How is language like a game of charades? According to a new book, quite a lot. Charades players and language users improvise and work together to create meaning in a situation, and they get better at it as they reuse elements and build up patterns.
Drs Morten Christiansen and Nick Chater explain their vision of language to Daniel and Hedvig on this episode of Because Language.
Listen to this episode
Patreon supporters
Huge thanks to all our great patrons! Your support means a lot to us. Special thanks to:
- Dustin
- Termy
- Chris B
- Elías
- Matt
- Whitney
- Chris L
- Helen
- Udo
- Jack
- PharaohKatt
- Lord Mortis
- Larry
- Kristofer
- Andy
- James
- Nigel
- Meredith
- Kate
- Nasrin
- Ayesha
- Moe
- Steele
- Manú
- James
- Rodger
- Rhian
- Colleen
- Ignacio
- Sonic Snejhog
- Kevin
- Jeff
- Andy from Logophilius
- Samantha
- Stan
- Kathy
- Rach
- Cheyenne
- Felicity
- Amir
- Canny
- O Tim
- Alyssa
and Kate B who contributed not via Patreon, but with the one-time donation button on our website, becauselanguage.com
And our newest patrons:
- At the Listener level: Sasha
- And at the Friend level: Elliott
Become a Patreon supporter yourself and get access to bonus episodes and more!
Become a Patron!Show notes
The Language Game: How Improvisation Created Language and Changed the World
by Morten H. Christiansen and Nick Chater
• Basic Books (US & Canada)
https://www.basicbooks.com/titles/morten-h-christiansen/the-language-game/9781541674981/
• Bantam Books (UK & Australia)
https://www.penguin.com.au/books/the-language-game-9781787633490
Morten H. Christiansen | Cornell Research
https://research.cornell.edu/researchers/morten-christiansen
Professor Nick Chater | Warwick Business School
https://www.wbs.ac.uk/about/person/nick-chater/
Man Charged With Threatening Merriam-Webster Over Gender Definitions
https://www.nytimes.com/2022/04/22/us/merriam-webster-threats-lgbtq.html
OC man arrested for allegedly threatening to bomb Merriam-Webster offices over definition of female
https://www.msn.com/en-us/news/us/oc-man-arrested-for-allegedly-threatening-to-bomb-merriam-webster-offices-over-definition-of-female/ar-AAWwxck
female | Merriam-Webster
https://www.merriam-webster.com/dictionary/female
What happens when a group of Fox News viewers watch CNN for a month?
https://www.theguardian.com/media/2022/apr/11/fox-news-viewers-watch-cnn-study
(2020) Consuming content from Fox News is associated with decreased knowledge of science and society
https://www.psypost.org/2020/07/consuming-content-from-foxnews-com-is-associated-with-decreased-knowledge-of-science-and-society-57499
(2012) STUDY: Watching Only Fox News Makes You Less Informed Than Watching No News At All
https://www.businessinsider.com/study-watching-fox-news-makes-you-less-informed-than-watching-no-news-at-all-2012-5
(2012) Survey: No News Is Better Than Fox News
https://www.theatlantic.com/culture/archive/2012/05/survey-no-news-better-fox-news/327757/
Meta announced that it will create a universal AI-translator for spoken language for all languages of the world
https://www.aroged.com/2022/02/23/meta-announced-that-it-will-create-a-universal-ai-translator-for-spoken-language-for-all-languages-%E2%80%8B%E2%80%8Bof-the-world/
Were Hurricane Katrina ‘Looting’ Photographs Captioned Differently Based on Race? (True)
https://www.snopes.com/fact-check/hurricane-katrina-looters/
Hugh Dundas | Wikipedia
https://en.wikipedia.org/wiki/Hugh_Dundas
Slobbing out and giving up: why are so many people going ‘goblin mode’?
https://www.theguardian.com/technology/2022/mar/14/slobbing-out-and-giving-up-why-are-so-many-people-going-goblin-mode
Aaron Sorkin showers up to eight times a day
https://www.hollywood.com/general/aaron-sorkin-showers-up-to-eight-times-a-day-59438552
Transcript
[Transcript provided by SpeechDocs Podcast Transcription]
HEDVIG: Daniel does my last name as best he can…
MORTEN CHRISTIANSEN: Hmm.
DANIEL: [CHUCKLES]
HEDVIG: …to the Swedish pronunciation, so he is capable.
MORTEN: /hiɽt gaɹd/. Or something like that.
HEDVIG: /ɧiɽ gɔɽd/.
MORTEN: Oh, /ɧiɽt/… Oh, right. Yeah, no, that’s the /ɧ/ sound.
HEDVIG: But not /ɧiʈ gɔɽd/ — /ɧiɽ gɔɽd/! [LAUGHTER]
MORTEN: I know it’s not /ɧiʈ/! No, that would be bad. Yes, that would be particularly bad.
DANIEL: Why? What does that mean?
HEDVIG: Shit guard.
MORTEN: Yeah.
DANIEL: Oh, wow, okay.
HEDVIG: Yeah.
[BECAUSE LANGUAGE INTRO]
DANIEL: Hello, and welcome to Because Language, a show about linguistics, the science of language. My name is Daniel Midgley. And with me now, it’s the linguist and channel swimmer — and when I say swimmer, I mean travels underneath it in a train — it’s Hedvig Skirgård.
HEDVIG: Thank you very much. Yes, when in Europe, I try to take trains and the tunnel train is fun. That is it. [CHUCKLES]
DANIEL: Hedvig, have you ever played charades?
HEDVIG: Yes.
DANIEL: Do you play often?
HEDVIG: Well… what is this? Do you want me to do, like, do “swimming”?
DANIEL: Could you please do “English Channel”?
HEDVIG: Well…
DANIEL: [OBSERVING] Okay, okay. We’re getting a teacup, drinking tea… all right… and then, one hand moving under another hand, sort of in a submerging gesture. That’s very good. Okay, I like it. I’m sure that if we played many times, we would get better at it. I feel somehow kind of certain. [LAUGHS]
HEDVIG: It’s probably true. That’s probably true. This is a wonderful segue.
DANIEL: Thank you. I worked on it very hard, because we’re joined by two very special guests, Dr Morten Christiansen and Dr Nick Chater. Thank you, both of you for being here.
MORTEN CHRISTIANSEN and NICK CHATER: Thank you for having us.
DANIEL: Dr Morten Christiansen is the William R. Kenan Jr Professor of Psychology, and the Director of the Cognitive Science of Language Lab at Cornell University in the US, as well as Professor in Cognitive Science of Language at Aarhus University, Denmark, and a Senior Scientist at the Haskins Labs. Morten, I’m really curious. We’ve been following your work for a long time, especially your work on sound symbolism, where you found some surprising similarities in the sounds that languages choose when they’re doing words. For example, the /ɹ/ sound seems to appear in round or circular things. How does that fit into your vision of language?
MORTEN: Well, it fits into sort of the broad idea of both language acquisition and language use, as involving multiple kinds of clues to figuring out what words mean, what syntactic constructions mean. And the relationship between sound and meaning is just one of these sources of information, and one that for a long time was sort of ignored or pushed out on a sidetrack. But then in the last decade or so, more and more work has been done within that area. So it’s exciting to see that, and I’m happy to have contributed to that as well.
DANIEL: All right. Thank you. And Dr. Nick Chater is a Professor of Behavioral Science at Warwick– /waɹwɪk/? /waɹɪk/.
NICK: Warrick. /waɹɪk/
DANIEL: Ah, at Warwick business school — just like here in Perth, there’s a Warwick in Perth — And he’s also the cofounder of the research consultancy, Decision Technology. Nick, I’m really fascinated about how you do both work on language and work on autonomous driving.
NICK: That does seem a bit of a stretch, but actually, it’s quite a close connection. The reason I’m interested in autonomous driving is the communicative aspect of it. So, if you’re a human driver, you wave at people, you flash headlights, you toot your horn, and the way you flash your headlights and the way you toot your horn, and the way you wave, these are really crucial things. And if you’re going to get an autonomous driving system to work well with people, it’s got to be able to understand those signals we send and it’s got to be able to send them back. Or, alternatively, if that’s just too hard, because that’s playing charades almost — we’ll come to that — but if you’re going to try and do that, you might just think, “Well, maybe it’s just too hard for machines. Maybe it is, maybe it isn’t.” And you might just decide you want to make the autonomous vehicle more like a dumb truck or a dumb tram. But that’s a reason I made that sort of connection really.
DANIEL: I can see that connection. And I can also see a connection in how if you’re trying to get a car to navigate a scenario, you want it to be able to do lots of things that we do in language, like plans and goals, keeping track of informational states, you want to be able to read the intentions of other drivers. So, I can really see a connection between those two things.
NICK: Yeah, well, that’s quite right. So, yes, human driving is just amazingly complicated when you think about it, but crucially, it’s a social activity, and of course, so is language. But the social element of driving, that’s the really hard part for autonomous systems, I think.
HEDVIG: I was driving around a lot in the countryside in Sweden recently and I was the navigator for my husband. And besides communicating to other drivers, communicating within the vehicle can be a challenge as well, especially because I have a tendency to entirely mix up left and right. So, I’ll tell him, “Oh, you take the next left,” and he’ll be like, “You must mean right in this scenario.” “Oh, yes, of course!” [LAUGHS] At least, the computers when they have a GPS don’t have that problem, I guess.
DANIEL: I don’t want you to be my navigatrix, okay?
HEDVIG: I’m okay! It’s just sometimes a bit much.
DANIEL: All right. We’re going to be talking about Morten and Nick’s new book, The Language Game: How Improvisation Created Language and Changed the World. But first, our latest bonus episode was an interview of us — me and Hedvig — by two wonderful up-and-coming linguists, Kitty Liu and Romany Amber of Cambridge University. Wasn’t that fun?
HEDVIG: It was really nice. I don’t know if, Nick and Morten, if you’re aware of this, but there’s like a British association for linguistics students. They asked us lots of clever questions about linguistics and career choices and academia and things. It was very nice. Really enjoyed it. They have a paper as well. They’re so talented.
DANIEL: It was lovely. For our next one, we’re going to be tackling a recent resolution to the LSA that claimed to be about freedom of expression, but what was really behind it? Sorry, Hedvig’s giving me a reaction, like, maybe you don’t want to be tackling that one for our next bonus episode.
HEDVIG: I just [SIGHS] I don’t know if our guests feel this way, but sometimes, America, it’s just a lot.
[LAUGHTER]
DANIEL: It’s a lot of work.
HEDVIG: It’s just a lot with America! Anyway.
MORTEN: A lot going on.
DANIEL: Can’t even!
HEDVIG: It is important. A lot of linguists are in America. LSA is not only an American institution, it’s also a lot of international people, etc. So like, it is of interest to the wider linguistics world, but I’m unsure how much I want… Yeah, anyway, we’ll see.
DANIEL: How much you want to wallow in that mud, right?
HEDVIG: Yeah.
DANIEL: Yeah, okay.
HEDVIG: Maybe my energy could be better spent elsewhere. I dunno.
DANIEL: Well, there are some things to say about freedom of expression. There are some atrocious takes on freedom of expression, including some of mine. So, that’s coming up, and we’re going to see what happens. You know, we have tackled these things before. It worked out. It was good, right, Hedvig? No? Okay.
HEDVIG: Yeah, we did an LSA kerfuffle complex episode before. It was interesting. It also included a lot of hedging. Often, you don’t know enough about what’s going on to make like, “Oh, this person is completely stupid,” statements. But sometimes you do. [LAUGHS]
DANIEL: We’ll get some help.
HEDVIG: Anyway, we’ll try and do that for our next bonus. Interesting. I’ll consume only European news the whole week before, and then maybe I can take in some America.
DANIEL: Very good.
HEDVIG: So, don’t do anything. No High… What’s it called? Supreme Court shenanigans.
DANIEL: No.
HEDVIG: I don’t want Biden or anyone to say anything. Everyone in America just stay quiet for weeks, so I can build up my American energy.
DANIEL: Just: shhhhh. That’s right. And afterwards, we’ll do non-American stuff again, including with our patrons who get to hang out with us on our Discord server. Some of them get to hear bonus episodes, or even get a shoutout. So, if you want to be a patron and support the show, you can do that: patreon.com/becauselangpod.
All right, let’s get to the news. And we’re going to a pretty dark place for the news. Usually when there’s nasty stuff, I try to ignore it. When there’s people doing bad things, sometimes I just try to leave it alone. Lately, it’s gotten to a level where it would be unwise to ignore because sometimes the violence comes to you. So, MrBobbyHunt on Discord has sent us a news story. We already know that language is political, and we learned from Dr Rebecca Shapiro on our Just Words episode that dictionaries are seen as especially political because people see dictionaries as having this gatekeeping function, and when a word appears in a dictionary, it is seen to have a kind of validity.
Well, it seems that a man in California has been arrested for allegedly threatening to bomb the office of Merriam-Webster, the dictionary people, and kill its employees over the dictionary’s definitions for women. Jeremy David Hanson of Rossmoor, California, sent anonymous messages to Merriam-Webster because they were changing their definitions, including BOY, GIRL, and TRANSWOMAN. He said the company was taking part in efforts to degrade the English language and deny reality. What are our thoughts on this so far? When you notice the story, what kind of comes up for you?
NICK: Well, for me, it seems… I mean, it’s a horrifying story. It also seems to go against the norms of how language always functions, and language is continually shifting. The meanings of words are very flexible and are continually being flexed. As society changes, the world changes, we need to change the way we use words appropriately. And that’s completely, completely always been the case. It’s completely normal. But, of course, I think there is this sense among many people that any change in language, whether or not it has some deep political significance — which it may particularly in this case, perhaps — but also just to change it at all is to degrade it, and the sense that that change can only be erosion is quite a powerful one, I think. And so you have people who are horrified by — probably not to this degree, I hope — by uses of LITERALLY not quite meaning “in a literal sense”, PATHETIC meaning generally feeble rather than worthy of sympathy, or AWESOME not being awe-inspiring as it were, but just being just generally great.
So, one can have a sense — and I have a bit of a sense with some of these words that: this is just slightly irritating! But, of course, that’s the classic error, right? The classic error is thinking, “The way I use the words is the correct way, or the way the dictionary I had when I was a child formulated these definitions is the correct way,” and to think everything else is collapsing. Rather than realising that words are just changing their definitions and new nuances are appearing in language all the time, too. The new words, new phrases, new ways of using terms are shifting. So, it’s not really a case of terrible decay. It’s just rebirth and regeneration. Languages are like human societies. Some things are going, some things are changing, some new things are being created all the time.
A final thing to sound out, I guess, is that language is much more shape shifting than we can imagine anyway, and we’ll come on to the charade nature of language, but it’s worth bearing in mind that even when we use everyday words, like for example, I was thinking about words like DOOR and WINDOW, they seem like completely solid things. I mean, a door is just a door and a window is a window. But if you think about going through the door versus opening the door, we’re talking about something different, because going through a door is going through the aperture, and opening the door is moving the actual physical sheet of wood. And that’s how you flip from one to the other. No one goes about saying, “Wow, this is a complete misuse of the word. We need to have a separate word for one or the other.” So even everyday words of the most concrete kind are being used very flexibly all the time.
HEDVIG: Yeah, exactly. And I think it’s also one of those questions where maybe people come to linguists and ask, but I’m not sure if linguists… this is just a symptom of a societal change. The dictionary tries to reflect societal change, they change the definitions of words. And people get irritated by the language and the dictionaries and point their anger there. But maybe also in these discussions, I wish people would interview, like, sociologists or historians or non-linguists more that can talk more to the societal change, because linguists… we’re very smart about language. And we can say, as Nick just said, and as many linguists would say, you know, it’s fluid, it’s going to change. Get with the program, or hide under your rock, and don’t engage with anyone else! [LAUGHS]
What the actual societal change is about is maybe not something that’s a little bit beyond linguistics, sometimes. But just a clarification question, Daniel. Merriam-Webster’s definition of WOMAN now also includes transwomen. And what was the other things?
DANIEL: The definition for a girl now includes not only a female child, but also a person whose gender identity is female. And this guy didn’t like that, because he thinks there’s no such thing as gender identity.
MORTEN: I agree with Hedvig there that yes, linguists and language scientists may not always have the best answers in this context. I mean we can just refer to, as Nick did, historically, meaning changes all the time, and that’s just the nature of how language works. And the problem is that there’s a lot of conservatism in terms of, “No, words should mean this and this and this.” But as far as we can go back in history, people have been complaining about language and language change all the time. And they’re going to continue doing that, and hopefully not as violently as what we saw in these sort of news clips here, which is terrible, right? That’s rather extreme. Most of the time, people are just rolling their eyes or saying, “This is terrible, the way they use LITERALLY, or PATHETIC,” and so on. But not sending bomb threats to dictionaries and so on. Dictionaries are just trying to keep up with the times. It must be frustrating in the old days, if you’re doing a dictionary, as soon as it was done, it was pretty much obsolete, at least partially, because the language just keeps on moving on. But these today with electronic dictionaries, they can update it sort of on a yearly basis, and it’s keeping more in time with the actual use of the language in the community that they’re covering.
DANIEL: I think there’s something else going on as well. I think we need to remember that this event is taking place in the context of some pretty heavy shenanigans that anti-trans folks and the right-wing authoritarians are trying to pull with this idea of the definition of WOMAN. Like, I noticed on Twitter that that some people were kind of throwing this in my face or in the face of other people I know: “What’s your definition of woman?” And I thought, “What’s that about?” And then, I noticed that during the confirmation hearing for Supreme Court nominee, Ketanji Brown Jackson, one of the senators, a Republican from Tennessee, tried to basically turn the proceedings into a circus by saying, “Can you provide a definition for the word WOMAN?” So, I think this is something that they’re having fun with these days. They want to be awful to trans people. And if somebody defends trans people, they want to get them on the record so that they can scream about it to their followers.
And this brings us to another Word of the Week from a long time ago, and that was STOCHASTIC TERRORISM, which is where bad media actors or political figures just keep pumping hate into the discourse, keep turning up the heat, turning up the heat, and they don’t have to commit a terrorist act, because somebody on the fringes will take up their cue and do it themselves.
MORTEN: It’s odd that dictionaries have become part of the culture war. Who would have thought? Somebody working on dictionary probably thought that, “I’m doing something really mundane,” and so on. But now, they some of the times end up at the front lines that the culture wars, which must not be fun for them.
DANIEL: No.
HEDVIG: Yeah, definitely.
NICK: Another thing to add to that is the sort of accusatory what’s-your-definition-of-X? is always an impossible question to answer, because words don’t really have definitions in a stable way. So, dictionaries don’t give you a sharp definition of anything. They will give you a bunch of examples, something like this, and here are a few examples. That’s the way meanings are. So, the inability to give a precise, crystalline definition is completely typical. We can’t give a precise, crystalline definition of, say, WINDOW. Now, what’s that? We won’t cover all the types of window in the world. It’s just not a reasonable thing to ask. But of course, one can fel somewhat cornered, if a definition is demanded, one shouldn’t. Just push back!
DANIEL: I want to change our focus now to some research about communication in the media, because I’ve been obsessed for the last little while about how opinions get spread, and if it’s possible to change people’s opinions, and how much the media influences opinion. So, previous research showed that the worst-informed people were viewers of Fox News. They gave people factual questions, like, about which party is in power in Congress or what’s the unemployment rate. And people who said that Fox News was their only source of news input, did worse on these factual questions than people who said that they watched no news at all. And what this kind of tells me is that going outside and banging your head against a rock is a better way of getting news input than watching Fox News. [LAUGHTER]
HEDVIG: I don’t think the other people were banging their head against the rock.
DANIEL: No, but if you did, you would probably know about the same or even slightly less than you did before, which would give you a better score probably. So, this story is about what happens when you pay a bunch of Fox News viewers to change for a while, to watch CNN for a little bit. What happens? Do they start changing their mind? This comes from political scientists David Broockman of UC Berkeley and Joshua Kalla of Yale. They took 304 Fox News viewers. They paid them $15 an hour — that number seems significant — to, instead of watching Fox, watching CNN for up to seven hours a week. This was in about September 2020. And then, they would give them opinion polls and see what they thought about things, versus a control group that just kept on watching Fox News.
Now at this time, a lot of the media attention was focused on a police shooting, which seriously injured Jacob Blake, who is black. After the end of the month, the people who watched CNN were less likely to say that “it is an overreaction to go out and protest in response to the police shooting”. And they were also less likely to believe “if Joe Biden is elected president, we’ll see many police get shot by Black Lives Matter activists”. The Fox News viewers were about 10 points more likely to believe that, whereas the CNN viewers tuned it down a little bit. There were a few other things as well. So, it looks like the media environment really influenced those people quite a bit.
HEDVIG: Yeah. I’m not terribly surprised.
MORTEN: Yeah, I mean, I think probably a lot of things is happening here. So one of the things that people have been looking into is how the kind of language input that we get in general from the media, from newspapers and so on, how that can affect sort of the stereotypical biases that people tend to have. Just one example, I remember seeing some pictures from after the Katrina hurricane hit New Orleans. And so there’s two pictures. One was of an African-American man coming out of a store with some provisions. I mean, he had taken them. And then a similar situation, but a white person. And so under the African-American, it said, “Somebody is looting.” Under the white person said, “Somebody is providing for their family.”
DANIEL: I noticed the term ‘finding’ as well.
MORTEN: And it was just like they were doing the same thing. But here, just the words matter, right? So, these are the kinds of ways that things are described in the media. And that over time has a huge impact on how people react to things like white versus black stereotypes and so on. So, I think if you change the input, like they were doing to some degree, by having them watch CNN, instead of Fox News, they’re going to get a different frequency of various kinds of stereotypes. They’re going to get less of certain stereotypes. They might get some other patterns instead. And we come to reflect those patterns to a certain degree, and of course, it’s probably a fairly subtle effect. But I think there’s a there’s a positive sort of a silver lining here is that we could potentially change people’s behaviour if we could get them to watch news that had less biases in. Of course, that’s not going to happen, because people like to live in their bubble. But nonetheless there is a possibility here.
HEDVIG: That’s the kind of thing that Facebook tries to do, right? Facebook has tried some of these experiments, with upvoting certain posts or certain ads or things, and they have shown change in people’s moods and attitudes. Just terrifying that they did, without anyone’s consent or anything.
MORTEN: Well, their excuse was that in the thing you say ‘yes’ to, the specifics whenever you sign up for any kind of services, that it’s a huge, long list. Nobody reads that. It’s incredibly small print. But in there, it probably said somewhere that you give your consent to this and this and this. So of course, they didn’t give consent for this specific experiment. When you form a new institutional review board, kind of IRB perspective, “No, this was not necessarily completely cogent,” but it was approved, that study, by an IRB.
HEDVIG: I always feel like, when I hear this about American media landscape, how lucky I am to live, for example, in Germany, where there are state-funded, semi fairly balanced, unbiased news reporting I can have access to. For me, English speaking, there’s a channel called Deutsche Welle, which also has a website, which has news about Germany, Europe in English for immigrants in Germany. In the UK, there’s the BBC. In Australia, there’s the ABC and the SBS. And these are really good institutions, and a lot of countries have these. Most Western European, most Eastern European countries have something like this. And the US doesn’t. All these, like, NPR stations, they’re all run on donations or weird stuff. I tried to cultivate a fun feed on TikTok and Instagram and Twitter or whatever, like research news and cute cats. But then, I try and have a core in my media consumption that is, like, I listened to the Swedish national news once a day, 10 minutes to just get a sense. And I know I have that stability. And in America, that just seems impossible, which must be very distressing.
DANIEL: And this is why many Australians demagogue the ABC relentlessly by… the hashtag #defundtheABC pops up every once in a while.
HEDVIG: That’s the same in the UK, right?
DANIEL: Yeah.
HEDVIG: BBC gets slammed for: it should be defunded, the same as Sweden, but it usually holds strong. And some of the people who produce content on those institutions say when you make a program and you get complaints from all sides of the spectrum, you know you did a good job. [LAUGHS]
DANIEL: Oh, no! See, that’s the worst, because here’s what happens. So, the ABC gets this as well. They’re really allergic to right-wing criticism. So when right-wing folks scream, they move rightward. But then, when lefties scream about that, then they say, “Well, now we’re copping it from the right and from the left, so we must be doing a pretty good job.” So, criticism from the right is taken as: we’re doing something wrong, but criticism from the left is: Oh, we must be doing something right. This is a huge problem, the false equivalency just drives me bananas, and it’s something the ABC and others really need to watch out for.
HEDVIG: Yeah, but at least, at least it’s better than the American situation! I think the BBC and the ABC and the Swedish Radio, Deutsche Welle, all of them seem to be… also, in the general public, it seems like they are read as less biased than for example, CNN is or something.
DANIEL: Oh, I agree. I just want to say, ABC, you need to ignore the right-wing screamers because you know what? They’re never going to love you. There’s never going to be a situation where they will say, “I think you’re doing a great job now.” They will always scream for your firing, your dismissal, and for the ABC to be broken down or BBC or whatever there is. So, please, folks, government agencies, you’re doing a great job. Please just focus on informing the public and tune out some of the noise because there are some bad people out there.
HEDVIG: Yeah. So I’m not surprised by this news that changing your media diet changes your opinion. That seems… may I say, almost a little bit tr… I don’t want to say the word ‘trivial research,’ because I think it’s good that they saw an effect. It brings weight to the argument, but I can’t say I’m surprised.
DANIEL: Can I say I feel a bit hopeful about it, because maybe it means that people that we don’t think are convincible actually are if they were swimming in a different media pool?
HEDVIG: Yeah. But they probably need change to their social networks and their social security as well. It’s not just about the media you consume. Right? i don’t know. Anyway.
DANIEL: Right. One of the worst things about social media is that we now get a lot of our news from our social network, and that is horrible. [LAUGHS] That is just the worse place to get news from, because when your information diet is tied to a group of people that you care about, and which gives you status, it’s really hard to change ideas. It’s really hard to break out of it. You have to basically pay somebody in an experiment to get them to change their diet. And in fact, all the Fox News viewers in this experiment went right back to Fox afterwards.
HEDVIG: Yeah, exactly. They’re not going to change unless their family and friends and everything. Yeah, use social media for like cute cats and fun TikTok dances and then go elsewhere for your news.
DANIEL: Agreed. [LAUGHS] Hedvig, you found this one. It looks like Meta…
HEDVIG: Yeah.
DANIEL: Do you want to take this one? Do you want to drive?
HEDVIG: Well, Meta, which is the parent company of Facebook and Instagram and WhatsApp and everything else, they’ve recently changed that name to Meta, which is… I have opinions about that as well. But maybe it’s a good name, I don’t know. They have made an announcement that they are going to make a universal artificial intelligence translator. So, for anyone who’s a sci-fi nerd, you might have seen shows like Star Trek, or… they should have a universal translator in Star Wars as well, surely, they just don’t talk about it.
MORTEN: But it’s also, I guess, it’s like the Babel Fish in The Hitchhiker’s Guide to the Galaxy, you sort of put this fish into the ear, and then you can understand everyone.
HEDVIG: Yeah, exactly. And sometimes it breaks down in certain situations, but generally it works. So Meta has announced that they are going to make a system for translating spoken language that can work worldwide in any language. Soo let’s just take that in. First of all, they’re restricting to spoken, so they’re not doing sign language stuff. Okay, good. Now, we know that. There are a lot of languages in the world. And very often when companies say these things, like, “This book has been translated to all languages,” they mean like 60 languages or something like that. And I think it would be great if their goal is actually around 7,000 we think we have, that’s super cool. But the material you need to make an artificial intelligence, which they do mention in this announcement, is you need some sentence or corpora, or some knowledge about the languages. And for a lot of these languages, we don’t have that. For many different reasons, we only have a decent grammatical description for about 4,000 of them. We have collections of texts for even fewer. So, they say that the first part here… what did they call the first part?
DANIEL: “No language left behind.”
HEDVIG: “No language left behind.”
MORTEN: That’s cute.
HEDVIG: “We’ll focus on creating AI models that can learn to translate languages using a limited number of training examples.” So, yeah, I guess we’ll see how limited it can be. But yeah, it sounds cool and sci-fi. I think they’re not going to be able to do it for all languages. So, why did they say that? I don’t get it. [CHUCKLES]
NICK: Yeah, it does sound really incredible, because the way that machine translation works and works well is through having really large corpora and also having parallel corpora, so documents which are in multiple languages. So when you’ve got that, you’re in good shape. And this is sort of technology that’s happening, or happened in some cases. But if you’re trying to translate languages where there’s small corpora — or none — but small corpora, and there’s not large amounts of parallel text in, say, English versus that language, it’s going to be… I mean, current techniques just can’t do it. And if they’re going to have to develop a whole new approach, then one becomes much more skeptical because all the approaches in the past which tried to be much more… less data intensive, but more knowledge heavy — so trying to process the meaning of sentences in some deep way and put them in some kind of underlying logical form or something of that kind — that’s proved to be really, really hard, and basically just hasn’t worked. So, the whole language technology has just moved in a direction of using big corpora, and without those big corpora, is there a solution they have? I doubt it. It’s at best an aspiration rather than based on any real technological insight. Although we may be proven wrong. That’d be fantastic if it’s doable.
HEDVIG: Yeah, it’d be really cool. I know that Microsoft and Google sometimes fund projects to try and create corpora for minority languages. But they focus on a couple of languages at a time, I think.
MORTEN: Yeah. It could be an amazing thing if they can do it, but I’m also like what Hedvig was saying is like, I imagine that this will be primarily for maybe 60 languages, and that’s what they call universal. There are systems now that can work fairly well in sort of limited narrow domains. For example, medical translation from Japanese to English and back in real time, and it worked reasonably well. However, getting it to work with all the many minority and Indigenous languages that have no corpora set up for which we actually know relatively little about. I mean, it could be an amazing thing, if they could actually do it. If they go out and actually document and record samples from those languages, insofar as it’s not going to be used to suppress people, but it could help languages that might be in danger of dying out and so on. And there’s quite a lot of languages that are in danger today. So, it could be a positive thing.
Now, given Meta’s track record, maybe that’s not what they’re thinking of, but certainly it could be a big boon for the language sciences if they actually could carry this out and doing it in a way that would be respectful of the people speaking those languages.
HEDVIG: Yeah. ‘Cause right now, like you’re saying, with the track record right now on Facebook, it tries to do translations. I don’t know if you’ve tried it out, when people make posts in a language, it’s so-so. I have a lot of friends who write Samoan on Facebook, and it tries constantly to think that that’s Italian. And it translates mumbo jumbo. And then, there’s a button I can click to say like, “This isn’t Italian.” But, of course, I can’t write Samoan. So, I just write it’s another language. And it’s like, “Okay, useless.” [CHUCKLES]
DANIEL: There’s been some interesting work in machine translation. I’m noticing one from Google. They’ve done work where they’ve dumped loads of sentences into the same corpus from all different languages without worrying about what language they’re from. And that can be a boost for low-resource languages. I thought that was interesting. And then, there has been another project where you can train on English-Korean, and then you can train on English-Japanese. And the machine translator can go Korean to Japanese, without going through English, and without any extra training, which boggles my mind. So, there are some…
HEDVIG: That’s very cool.
DANIEL: It’s really cool, but there are some bootstrappy ways to help low resource languages. But of course, my concern is that the digital tech revolution will benefit a lot of people, but it won’t benefit everyone equally.
HEDVIG: You need material. They ain’t got material for languages in the world. That’s the main thing that’s going to happen. If they decide to fund… like, I know, Microsoft and Google sometimes does, if they decide to fund like community-driven field projects which will gather material, I’m going to be the biggest Meta fan there ever was. I doubt it.
DANIEL: [CHUCKLES] Well, this is a bit of a Hail Mary move for Meta, I think. They need to do something. They are banking hard on the Metaverse, which is very speculative and not a done thing yet. This could be something that could help them gain in popularity. This could give them a technical edge if they can do it. But it sounds like we’re all a bit skeptical at this stage.
HEDVIG: Why can’t they just say 60 languages? That’s really cool, if they did that.
DANIEL: No, Hedvig, it has to be all. It has to be all the languages.
HEDVIG: Hmm. Yeah.
DANIEL: All right. [LAUGHS] Thank you. Thank you for your insights on the news. Mmm — there’s some ouchy bits, but there are some glimmers of hope.
[TRANSITIONAL MUSIC]
DANIEL: We’re here with Dr Morten Christiansen and Dr Nick Chater. They’re the authors of the new book, The Language Game: How Improvisation Created Language and Changed the World. Let me start with the big picture. You used the analogy of language being like a game of charades. How is this so? How are those two things alike?
MORTEN: The basic idea is that both language and the game of charades are all about improvisation and active collaboration as you’re trying to get your ideas across and how you’re trying to communicate. So, you have to do this in the here and now. And what we’re doing in both cases is we’re improvising to provide cues to help each other understand what it is we’re trying to say. So the crucial point, from our perspective, is that linguistic communication is fundamentally collaborative. It’s often been treated in the language sciences as monologue. So, many experiments that we do, including many of my own, it’s all about just having one person responding to something rather than actually engaging in dialogue. We think this is missing an important point, and that’s what charades brings back in, is that as a listener, we’re not passively waiting to receive a message, like a computer might do. But what we are trying to do is we are actively using everything we know about each other, about what was said before, what we know about the world in order to sort of make sense of what the other person is saying.
And so what we’re doing is that we work together to build a shared understanding just one improvisation after another, just like we do in a game of charades. For our perspective then, language is fundamentally improvised communication. This contrasts from the more general view of looking at language as kind of like a fixed code that would allow us to bottle our thoughts into a stream of words, to be uncorked and decoded at the other end by the listener. So from viewing language as charades, essentially, we communicate as best as we can, using hints and clues which are the words and sentences. These are created, understood — sometimes not, sometimes they are — through basic human powers of human ingenuity.
DANIEL: So we’re playing a game right now. We’re watching each other on Zoom, and we’re listening to each other as well. If we weren’t watching each other on Zoom, what would we be missing? How would that change the game? What cues are you picking up right now in this game of charades that we’re all playing?
MORTEN: First of all, we’re looking at each other’s faces. So, there’s all these conversational devices we’re using when we’re interacting. Some of them are verbal, some of them are… I’m waving my hands around here, and so on. We’re using our bodies. These are all providing clues to how we communicate. And in a normal conversation, in a normal dialogue, we use, for example, backchanneling, so you’re nodding your head, or you might say, “Mm-hmm,” and so on. That happens about every few seconds. And sometimes, of course, things go awry, so we might go, “Okay, hang on, what did you just say?” People might say, “Huh?”, and so on. These are ways in which we can help move the conversation along so we both understand. So, we can use certain kind of conversational devices to make sure, to indicate that we are on the right track. And then, we can use others to sort of indicate, “Oh, well, you need to repair this. That’s something I didn’t understand.”
And so we’re doing this all the time, both when we’re doing charades, we do the same thing as well. We’re like, “That’s not the right thing,” or it’s like [GESTURE] when we don’t understand what’s going on.
[LAUGHTER]
MORTEN: But of course, people can’t actually hear what I just did. But I was just making some…
HEDVIG: He’s waving his arm around in various ways.
DANIEL: But I saw it, right? [LAUGHS]
MORTEN: You did see it. Yeah. So essentially, this is what we do, and we have to do it on the fly, right?
HEDVIG: It’s very unsettling when you meet people who speak the same language as you but don’t have the same conversational pragmatics when it comes to, for example, backchanneling. So I’ve met people who I speak to, who don’t do almost any backchanneling. And I’m like, “Are you hearing…? Am I being understood? Is my voice coming across?” It’s like, “Oh, no, I hear you. I’m just waiting for you to finish and then I’ll speak.” And I’m like, “Oh, okay.”
[LAUGHTER]
HEDVIG: That can be a bit disorienting.
MORTEN: Yes.
HEDVIG: I know there’s also research from the Nijmegen folks with Steve Levinson’s lab, that there’s evidence to show that the time it takes for someone to plan an utterance is such that the dialogue material they have must mean that people are actually starting to plan their utterance before the other person has stopped speaking. Which means that we must be able to guess and predict the end of that sentence, in order to have material such that we can respond. Which is amazing to me, which means that we’re running little predictions in our head of like, “What could they be saying next to fill in gaps?” So, yeah, it’s really a back and forth. We’re not, like, passive at all.
MORTEN: Yes, unfortunately, that… again, as mentioned earlier, that oftentimes language has been treating like monologue rather than dialogue. And that has, I think, deep consequences for how people have approached the study of language and even theories of language.
NICK: Just adding to Hedvig’s point, this idea that we’re continually essentially regenerating and predicting what other people are saying, this is an idea that goes back to the old idea of analysis by synthesis, which is a very early idea in language processing. When you’re trying to understand something, you’re actually going to recreate it. But that was very interesting in itself a general point about the way perception works, actually, in general, not just language. But it’s also very interesting in the context of charades, because it’s clear that if we’re going to understand each other, we’ve got to be regenerating the same thing. So, here you are making a signal, and you have this in mind. I’m watching you, I’m trying to recreate that signal, what are you doing, and what are you going to do next? And I’m also going to try and have the same thing in mind that you have. So, we’re trying to sort of coordinate and synchronise both the signals and the meanings. So, it really is this joint activity. It’s not that, as Morten was saying, I just fire a message at you, and then you figure out what’s that mean, decode it, fire me a message back. Really, our brains, when we are engaged in conversation, we’re always in the process of joint understanding.
HEDVIG: And you have to put yourself in the other person’s shoes at the center as well to guess what could they be meaning. Like, if I wanted to communicate this thing, maybe I would express myself this way. But the person I’m listening to is not me. So they’re doing something else. So, you have to do several levels. And then you might have to be like, “Oh, but they know that they’re speaking to me.” [LAUGHS] So, they’re trying to fit what they’re doing to me, and you have to like…. But the more you spend time with people, obviously, the better communication goes until you get to a point where you can like grunt. My brother used to just grunt at the breakfast table, and we knew he wanted more buttermilk.
DANIEL: That reminds me so much of how we’ve seen experiments with language games, where people are drawing pictures to try to… there are eight people in a group and two of them are being paired off at a time, and they’ve got to come up with a symbol or something to guess the object, but they’re not playing with each other. But eventually, down the line, they do standardise on things. So, I guess in this view, a language like English is basically just the result of a bunch of people playing charades, conversational games with each other, building up patterns. And that’s what becomes the language. Have I got that right?
NICK: Yeah, that’s totally right. From our point of view, you start playing without a set of conventions. But after a little while, as with Hedvig’s English cup of tea drinking, we start to think, “Well, that was a pretty good way of conjuring up England, or English things, or things that people do in England. So, we’ll use that again.” And that will become more and more standardised and get more and more used. And it’ll get simplified. So, the actual signal will get stripped down, you don’t have to do so much detailed miming of the cup-of-tea-drinking action, so the signals become pared down. And also, the meaning will become both more… also become used in wider context. It’s start off being just for the channel, but then we’ll start to use it for the Queen. And we’ll start use it for the country of England, and maybe for the English language, and so on and so on.
So yes, you start with these developments, but they’re continually becoming more and more eroded and simplified, but also more general. And, of course, you’re continually making new ones. But equally, you’re putting them together. So, you’re combining them in new ways. And that combination is gradually going to be giving you syntax. So, instead of just a sequence of charades, you’re acrtually going to get phrases, patterns, which will ultimately lead to grammar.
HEDVIG: Yeah. I read one paper from Simon Kirby’s lab in Edinburgh where they had people… I think it was like Pictionary, they had to draw things. And after a while, a group settled on… if you know the Teletubbies, the Teletubbies have different colors. There’s a children’s show that’s popular in Britain and in Europe. And the Teletubbies have different colors, and they have different shapes on their head. And one of them has a triangle, and it’s purple. I forget the name. it’s not Laa-Laa, it’s…
DANIEL: You’re thinking of Tinky-Winky. I know it well.
HEDVIG: Is Tinky-Winky purple?
DANIEL: I believe so. If I’m wrong, I’ll eat a coat hanger.
HEDVIG: Oh, don’t do that.
DANIEL: [LAUGHS]
HEDVIG: But anyway, after a while, everyone settled on a triangle being purple, the color purple. And I don’t think any of the other groups did it. But it’s just amazing how, when someone comes up with a smart idea — like my teacup is very smart! — it can just spread, even though the connection can get quite arbitrary in general after a while.
NICK: Yeah, that’s a great example.
HEDVIG: And you can also tell that — speaking of what we talked about earlier with WOMAN — some of these conservatives have invented a new word for what they want women to be, TRADWIFE, which stands for traditional wife.
DANIEL: I have not heard this. That sounds kind of odd.
HEDVIG: Yeah. It’s someone who stays at home and enjoys cooking and cleaning. It has nontoxic elements to it. Some women choose that and enjoy that, and they share like baking tips online.
DANIEL: Yeah, cool.
HEDVIG: And then it has like some super toxic part of that community. So, they’ve started to feel… yeah, that maybe WOMAN is no longer a term. Anyway. Sorry! I’m getting off track!
DANIEL: I feel this idea of language games touches on a lot of things, like not only how language began – that’s the first thing I think of — but it also reminds me of how children acquire language. Could you tell us a little bit about child language acquisition? In the book, you mention construction grammars, but I’m not really sure I’m on top of that.
NICK: Do you want to pick that up, Morten, or shall I?
MORTEN: Well, you can start, then I’ll…
NICK: So the thing about construction grammars, they are an approach to understanding the structure of language, which work… where they basically use small chunks, so they’re essentially chunks which have a both a sound and a meaning. So, they might be just the individual words or morphemes — parts of words which are meaningful — or they might be ways of putting words together, as they might be ways of putting together some nouns or verbs or determiners and nouns and so on. But they’re little chunks. So, the idea is to think that each construction is just a little package, which helps you connect sounds and meanings. But then, the language itself is built up by a very large number of these, which then interact in complicated ways. So, rather than thinking — as Chomsky used to think that there was… and still does — that grammar has a fundamental underlying deep abstract structure. It’s like sort of mathematical system of great complexity. And then, individual languages are kind of variations on that structure.
The construction grammar approach goes the opposite way. It says: No, no. Rather like the charade story, you build up a language by learning the pieces. But then as you build more pieces, you can interlock and interact some more and build further pieces. So, you create this thing gradually. The research on child language acquisition seems to very much fit with the construction grammar perspective. Children seem to learn these little patterns one by one. There’s no point where they suddenly master some highly abstract grammatical rule. They seem to learn the elements of language in a kind of piecemeal way. And second language acquisition seems to be the same too.
MORTEN: And it’s worth noting that, for example, with Stewart McCauley, I built sort of a computational model that takes small chunks of language and can use that both to capture aspects of comprehension, but also to do production. We’re able to model the acquisition of language as essentially learning to pick up on these kinds of chunks or constructions across 29 old world languages. Now, we’re not claiming it to be universal. We’re just saying 29 old world languages in this case, but it’s a basic idea.
And secondly, when we look at adult language use, and so there’s been a lot of corpus analysis of how we interact with one another and we can actually account for about at least 50% of almost everything we say, just by looking at small chunks that are reused, little formulaic sequences that we use again and again. So this notion we have that everything we say is new and fresh, and, of course, every now and then we say something that’s new and fresh. But oftentimes, we’re just reusing all [HEDVIG LAUGHS] sorts of different chunks. What we’re doing is we’re picking up on these basic construction or patterns, and then we’re able to reuse them in the same way as we can do when we’re playing charades. And that’s where we think there’s this nice connection between the two.
HEDVIG: That’s really cool.
DANIEL: Hedvig, you’ve said this before, that language is potentially infinite, but we mostly just use it to say the same things over and over again, like, “How are you doing?” and “Pass the salt.”
HEDVIG: I know, I just don’t think we’re that special. When people get creeped out that, like, companies like Meta are really good at making targeted ads to like, “Oh, they’re listening to me.” And it’s not, it’s because you’re really similar to the rest of your demographic. You’re not that special, and not being that special is okay. And it’s like what Morten said, probably most of sentence you say, you’ve said something quite similar to it before. You’re not saying unique and you never uttered before sentences all the time. Like Chomsky says, “You can. I can say…” No, I can’t. I’ve got to think something.
[LAUGHTER]
HEDVIG: I can say entirely absurd things. But why would I? I usually don’t need to communicate very absurd things. [CHUCKLES] I usually just need to communicate simpler things!
MORTEN: I think that’s an important point, because most of the time, we are using language to communicate, to make ourselves understood, and it doesn’t necessarily make sense to try to do that in a completely new novel way that the other person would have a hard time understanding. If previously we have communicated, and we’ve used certain combination of words to do that, and we want to say the same thing again, it makes sense to reuse the same thing again, because that’s what language is for, is for communication. Yes, we can use it to say new things. We, of course, come up with new ideas all the time, but they’re typically phrased in terms of what has come before. So, even when there are new things, we are building it out of what came before.
HEDVIG: What about this other use that some language theorists like to say language has, which is for structuring our internal thought, not for communication? What do you think about that as a…?
NICK: Yes, I think that’s a pretty bizarre thought really. Apart from anything else, all the things that are particular for individual languages, we seem to have roughly the same thoughts, but we can express them in a very different way, in different languages. The idea that by having a phonology, say, and a particular set of syntactic constructions is somehow helping shape your thoughts or is allowing thought to be possible is pretty peculiar. One thing people think naturlly is to think that languages are just like other ways of conveying information, so notations, for example, being one for, say, mathematics. So, having notations is very helpful. If you didn’t have decimal notation, doing mathematics, doing arithmetic could be much tougher. So, having good notations is a great thing. And that’s like having the right words, having a sufficiently powerful set of communicative tools, but we’re inventing them all the time. So, it’s not the language that fixes some boundaries, and you say, “Well, but this language, you need to say these things, you can’t say those things.” This is where the charades story is so useful. Of course, we can say anything, and we just have to create the charades in the moment to do it.
The idea that having the right language is somehow… its fundamental purpose is allowing us to think is peculiar, I think. Which doesn’t mean to say that the ability to communicate with each other doesn’t transform completely our ability to think because it allows us to connect with other minds and that is a big deal. But that’s the transformation of thought via communication. It’s not viewing thoughts as somehow primary purpose of language.
HEDVIG: But you’re right that we can say anything in any language. But we still have different grammars and words in different languages that allow us to express certain ideas more or less succinctly. And also, grammar can also dictate that you need to make a certain kind of distinction every time you speak a certain language and not in others. Most Austronesian languages have an inclusive/exclusive way, whether you include the speaker or the hearer, which presumably means that they’re more aware of that distinction maybe than other speakers, right? They’re not all exactly the same.
NICK: yeah, I think it’s not that language is irrelevant. It’s not irrelevant to thought, it just biases you in one direction or another.
MORTEN: Yeah. Just like when you’re playing charades, if you’ve found certain ways of doing it that might be easier to express certain kinds of, say, movie titles or book titles, based on what you’ve played with before, the same is true for language that speaking a particular language might make it easier for you to talk about certain things, but it doesn’t make it impossible to talk about other things. It just is going to make it slightly easier or slightly harder. So, at least that’s our take on the impact on language on thinking: that it doesn’t determine it, but it can make it easier or harder. And this may be why certain languages, because it might be important for their culture for their whatever that it’s relevant to their environment, to pay attention to certain distinctions, whereas for others, maybe it’s not so important, and then becomes entrenched in the grammar of that language.
I mean, again, if you look at all the languages across the world, we are looking at language change and language evolution is through cultural evolution. So, different languages can end up in different parts of that space. They are constrained by the way our bodies work, the way our thought processes work, how we interact through social pragmatic interactions, and also limitations on cognition, learning, memory and processing. But all these constraints, they don’t determine the cultural evolution so that it only can take one direction. You get the history of the group of people that speak a particular language. As they move, there’s going to be certain changes that are going to be easier for them to adopt in that language versus another language. Over time, we get this gradual evolution of different kinds of systems, different ways of interacting through language, but they’re all constrained by these different patterns… sorry, with these different constraints, like from the body, from how we interact, from how our minds work, and so on.
HEDVIG: And are those constraints the reason why certain structures or certain ways for languages to be are more common than others? We know, for example, when it comes to basic word ordering languages, we know that like SVO — subject, verb, object — and SOV are more common than almost any of the other ones, roughly. Does that mean that our general human constraint that you spoke of, our cognitive processes, our bodies, our minds are so similar across our species that certain things are just universally a little bit easier?
MORTEN: Yes, I would think that’d be the case there, that it forces us in certain directions, but it’s unlike this notion that, for example, Noam Chomsky or Steven Pinker talks about, sort of like Universal Grammar or language instinct that’s very specific in nature, that they are constrained specific to language. But we’re suggesting the constraints on language evolution really are more broad. They’re based on the cognitive system, the way our bodies work, but they’re still constraints nonetheless. And so, there are going to be certain things that are going to be easier for us to do, based on those constraints than others. And that becomes sort of… might come into these languages, such as the example you mentioned with the word order patterns that we see.
HEDVIG: That makes a lot of sense to me. I think a lot of discussions we’ve been having over this show for a couple of years, we often just pull it out to, but what beyond language? What is the communicative context of what you’re trying to do? And how could that shape, explain what’s going on, instead of just focusing in on language as a sole sort of phenomenon. I think that makes a lot of sense.
DANIEL: Since you mentioned Chomsky, how would you respond to poverty-of-the-stimulus arguments? We’ve had chats with generativists, and one thing they’ve said is: you could look at a whole bunch of languages, and there are certain incredibly specific things that you just never see. And the generative explanation is because that violates the Universal Grammar that is born in our brains, and is part of our genetic human inheritance. Why do you think that we never see certain things in language? And I can be more specific if you’d like.
MORTEN: Let me start by saying that I think it’s actually time to retire the poverty of the stimulus argument. So, I’m going to be bold and say that here. I think that argument made sense when it was first proposed back in the 1960s, because at that time, we knew relatively little about how kids actually acquire language. There was very little experimental data about how kids acquire language. And also, there was not a lot of knowledge about what’s available in the inputs. Most of it was based on intuitions about it, “Well, kids don’t say this,” or “They don’t hear that.” But really, there was very few analyses on what actually kids hear. And then on top of that, there was actually not a lot of knowledge about what simple computational mechanisms could do, that mechanism that did not have any kind of built in, say, linguistic constraints.
But now, we know that kids are really very good learners. They’re excellent learners. As people have been doing lots and lots of experiments with infants and young kids, we can see they’re able to pick up on all sorts of kinds of statistical and all kinds of regularities. And on top of that, we now have the availability of large databases of speech directed to children and so on, or in the background that people have been analysing, and we’re finding out that there’s a lot of information in what we say to kids that they can potentially use for learning their native tongue, without having any built-in linguistic constraints, such as in Universal Grammars.
And then, of course, through the advance of computers, we’ve been able to build complex models of language acquisition, but actually not necessarily that complex. And all these models do not necessarily have sort of innate linguistic knowledge. I just want to mention just one example, I think it’s interesting existence proof. So, there is this model out there called GPT-3. It’s a deep learning, large-scale language model. It’s a gigantic model. It’s trained on billions of words and so on. It’s been criticised for producing biased speech sometimes… or not speech, but text that might be racist, or sexist, and so on. Sometimes, it says things that are just silly, nonsensical. You can ask it questions, and it can sort of say… you ask it, “How many rainbows does it take to get from here to Hawaii?” And it might say “Two.” The question is nonsensical, and so is the answer.
But I think what people have overlooked is that its output is almost always grammatical. And so, what you have here… and this model is trained up on just trying to predict the next word in a sequence, that’s all it’s been trained up on, of course, on billions and billions of words. But this is an existence proof that here you have a learner — a silicon one — that can learn language and produce very passible human text. So, it’s been tested on producing news articles, and all sort of other articles. And we’re actually running a project where we actually making it produce poetry here at Cornell. And it can axtually do… the newest version can do a very passable version of, say, Emily Dickinson or other authors.
HEDVIG: Interesting.
MORTEN: And so the point here is not this kind of model is learning language like human children are, but we have an existence proof that it’s possible to learn language and produce essentially loads and loads of new language that these models can produce without actually having any Universal Grammar in there. There’s no grammar rules in there, there’s no Universal Grammar in there. Yet, it can learn language. Now, so far, of course, most of the discussion about the poverty of stimulus have been: either you believe in Universal Grammar, or you argue for something else. But here we actually have a thing, a system that can actually use language in a way that’s not unlike the way people use it. But, of course, it doesn’t understand anything. So, that’s important to keep in mind. And clearly kids have all sorts of other information when they’re learning language through interaction, through playing charades, and so on.
So getting to the second part of the question, which was about why it is that there are certain patterns we don’t see and others we see, this goes back to the kind of constraints I talked about earlier, about how the way our bodies work, the way our minds, our thought processes work. Those are the kinds of things that are going to make certain things possible and others impossible. So I think, instead of moving away from saying that you need specifically linguistic constraints to explain these patterns or absence of patterns, as opposed to having, say, that these derived from a suite of other more general skills that are not necessarily dedicated to language.
NICK: Just to add one tiny thing to that. It’s quite common for people hearing this kind of argument to say, well, in any particular case, how are you going to explain this particular constraint using these pragmatic, or cognitive, or perceptual factors? And often one can have a go at that. But if you say, “Well, what’s the alternative?” I mean, if we’re going to postulate a gene, or some kind of constraint, some instinctual principle, what’s that? The idea that there’s some gene that’ll explain why some particular form is possible, and another one isn’t, it’s completely impossible to create any concrete example at all. And, of course, we know things like writing systems obviously have lots of constraints in writing. You could use pictures or things that can be made with a pencil, which can have any form at all, and they could convey information, all kinds of crazy ways. But actually, we tend to use nice natural strokes, and quite often use alphabets. But clearly, that’s not a product of some writing instinct, because writing’s only been around for a few thousand years.
HEDVIG: And it’s also a funny view of genetics, I think. We were talking about this in my department the other week that if you look at how geneticists study other parts of our genome in our body, it’s very rare that you have one gene that does one thing like that. Maybe there are certain gene mutations that make you more likely to have certain diseases or cancer, or something, people have the breast cancer gene and things. But for general larger systems, it’s not even one area of our brain. It’s not one gene that’s doing it. So, why would that be true of language? Especially because language we know from the unfortunate studies we have from children who have been brought up with very little language input, that you need a cultural context to develop a language fully. And it makes sense… if I was evolution, cultural evolution is a lot cheaper than genetic evolution. It’s a lot easier to hope that the parents transmit certain information, that the community transmits certain with generation and generation by speaking to each other, without hardcoding it selection-wise into the gene. It takes a lot longer and it’s a lot more expensive.
DANIEL: That’s where I was going, because it feels like the folks who postulate a genetic origin of language are just really looking in the wrong place, because if you postulate that language is a part of our genetic human inheritance, it becomes very difficult to explain why language would, number one, change over time. And number two, why it would vary between communities? Why would it? Why would language change any faster than human genes change?
HEDVIG: But they don’t think it does.
DANIEL: Yes, I know, [HEDVIG LAUGHS] not in ways that they think are interesting, but I think that they are interesting. And notice, by the way, that if you want to take this view, you have to become a denier of language change and language variation. So, if you viewed language as a series of improvisational games, then it becomes easy to explain why language changes. It becomes easy to explain why language varies. It’s very hard otherwise.
NICK: Yeah. And of course, animal communication systems are so stable, aren’t they? The bee waggle dance is just the same for every bee, which is genetically from the same species. They all do the same waggle dance with the same meaning. And that will be true across communicative systems in the animal kingdom. So one of the things Morten and I are always quite fond of is that the language instinct story works pretty well for bees, but it’s not so good for people!
[LAUGHTER]
HEDVIG: That’s funny. I like that.
DANIEL: I love the book. I’ve been reading it, and the thought that I had when I started reading it from the very first bit was a sense of relief that someone was actually saying the things that you two are saying. Like, the thought that I had was, “Finally, somebody has said it.”
NICK: I think one of the odd things in writing this book is that in some ways, it feels like something that’s been building for a long time. It is a kind of… well, there’s lots and lots of… this is an accumulation of a lot of work that we’ve done it for years. But, of course, it’s not really that. It’s accumulation of work that’s been going on for decades by a wide range of people who are working with a variety of different perspectives, but by breaking away from sort of the Chomsky-Pinker view of how language works. and in some sense, lots of people who would read this book and think, “Well, I kind of thought that anyway.” or “Oh, now it’s sort of set down, but that’s kind of how I was seeing it.” And, of course, we’ve been incredibly inspired by huge numbers of those people. So, the hope, I think, is if people do read this and think, “Yeah, that’s a good crystallisation of what I kind of knew anyway,” that would be great.
[LAUGHTER]
HEDVIG: I mean, that’s a little bit how I felt. So this book, who do you imagine the audience is? Who should read this book?
MORTEN: We wrote this book to be hopefully accessible to a general reader. It’s not an academic book. We’ve done that before. But this is really meant to be something that the average person who might be interested in language, how language works, why we have language, why are other animal communication systems different, and so on, and many other questions. But also, they’re understanding the importance of language in explaining cultures and broader societal complexity and so on. And so, it’s really meant for everyone. But also, we are hoping that academics too, will find it of interest. So, those who actually work with language also.
And so, in working on this book, we have developed new ideas we think that… of course, building on previous ideas, but new ideas about how to think of language as charades, and so on. So, we’re hoping that will also inspire people working within the language sciences, broadly speaking. And so, it’s meant both for these two kinds of audiences.
HEDVIG: That’s really cool. It’s nice with a book that is directed at a nonacademic audience as well. I think that’s very important, especially as a counterpoint maybe to Language Instinct.
MORTEN: Indeed.
DANIEL: The book is The Language Game: How Improvisation Created Language and Changed the World. It’s by Dr Morten Christiansen and Dr Nick Chater, available from Basic Books in the US and Canada. But if you’re in the UK and Australia, it’s getting a release from Penguin. Check under the imprint Bantam Books. That’s The Language Game, be sure to check it out. Thanks so much for telling us all about it.
NICK: Well, thank you for the opportunity.
MORTEN: Thanks for having us.
[TRANSITIONAL MUSIC]
DANIEL: You’re not done yet though, because we’re doing Words of the Week.
HEDVIG: Yay.
DANIEL: Morten, Nick, do you have a word for us? Is there a word that you’ve noticed or that you’ve thought was kind of interesting that you wanted to throw at us?
NICK: Well, I was going to go backwards in time to a word from my mother. So, this is a word that always struck with me particularly strange, which is the word COCKY DUNDAS. A COCKY DUNDAS is somebody who is…
HEDVIG: Excuse me??
NICK: A COCKY DUNDAS. I think, it’s one word, it might be two. A COCKY DUNDAS who is someone who’s very arrogant and full of themselves. So, my mother would say, “Oh, he’s a bit of a Cocky Dundas.” [LAUGHS] And this is a very strange fringe word. There are no other Dundas. It’s just Cocky Dundas, but it was in fact the nickname of a pilot during the Second World War, so I assume his second name is Dundas. So, I assumed that this pilot was known as because his nickname was Cocky, I assume, because the word Cocky Dundas was also out there at the time, because my mother as a child in the 30s was picking this word up before. But it’s a lovely word, I think. You just sort of feel the sense of just slightly irritating puffed up importance of someone who’s a Cocky Dundas, who needs to be taken down a peg or two.
[LAUGHTER]
HEDVIG: I like that. Where in England is your mother from?
NICK: She’s from Kent. She was brought up in Tunbridge Wells, which is on the border of Kent and Sussex, which is just about 30 miles south of London. So, I have no idea how geographically specific that is. I have no clue.
HEDVIG: I was just curious.
DANIEL: Do you feel like a lot of people would have known this word?
NICK: No. I don’t think many people do.
HEDVIG: Shall I go see with my Northern husband? What you call it? Cocky Dundas?
NICK: I would. Cocky Dundas. Check it out. I’d be very interested.
DANIEL: I see here that cocky — I’m checking this out at Etymonline, the Online Etymology Dictionary — calling somebody cocky dates from 1768. And it does appear to be from the bird, the cock, who is a very proud sort of bird.
NICK: Yeah.
DANIEL: Hedvig, what’s the verdict?
HEDVIG: He doesn’t know it [CHUCKLES] and his guess is really funny.
[LAUGHTER]
DANIEL: Oh, right. What’d he say?
HEDVIG: He thought Cocky Dundas meant underpants.
DANIEL: You know, the Dundas is reminding me of like a dunderhead. Somebody who’s a bit silly.
NICK: Yeah. Maybe something to do with it, yeah.
DANIEL: Maybe that plays into it, I have no idea. Well, thank you. COCKY DUNDAS.
HEDVIG: He smiled, he thought it was funny. So, you know.
DANIEL: Aristemo on our Discord server has given us one. GOBLIN MODE.
HEDVIG: Hmm.
DANIEL: This comes from a Guardian article by Kari Paul – Slobbing out and giving up: Why are so many people going goblin mode? Obviously, the times are a little different now. We spend a little more time indoors, and sometimes we sit in our chairs doing whatever, answering emails, and maybe going to the fridge or maybe we go out for a little while in soft pants, but we don’t dress up or indeed have very good hygiene. And when that’s you, you are in GOBLIN MODE. Self-described goblin, Dave McNamee, says, “Goblin mode is like when you wake up at 2am and shuffle into the kitchen wearing nothing but a long t-shirt to make a weird snack, like melted cheese on saltines.” That is not weird. I will defend the melted cheese on saltines.
HEDVIG: I was going to say, a lot of things here isn’t that weird.
DANIEL: No. But then, are you in goblin mode? Come on, Hedvig, be honest.
HEDVIG: I don’t know. I’m wearing the same shirt and pants that I wore yesterday, and I think that’s fine. And I haven’t been outside today, I think. I think that’s fine. I resent being called a goblin.
DANIEL: I think it’s something you have to embrace. I don’t think you’re a goblin until you self-identify.
HEDVIG: But the Guardian title has the word “giving up” on it.
DANIEL: Yeah.
HEDVIG: I’m giving up on life or anything like that. I’m just, like, giving up on certain society standards!
DANIEL: Some of the most wonderful things in my life are things that I gave up on. It’s not a criticism. Or is it?
HEDVIG: I think when the Guardian calls it giving up…
DANIEL: Sorry.
MORTEN: I also have a possible Word of the Week. It’s KAMELÅSÅ.
HEDVIG: [GASPS]
DANIEL: I know this.
MORTEN: This is a nonword, it’s not a real word. It’s a made-up word that some Norwegian comedians came up with in making fun of Danes. The difficulty that loads of people have in understanding Danish, including Danes themselves. I’m a head of a project in Denmark called the Puzzle of Danish Project, where we’re looking into how the opaque nature of the sound structure of Danish seems to be influencing not only how Danish kids acquire their language, they seem to be doing it more slowly than, say, Norwegian kids and Swedish kids. But also, it seems to be affecting how adults process language from both speech processing, all the way up to the way they do dialogue. So this KAMELÅSÅ is just a wonderful word. People should look up KAMELÅSÅ on the web, and they can find a video on YouTube by these comedians, and it’s incredibly funny. I’ll highly recommend it.
DANIEL: It is a really funny sketch. Two Danish speakers don’t want to admit that they can’t understand each other. So, they just sort of make up these words, and one of them is KAMELÅSÅ.
MORTEN: So, if you want to say something, but you don’t know what the right word would be, you just say KAMELÅSÅ.
HEDVIG: I’m trying to google because we had one of your postdocs, I believe, on the show talk about the project.
MORTEN: Oh, yes, yeah.
DANIEL: It was Fabio Trecca.
MORTEN: Fabio Trecca was here.
HEDVIG: Yes!
MORTEN: He would talk about this, for sure.
DANIEL: [CHUCKLES] We were very grateful to Fabio for telling us about The Puzzle of Danish.
HEDVIG: It was really fun.
DANIEL: It was great. We’ll be sure to put up a link on our website becauselanguage.com.
There’s just one more. We need a word for this. It seems that Oscar-winning screenwriter, Aaron Sorkin, takes up to eight showers a day whenever he suffers from writer’s block to help revive his creativity. If he’s writing, and he’s writing and he’s struggling and he’s getting nowhere, he stops, he goes, there’s a shower in his office. He takes a shower, gets dressed in entirely new clothes, pretends that the day is new, pretends that he’s getting up, ready to write, and he tries again, and again, and again. Do you have…
HEDVIG: Groundhogging.
DANIEL: What?
HEDVIG: Groundhogging!
DANIEL: Oh, that is good. Oh, oh. Okay, now I really liked that. Sorry, Morten, Nick, you understand the reference, I’m sure.
NICK: Yeah.
HEDVIG: I haven’t even watched the movie.
DANIEL: It’s a fun one.
HEDVIG: You could call it Russian Dolling.
DANIEL: You could. But groundhogging is very good, because now that takes Groundhog Day, which has already syntactically changed. There you go. It used to be a holiday, then it’s a sensation of deja vu. But now we can take it and turn it into a… just we could strip off the Day, just make it a verb and it’s… starting over. It’s starting over afresh.
MORTEN: And also, it captures the notion that you keep on doing it until you get it right. So here, he would keep on taking the shower until he can break the writer’s block, so to speak, so it kind of makes sense from that perspective. That’s great, Hedvig!
DANIEL: That was good. I was going to go with SORKIN RESET, but I really like GROUNDHOGGING much, much better. And I love how it’s an illustration of the language game where you can take old stuff that has a meaning and you can just sort of pick it up and use it again for a different purpose. That’s such a clear example.
HEDVIG: I don’t know how often, Morten, you listen to this show, but as a Swedish person, I may or may not have joked about what Danish is like.
MORTEN: Ah. [LAUGHS]
DANIEL: Totally has.
HEDVIG: And it’s very funny to me that you chose KAMELÅSÅ. You also are able to have a society and a parliament, and so clearly the language works enough.
DANIEL: It works!
MORTEN: It works well enough, but it is interesting. We’ve been doing analysis, for example, how people interacting in dialogue… so we recorded dialogue of people either doing some tasks together or just shooting the breeze. And we’re seeing some interesting differences between Norwegians and Danes in that regard. So, one of the things we thought Danes would do would actually be engaging in more repair. So, given that Danish has fairly opaque sound structure, we might think that you will just keep on saying like, “Oh, why did you say?” “What? Huh?” But, no, Norwegians do more. What Danes do is actually, seemingly… this is our interpretation. They appear to be preventing that by doing entrainment or alignment. So, they repeat the words more often that the other person is doing. They use more similar syntactic constructions or semantic topics. So, by doing that, that actually facilitates communication, and so we think might result in having fewer repairs. These are data. Well, they’re not published yet. They’re about to be submitted, so soon you can read about them. But it’s kinda cool.
HEDVIG: That’s really cool!
DANIEL: Oh, we get a special preview on Because Language. Awesome.
MORTEN: Yeah!
DANIEL: All right. Well, in that case, COCKY DUNDAS, GOBLIN MODE, KAMELÅSÅ, and GROUNDHOGGING: our Words of the Week. Nick Chater and Morten Christiansen, thank you so much for coming on the show and telling us about your stuff. It’s been great having you.
MORTEN: Thank you for having us.
NICK: It’s been real fun.
DANIEL: All right, we have some comments. One is from PharaohKatt. Now, the background is that we defined ALLISTIC as non-autistic but we weren’t sure exactly how that Greek ‘allos’ meaning ‘other’ fit in. So PharohKatt says, “Hello, autist here. Allistic is specifically not autistic and is used instead of neurotypical for the sake of accuracy, not in some separate sense. Also, the term ‘autistic’ comes from “autos” meaning self, the idea that we are solely self-focused. Allistic is the opposite in the sense that allistic people are other-focused. Basically, we are reclaiming it from the idea that we’re all selfish.” How about that?
HEDVIG: Right. Cool. So, ALLISTIC is formed… so even though maybe not everyone who’s autistic sees themselves as self-focused, they still picked up on that contrast in Greek to form the term for people who aren’t autistic, as allistic.
DANIEL: And that is the tie-in to Greek ‘allos’. I noticed also that one of the definitions, we said allistic means non-autistic, but we also said not “affected by autism””. That gave me weird feels a little bit. At least, we didn’t say not “suffering” from autism. But the more I thought about it, the less I liked, not affected– the word ‘affected’ just did something to me, so “non-autistic” or allistic. I just found I like that better.
HEDVIG: I think the reason maybe — if I was one who said that — is that I know there are people who don’t like the idea of being called, for example, autistic person as if that is like their entire personality, but that they have a lot of other traits as well. And one of them is that other people have said that they are different in this way and the term ‘autistic’ applies to them. And then, people say, yeah, just to flip it a bit instead of… because when you say autistic, some people might think, “Oh, that means everything in my life is determined by me being autistic.” So, I think that’s why, but I agree that ‘affected’ has a negative connotation, and that’s not something everyone feels. If PharaohKatt or someone else could… they describe themselves as autist. “Hello, autist here,” the message is. So, that’s as least what some people prefer. In that case, am I an allist?
DANIEL: I suppose we could be.
HEDVIG: Okay!
DANIEL: I’ll be that.
HEDVIG: Neat.
DANIEL: Our next comment comes from Nicole. In our last episode, our listener Nicole pointed our attention to my mention of Hedvig’s “relatability”. Would you like to read this one? We had a big discussion about it, and then she got back in touch. So, yeah.
HEDVIG: We did. We talked a lot about it and how media representation and visibility is different for women and men. And the thing that we talked about last thing that I wanted to highlight, again, is maybe what Ben said, which is that I am in a different place in my career and my life, then, for example, you two. There are consequences for me if I am disagreeable. [NERVOUS CHUCKLE] But also, in general, in my professional life, and also in a lot of personal relationships, I try to be not edgy with people. [LAUGHS] I think that makes it a nicer place for everyone.
DANIEL: And I tried to do that too. But I think what we really hit on was the lack of a parallel expectation for men versus women in that area, and how women bear a disproportionate share of risk by speaking out. And that kind of sucks.
HEDVIG: Maybe that’s true. I wonder if that’s the case in our dynamic though, or if it is our different life situations, rather than specifically that I’m a woman, but I don’t know.
DANIEL: Well, can I get you to read Nicole’s comment?
HEDVIG: I’ll try and think of some edgy views. [DANIEL LAUGHS] Nicole said, “I just heard your response to my comments, and a heartfelt thank you for your graciousness, sincere reflection, and interesting discussion.” I’ve never been called graceful, gracious before.
DANIEL: I like gracious.
HEDVIG: “In addition, I have a word to contribute to your follow-up question about words and posts on women to make us more nonthreatening. In graduate school, a male visiting professor called my analysis of WH-exclamative constructions ‘cute’. How is that for a diminishing compliment? That would never be made to a male theoretical linguist. I look forward to hearing what other listeners recount.” Yeah, don’t call people’s efforts cute, anyone. You can maybe call their outfits cute if you are enough friends with them for that to be okay. But don’t call other people’s analysis cute. Unless it is… I have friends who analyse Pokémon names.
DANIEL: That could be cute.
HEDVIG: They like to say that the data is cute, at least.
[LAUGHTER]
HEDVIG: That’s fairly noncontroversial, that Pokémons are cute, or at least some of them. But yeah, let’s not talk about each other’s analysis like that. I would argue especially not to women, because it is something that happens more often. Yeah.
DANIEL: This is not to take anything away from Nicole’s point. But I did have a supervisor call a big result that I had “cute” even though I’m male. It was atypical, and yeah, it wasn’t very nice. I think it was dismissive. In fact, this was the supervisor that dismissed my work and [LAUGHS] wasn’t very supportive! [SIGH]
HEDVIG: That’s not very nice.
DANIEL: Yeah, it’s not very nice. But that’s not to take away from Nicole’s point that, yeah, it is dismissive and kind of terrible.
HEDVIG: Yeah. Unlike sometimes when people say things that hurt you, you can in your head be like, “Maybe they meant it like this.” I had a friend who one time called me pathetic, because she thought that the word PATHETIC meant full of emotions.
DANIEL: [CHUCKLES] Oh, right. Why did she call you that anyway?
HEDVIG: She was like, “Ah, Hedvig, I love you. You’re so pathetic.” And I’m like…
DANIEL: [LAUGHS]
HEDVIG: “What? I don’t think that means what you think it means.”
DANIEL: No, I don’t think so.
HEDVIG: But at least then, my face and my reaction told her that something was wrong. I was like, “What do you think that means?” She’s like, “Full of emotions. Pathos!”
DANIEL: Pathos.
HEDVIG: Yeah, no. That’s not…
DANIEL: Newp. Semantic shift again, right?
HEDVIG: That’s not how that works, at least for the last 800 years, you know? But then, I could understand where she was coming from. I had another male senior person say that one of my qualities as a scientist is that I’m very sociable.
DANIEL: Hmm. I’m getting flags there. I’m getting some red flags.
HEDVIG: Yeah. It was not fun. And it made me think I shouldn’t be a scientist for over a year.
DANIEL: [LAUGHS] You’re a real team player.
HEDVIG: But I later realised that was something that he thought made me stand out, that other people in my class were smart, but that you needed to be sociable to be a scientist and that was a unique skill that I had.
DANIEL: Okay!
HEDVIG: So he meant it in a sort of…
DANIEL: Yeah, I guess so.
HEDVIG: Yeah. yeah. But when someone calls your analysis cute, there’s not many excuses you can think of.
DANIEL: No, not really. Even though we try, right? We try to explain away bad behavior, “Oh, perhaps they meant.” Instead of saying…
HEDVIG: Yeah. But that one, I don’t think so.
DANIEL: No.
HEDVIG: I don’t think so.
DANIEL: Well, Hedvig, thank you for being here on this show with me today. This has been a lot of fun. I’m so glad we got to do this.
HEDVIG: I’m glad you’re here.
DANIEL: Mm-hmm. I’m glad I’m here too! [LAUGHS]
[END THEME]
HEDVIG: If you like our show, which I believe you do, if you’ve listened for this long into the recording, because then we come to the end.
DANIEL: And how could you not?
HEDVIG: Well, I mean, maybe you’re one of those people who always skip the end. But if you do like our show, we’re very happy that you do, and we would love to know about it. And you can get in touch with us and tell us why you like our show. I think very often in the world, people only get in touch when something is wrong. And it’s very nice when someone gets in touch… even if you just like us like a normal amount, even if we’re one of your, like, decent podcasts in your feed. Even if we’re not like your absolute favorite one, that still means a lot to me that we met your mean criteria of quality. So, if you believe that we are decent, we would like you to get in touch with us. You can also give us ideas for segments like Words of the Week, or news items or other thoughts you have about our content. You can get in touch with us in many places. We are @becauselangpod, essentially, in a lot of places, not on Spotify. You can leave us a message on SpeakPipe. That means that you can record a little audio snippet that will be sent to us, and we can play in the show. Or, for example, if you have some thoughts, or if you have a Word of the Week, and you want to tell us how to pronounce it, and we would love to hear your lovely sweet voices, you can do that on our website, becauselanguage.com. You can also send us a good old-fashioned email hello@becauselanguage.com. And that’s all the ways you can get in touch with us.
If you want to help and spread the good gospel, you can also [DANIEL LAUGHS] tell other people about us. That’s really nice. Like Dustin of Sandman Stories does all the time on Twitter. And you can leave a review for us and we read reviews. And if you leave nice content in the reviews, we’ll maybe read it out in the show as well. We love a good review.
DANIEL: Hey, Hedvig, there’s a new review! Do you want to read it?
HEDVIG: We have a review. Yes! Okay. The title is “No, because.”
DANIEL: That’s a good title.
HEDVIG: Which was a Word of the Week for a while ago. “This is our go-to podcast when we are traveling hosted by Hedvig, Daniel, and Ben. We 😀, 😳, 🤔, 🥰” a lot of smileys. Okay. I’m going to try… We can do this as video clip. There’s one smiley that’s like this 😀. And then, there’s one that’s like… no, that one has like blush cheeks 😳.
DANIEL: Eyes open.
HEDVIG: And then there’s a thoughtful one.
DANIEL: Yep, thinking. Yep.
HEDVIG: Blush cheek, and then there’s a heart one, with all the hearts 🥰. So, we’re generating a lot of emotions in this group of people, which is nice.
DANIEL: [CHUCKLES] Yes.
HEDVIG: “You get the sense there’s dark academia at play, and I just love the Word of the Week.” Dark Academia is an aesthetic that you can have in music and fashion. That was from Scibidoo via Apple Podcasts.
DANIEL: Have you read the Dark Academia page on the AestheticsWiki, by the way?
HEDVIG: Aesthetics fandom? I have. Yeah.
DANIEL: It’s so bullshit, because somebody obviously who knew me sat down and just wrote all the stuff that I do, because…
HEDVIG: But you almost never wear Dark Academia aesthetic clothing.
DANIEL: Okay. I very often wear Dark Academia clothing, especially when I was teaching, but come on… wearing Doc Martens? Listening to Eric Satie? Come on. They knew me. I felt seen in all the ways. It’s such an invasion of my privacy. Anyway, thank you Scibidoo for that wonderful review. Our patrons make it possible for us to do things, like make transcripts of the show, so that you can read them and search them, do corpus-based analysis on the stuff we say. Thank you to the team at SpeechDocs. They are doing a great job on our transcripts. They really go the extra bit for us. And I think they even check out the show notes page when they’re not sure about, like, a source or how something is spelled. It’s that good.
HEDVIG: Oh, nice.
DANIEL: Yeah. If you think that accessibility matters, you should probably get transcripts from somewhere, and SpeechDocs is a pretty good way to do it.
The people that pay for that is our patrons. So, here are some of them. Dustin, Termy, Chris B, Elías, Matt, Whitney, Chris L, Helen, Udo, Jack, PharaohKatt, Lord Mortis, Larry, Kristofer, Andy, James, Nigel, Meredith, Kate, Nasrin, Ayesha, Moe, Steele, Manú, James, Rodger, Rhian, Colleen, Ignacio, Sonic Snejhog, Kevin, Jeff, Andy from Logophilius, Samantha, Stan, Kathy, Rach, Cheyenne, Felicity, Amir, the Canny Archer, O Tim, and Alyssa. And also, Kate B, who did not contribute via Patreon, but with the one-time donation button on our blog, becauselanguage.com. You can smash that button too. And our newest patrons at the Listener level, Sasha, and at the Friend level, Elliot. Thank you to all of our amazing patrons.
Our theme music has been written and performed by Drew Krapljanov, who’s a member of Ryan Beno and of Didion’s Bible. Thanks for listening. We will catch you next time. Because Language.
[BOOP]
DANIEL: Are you showing me a cat? Is that what I’m seeing here?
HEDVIG: Yeah. I’m trying to give you a zoomed-in version of my cat.
DANIEL: It’s a great cat.
HEDVIG: She’s good. She’s a good one. She’s very asleep.
DANIEL: I make no claims as to the relative cuteness of your cat versus Ben’s cat.
HEDVIG: Well. Be that as it may.
[LAUGHTER]
DANIEL: I didn’t make any claims! Hello, Cattums. If you really want to see some attractive cats and other animals, you can head to our Cats and Other Animals channel on our Discord, patrons.
HEDVIG: That is true. I should put up more pictures of them there.
DANIEL: Why don’t you? Because people would like it.
HEDVIG: That’s a good point. I like our Discord, but I also get a bit overwhelmed by it because it’s so active. [CHUCKLES]
DANIEL: There’s a lot of stuff.
HEDVIG: Yeah.
DANIEL: I’ve been posting my Redactle scores. Redactle is awesome. Are you playing Redactle?
HEDVIG: Oh, my god.
DANIEL: It’s the latest -le. So, they give you a Wikipedia article, but all the words are blanked out except for prepositions like OF and TO and then articles like A, AN, and THE. And you have to fill them in and guess what the article title is. It’s so good.
HEDVIG: It’s so weird! [LAUGHS]
DANIEL: Yep.
HEDVIG: But I like it.
DANIEL: I love it too.
HEDVIG: It’s a good game. I’ll be very disappointed if for this year’s Linguistic Olympiad problems worldwide, there isn’t one that’s Wordle inspired.
DANIEL: Hmm, I will too. I’ll tell Henry.
[Transcript provided by SpeechDocs Podcast Transcription]