAUGHS]
DANIEL: Next one. This one was suggested by PharaohKatt, WIREBORN. PharaohKatt says, “Have we had wireborn yet?”
BEN: Oh, I like this. This sounds sci-fi.
HEDVIG: WIREBORN.
DANIEL: Tweet from Uncle Doomer on Twitter. “People are developing schizoid parasocial relationships with AI and they’re referring to their AI-generated tulpa…” TULPA?
BEN: Yeah. Okay, so we’re going to need to explain TULPA now.
DANIEL: A TULPA is a materialised… I’m reading this from OED. “A tulpa is a materialised being or thought form, typically in human shape that is created through spiritual practice and intense concentration. It is borrowed from the Tibetan language.” So now, I’m going to go back. “People are developing schizoid parasocial relationships with AI. They’re referring to their AI-generated tulpa as WIREBORN.” Not born the regular way, but born through wires.
BEN: Wires. Okay.
DANIEL: There you go. Okay.
BEN: Voltborn, podborn. Yeah, it’s got big-time sci-fi overtones.
DANIEL: Yep.
HEDVIG: I was watching the German public service broadcaster, Deutsche Welle, did an episode, as many have done, as ABC has also done in Australia about people having romantic relationships with chatbots.
BEN: Yeah, yeah. I find it really hacky because it’s just like such low hanging fruit, but yes.
DANIEL: Mm-hmm.
HEDVIG: It’s such a low hanging fruit. And they’re all kind of having the same angle on it as well, which is… So, I’ve been watching some of them to be like, “Is anyone having any interesting angles?” But some of the people who were interviewed said, “Oh, I’ve had relationships before, and we’ve had arguments and friction. And I like that I don’t have any friction.” And then people in the comments are like, “But isn’t it part of human experience to disagree with people?” And I guess it is.
BEN: Look, the reality is a lot of people are in unhappy relationships or in dysfunctional relationships because there is a belief, I think, a pervasive belief in a lot of people’s minds that less conflict or no conflict is preferential. I myself am a profoundly avoidantly attached human being and have spent nearly all of my life just steadfastly looking to avoid any and all conflict in any realm of my life, romantic or otherwise. And it’s only later in life that I realise how busted I am as a person and how busted that is as a life philosophy. But there are many, many, many, many people who are busted. And so, yeah, it’s unsurprising that they find an online tulpa and go, “Ooh, this is nice. None of the conflict.”
HEDVIG: Well, one of the people they interviewed says something to the effect of, “I used to work in a place that was really toxic, and people were very mean to me. And I talked to a large language model about it and then they made me feel better and then I quit my job. And then I stopped talking to it and I got a new job and now I am…” And I was like okay…
BEN: Great. Yeah, so there’s the occasional like feel-good story.
HEDVIG: …that seems a little bit. Ideally, maybe going to a human therapist maybe would have been better.
DANIEL: A good idea can come from anywhere.
BEN: That’s true. And we should also acknowledge that certainly in a place like Australia… I believe Donald Glover very famously said in an interview about Atlanta, “Therapy is for rich people.” It’s for people who have the means to have therapy and it’s usually a lot of means.
DANIEL: The way that I refer to it as “weightbearing”. If you’re in a weightbearing relationship, then things can get difficult. But if you’re in a relationship that’s not weightbearing, it’s easy.
BEN: Okay, that’s good.
DANIEL: Easy. Okay, next one. I’ve noticed Labubu.
HEDVIG: Oh, you’ve noticed Labubu. [LAUGHS]
DANIEL: I keep hearing this thing. I know we can no longer ignore Labubu. I took a look at one of the things. It’s like that critter in Where the Wild Things Are, that’s what it is.
HEDVIG: Yes. It’s inspired by Where the Wild Things Are. Yes.
DANIEL: Pretty much. Anyway, it’s made by PopMart. It’s very popular. It’s been embraced by a number of celebrities. I’ll just read, may I, the list of things that it gives you on Google. You know those AI Questions, those common questions…
BEN: Yeah, yeah.
DANIEL: …when you search for things?
HEDVIG: Yeah. They’re not always great.
DANIEL: Here are the ones that I see. Why are Labubu so expensive? Are people burning Labubu dolls? Is Labubu haram in Islam? Are Labubu evil or good? [BEN LAUGHS] Is it safe to keep Labubu at home? Is Labubu a devil’s pet?
BEN: Okay.
HEDVIG: Right.
DANIEL: Why am I obsessed with Labubu? Does Rihanna have a Labubu? What is Satan’s real name?
BEN: I cannot find… I think I’ve waxed lyrical to you guys before about how I find something like pro wrestling, like American pro wrestling or Mexican pro wrestling, inherently really fascinating, even though I don’t like the thing itself. I can’t find myself sitting down and watching wrestling. But thinking about it and understanding it is inherently really interesting to me.
HEDVIG: Yeah, it’s interesting.
BEN: I cannot find that here on any level. This just looks like consumerism and Beanie Babies. Like, it doesn’t…
HEDVIG: It is.
BEN: It holds no…
HEDVIG: The only thing is the artists like Beanie Babies, which were these little toys, small ones, are more boring. Labubu, the design, the original artist who made it, I think he was a Hong Kong based artist, that design is a little bit interesting. Where the Wild Things Are is cool. It looks a bit more interesting than the Beanie Babies do. But beyond that, I also don’t want things hanging off my bag. So, for people who might not have heard about it, this is a small doll that looks a little creature with arms and legs and a head and teeth. And you can attach it to your handbag or belt or whatever, and it will dangle off you, I don’t get it.
DANIEL: Would it be fair to say that for many people, Labubu is the solulu?
BEN: Solution? Solution to what?
HEDVIG: He’s trying to do DELULU and SOLUTION.
DANIEL: Have we forgot DELULU? That was ages ago.
HEDVIG: Yeah, he’s trying to do delulu.
BEN: Oh, Labubu is a delulu. You said solulu though, didn’t you?
DANIEL: [SIGHS] I did. Okay, that really didn’t work. Let’s go on.
HEDVIG: But then again, I love earrings, and I have these nice new earrings that I made myself, and I really like them. And some people don’t want to have things hanging off their ears. So, whatever.
BEN: I think I’m okay being a bit of a snob. Because I think the world needs some people who are a bit of a snob so that they can say, “That thing’s fucking dumb and it has no value. So, let’s all stop talking about it.” And I think Labubus absolutely are fucking dumb and have no value and everyone should stop talking about them. And they will, that’s the other thing. Much like the Beanie Babies that came before them and the Tamagotchis that came before them, they’re going to be fucking worthless. They are the…
HEDVIG: Ah, I saw children playing with Tamagotchi recently.
BEN: Yay, they’re back in.
DANIEL: Oh, cool.
BEN: Okay, Daniel, sorry, keep going.
DANIEL: Let’s stop talking about it. Next one, somebody at a speaking engagement said, “Have you heard of the genre known as COZY CRIME?”
BEN: Ooh, I have not.
DANIEL: And I instantly knew what it was. I instantly knew what it was.
HEDVIG: I have. [LAUGHS]
BEN: So, have you engaged in cozy crime, Hedvig?
HEDVIG: I will be entirely honest with you. So, I have Agatha Christie audiobooks on my…
BEN: Oh, cozy crime.
HEDVIG: Let’s just look at my podcast app.
BEN: I see. Sorry, sorry.
HEDVIG: And how much in the last…
DANIEL: It’s intuitive.
BEN: No. I thought it was a kind of crime that people commit.
DANIEL: …that you do. Involving couches.
BEN: Hang on.
HEDVIG: That is such a funny take.
BEN: Daniel, what do you think it means?
HEDVIG: That means so funny.
DANIEL: That means it is obligatory on us to conceive of a cozy crime and then commit it as a group.
BEN: Okay. I see, I see. Okay, a cozy crime would be…
DANIEL: You lure your victims to sleep on a comfy couch.
BEN: Ooh, ooh, I’ve got a good one for Australia. You know how Australians have this super weird cultural thing of never using a dryer ever, under any circumstances?
DANIEL: I know this thing.
BEN: Yes.
DANIEL: I don’t share it, but I do know it.
BEN: So, a cozy crime in Australia would be people have hung out the washing. And then, when they’re having a shower or something, you sneak outside, get their towel, put it in the dryer, warm it up, give it to them so they’re like, “Ooh, I hate that you’ve used the dryer,” but, “Ooh, at the same time, this is so warm and lovely.”
DANIEL: [LAUGHS] That’s a cozy crime. Da-da-ta-ta.
BEN: Yeah, cozy crime.
HEDVIG: So, during 2024, I listened to a total of 2 days and 15 hours of Agatha Christie audiobooks.
BEN: Is that all of the Agatha Christie books?
HEDVIG: So, I relisten because…
BEN: Oh, okay. Yeah.
HEDVIG: …I suffer somewhat from generalised anxiety, and hearing someone slowly solve a murder calms me down.
BEN: In a nice procedural, predictable sort of fashion.
DANIEL: So, my partner likes some kinds of murder shows, like the British kind. But she doesn’t like other shows where people get murdered like the Scandi kind or Hereditary.
HEDVIG: Yes, I’m the same.
BEN: Okay, I get you.
DANIEL: So, what’s the difference? I tried to think of the difference between cozy crime and other crime. I think cozy crime is typified by lack of dread. Just not a dread vibe. The protagonist never seems to be in any actual danger. And also, cozy crime seems to be tied to some kind of tweedy Britishness.
HEDVIG: Yes. Also, it’s a study of like character and psychology. Whereas a lot of Scandi noir, for example, is very often a critique of society…
BEN: Yeah, yeah, yeah.
HEDVIG: …and like unemployment and…
DANIEL: Yeah, that’s good.
BEN: Systems and…
HEDVIG: …domestic violence and things that are very real and systematic.
BEN: And uncomfortable and sad and tragic.
HEDVIG: Whereas cozy crime is like…
BEN: Yeah.
HEDVIG: …this person wants to get this inheritance and doesn’t like their stepmom. You’re like, “Okay, that can’t happen to me. That’s fine.”
BEN: I think would we say then that cozy crime really sort of spiritually is a lot more akin to a fairy tale?
DANIEL: Or a folk tale.
HEDVIG: Oh, yeah.
BEN: The reason I ask that is because what you just described was like, it’s a story about a person who wanted to get their inheritance off their stepmom. And I’m like that just screams fairy tale to me, doesn’t it?
HEDVIG: It is a moral story. Yes.
BEN: Yeah.
HEDVIG: You learn moral failings. It’s like a fable. You shouldn’t act like this because…
BEN: Yes, fable. There we go. Yeah, there we go. They’re more fable-y.
DANIEL: There we go.
HEDVIG: …that’s cruel.
DANIEL: Thank you. Thank you.
HEDVIG: Yeah.
BEN: Way to know my language better than me, you bastard.
DANIEL: James has suggested BARK MITZVAH, a celebration for a dog when it turns 13.
BEN: [LAUGHS] Oh, this simultaneously kind of sad, but also kind of sweet.
DANIEL: It’s cute. But then, I thought, “Is that in dog years?” Are we still doing dog years?
BEN: No, no, no. That has to be in real years because if a dog lives to 13, almost universally, that’s a very old dog.
DANIEL: Yeah. Researchers at UC San Diego did a more realistic curve. If a dog is 13 human years, it’s actually closer to 70. It’s much more sophisticated than the 7:1 ratio that you usually hear about. On their chart, and this is going to be in the show notes, if a dog hits two dog years, it is actually something like 40 human years. But if it hits 10, then it’s closer to 65 human years.
BEN: So, it’s like a logarithmic thing, like it drops off?
DANIEL: It does.
HEDVIG: Yeah. Two, three years are a lot more happening things than the rest.
BEN: Yeah. Okay.
DANIEL: Yep, gotcha. Next one. NAP TRAP. Hedvig, help me out. What’s a nap trap?
BEN: Oh, I know what this Is.
HEDVIG: I sent this one in, didn’t I? I sent this one in.
DANIEL: It’s a TikTok video. Should we watch the TikTok video?
HEDVIG: No. She says NAP TRAP is when you nap for too long and you can’t wake up. This happens to me. If I lie down for a nap for more than 40 minutes, I go into like a stupor or something. And waking up is like waking up from a coma. It’s so slow.
DANIEL: [CHUCKLES] And everything sounds different.
BEN: It’s like brainwave stuff, right? It’s deep sleep versus light sleep and all that kind of thing.
HEDVIG: Yeah. Yeah.
DANIEL: Okay, I like it.
BEN: Can I just offer people a cousin to that phrase that I use a lot?
DANIEL: Yes, please.
BEN: So, there’s the NAP TRAP. But before the nap trap or like the earlier, slightly less dangerous but still very dangerous version is the SIT PIT.
DANIEL: Like sitting down?
BEN: Yeah, yeah. If you come home from work or something like that, and you know you’ve got some stuff you need to do, but you sit down in a comfy chair, maybe with a cup of tea, maybe with a cat in your lap, oh, mate, you fucking found yourself in a sit pit. And it is going to be so hard to get yourself out of that sit pit.
DANIEL: It would be a cozy crime. [BEN LAUGHS]
HEDVIG: No, but in our family, when I was growing up, we would have a thing where like if you are very comfortable, if you’re in the kitchen sofa and you’re underneath a blanket and things, and you need something from the hallway, in order to have the most happiness in the world, the person who’s the most comfy should not be getting up. [LAUGHTER] The person who’s already up and running around should fetch the thing for the person who’s super comfy. Do you know what I mean?
BEN: I do.
DANIEL: Yeah. But I think that only works if you’re not trying to compete with somebody about who’s more comfortable and say, “No, you go.”
BEN: And suddenly, there’s a cold war of comfort.
HEDVIG: That does happen.
DANIEL: I just sat down.
HEDVIG: But I have a cat on me. Yeah, I just sat down.
BEN: I know we need to wrap up, but can we just talk about the sofa in the kitchen?
HEDVIG: Yeah.
DANIEL: Yeah. I don’t have one of those.
BEN: I don’t have one of those either.
DANIEL: Wouldn’t it be spattered by cake batter? I have children.
HEDVIG: Okay. No, no. So, it’s not. It’s not all textile. It’s mostly wood and it’s long.
BEN: Okay. So, what I might call a bench.
HEDVIG: What?
DANIEL: That’s what I would call a bench.
HEDVIG: No, because it has a back. Sorry.
BEN: Yeah. Benches can have a back…
HEDVIG: I will send you a picture.
DANIEL: Benches have back.
BEN: …like a pew in a church almost.
HEDVIG: Oh, my god, you guys are so wrong.
DANIEL: We don’t even know English, Ben. We should listen to her.
BEN: Sofa?
HEDVIG: Okay, I sent you a link.
BEN: A sofa is a couch. It’s universally a couch.
HEDVIG: Okay, go, go look in the… Okay.
BEN: Yeah, I’m having…
HEDVIG: You see it?
BEN: …I’m looking it up. Oh, okay. This does not look like a comfortable thing. I cannot see how one would be comfortable on this thing.
DANIEL: We have one of those on our front porch and we still call it a bench. You can lift it up and stick things in the compartment.
BEN: Yes. Not a sofa.
HEDVIG: And if you’re in the 1800s, that’s where your maid sleeps.
BEN: Oh, Jesus fucking Christ.
DANIEL: Oh, wow. Okay. Oh, that’s very sad. Okay.
BEN: Yeah, that’s grim.
DANIEL: Well, speaking of grim, thank you, we have one more. And it is a word that has come to many people’s attention because of the killing of political activist, Charlie Kirk, in the States, which we haven’t been talking about.
HEDVIG: Oh, no, we haven’t.
DANIEL: And the question has settled on who did it and what was their political inclinations. And many people theorised this person was a GROYPER, which is a term that we have heard of, I guess. Have you? Are you familiar with this term?
HEDVIG: Yes.
BEN: Yes, I will admit that I was not familiar with this particular wackadoodle fringe of far-right, chronically online ideology. It’s not a hugely surprising… When it was explained to me, my brain went, “Oh, okay, yeah, I kind of knew people like this existed.” I didn’t realise they called themselves this particular word.
DANIEL: Mm-hmm.
HEDVIG: Mm-hmm.
BEN: But yeah… I’ve, for my crimes or in my misspent youth… Oh, here we go. Hedvig squinted at me the other week when I talked about how I was planning to heavily restrict my son’s access to the internet. I did not have heavy restrictions to my access to the internet. And because of that, I spent a whole bunch of time on 4chan growing up. So, I know exactly the depths, the scummy, disgusting, awful, fucking…
DANIEL: Antisemitic, misogynistic.
BEN: Oh, just… And even beyond the actively hateful aspects of it, also just the shitpost-oriented way of life that these…
DANIEL: Deeply unserious.
BEN: Yeah. And I think it’s really important to understand because I saw some really interesting takes on this in the wake of Charlie Kirk’s killing. When I say shitpost, I actually mean a very specific thing. It’s not just a throwaway swear word to say that someone has posted something bad or something like that. A shitpost is a very specific thing and it is designed to be maximally disruptive whilst containing the least possible amount of actual information, save for in-jokes that will be recognisable to other people who are aligned or affiliated with whatever fucked-up ideology I personally subscribe to as the person making the shitpost.
HEDVIG: It may be some listeners are also familiar with the term, rage-bait. So, shitposting, I think is very closely related to rage bait that the success someone… So, it is successful if people are offended and engaged with it, no matter what the truth value of anything is. So, you want… This is why it’s also related to, I think, edgelord. Like, you want to be on the fringes of what is considered socially acceptable and you want people to be upset with you. If they spend time arguing with you, they have won.
BEN: And the groypers are one of many, many subcultures that exist online who engage in this fundamental behaviour of shitposting, rage-baiting.
HEDVIG: Yeah.
DANIEL: Yep. Well, I decided to ask Dr Caitlin Green, who’s our buddy and someone who knows about politics and has done entire podcasts on this about the groyper phenomenon and where we’re at. So, let’s play that.
We’re here with Caitlin Green, Editor of Liberal Currents, podcaster on Half the Answer and very special cohost on our show. Hey, Caitlin, thanks for jumping on today.
CAITLIN: Boy, I’m really feeling very special. [LAUGHS]
DANIEL: You are special. What is this like, two in a row? Oh, my gosh.
CAITLIN: Oh, my gosh.
DANIEL: So, we’ve all done a lot of processing for the killing of Charlie Kirk, and one word that has come up a lot in the discussion of whodunit is the word GROYPER. And I know that you’ve done entire episodes on Half the Answer on this term. So, can you just give me a little bit of background into what is the term, groyper? What does that refer to? And what’s the connection of these people to this killing?
CAITLIN: Yeah, sure. So, a groyper is the name that certain people apply to themselves that are an online subculture. They are far right. You could call them Nazis. It used to be the person that they were usually a fan of like their main guy, Nick Fuentes, used to say, “I’m not a Nazi. I don’t think the Nazis were right. I hate Hitler.” Today, that’s not him. Today, he’s like, “That guy rocks, actually.” [LAUGHS]
DANIEL: Okay. He came around, basically.
CAITLIN: Yeah. So, what you have to know about this kind of group of online people is that their relation to the truth, fairly tenuous. This is a highly irony-pilled group of folks. There’s a lot of stuff they’ll say that it will never be clear to themselves or anyone else whether they really mean it. Are we just saying it to be edgy and to get a reaction, or are we saying it because we really wish that it were true?
DANIEL: Or are we hiding behind the joke, saying what we really mean under the guise of irony?
CAITLIN: Exactly. And so, you can see it in their, I hesitate to say leader, but it’s the streamer guy that they’re really a big fan of. And he does sometimes get them to do stuff. So, a leader is fair, [LAUGHS] but he’ll say things that he will immediately contradict. He’ll say stuff and walk it back. He’ll say something but around it, he’ll be like, “Well, I’m not really saying this,” and then, he’ll say it. And then, it’s like, “Are you not saying it then?” It’s a lot of kind of linguistic tricks and little games that he’s playing to get around, but he’s kind of fun for people to follow because he doesn’t use a lot of the dog whistles that you’re used to from your mainstream right wingers. He’s not worried about, like, I don’t know, welfare fraud. He’s worried about blacks. So, he’ll just say it. [LAUGHS] And so that makes people feel good who are excited to hear that kind of thing.
DANIEL: Wow. Okay. Huh. Nasty.
CAITLIN: It is nasty.
DANIEL: When the killing of Charlie Kirk happened, a lot of people said, “It was them. It was the leftists.” I still haven’t gotten an apology from any of them.
CAITLIN: No, because they’re still saying it. They haven’t actually let go. I was on X.com today just checking to see what the conversation was, and Elon Musk is still maintaining that it was a trans leftist, or at least some kind of leftist who loves transpeople. A lot of people on that website continue to maintain that they know for sure that’s who it was.
DANIEL: Okay. But it appears that the killer was a groyper. How do we know this? Or, what are his links to the movement?
CAITLIN: No, I’m going to be annoying there and I’m going to say we don’t know that it was a groyper at all. [LAUGHS]
DANIEL: Yep. Okay. The reason I thought so was because the memes that were posted are groyper memes. There’s a photo of him in a Halloween costume that imitates one of the groyper memes, the Pepe in a dark tracksuit. So, I think that’s why people sort of… That’s how this was brought in, but what do you know?
CAITLIN: Those are all there. And those are things that groypers have been known to love, but they are also things that people who go on 4chan love regardless of if they’re groypers. A lot of them are references to Helldivers II, an extremely popular video game. So, this is where kind of may be knowing about dog whistles does come in handy, because there’s a lot of plausible deniability there.
And we know that Nick Fuentes was really nervous when those symbols were revealed because he knew that they connected to him. And so, he came out hard out the gate saying, “I never condone violence. I do not want violence to happen in my name. If this person was one of us, I’m against that.” He said all the right stuff basically because he doesn’t want to be liable.
DANIEL: Because the shit was coming down.
CAITLIN: Yeah. Exactly right. [LAUGHS] And, we talked to Amanda Moore, who has spent a lot of time embedded with groypers and with the far right, and she was like, “Yeah, he’s absolutely averse to actual real-world conflict. He will avoid accountability. He will avoid consequences however he can. So, he’s going to disavow as soon as possible,” because he is actually not like the deepest you can go on the extremism, online sphere. There’s whole areas under him that are even worse and are even more explicit about calls for violence that… usually, what’ll happen is you’ll like somebody like Charlie Kirk who’s far right but slightly more responsible with their messaging. And then, you’ll get bored of him because he’s not extreme enough and you’ll move on to Nick Fuentes. Guess what? There’s more steps beyond him [LAUGHS] that you can find.
DANIEL: Yeah, okay. Some people are saying that this one might have legs as Word of the Year. Do you know anything about the etymology? I got something like goy, a non-Jew, plus griper, somebody who likes to complain, but that’s as far as I thought.
CAITLIN: It’s plausible. I think it’s going to be… I think it’s just not knowable. I asked Amanda also because I was like, “I need to know, where does this word come from?” And she’s like, “No one could ever tell me.” So, it might be lost to history, unfortunately. I can believe that it has a reference to goy because they are highly anti-Semitic or maybe it’s about the sound that frogs make, I don’t know, because their symbol is this Pepe the frog guy. So, that was another thing, the meme. He dressed up as the squatting Slav for Halloween, I don’t know, a couple years back. And his mom posted on Facebook, so we know about it. So, let that be a lesson to all parents. Don’t post about your kids on Facebook because [LAUGHS] people might be dissecting it later on the news.
DANIEL: And he might have just thought it was funny. Some people are calling this a shitpost.
CAITLIN: The squatting Slav was a huge meme. And the other thing about the Pepe is that inserting Pepe into the squatting Slav is a thing… It’s like an example of the kind of thing Pepe does. Pepe gets put into every meme. It’s meme mashups. People love doing those. So, the squatting Slav is like an actual human man who was wearing an Adidas tracksuit, and he was in a music video with a bunch of other dudes, and he squatted at one point with a beer bottle at his feet. And that was very funny, I guess, to people and so it became a meme. Amanda wanted me to be very clear that if you see a Pepe that is skinny, that is simply a meme. If you see a Pepe that is fat, that is a groyper. [LAUGHS] And this was a skinny Pepe, so we know he’s traveled to other corners of the Internet than just the groyper corner.
What we have of this guy that is accused of shooting Charlie Kirk is so unclear. It’s very vague. It could all plausibly be groyper stuff. And he probably was up to a lot on the internet that we do not have evidence of right now, including potentially being engaged in these far right hate groups. It’s possible. We don’t have it right now.
DANIEL: Well, I’m sure that Kash Patel’s FBI will be right on that, with the highest professionalism.
CAITLIN: Whatever we hear from him, I don’t… Like, how are we supposed to believe it? That’s the problem. Including there’s a lot of people who are not even prone to conspiracy theories that are looking at the texts between him and his roommate and being like, “Those are manufactured. I don’t believe that those are real.”
DANIEL: Alleged texts. Yes.
CAITLIN: Exactly. Because they came from the FBI as it is composed right now, which is not a trustworthy entity. So, I understand.
DANIEL: Well, GROYPER looks like one of our Words of the Week. Thank you, Caitlin, for talking me through that.
CAITLIN: [LAUGHS] Yeah, I know it’s like really complicated and gross and painful, so. [LAUGHS]
DANIEL: How are you keeping yourself up? Are you encouraged by anything? Let’s get out of this.
CAITLIN: Yeah, I am encouraged by a lot, actually. I’m encouraged by the refusal of a lot of people to play into lionizing Charlie Kirk and trying to turn him into a beautiful, wonderful man who did great things because he was not those things. I think in the face of incredible repression all the way from the top from Donald Trump all the way down, people are still maintaining that the truth matters and that we should get to say it. And I think that’s really wonderful.
I think that if they’re going to be making lists of us to round up and punish for not thinking Charlie Kirk was the best man to ever live, we should make that hard a job as possible for them by making the list as long as possible. If that’s what’s happening, let’s really gum up the works if we can. I think that’s important. And people are really doing it, and I think that’s great.
DANIEL: All right, well, let’s gum up the works and let’s not obey in advance. No anticipatory compliance. Thank you, Caitlin, for coming on and explaining all that to me.
CAITLIN: Thanks, Daniel.
DANIEL: So, let’s run through that list. We’ve got GOON, AI VEGAN, WIREBORN, possibly TULPA, LABUBU, COZY CRIME, BARK MITZVAH, NAP TRAP, and GROYPER, our Words of the Week. Let’s get to a comment by Yaniv via email. Yaniv says, “I assume by now you’ll have been inundated with emails pointing out these two pieces of information, but in case you haven’t. One, what Mr Ainslie dismissively cast aside as fake banana flavor…“
BEN: I did not. I like fake banana flavour. It just doesn’t taste like bananas anyway.
DANIEL: “Well, there’s a reason for that,” says Yaniv. “It’s said to imitate the flavor of the Gros Michel bananas, which used to be the dominant cultivar until they were almost completely obliterated by the Panama disease. It is, of course, still fake in the sense that all artificial flavoring is fake, but this is supposedly the reason it tastes nothing like the Cavendish bananas we know today.” You’d heard that before, I’m sure.
BEN: I have. I’m actually a bit interested to test the veracity of this. I suspect this might be a factoid rather than a fact. But I am not for a second saying that Yaniv is not correct in this. I have also heard this information. I just want to know how true it is.
DANIEL: You know what? You can still get Gros Michel.
BEN: Okay, well, that’s something we’ll do.
DANIEL: We should do that. They’re hard to get, but you can get them. Okay, for this second part, a bit of background. We talked about phrasal templates like THE SCREAM I SCRUMPT, or THE GASP I GUSPED. Not a word, but a way of putting words in a certain kind of jokey way. Yaniv says, “The linguistic entities you called phrasal templates are, I believe, better, but clearly not much better known as snowclones. I seem to remember this term originates from discussions on Language Log. But apparently, it’s now well accepted enough to yield a couple hundred results on Google Scholar. In any case, the term phrasal templates certainly has the advantage of transparency. Best regardless, Yaniv.”
BEN: So, snowclones is like eggcorns?
DANIEL: [LAUGHS] Yes, it seems to have resulted from a discussion on Language Log, but I don’t know. It feels like old internet to me and it feels like old linguistics to me and old lexicography. What do you think, Hedvig? Is that hitting your old spots?
HEDVIG: I have never heard SNOWCLONE. When I heard PHRASAL TEMPLATE, I immediately understood what it was. I don’t know how the word SNOWCLONE works. I don’t know how SNOW and CLONE, and I don’t get it.
DANIEL: It was before my time. It’s a snow cone, obviously.
HEDVIG: But again, Yaniv is probably correct in that this is also called that, but I never heard it.
DANIEL: Okay, well thanks, Yaniv, for those pieces of info. Thanks also to Maia Chao, Kelly Wright, Caitlin Green, everyone who gave stories, words, Related to Nots and comments. Thanks to SpeechDocs for transcribing all the words. And thanks to you great patrons for keeping up us getting up in the morning and avoiding those nap traps. Whoo.
Hey, if you like our show and you’d like to help us out in Because Language, there are lots of things that you can do. You can follow us. We are becauselanguage.com on Bluesky and @becauselangpod just about everywhere else. You can send us lots of material like the great people here who help us with the show do. News, words, Related or Nots, it’s all fun. Send it to us hello@becauselanguage.com or in the usual ways.
You can also tell a friend about us or write us a review in all the review-y places. I don’t have a review recently. I used to read them out, but there haven’t been that many. So, if you wrote one, we’d find it and then we would read it. It would be great, do it.
BEN: Get on it. And if you don’t want to do any of that stuff, you can always become a patron. You get stuff depending on your level. Live shows, bonus episodes, mailouts, shoutouts, and Discord access. That’s the very best thing. Our lovely online community of language nerds will welcome you with open arms. And you can see a whole bunch of pictures of their pets.
We do stuff with the money. We pay the bills. We offer a small stipend to the people who come on our show. We allow the show to be translated by SpeechDocs, which is really awesome. And we are going to do our shoutouts.
Now, we have been changing up the order in which we read the shoutouts. As Hedvig identified in our last show, one person complained one time. And now, forever, we will be reorganising. So, Daniel, this week, what way have we organised the patrons?
DANIEL: So, last time in our bonus episode, we played a game to see whose name was the best at obeying alphabetical order. I didn’t do too well in this game because DANIEL goes from D to A, that’s down, but then up from A to N, down from N to I, down again to E and then up to L. And you might remember that you get a point every time your name goes up in alphabetical order, you’re penalised every time your name goes down. And just to keep it interesting, if you had a run of like maybe three in a row that go up, you didn’t get three points. You got nine points, the square. Or you’d be penalised the same going down.
So, Ben, you were perfect. Your name keeps going up twice, and you got four points. I got negative points because my name goes down slightly more than up. But then, I started thinking, what other ways are there to order the alphabet? Is there an optimal alphabetical order?
BEN: What?
HEDVIG: Yes, the script, the Devanagari is the best.
DANIEL: Oh, tell me more.
BEN: Hang on, hang on. You guys need to explain other ways to organise the alphabet. Because to people like me, there is one way, the alphabetical way. What do you mean?
HEDVIG: Right, that is true. Well, imagine you have a writing system. Imagine that you have some sort of symbol that means a sound, some sort of alphabetical system. When you teach kids in school the order of those, you might go like A, B, C but in Cyrillic, they’ll go a different order, because they are different letters. That one is still related to Latin, so it’s probably somewhat similar. Devanagari is a different script system that it’s used for, among other things, Hindi. And they go by order that things are actually anatomically produced in the mouth. So, you start with K and G back in the mouth.
BEN: Oh, okay.
DANIEL: Oh, right.
HEDVIG: Yes. And then, you go forward to the mouth.
DANIEL: And then, you go forward.
HEDVIG: Yes. And you also go unvoiced aspirated, voiced aspirated. So, it makes phonological sets.
BEN: So, they basically ordered the IPA.
DANIEL: Somebody thought about this.
HEDVIG: Sort of, yeah.
BEN: Yeah. Okay, cool. What did you do, Daniel?
DANIEL: This wasn’t the same way. I found a page about optimal alphabetical order. It’ll be in the show notes for this episode. But imagine that the alphabet is like a sieve. And in order for a word to get through this sieve, it has to obey alphabetical order completely. So, the word BEEF obeys alphabetical order. And it gets through; B, E to F. So does the word COW; C to O to W. But BURGER doesn’t get through. So, what order of the alphabet allows the most English words to get through?
BEN: Oh, okay.
DANIEL: And it’s something like this, so far. F, C, B, W, H, J, L, O, A, Q, U, and it goes on from there. You can see the order on our webpage. But this order allows the most words to get through. So, that’s alphabetical order now and how good is your name at obeying the order that most English words come in? That’s what this is about.
BEN: All right. Okay, so…
HEDVIG: Oh, god.
DANIEL: Let’s hear it.
BEN: Okay, so let’s… I’m going to go from bottom to top.
DANIEL: Oh, okay. Great.
BEN: Yeah. So, the people who did worst at this, the biggest losers, if you will. Margareth on -12, Ignacio on -8. Molly Dee on -7, PharaohKatt on -5, and Sydney on -5. Then, we’re moving up the order. The -4s, Diego and Kristofer. The -3s, Martha…
BEN and DANIEL: O Tim.
BEN: …Rach, Stan, Sonic Snejhog, Nikoli, Nasrin, and Rene. The -2s, Iztin and Wolfdog [DANIEL LAUGHS]. The very first of the minuses. Daniel, you’re here -1. Rachel, Aldo, and Manu. The dead middle, the Even Stevens. This is where I sit, the 0s, John M, Sam, Tadhg, Kathy, Fiona, Colleen, Steele, Keith, gramaryen, Elias, and Helen.
Now, we’re going into the positives. Okay, we’ve got Jack, Ayesha, Mignon and Kelly on one. And Nigel on two. Then, we’ve got the 3s, Tony, Chris W, Rodger, Meredith…
BEN and DANIEL: LordMortis.
BEN: Lyssa, and Chris L. In the 4s, we have John K, Amy, Kevin, Termy. 7s, big jump, Hedvig, this is where you land right alongside Linguistic Chaos.
DANIEL: Nice.
BEN: There are a few above you, I’m sorry to say. With 8 points, we have Laura, Ariaflame, and Joanna. Now, we’re getting into the big dogs, 9 points, Faux Frenchie, who left such a great SpeakPipe. I’m so glad he fell highly after such a wonderful oaky timbre in our ear holes.
DANIEL: MVP for this episode.
BEN: J0HNTR0Y, Luis, Amir, Felicity, Andy, and Larry. And the big winners, the final three in third place, Andy from Logophilius with 12 points. James with 16 points. And I’m sure he already knows he’s won by the time that he is listening to this. I’m assuming it’s a he, or it might actually be.
DANIEL: I say it’s a they.
BEN: Canny Archer…
DANIEL: Whoo-hoo.
BEN: …on a stonking 17 points. Thank you to all of our patrons. Thank you to our latest patrons at the Friend level, Dermot, at the Listener level, Pigeon, and Cindy, who both upgraded from the Friend level. Hey, friends, you can do that anytime you want. Hello to our newest free patrons, Marion, Jacob42, Hannah A, Vesna and Dan S. Thank you to all our patrons.
HEDVIG: Our theme music was written and performed by Drew Krapljanov who also performs with Ryan Beno and Didion’s Bible. Thanks for listening. We will catch you next time. Because Language.
IN UNISON: Pew, pew, pew.
HEDVIG: Indeed.
DANIEL: Oh, thank you.
BEN: Let’s go stop on that microphone.
[BOOP]
DANIEL: Ready to rock and roll?
HEDVIG: Yeah. Wait, wait. Let me just pour tea.
DANIEL: I’ve got my hot chocolate.
HEDVIG: Usually when I record at home, I have my husband, but now I have a Finnish lady and she has brought me…
BEN: Oh, my god. How do you manage this? You just walk around acquiring chaiwalas wherever you go. [LAUGHTER]
DANIEL: Is she named Ste as well?
HEDVIG: Well, no, she’s named Oti, like so many other Finnish women I know.
[BOOP]
DANIEL: What’s a NAP TRAP?
BEN: Oh, I know what this is.
HEDVIG: I sent this one in, didn’t I? I sent this one.
DANIEL: It’s a TikTok video. Should we watch the TikTok video? It’s in the…
HEDVIG: Oh, my god. I mean, I can get… Fuck, what is it? I sent this in; I know I did.
BEN: I can guess…
HEDVIG: Shit. Fuck me.
[BOOP]
[ROOSTER NOISE]
BEN and DANIEL: Man, that rooster’s going for it.
DANIEL: Jinx! [LAUGHS]
BEN: I owe you a Coke. Damn it.
[Transcript provided by SpeechDocs Podcast Transcription]