The sciences are facing a replicability crisis. Some landmark studies were once considered settled, but then failed when they were retested. So have any linguistic experiments been toppled? And how do we fix this?
Dr Martine Grice and Dr Bodo Winter have contributed to a special issue of Linguistics, and they join us for this episode.
Listen to this episode
Patreon supporters
Our patrons keep us talking! And they make our transcripts possible. So big thanks to our top patrons:
- Dustin
- Termy
- Chris B
- Matt
- Whitney
- Helen
- Udo
- Jack
- Kitty
- Lord Mortis
- Elías
- Michael
- Larry
- Kristofer
- Andy
- James
- Nigel
- Kate
- Nasrin
- River
- Nikoli
- Ayesha
- sneakylemur
- Moe
- Steele
- Andrew
- Manú
- James
- Rodger
- Rhian
- Colleen
- glyph
- Ignacio
- Kevin
- Jeff
- Dave H
- Andy from Logophilius
- Samantha
- zo
- Kathy
- Rach
- Taylor
- Cheyenne
- Felicity
- Amir
- and national treasure Kate B
Become a Patreon supporter yourself and get access to bonus episodes and more!
Become a Patron!Show notes
Prof. Dr. Martine Grice
https://ifl.phil-fak.uni-koeln.de/phonetik/institut/personen/prof-dr-martine-grice
Dr Bodo Winter, University of Birmingham
https://www.birmingham.ac.uk/staff/profiles/elal/winter-bodo.aspx
TikTok: Missing girl found after using viral call for help sign
https://www.bbc.com/news/technology-59206549
Why every parent should teach their child this viral hand signal for ‘help’
https://www.today.com/parents/how-danger-hand-signal-could-help-save-your-child-s-t238606
Study offers new insights into how we comprehend and evaluate pronunciation
https://phys.org/news/2021-11-insights-comprehend-pronunciation.html
[$$] Adriana Hanulíková (2021) Do faces speak volumes? Social expectations in speech comprehension and evaluation across three age groups
https://doi.org/10.1371/journal.pone.0259230
[$$] Jennifer Hay and Katie Drager (2010) Stuffed toys and speech perception
https://doi.org/10.1515/ling.2010.027
Which is “Bouba” and Which Is “Kiki”?
https://allthingslinguistic.com/post/627371292593651712/which-is-bouba-and-which-is-kiki-its-a-new
Bouba & Kiki in the Himalayas
https://blogs.ntu.edu.sg/blip/bites/bouba-kiki-in-the-himalayas/
What is the Bouba/Kiki effect and what does it mean for the origins of language?
https://www.euronews.com/culture/2021/11/28/what-is-the-bouba-kiki-effect-and-what-does-it-mean-for-the-origins-of-language
Regardless of The Language You Speak, You’ll Likely Call One of These Shapes ‘Bouba’
https://www.sciencealert.com/nonsense-words-make-people-who-speak-different-languages-think-of-the-same-shapes
Perceptual links between sound and shape may unlock origins of spoken words
https://www.sciencedaily.com/releases/2021/11/211117100114.htm
Aleksandra Ćwiek (2021) The bouba/kiki effect is robust across cultures and writing systems
https://royalsocietypublishing.org/doi/10.1098/rstb.2020.0390
Pamela Perniss, Robin L. Thompson and Gabriella Vigliocco (2010) Iconicity as a General Property of Language: Evidence from Spoken and Signed Languages
https://www.frontiersin.org/articles/10.3389/fpsyg.2010.00227/full
English-language proofreading for non-native authors in the language sciences
https://docs.google.com/forms/d/e/1FAIpQLSeksfoZR20t1OTSz03izaHT-iC9vdYCew5eecl_Nf7OlB60-A/viewform
[$$] Volume 59 Issue 5 – Special Issue: The replication crisis: Implications for linguistics; Guest Editors: Lukas Sönning and Valentin Werner
https://www.degruyter.com/journal/key/ling/59/5/html?lang=en
Of 2 Minds: How Fast and Slow Thinking Shape Perception and Choice [Excerpt] | Scientific American
https://www.scientificamerican.com/article/kahneman-excerpt-thinking-fast-and-slow/
The W.H.O. skips forward two Greek letters, avoiding a Xi variant.
https://www.nytimes.com/2021/11/28/world/asia/omicron-variant-name-covid.html
The Facebook Meta presentation
https://www.facebook.com/RealityLabs/videos/561535698440683/
The Real Reason Facebook Changed Its Name
https://www.theatlantic.com/ideas/archive/2021/10/facebook-metaverse-mark-zuckerberg/620538/
Hey, Facebook, I Made a Metaverse 27 Years Ago
https://www.theatlantic.com/technology/archive/2021/10/facebook-metaverse-was-always-terrible/620546/
Meet skimpflation: A reason inflation is worse than the government says it is
https://www.npr.org/sections/money/2021/10/26/1048892388/meet-skimpflation-a-reason-inflation-is-worse-than-the-government-says-it-is
Why McDonald’s Fries Used to Taste Better
https://blog.cheapism.com/why-mcdonalds-fries-used-to-be-better/
The Real Reason McDonald’s French Fries Taste Different Now
https://www.mashed.com/234662/the-real-reason-mcdonalds-french-fries-taste-different-now/
List of words ending with FLATION
https://lotsofwords.com/*flation
jongraz onn Tiktok
https://www.tiktok.com/@jongraz?lang=en
TikTokers Are Skipping School And Work Because It’s ‘No Bones Day’, Here’s WTF That Means
https://www.pedestrian.tv/online/bones-day-no-bones-day-tiktok/
Meet BreadTube, the YouTube activists trying to beat the far-right at their own game
https://theconversation.com/meet-breadtube-the-youtube-activists-trying-to-beat-the-far-right-at-their-own-game-156125
Pëtr Kropotkin: The Conquest of Bread | The Anarchist Library
https://theanarchistlibrary.org/library/petr-kropotkin-the-conquest-of-bread
Transcript
[Transcript provided by SpeechDocs Podcast Transcription]
Daniel: For my part, I’ve been editing this show for so long that I can actually tell when somebody is saying um, because the word um looks a certain way.
Hedvig: Um.
Daniel: Not the same for everybody.
Hedvig: Um, I’m looking at my waveform now saying um.
Daniel: Um.
Hedvig: Um.
[Because Language theme]
Daniel: Hello, and welcome to Because Language, a show about linguistics, the science of language. I’m Daniel Midgley. And with me now, she recently ranked a perfect 10 out of 10 on Room Rater, it’s Hedvig Skirgård.
Hedvig: I did.
Daniel: Hedvig, your room game is strong.
Hedvig: Yeah. I had basically given up on going into the office-office by now. So, I’m home officing. So, I’ve just tried to make my home office nice. Put plants in it, plants that don’t kill my cats.
Daniel: Mm-hmm.
Hedvig: Yeah.
Daniel: We’re going to need a retronym for actually going into a place to work.
Hedvig: Office-office, I thought, as opposed to home office? That’s how reduplication works in English, no?
Daniel: Office-office. I love a bit of reduplication-reduplication. Although you’re making it sound like the office-office is the real office– Well, it is the real office, isn’t it?
Hedvig: Yeah.
Daniel: But now home office is the real office.
Hedvig: Yeah, when I do go into the office-office, I frequently forget to like bring my mouse that I like or bring something or bring a dongle, and I’m like, “My setup at home is so good. Now I’m in this deprived environment.”
Daniel: Well, we are all the richer for your tasty taste in room decor.
Hedvig: Thank you. I like your cupboard as well.
[laughter]
Daniel: By the way, I tidied this pantry, and I threw out a bunch of junk that was cluttering up the place. It doesn’t show from this angle though, really. Maybe I’ll do a 360 in a bit.
Hedvig: Okay.
Daniel: That’s quite enough of that. We have two very special guest cohosts. From the University of Birmingham, it’s Dr. Bodo Winter. Hello, Dr. Winter.
Bodo: Hello. Happy to be here.
Daniel: Excellent. From the phonetics lab at the University of Cologne, it’s Dr. Martine Grice. Hello.
Martine: Hello. I’m happy to be here too.
Daniel: Yay. We’re going to be talking about some of your work in a very important special edition of linguistics about the replication crisis that has gripped the sciences, and linguistics in particular. So, we’re really looking forward to talking to you about that. Just quickly, how did you both get on this topic and how did you come to appear in this special issue of linguistics? What got you here?
Bodo: Yeah, that’s a good question. Martine and I, we’ve had a long history of thinking about research methods together. We’ve worked on several projects together where stats have been very important. So, when this opportunity of the special issue came about, we’re thinking, “Hey, let’s actually put some of those thoughts into a paper,” because we had some strong opinions, perhaps, on how stats and research methods are being done in the field, some of the rituals that exist in the field and what could be done better also pedagogically.
Martine: Yeah. We started talking about this in Hong Kong at Phonetics Congress way back then when you went to conferences. This was 2011, and we’ve been talking to each other ever since.
Daniel: Wow. So, this is the distillation of a long history of thought about this topic.
Martine: Yeah.
Bodo: Yeah. So, I’m at the University of Birmingham right now, but I visit USC of Cologne, which also happens to be my hometown. So, it’s like a nice opportunity to visit Martine and do work together, but also visit my family. I’m actually there almost every summer. So, we’ve continued this tradition, and also, I’ve been going there to do workshops with Martine’s students, and it’s been really nice to keep that connection going.
Martine: Yeah, it’s been fantastic.
Daniel: Wow. Okay, this is going to be a lot of fun. Let’s first have a bit of housekeeping. What’s coming up for the rest of this year, we’ve got a Journal Club episode. I think that might be a Zoom episode with some of our best friends on our Discord who bring us stories. We’re also going to be having our Word of the Week of the Year. So, we want to hear your votes. We’re going to have all the words that we’ve done, all of our Words of the Week on Twitter and on Facebook, and you can vote for them by giving them a thumbs up or a plus one or whatever you do on Facebook and Twitter these days. The winner will be our Word of the Week of the Year 2021. Hedvig, do you remember last year’s Word of the Week of the Year?
Hedvig: Yes. Wasn’t it the thing with citrus?
Daniel: [laughs]
Hedvig: The what’s-it-called?
Daniel: Yes, it was.
Hedvig: It was–
Daniel: Oh, [crosstalk] but you can’t remember, that means it might not be that great of a word.
Hedvig: It was about the event that happens when you’re peeling or dealing with a citrus fruit and the little cells on the peel burst and squirts into your eye.
Daniel: Yes.
Hedvig: It had the word orb and fuscate in it.
Daniel: Orbisculate.
Hedvig: Yes, orbisculate.
Daniel: Orbisculate. Man, I love that word.
Bodo: What a weird word.
Hedvig: Yeah, it has a very nice story to it.
Daniel: I know.
Hedvig: It was a very special story attached to that word. So, I really recommend listening or reading the transcript of why we chose that word. But I think we don’t have time to treat at all now, but it was very nice.
Daniel: Also, we’ve got a couple of new designs in our shop. Yes, we have a shop. There’s the “Language is involuntary telepathy.” Also, the “Chomsky is not the devil” design, which– [laughs]
Hedvig: Which I pushed for very hard. Yeah.
Daniel: You liked it as soon as I suggested it. And then, it just flowed naturally. I grabbed an old photo of young Chomsky– wasn’t he dreamy, by the way?
[laughter]
Daniel: Back in the day.
Hedvig: Maybe.
Daniel: And then, I put some horns on him and that was it. So those are on our shop.
Hedvig: Good job. Maybe for Bodo and Martine, we should say. So, we did two episodes about generativism, and we’ve been trying to build some bridge or make people more understanding of each other, because we feel–
Martine: Really?
Daniel: [laughs]
Hedvig: Yes, we’re trying to be good citizens and people. Because we feel we come from the non-generativist, I at least feel the responsibility for telling people on that camp that Chomsky is not the devil. So, yeah.
Daniel: Not actually evil was the best we can do. [laughs]
Martine: Not actually, right.
Hedvig: Not anti-Christ.
Daniel: Yeah. Contrary to your expectations, yeah.
Bodo: That’s nice.
Daniel: Yeah.
Hedvig: We’re nice.
Daniel: We try. What’s happening now is I’m getting our patron mailout together. All the items have arrived. I’m addressing envelopes, all patrons get it regardless of level. So, please make sure that your mailing address is correct on Patreon, or add it. If you haven’t, and if you’re not a patron, you probably want to be because our latest bonus episode was a mailbag with Dr. Caitlin Green. If you want to hear it, become a patron on Patreon at the listener level. And if you like what we do, please become a patron. You’ll be supporting the show. You can hang with us on our Discord server. We’re currently debating whether Australians use the ‘R’ sound when saying nor.
Hedvig: Oh, they definitely do. That was the first thing I ever learned about the Australian accent was Natalie Tran on community channel on YouTube saying no.
Daniel: Nor. Well, I say yes, but others are saying nor, including my partner who says, “No, we don’t say that.”
Hedvig: They’re just willfully deaf.
Daniel: Well, it’s only certain kinds. She doesn’t have that variety. Anyway, become a patron at patreon.com/becauselangpod. Thank you to all our patrons. Okay, that’s done.
Hedvig: Okay. Now, we get to the news– I’m going to have to be Ben, because Ben is not here today. So, now is the news segment. Daniel, please take it away.
Daniel: Thank you very much. We do have a bit of an update. We’ve been attempting to rename the euphemism cycle, because we’re trying to get rid of all the neologisms from Steuben Planker. So, this one came to us from Timothy via email. Are you ready?
Hedvig: No.
[laughter]
Daniel: Okay, how about now?
Hedvig: I mean I’m still– I thought the cycle wasn’t too bad. We had treadmill.
Daniel: It was good.
Hedvig: What does this person have?
Daniel: Bodo, Martine, you know what we’re talking about, with the euphemism cycle?
Bodo: When a word becomes used too often that it becomes more euphemistic?
Hedvig: Yeah, especially becomes derogatory or bad. You have a word about a people or profession or a gender and then suddenly, by just be associated with someone who is stigmatized in the community, that word in itself becomes stigmatized, then you have to make up a new one. So, you get things like facilities manager instead of janitor.
Daniel: You picked a good one.
Martine: I see.
Hedvig: Yeah.
Daniel: I think this one’s a bit Terry Pratchett influenced. Timothy says, “We should call it the circumlocution paternoster.”
Hedvig: No.
Daniel: [laughs] Which I loved.
[crosstalk]
Hedvig: I love all of our listeners equally, except when I hate them when they’re asking too smart questions, but we need a simple word.
Daniel: Okay, but hang on. I love this because I know what a paternoster is, and I’m not sure everyone does.
Hedvig: Exactly.
Martine: Yeah. The thought of jumping on and jumping off seems pretty good to me.
Daniel: Martine, help us out. What’s a paternoster?
Martine: It’s kind of lift where you can jump in and jump out, and you have to be very quick.
Daniel: There’s no doors.
Martine: There’s no doors.
Daniel: It doesn’t stop. [laughs]
Martine: And it just keeps going round and round and round.
Hedvig: Is it paternoster as in ‘Our Father’?
Martine: Yes.
Daniel: Yes.
Martine: Yeah. I found the name a bit strange myself, but I’ve been in one and jumped in and out of one. It’s really fun.
[laughter]
Daniel: Okay, did you did you feel safe?
Martine: Ah, no. [laughs]
Hedvig: [crosstalk] -the fun of it. Okay.
Daniel: I’m here with a bunch of Europeans. I believe that Europe is the only place where you still see these things.
Hedvig: Where?
Daniel: I’ve never seen one.
Bodo: I’ve never seen one.
Hedvig: We don’t have them in Leipzig. Does Cologne have them?
Martine: We have one in Cologne.
Hedvig: You do?
Martine: Yes, one. But it’s in a private building, but I’ve been in there.
Hedvig: It’s in a private building?
Martine: Yeah. Well, I think it’s the building of some bank or something like that. I can’t remember, I’m sorry.
Hedvig: Sorry, I thought this was public transport.
Martine: No. It’s a building with a lift. But instead of an elevator, but instead of having doors that open and close, you just see these cubicle-sized things going past you from bottom to top, and you have to jump on and then–
Hedvig: And you just step in and step out?
Martine: Yes.
Daniel: Yeah. It seems staggeringly dangerous.
Bodo: Yeah, very scary.
Hedvig: Yeah.
Daniel: And the reason they’re called paternosters is that they just keep going in a circle and they’re like beads on a string, and that’s what you do when you say the rosary. Apparently, you’ve got these beads on a string and you say, “Our Father,” and that’s paternoster. So, the whole thing is– Timothy says, “Both words are circular in one sense or another. I’ll grant you, they’re a tad sesquipedalian, but what’s a few syllables between linguists?” Oh, Timothy, you cut me up. All right.
So, let’s go on to our first news item. This one was suggested by aristemo on our Discord channel, and it involves a hand sign that a missing girl used to indicate that she was in trouble, she wanted to get help. I had not seen this hand signal before I saw the article. Had anyone? You’ve seen the story of about–?
Hedvig: I haven’t. I saw the story, but I hadn’t seen the signal. For our listeners, because this is an audio medium, you hold your palm up, and you fold in your thumb, and then you fold over your fingers over the thumb, and then up and down. I think that’s it. I don’t think you move the thumb.
Martine: I think you’re meant to hold your hand with your thumb out, and then you move it in, and then you cover your thumb. So, the thumb is trapped.
Hedvig: Okay. So, you move the thumb as well, okay. But the idea is that if you are being held against your will, and you’re in a car or in a public space somewhere, and you don’t feel like you can say help to someone because the person who’s holding you against your will notice that and act on that towards you, you can do this, and hopefully, it’s so subtle that maybe they won’t notice and even if they do, maybe they don’t understand what that’s that.
Bodo: I’m curious about where does that convention come from? It seems like in order for people to actually know that that’s the meaning and it’s not obvious from the gesture. If I’m looking at this, like, it’s not immediately clear what it means. So, where does it come from?
Hedvig: I don’t know. I think it’s spread on American TikTok in particular. And then, it happened to get to this girl who was in the situation, and it saved her, someone noticed this. So, you’re right, because the recipient also needs to know what this is about because probably, “Oh, she’s waving at me strangely,” is the first reaction you might have.
Daniel: Well, it was created in 2020 by the good folks at the Canadian Women’s Foundation, because they were aware that a lot of people would be trapped and locked down with perhaps someone who was abusive. So, it spread around the world since then.
Hedvig: Yeah. It’s a bit of a double-edged sword as well, because the more it spreads, the more likely the person who’s holding you is to know about it. So, yeah, it’s tricky.
Bodo: But, I guess at least it’s something that can be done silently, isn’t it?
Hedvig: Yeah.
Bodo: If you are on Zoom, for example, the other person might not be watching you the entire time, so a silent gesture might be more effective than any code word.
Hedvig: Yeah, exactly.
Daniel: I wanted to spread this one because I like conventionalized gestures. I hadn’t seen this one. Potentially, it could be of use to somebody, so I wanted to get it out there.
Next, this story was suggested to us by a Wolf of the Wisp, and this one has one of my favorite experiments. The Hallucinating Accent Experiment. Well, actually, it’s two experiments at least, decades apart, one replicating the other. Here was the experiment, the Hallucinating Accent Experiment. You take two groups of students, you play them both the same audio of a lecture. But to one group, you show a photo of a white woman, who you say is the lecturer. And to the other group, you show a photo of an Asian woman who you say is the lecturer, exact same audio. But what happens is that the students who see the Asian lecturer, not only think that the lecturer had more of an accent and was harder to understand, but they also score worse on a comprehension test, which is an amazing result.
Bodo: That’s shocking.
Hedvig: Yeah. I think we have other experiments also showing that intelligibility is not even just about what we talked about with the McGurk effect with the visual and audio, but also about your stance, your attitudes towards the speaker. We’ve talked before about– because I’m Swedish, and we love to pick on Danish people, but if you–
[chuckles]
Hedvig: I think there’s been some experiments showing that Danes or Swedes who have a worse– if you ask them, like, “What do you think about Swedish people? What do you think about Danish people?” And if they think badly of them, they also understand them less well. And I’m not–
Daniel: This stuff runs so deep.
Hedvig: Yeah. I think it does–[crosstalk]
Martine: I think though we would have to have a control condition, because it could be that you’re looking at the photograph of someone, and the way they speak doesn’t match what you expect. Then, you find it more difficult to understand. So, we would have to have, for example, a person with an Asian accent and looking at a picture of a white woman, and seeing whether that causes the same problem, because it could be just– well, it’s not that I’m saying that people’s attitudes don’t affect the way they understand things, because that would be wrong for me to say that. But in order to report on these things, I think we have to always remember the control condition to be really solid in our results.
Daniel: Well, Martine, it appears that this research that you’re describing has just been done.
Martine: Good. [laughs]
Daniel: And that’s the subject of this news story. This word comes from Dr. Adriana Hanuliková of the University of Freiburg. This was published in PLOS One. The language of this study was German and subjects were asked to listen to words spoken with three different accents, a standard German accent, a regional German accent, and then Korean-accented German. So, the two things they had to do were repeat the words they heard, and then say how strong an accent– strong, I hate that terminology, but that’s what a lot of people use it, how accent that they felt like it was. And there was background noise, so that made it tricky. Also, sometimes the group saw a photo of an Asian woman, but also sometimes they would see a white European woman who were supposed to be saying these things. The results were everyone, whether they were teens, young adults, older adults, said that it was harder to understand when they saw an Asian woman. Okay.
Hedvig: No matter what audio they heard?
Daniel: No matter what audio they heard.
Hedvig: Right. So, even if they heard the standard German audio, and they saw the picture of the Asian woman, they said that it was harder to understand than if they saw the standard German with a white woman.
Daniel: That’s right. When contrary to expectation, it causes confusion. Hanuliková says, “The more closely our expectations match the pronunciation, the better we understand the person.” There we go. It looks like we got our control condition.
[chuckles]
Martine: Yeah.
Daniel: I love it when that happens.
Hedvig: What did they say when they saw a picture of a white European woman and heard a Korean accent in German?
Daniel: Okay. Well, there was one group in particular, that had a hard time, and I’m going to get you to guess. Do you think it was the teens, the young adults or the older adults?
Hedvig: I think it was the older folks.
Bodo: I think older.
Daniel: Okay, how come?
Bodo: I guess maybe– I think this is perhaps a bit there might be a bit more stereotyped and maybe younger people are a bit more cosmopolitan, that that will be my thinking.
Daniel: Yeah, okay.
Hedvig: Yeah. Younger people are more likely to have met Germans who don’t look like white European people, and have been accustomed to that. Yeah, that’s what my guess as well.
Martine: But it depends how old the teens are, I suppose, because if you’ve had less exposure in general in your life, it might be more difficult.
Hedvig: Yes, good point.
Martine: [sighs] Tricky one.
Daniel: Okay. Well, what they found was that the group that had the hardest time when they saw a photo of an Asian woman and heard standard German was the older adults. They reckon that it’s because they had less exposure. So, what we think is happening here is it’s not just that we have a hard time understanding when we see someone that we think is foreign. In fact, they actually understood better if the speech matched the photo. Bias is still totally a thing but what this tells us is that when we’re trying to comprehend speech, we use a lot of factors, including how we expect the person to sound. And if there’s a mismatch, then it can muck us up.
Hedvig: There’s a large amount of variability in how individuals speak. Computers can have a hard time managing with individual differences. Humans use biases and assumptions about the world to rein in, so if you have the perception that a person looks a certain way and speak a certain way, when they start speaking, they do something else, you have to recalibrate your brain a bit to go like, [onomatopoeia]. But, usually, you can do that with a few sentences. I don’t know how much content this was, because you can usually settle in after a little bit.
Daniel: The fact that it was isolated words makes it difficult.
Bodo: The study, like you mentioned, how it shows that our speech is really malleable and variable depending on context and who are listening to you, it reminds me of this fascinating study by Jennifer Hay and Katie Drager. I don’t know whether you know, the stuffed toy and speech perception study?
Hedvig: Yeah.
Daniel: No.
Bodo: It’s fascinating. The participant comes into the lab– I think it was done in New Zealand, if I recall. New Zealand speakers have some awareness of both Australian English and New Zealand English accents. The participant comes to the lab and is greeted by an experimenter who’s a bit disorganized and can’t find the consent form.
Daniel: Sure they can’t.
Bodo: Of course, they can because it’s the setup, and then they’re looking for the consent form. And in the process, they find a stuffed toy. They look at the stuff toy, and say, “Oh, there you are.” So, that the participant notices the toy. And then, they go on to do a speech perception experiment. The critical manipulation is that the stuffed toy is either a kangaroo or a kiwi. And so that primes people to think about Australia and New Zealand. And then, their vowel perception has actually shifted towards Australian English, or New Zealand English. And so that really shows that like these very subtle cues. Yeah, it’s funny, isn’t it?
Daniel: This is great. [laughs]
Hedvig: That’s insane. That’s so much fun.
Daniel: Oh, my gosh. Oh, my goodness.
Bodo: It just shows like how speech just takes in any information that you can get. Even just briefly seeing a stuffed toy makes you listen to speech differently. So, [crosstalk] it’s fascinating.
Hedvig: It is fascinating how subtly you can influence people in psychology experiments overall. There are these experiments where you ask people to do a certain test on rate people’s faces on how much they like them. And then beforehand, you just gently remind them somehow of their own mortality. And when you do that, people have a tendency to rate people who look like them as more likeable. There’s this idea that the more you’re aware of your own mortality, the more biased you become to people who aren’t like you, which is terrifying. But they just ask some sort of question, they say something– I forget exactly what they have, but I always wanted to like, how do you prime people to think about their mortality but you can apparently. [laughs]
Bodo: Yeah. These days, the only thing you need to do is watch the news to be reminded of your own mortality, isn’t it?
Hedvig: Yeah. We’re going to travel. We’re going to do international travel soon, so I’m keeping more up to date with the news than I usually do to just keep track if we can travel and it is bringing me down.
Daniel: Lovely.
Hedvig: Yeah.
Daniel: Did that study get replicated, by the way?
Bodo: That’s a good question.
Martine: Yeah. I was wondering–
[crosstalk]
[laughter]
Bodo: Yes, that will be good to know.
Daniel: Let’s go on to our next one. I am fascinated by this topic. This is about Kiki and Bouba. Everybody knows that if you have a blobby shape, and you have a spiky shape, that Kiki is the spiky shape, and Bouba is the blobby shape. But there have been some questions about how robust this effect is across languages or even across writing systems. There’s been some work from our friends, Lauren Gawne and Susie Styles, which tells us that it can depend on whether Kiki and Bouba are legal words according to the phonotactics of the language that you’re studying. But I noticed that there is a new study, and we have one of the authors here because, Bodo, you’re on this one.
Bodo: Yes.
Daniel: Tell us a little bit about the Kiki and Bouba effect as you’ve studied it across languages and across writing systems.
Bodo: Yeah. So, Kiki/Bouba effect, of course, has this long history. It’s been done for centuries now–Not centuries, decades. Sorry.
[laughter]
Bodo: Not that long. People have always been concerned that if you’re doing this with particularly literate German or English speakers, and they know the writing system, that the way Kiki is spelt, the letters themselves are actually spikier.
Hedvig: Yeah.
Daniel: I haven’t thought of that.
Bodo: Yeah. Like K looks spikier than O and OU of Bouba. So, it could be that people are not actually matching sound to shapes, which is the more interesting case, arguably, but they’re actually just matching shapes to shapes. And if that’s the case, then this effect is not that interesting to things, I don’t know the origin of language and these sorts of things where if we want to imitate something, like we want to imitate a shape with our speech, if it depends on writing, it will be less interesting for these sorts of questions. What we did is we had an opportunity to do the Kiki/Bouba task. People, they listen to Kiki, and then they have to choose one of those shapes. We had the opportunity to do that in more than 20 different languages by just collaborating with lots of linguists together.
What our results show is that the effect is very robust across languages. Even if you have a different writing system, so even if you don’t have the Roman alphabet, you would still actually find a Kiki/Bouba effect, but it will be a bit weaker. What we conclude then is that having the right letter shapes sort of gives you a boost and makes the effect stronger, but it does exist in cultures even if you don’t have the Roman script.
Hedvig: Right. What other scripts did you use? I can imagine Burmese or Thai maybe would–?
Bodo: Yeah. We had, for example, Georgian. We had Japanese. We had Korean in the study. I think we had a total of nine different scripts, if I recall. One big concern of the study– while this is, so far, the biggest Kiki/Bouba study in terms of the number of speakers and the number of cultures that have been tested, there’s one big concern, and an obvious limitation that we have to just acknowledge, which is that the way we administered the test is online. So, it’s a web-based survey. And of course, this gives you a very biased sample of people who live in industrialized societies. So, these are all major big languages, we don’t have many language or we have no language in this particular sample that actually don’t have access to computers, don’t have access to any writing system. So, more definitely needs to be done, but it seems that at least in this population of these very big languages, we consistently find the Kiki/Bouba effect.
Hedvig: Cool. For a long time, people have been using the [unintelligible [00:27:50] as a mascot for linguistics. When there was a lot of kerfuffle regarding the copyright status of that, people were like, “Oh, what should we use instead?” I have seen people suggest Kiki/Bouba, some sort of anthropomorphized or combination of the Kiki/Bouba shapes, so a spiky speech bubble and a bulbous speech bubble is the easiest way to explain it in an audio medium. This puts another, how do you say, another feather in the hat of that camp. That’s nice. I like that.
Bodo: Yeah.
Daniel: If I remember correctly, Georgian script is the one that’s super-duper round, like all the shapes look really round to me.
Bodo: It looks really beautiful, actually.
Daniel: It’s gorgeous. But nothing looks spiky at all. I guess that’s still held for Georgian.
Bodo: Yeah. What we actually did is we also had a separate task, where we took the words in the way they were written in different scripts. Let’s say you have the Korean or the Georgian rendition of Kiki and of Bouba, if you see those scripts, so if you see in Korean, Kiki, and in Korean, Bouba, can you tell which one goes with which shape?
Daniel: [gasps] That’s cool.
Hedvig: Oh, for non-Japanese, non-Georgian speakers.
Bodo: Exactly, yes.
Daniel: Oh, my God.
Bodo: It turns out that, in at least in our data, that this did not work out. So, if you’re seeing the Kiki in Korean or you seeing it in Georgian, there isn’t a strong tendency to match it with these shapes. The degree to which the letters are spiky or round does not predict whether the language has a stronger Kiki/Bouba effect. The ultimate argument is, it’s less dependent on writing systems than people thought.
Hedvig: Nice. And that’s nice that you did that test as well. That really shows your thorough methods thinker that you thought of doing that while you’re at it. Yeah, that’s really cool. That’s a good idea.
Martine: Do you think the writing system for German and English has been affected a little bit by the Kiki/Bouba effect? There seems to be some correlation, do you think there’s a direction to that?
Bodo: That’s a good question. I really don’t know. I’m not an expert in the evolution of scripts, on the history of how they have evolved. But it’s certainly the case. It looks like for the four round vowels and for voiced consonants, that they’re more likely to be rounded letter shapes. I don’t know what the direction of causality here is, whether people have changed the script slowly in response to this just feeling right for those sounds. But yeah, in and of itself, the fact that this script actually mirrors the speech sounds a bit is really fascinating.
Daniel: Yeah, because all your low vowels are really round, like ah, ae, and o. But all your high ones, the e and the oo, they have points on them. [crosstalk] -it’s right rounded at the bottom. Oh, my gosh, we’re onto something here, folks. Let’s make a theory.
Hedvig: Yeah. We think that Latin script probably evolved from something where maybe round shapes weren’t possible to make, because you were chiseling things in. So, they would have to evolve over time. If they all sorted out as more pointy and some of them became more rounded, and maybe there was some pressure to– yeah.
Bodo: It’s possible that this is just a small bias that if compounded over time, over hundreds of years of language evolution amplifies. There’s a cool Kiki/Bouba study that’s been in Gabriella Vigliocco’s lab. She’s a professor at University College London. They have shown people the Kiki/Bouba shapes, and it gave people random words that they had to learn, like you see the spiky shape, and then you have to learn a random word like, I don’t know, work, for example. And then you have to repeat that word, and to a new participant who has to also learn this word from you. People invariably introduce small changes. It turns out that over times, as you’re playing these Chinese whispers game, the wordforms will converge on the Kiki/Bouba-like shapes. So, the words will become more Kiki like for the angular shape. So, people slowly introduced these– making the sounds match slowly more to the shape. I think it could be the same with the writing systems.
Hedvig: Right. That’s a really important point you made there as well that we could be talking about small biases that accumulate over time. If you let evolution run in some Brownian random way, even if you have a small bias, if that is consistent over time, that can build up to a larger effect. We’ve talked about this in multiple language evolution related research that it doesn’t mean that you’re predestined to always do this so that’s a huge factor, but even a little tiny effect can compound over time. I think that’s really interesting.
Bodo: 100%. Yeah.
Daniel: Well, thanks so much for taking us through this part of your work. That was really interesting.
Hedvig: Yeah.
Bodo: Cool. It was fun to do.
Daniel: Just to round up the news, there is an action item if you would like to help. As we know, English is often used as the language of science, something like 80% to 90% of scientific journals on planet Earth are in English, depending on your field. What that means is that many researchers have to take time away from their research to work out how to write their research up in English, and not just English, but in academic English, a very unusual written register. It’s not fair that they have to do that. One thing that we can do as English speakers to level that playing field is to volunteer as a proofreader for people who are learning English, but I haven’t really found a good way to do that, until now.
There’s an effort from Mora Maldonado, Alexander Martin and Carmen Saldana, English language proofreading for nonnative authors in the Language Sciences. They are compiling a list of researchers who are willing to help out. You don’t need to be an expert in the particular linguistic area. But if you sign up, they will in future days, I hope, send you some work that you could proofread. They say, this is on a Google form, “Please select below the areas that you would be comfortable proofreading for. Note that proofreading does not include a critical review the content, only suggestions for the language. So, it is reasonable to select subfields out with, with, beyond your area of immediate expertise.” So, if you want to sign up, you could do that. There’ll be a link on the show notes page for this episode. Or, you can look up English Language Proofreading for Nonnative Authors in the Language Sciences. What do you think?
Hedvig: I think that’s really cool.
Bodo: It’s cool. Yeah.
Hedvig: I think you’re very right. It is an extra amount of effort for us. I think, three out of four people on this call right now aren’t native speakers in English, correct?
Daniel: Yeah, I’m just soaking it in.
Hedvig: And like you said, academic English is on top of that, like a special register. Lately, I’ve been reviewing articles and grants and abstracts and stuff. I try to keep in mind that– because sometimes I read things where I’m like, “Okay, this isn’t a native speaker has written this,” but it is still clear and it is still good research. Just keep in mind that just try and see beyond that, and maybe also not leave reviewer comments like, “It’s better if you hand this over to a native speaker,” because sometimes that isn’t even necessary. You can be clear enough without being entirely correct by native standards.
Martine: We do some community proofreading in an online, free, and fair open access book series called Language Science Press, and this is based in Berlin. I have a book series called Studies in Laboratory Phonology. What happens there is the manuscripts are reviewed for content, but then before they go to press, they’re– what do you call it, proofread or copy edited by people who volunteer to do that. And then, there’s webpage where we can acknowledge the work that people have done on this. So, there’s lots of more advanced students, young researchers who are offered to do this work for free, and it’s really good.
Daniel: That’s cool.
Bodo: Great. I was just thinking. One of the things, Hedvig, you also mentioned, biases that come in during the review process, and the fact that it’s harder for people, the journal that I work at, we actually have a policy where we say that reviewers should not say, “Oh, this is not native speaker English,” or “They should consult a native speaker,” which could be very insulting if it’s anonymous, and maybe actually, the person is a native speaker, you don’t know. But also, not making native speakerism as this quality stem and just saying, you can say this without introducing this bias, you can just simply say, “Hey, this should be proofread more,” without saying, “Hey, this should be proofread more because you’re not a native speaker,” which is just not nice.
Hedvig: Yeah, exactly. There is also I’ve read things by native speakers in English that should be proofread for increasing clarity, which is a thing we all struggle with, and we all need help with. Often, that is more crucial than the– I for example, as my husband knows, because he’s a native speaker of English, I really suck at verb agreement in English. [chuckles] All of my supervisors know this, and everything I write, there is, there are, my brain is just like, I couldn’t care less apparently.
[chuckles]
Hedvig: I’m actually curious to know, because sometimes when I look at the transcripts, or listen to this podcast, I noticed that I make them while speaking to, and no one seems to care at all. No one’s ever pointed out to me. But in writing, obviously they do.
Daniel: We’re talking about a couple of different things here. This effort is designed to help someone over the bar, over the gate, but we should also be working on just breaking down that gate. There are two ways to reduce the unfairness here. I like both approaches to reducing the unfairness, but if there’s– I can be more relaxed about academic English, but I can also maybe help somebody every once in a while with clarity, which again, is not the exclusive province of people who have learned English as a first language.
Hedvig: Yeah, exactly.
[music]
Daniel: We move on. We are here with Dr. Martine Grice and Dr. Bodo Winter talking about the replicability crisis in science and also in linguistics. There have been some landmark studies that were once considered more or less settled, and then failed when they’ve been retested. We’re going to talk a little bit about their contribution to a special issue of linguistics, dealing with the replicability crisis. Their chapter is, “Independence and Generalizability in Linguistics.” Thanks once again for being here with us and talking about your work.
Bodo: Thank you.
Daniel: Tell us about the replication crisis. Maybe we could just start generally in the sciences. I know that there has been some solid work that’s been toppled. What have you noticed with replicating experiments, not just in the sciences, but specifically in linguistics? What’s going on?
Bodo: Yeah. Martine, should I go first?
Martine: Yep.
Bodo: Okay, cool. The replication crisis is currently taking all sorts of fields by storm. It’s targeted at the core of what we do in research and it’s fundamentally transforming everything. It’s very much focused, I think, that the most relevant to linguistics is the debates that are happening in psychology, where there have been a lot of large-scale attempts to replicate experiments. For example, one attempt was 100 studies from a particular year of a major psychological journal, and out of those only 30% to 40% replicated. So, they did the study again, and they couldn’t find the same results. And that’s just hugely troubling, now we’re not certain about these findings anymore.
Linguistics hasn’t had its replication crisis yet but given that we’re using the same methods that psychology does and we’re having the same incentive structures in academia, it’s quite likely that we have the same issues. Actually, they’re just not as obvious because we haven’t done these big replication checks to see whether linguistic studies actually replicate.
Daniel: Is anyone doing this now?
Martine: Well, there is some replication going on. For example, Timo Roettger and Dinah Baer-Henney wrote a paper just recently where they advocate replicating studies in the classroom, which I think is a really good idea. Of course, the problem with linguistics, and especially phonetics which is my area, is that it’s quite hard to make recordings of 500 people, let’s say. So, we tend to make recordings of something more like 10-15, something like that. This is a very small number. So, of course, you can imagine when you want to replicate something, you take a different 15, that’s quite likely not to work. I think that this is something that that we really need to do more. I think one way to change things is to start with the people who are now learning in the field, so to get students onto that, and to train them much better in statistics than people in the previous generations were trained, so that we can then move forward.
Hedvig: And maybe we’re seeing also that there’s different ways of replicating things as well. One of the core things that’s needed in order to replicate is that the original experiment is enough described or that the data is accompanying the publication and even better, that old parts of the method are spelled out, like exactly what priors or what details of how they ran things. That has been a convention that has developed and that isn’t as much true for, I think, older publications. Sometimes, it might be that it is– I’ve heard some, some more senior people say, “Oh, no, but this could be replicated. It’s just that you didn’t do it the exact way I did it.” And then, people say, “But you didn’t write down the way you did it. What am I supposed to–? I have to make some decisions when I replicate.” That can be a cause of conflict as well.
Bodo: You’re mentioning all the decisions that you’re making in the research process, so it’s important to realize that as you’re designing a study, there’s just myriads of knobs to turn and they could massively determine study outcomes. If you don’t know about some of them, some of those knobs that have been turned in the background, then you can’t replicate that study. I think it’s just important, I’m also harkening back to what Martine is saying, training the new generation of linguists to think about the idea that making all of this transparent, making all of these decisions so that they appear in the paper, in the published work, and that the data is shared, hopefully this will be natural to the next generation of linguists that they grew up with the idea that it’s normal to share data.
Martine: But we also have a problem, of course, with phonetics, in that if you make recordings of people’s voices, it’s not as easy to share as it would be to share, for example, a transcript. But we can still share measurements or maybe alter the voice in some way. There are various things that we’re working on at the moment to be able to share more data.
Daniel: Okay. One of the things that you mentioned, in fact, a big chunk of your paper really is about the independence problem, which I understand as essentially what you’re doing is you’re saying, “Hey, I looked at 17 different languages, and I found a pattern. When languages do this, they also do that.” I looked at 17– except you weren’t looking at 17 different languages, they were all kind of related. So, you were actually looking at one thing.
Bodo: Exactly.
Daniel: And thinking you found a pattern. Is this Galton’s problem? We keep talking about Galton’s problem.
Hedvig: That’s Galton’s Problem for phylogeny, but what they’re addressing in their paper is also that this can happen for not crosslinguistic studies as well. This can happen for speakers or genre of text and things. And Galton’s Problem is specifically about shared dissent of community groups. So, like, you can’t say that Portuguese and Spanish are completely independent data points. But you also probably can’t say that, like– what’s another good example? [chuckles]
Martine: Maybe Basque and Spanish. They might be not related, but they’re nearby.
Hedvig: Yeah. You could have spatial controls as well. But I was thinking of you can’t say that two speakers that are of the same gender and of the same age and attending similar educational institutions, are also not independent. But that’s technically not Galton’s problem. There’s also something called spatial autocorrelation, which is that you can’t assume maybe the things that are next to each other independent. But this, like Bodo and Martine write about in their paper, non-independent data points is a concern for a lot of other type of data as well including things you collect on speakers or genre, etc. So, yeah, it’s a general problem.
Bodo: For example, if you were to design a study– and actually almost all studies that we do in linguistics, have some sort of clusters of data in them. So, if I get multiple datapoints from you, Daniel, and multiple datapoints from Hedvig, then each one of those datapoints that come from one person, they’re going to be more similar to each other, because we have idiosyncratic differences. There’s variation across speakers. The key point of our paper is that this source of variation, the fact that these datapoints are not independent anymore, by virtue of coming from the same person, you need to incorporate that into your statistical analysis.
The basic idea is that still to this point, even though linguistics has done massive advances in stats over the last few years, still to this point, we think that there are certain sources of variation that exists in the data, certain ways in which languages vary, in which speakers vary, that are not accounted for in the statistical analysis. And if you’re hiding variation from your model, then you’re not using that to quantify your uncertainty in a claim. So, it’s easy to make a bold claim if you are omitting part of the variation that exists. You’re saying, “Oh, this is really strong pattern in the data,” but actually, I’m looking at this pattern while ignoring the fact that there’s all of this stuff going on that I’m hiding from my model.
Daniel: Is this a serious problem? Or is this just a minor methodological flaw? Is there anything that’s going to be overturned as a result of this?
Hedvig: Could be.
Bodo: Massively, potentially. One actual example of this is the correlation between grammatical tense and savings behavior.
Daniel: Ah, I knew it.
Hedvig: We talked about this. We’ve had Sean on. Yeah.
Daniel: [laughs]
Bodo: Okay, that’s a good example of that, where this analysis treated languages as independent datapoints even though languages aren’t independent datapoints. So, as was just explained by Hedvig, they are related by virtue of dissent, of contact, and all of this stuff. If you account for that in the analysis, the correlation is much weaker and you would be much less confident in this correlation to the point where it’s not confident enough for us to boldly claim that this is true anymore.
Daniel: I also noticed the one with Quentin Atkinson and the team where they found that there were lots of phonemes for languages in Africa. But as you get farther away from Africa, then they start shedding phonemes. It was something about the founder effect.
Bodo: That’s another one. There’s actually a paper specifically on that. It’s been a while since I read it but it talks specifically about how also in that analysis, not all sources of variation were adequately controlled for, and that the actual pattern is much weaker once you control for those sources of variation.
Daniel: Yikes.
Hedvig: And that brings me to– because I should say also that, I collaborate with Quentin, and one of the other things besides that we talked about that you should control your samples and you should make your data available, you should also cultivate an atmosphere and a culture within academia where you are open to criticism and not ideologically keeping and defending yourself when that’s not worthy. And yeah, I think most of researchers that I’ve collaborated with or work with tend to have this attitude more, but I’ve noticed it can be a bit lacking maybe in some generations of older scholars in certain fields. And the less transparent you are about how you can about your research, the more likely you seem to be to get very defensive when you’re critiqued, which is– I don’t know, this is just my impression. I don’t know if you have thought about that as well.
Bodo: Well, I think in general, people, when it comes to methods and stats in particular, they follow a lot of rituals. There are some deeply ingrained things that are mostly being done, because of the way they’ve been done in the past. For example, significance tests, which are still widely used in linguistics, there’s been lots of critique of this idea that, “Oh, I get a simple result that tells me, oh, it’s significant, there’s an effect or it’s not significant, there’s no effect.” Yet, this tradition persists, and it’s persisted over decades. It’s a lot because people just do the methods that they’ve been taught rather than, as you say, having a debate and discussion about it. I think that’s the important bit of the special issues, just to raise these issues and start talking and developing a methodological debate.
Daniel: Yeah. I’ve used chi-square tests lots of times, but the chi-square test assumes independence. And so, my results could just be flat wrong.
Bodo: Exactly.
Martine: Yeah. [laughs] I mean, there’s another thing is, especially in the research I do, because I work a lot on intonation, melody, and rhythm of speech. Of course, it doesn’t work to get a person to just talk on their own. You have to have someone to talk to. Normally, we record two people talking or groups of people talking, but mainly dyads. The fact that in most of the studies, we have to get people to talk, so we invent games and tasks for people to do. For example, making people talk about a map. One person has a route on the map, and the other person has to find out about that route. And then, typically and probably to make sure that we get enough people to do the recordings and we get enough time recording, we ask the people to be the instruction giver. And then, they swap maps, and then they do the task again.
Then, if we’re looking at their behavior, for example, we may be looking at how often they say, “Mm-hmm,” “yeah,” any kind of back channeling behavior, then obviously, in the second task, the person who’s listening more, listening and doing the actual task, is going to behave in a way that might be affected by the first task. So, you really can’t treat those two people as being independent. But so far, people have been doing that all the time.
That’s one of the problems and some people have acknowledged that and then they have a so-called confederate. So, they have one person who plays the game with everybody, and just pretends to not know about the game. That’s less natural for one. And, of course, this confederate can be influenced by all of the different speakers that he or she has been talking to. So, you still don’t get away from that. I think the key is to acknowledge where the influence is and where the lack of independence is, and just deal with it statistically.
Bodo: Martine’s example for phonetics just highlights how easy it is for these, the fact that datapoints are actually mostly connected. They’re almost always not independent. It’s really very widespread across linguistics. Just to give one more example, corpus linguistics, so the analysis of text databases, there’s rampant nonindependence. Sentences that are adjacent to each other are more similar to each other because, for example, it might be on the same topic, or multiple sentences from the same author, multiple sentences from the same type of texts. There are all of these nested structures. Unfortunately, a lot of the stats that’s being taught in corpus linguistics, actually still to this day, ignores these different clusters, these sources of variation. The point that we make in the paper is to actually emphasize how important it is to bring that structure into the stats and make it so that the stats actually pay justice to the variability of language.
Hedvig: Maybe it’s worth also saying that by doing that, you’re not necessarily going to not find the result you previously found. You could find the same outcome, but you would be much more confident and sure about it, and you would be much more defended and future proofed for critique, if that’s something you’re concerned about. So, it’s not the same as saying as none of these studies weren’t– they could still be finding real effects. It’s just we want to be more sure about those effects, right?
Martine: Yes.
Bodo: Exactly. So, if you’re ignoring the independence assumption and if you’re not putting that variation into your statistical model, then you’re much more likely to have a spuriously significant result. So, you’re much more likely to claim that you have found an effect, even though it actually isn’t there. So, it’s really important to sort of– this is one source of replication for your failures out of many other sources, of not doing the right stats for the type of data that you need.
Hedvig: There’s also something maybe we should talk about, which is where you guys are brought up, which is that the training and the knowledge of which stats to apply, because it is also maybe not reasonable to assume that anyone who wants to become a linguist is necessarily going to have a great love of statistics and mathematics. That’s why we try and have stats courses and tools and tests that people can run, maybe without doing also another degree in stats at the same time. So, there’s also– usually in linguistics, if you get stats, you get taught about some student T-test and chi-square and a couple of other things. So, by reforming the kind of tests that people do and what they should be on the lookout for, maybe it isn’t necessary that everyone gets a full statistics degree.
There’s also the possibility– I was very fortunate when I was doing my PhD at the Australian National University, they had a stats consultation unit, which was separated from any faculty. As a grad student or staff, you could go there and sit and talk to them and be like, “I want to do this thing. What do you think is the appropriate test?” I went there and spoke to them for a long time, and I found that really useful as well.
Martine: Actually, Reading University where I started my PhD, they also had a stats consultation unit and I found that really helpful. But unfortunately, it gave me the impression that I didn’t have to learn it myself. And then later on, I had to catch up.
Hedvig: Right.
Bodo: I think, in general, linguistics departments seem to struggle with where to incorporate statistics into their curriculum. In particular, when it comes to something like, let’s say, MA programs, which are often very short, I often hear the argument that, “Oh, all of the linguistics content that we have to teach, we don’t have time to teach the stats.” I think it’s sad because the stats are treated as something that is external to linguistics. They’re treated almost as an afterthought. A lot of people design their studies, and then do the stats at the end. I think it should be taught and from the beginning, something much more integral to the entire process, and is actually a core part of linguistics.
Hedvig: Yeah. Like you were saying, something you can think about already when you design your study. So, I went to talk to stats consultation unit before I went and gathered data, and they said, “Okay, it’d be good if you’ve got at least these many recordings. And if you thought about this, and that and that, and that” so that when I went and collected things, I was a bit aware of things to look out for.
Bodo: Yeah, 100%.
Daniel: Well, it sounds then like the solution here is we need to get right down to the pedagogy when people are becoming linguists to make sure that this is addressed and have better tests and talk about the kind of things you mention in the paper, but also have a culture of replication, where that is something that is considered normal, that is non-threatening, that is just part of upholding good standards. Is there anything else that we need to do to be fixing this problem?
Bodo: Yeah.
[laughter]
Bodo: Because the replication issue or the replication crisis is such a massive problem, it’s important to realize that it’s not just one problem. It’s lots of small, different practices that are questionable and things that we’ve done in the past and research that we now know are not good anymore. Because of that it’s not going to be a simple fix. And, yeah, it’s good to target pedagogy and these sorts of things but we also need to do things, like, convincing journal editors, that open data is a good thing. More journals will have open data policies. So, there’s not just one thing that can be done, I think we need to work both in pedagogy, but also from top down focusing on the publication process.
Hedvig: Yeah. Also, that journals– a lot of people don’t want to do replication studies or publish replication studies, because they don’t think it’s going to be enough cred on their CV or that it will even be accepted by a journal that is enough cred-y for them. That’s another problem. At the same time, if something has been proven well in a study and in many studies, then there’s a limit to how much replication– We’re not there yet. So, maybe we shouldn’t worry about that. [chuckles] We’re not there at all.
Martine: One thing that is starting is that some journals are now prepared to take registered reports so that you can preregister a study. That’s really good, because if you don’t get any results, which you would normally be able to report, then you can still publish your study.
Hedvig: Yes.
Bodo: Yeah.
Martine: Because otherwise, no results are just not interesting for journals and we have to change that because probably people have replicated studies but if they didn’t get anywhere, they couldn’t put that information out there.
Hedvig: Yeah, this point about preregistration is really important. The idea there is that you contact a data registry or a journal or something, you say, “I am going to study this with this data in this way. And I’m going to test it in this way and I’m going to evaluate the result of my results, my study in this particular way.” Then, that binds you to doing it by that approach so that it avoids people doing an experiment and then not finding what they were looking for and then trying another test, and then finding what they were looking for and publishing that. So, it’s trying to avoid things like that.
I’ve always had this daydream that sometime when I’m like a senior researcher, and I can do whatever I want, and I have all the cred or whatever, I don’t know if that’s ever going to happen, but it’d be fun to do a journal where reviewers review papers without reading the results section. So, you get pre-reviewed by only getting evaluated on your research question and your methods. And if the research question is interesting enough and your methods are good enough, you get accepted, no matter the outcome. That’d be really cool.
Daniel: Wow.
Martine: Well, that is what happens actually in, for example, language and speech, you can submit-
Hedvig: Ah, sweet.
Martine: -something without the results, and then it’s reviewed. I think also General Phonetics have been talking about doing that. I can’t remember now whether they’ve done that or not.
Hedvig: That’s so cool. I’m so glad.
Bodo: Martine, since you mentioned this, I just want to give a quick shoutout to Language and Cognition, which is the journal that I work at. We also just have this results-free reviewing process by registered reports, and I think more journals– I think there’s like five, six journals now that have it in linguistics.
Martine: Yes.
Hedvig: Sweet, I love it. That’s so exciting.
[chuckles]
Daniel: It sounds like we’re getting there slowly, but it’s still a massive problem.
Bodo: Yeah. It’s going to take a lot of time.
Daniel: Dr. Bodo Winter and Dr. Martin Grice, thanks so much for telling us about your work.
Martine: Thank you.
Bodo: Thank you.
[music]
Daniel: We’ve got some Words of the Week, my favorite, favorite part. Well, we’ve got a lot of words this time. Martine, I believe you’ve got us a word that you have had on your mind recently.
Martine: Yes, it’s slowness. Maybe you don’t think it’s a particularly new word, because it’s been around a while, but it’s definitely in my head. It also relates a bit to what we’re talking about, because if you did slow science, then you would spend some time thinking about your design, and really spend more time in the conceptualization of things. I think that would be really something that I would like to do more. I wish I were slower, actually.
[chuckles]
Hedvig: Yeah, I think that’s important. It’s like slow TV, but for science.
Daniel: It reminds me of the work by Daniel Kahneman and team who talks about two kinds of thinking that we do. The fast thinking, which helps you reach conclusions quickly and cut through lots of what might be extraneous noise. But then, the slow thinking, which is the more deliberate, which we don’t always feel like we need to do because we make fast decisions. But the slow thinking, that’s where it happens. [pause] All right. Thank you for the slowness.
[laughter]
Hedvig: Appreciate it.
Daniel: It can’t get much slower than the way I do science.
[laughter]
Daniel: Well, let’s see. Obviously, Omicron is on everyone’s mind, and because we are such an authoritative podcast, we have the power to say how people should say the name of this COVID variant. Is it Omee-cron, Om-i-cron, Omic-ron? What do you like?
Hedvig: Oh.
Martine: Oh, I thought it was Omee-cron.
Hedvig: I also thought it was Omee-cron. That’s the way the German media says that, I think, Omee-cron.
Bodo: Omee-cron.
Hedvig: Yeah.
Daniel: Omicron. I like it. Okay.
Hedvig: We don’t have to pick it because of that.
Daniel: No, but I like that. I was going for Om-i-cron, but, okay, that’s cool.
Bodo: That sounds like Macron, like–
Daniel: Oh, Macron.
Hedvig: [laughs]
Bodo: [laughs]
Daniel: You scam.
[laughter]
Daniel: Okay, we won’t do that one, because we don’t want to sound too much like the name of a world leader. And coincidentally, The New York Times reports that two Greek letters were avoided. The letter Nu because it sounds like the new variant and that would be confusing. The one that spelled X-I in Roman letters, whether that’s chi or she, or he, however you pronounce that. That was avoided because of its perceived similarity to a certain Chinese leader.
Bodo: Oh.
Hedvig: Oh, like Xi Jinping.
Daniel: Yes.
Hedvig: Oh.
Bodo: Oh, wow.
Hedvig: Oh. I didn’t know that. Okay.
Daniel: But Florian Krammer, @florian_krammer on Twitter, gives us another word, and that is scariant. And he says, “There are a lot of unknowns and many scariants have come and gone.” And of course, that would be a blend of scary and variant. There have been a lot of scary variants that have come and we thought they were going to be a thing and then they haven’t. They’ve turned into nothingburgers. Scariant, do you think it’s got legs? Do you think this is going to work?
Hedvig: What does it do that variant doesn’t do?
Daniel: Yeah, variant is pretty scary all of its own. Yeah.
Bodo: Harkening back to the beginning of the podcast, you’re doing the thing like reminding us of our mortality right now. That’s what I’m thinking about.
[laughter]
Hedvig: Yeah.
Daniel: Little close to the bone?
Hedvig: Yeah.
Daniel: Okay, cool. Maybe we’ll just move on to the next one. This one comes from pharoahcat on Discord. “Hate to say it, but Meta and Metaverse as Words of the Week.” I know, I know. We all hate Facebook. Everyone hates Facebook. Facebook is trying to overcome controversy with a ridiculous rebrand. Also, from Sam on Twitter, “Is Meta going to be a Word of the Week?” Now, it seems that the brand of Facebook has become synonymous with awful and toxic. So, the meta rebrand it is. What do you reckon?
Hedvig: I mean, they’re not changing the name of the website, the specific tool, Facebook. I’ve only seen Meta when I load up my Instagram, it says, “Brought to you by Meta,” and it used to say, “Brought to you by Facebook.” So, it’s like if if Alphabet changed their name to something else, it’s like people are still going to say Google, people are still going to say Facebook.
Daniel: Well, isn’t this the same as Alphabet?
Hedvig: Exactly.
Daniel: [crosstalk] -Alphabet has Google.
Hedvig: A lot of people don’t know that Alphabet exists. A lot of people will not notice that Meta exists. I don’t think it’s going to be that much of a rebrand. The video they did was silly, and yeah, I don’t.
Bodo: It’s just weird to me that both Google and Facebook picked those particular rather boring names. Also, they’re already existing words of the English language. So, it’s just like now putting you into this weird position where you have to– like the word ‘meta’, every time that you say it, you have to think about the company. It’s actually quite sad. Like, meta-analysis is a particular type of stats. It just sounds weird. I’m not analyzing Facebook necessarily.
Daniel: That’s so meta.
Bodo: That’s very meta.
[chuckles]
Hedvig: Do you feel that is going to happen? When you alphabetize, an author list, do you think of Google? I don’t.
Bodo: Probably not.
Hedvig: Yeah, I’m not so sure.
Daniel: I think the idea of a metaverse, that is a universe with metadata. It’s an interesting idea and it’s been around for a really long time. Imagine special glasses that you wear, hello Google Glass, and then you see an augmented reality view of your surroundings or you can see digital objects that you can interact with. Imagine going to a gig or a concert or something, imagine that at all. [chuckles] But imagine going to one, and then you can see maybe a glowing green arrow above everyone who’s chosen to make themselves discoverable to either everyone or to their friends’ network, or to only certain people, it’s like saying, “Hey, I’m here to come talk to me.” I think that’s interesting.
Hedvig: But is it going to work? I took the Eurostar through the Channel Tunnel a couple years ago, and they advertised an AR thing that I could use with my smartphone, so that I could be in the tunnel and the train in the English Channel and then wave my phone around, and then see fishes and see the ocean, does that make sense?
Daniel: Mm-hmm.
Hedvig: And then, they would use the gyroscope and the GPS on my phone to show me particular things as I was going through. The thing would just glitch and not work, the connection would not work. I’m not sure we’re practically there, is what I’m trying to say.
Daniel: I see why Facebook wants to do this, because it’ll probably be happening somewhere. If Facebook can become a leader in the AR space, then number one, it can undo the tailspin it’s in because younger users are not using Facebook. But the other thing is that Facebook has a problem. They don’t make the devices. So, Apple, for example, can act as a chokepoint, like it did when they said, “Do you want this app to be tracking you across sites?” No, I do not. And so, suddenly Facebook just lost a lot of business because their advertisers had no idea what their users were buying or looking at. So, this is a way for Facebook to control the device so that they won’t have to rely on Google or Apple as a chokepoint.
Hedvig: Yeah.
Daniel: I’m not happy about it because Facebook sucks. Is Facebook, the most ethically challenged company on Earth? I think it might be.
Hedvig: I mean I think Apple–[crosstalk]
Daniel: Nestlé?
Hedvig: -in China, not having great labor conditions, and Nestlé and Unilever. I think we’re not lacking of companies with ethical concerns.
[chuckles]
[crosstalk]
Hedvig: But, yeah, Facebook is definitely not. I don’t think this Metaverse thing is going to matter much. I think it was just a blip.
Daniel: Well, I think you might be right. There’s an article by Brian Merchant in The Atlantic, and this is a great quote, “There is not a single person in existence who has scanned Facebook’s newsfeed and said, ‘Yes, immerse me in this reality. I want to feel my uncle’s meme about hot pockets on my face.’”
[laughter]
Hedvig: I’m a big believer in that you should try and discipline your feed. There are things you can do with third-party apps and things, and you can do things with hiding words on Twitter so that you don’t see those sponsored or recommended “for you” things. We are very much living in our feeds and you should follow accounts and things that make you feel better or better informed. Use the unfollow button on both Twitter and Facebook. Unfollow your racist uncle. You should do that. It’s good for you. There’s no need for them to be in your feed, and you can do things–
Daniel: But, Hedvig, you’re siloing yourself, hearing things that only agree with you. [chuckles]
Hedvig: And then you should subscribe to massive good public news sources like, I don’t know, the ABC or Swedish Radio or BBC or Deutsche Welle or these things. You should subscribe to those, so that you get a dose of fairly balanced, well-edited news.
Daniel: There you go. Now, the cool thing about Meta is that it owes its current meaning to a mistake. Meta in the original sense was after or behind, or among or between, something about place. So, for example, a metaphor, meta would be a “cross” or “over,” and then the ‘phor’ is the Greek root meaning “carry.” A metaphor is when you say, “This thing is like that thing.” So, you’re taking knowledge from one area, and you’re carrying it across to another area. Metathesis is when sounds swap places in a word they move between them.
Okay, then something happened. Somebody coined the word metaphysics, because they were talking about Botton philosophy about what caused the universe, meta meant after. Metaphysics are the books that came after physics, that was how you arrange them on the shelf, but then, at least in one very popular volume, but then people looked at meta and thought that it meant the overarching principles behind physics, like something transcendent, like a layer on top of another layer.
Now what we have is metalanguage being language about language, metagaming is using knowledge about the game to beat the game, like gaming the game. And of course, as we said before, meta is jumping off and forming its own word, that’s so meta. So, we have the metaverse, but this was actually because of one of those little common shifts that happened in language. A little mistake that became a thing.
Hedvig: That’s fun. I like that.
Daniel: Yeah. Let’s go to our next one, skimpflation. This one comes from aristemo on our Discord channel. We’re all printing money, because we’re good Keynesians. We’ve moved from a production-based system, to a consumption-based system. And if money in the system is out there, then it bounces around and multiplies and goes through a lot of hands and things are good. But if the economy heats up too much, you get too many dollars, chasing too few goods, the cost of everything rises and you get inflation, which acts like kind of a tax on the least well off. So, one way that producers avoid inflation is skimpflation. What does that bring to mind for you?
Hedvig: I think on The Daily Show, they exemplified it with pepper. If you’re have a jar of peppercorns, and it used to be 12 ounces, and now it’s 6 ounces, but it’s the same size jar and the same price. You’re selling less product, isn’t that?
Daniel: Okay, so here’s the thing. We’re bumping up against two closely related concepts. The one that we’ve had for a while is shrinkflation. That’s the one where you get fewer chips in the bag and they still charge it the same price. But in an NPR article, Greg Rosalsky suggests that something else is happening as well. They’re not just shrinking the goods and the services. They’re are actually skimping on them, hiring less people. Not just getting a smaller amount, but the whole thing is becoming worse. Taking longer to deliver pizzas, putting you on hold for hours, and making everything kind of suck.
Hedvig: I’ve heard people say that about McDonald’s fries. I don’t know. I don’t go to McDonald’s very often, but apparently their fries used to be better for some reason.
Daniel: Really?
Hedvig: Yeah. People are like saying, “Oh, it’s a notable difference. They’re not as good now.” I don’t know. I haven’t tracked this, but–
Daniel: Oh, wow. Okay. Let me research this, and I’ll slap some links onto our show notes. What do you think? Is this going to travel?
Hedvig: Um, [chuckles] I think the silence from Bodo and Martine is telling.
[laughter]
Daniel: Yeah, that’s right. We’re test marketing these on you. It sounds like it’s not going well for skimpflation, no.
Martine: Doesn’t roll of the tongue.
Bodo: Hmm. But it’s certainly something that happens.
Hedvig: Yes, it is a phenomenon.
Bodo: I guess it depends on how aware people are of the phenomenon. I’m actually just reminded of– ages ago, I read a newspaper article, how in Germany, I think it was Nestlé, if I recall, but for one of their coffees, they were over the year, slowly reducing the roast time. It’s just so evil, because what they actually did is they’re almost working with the cognition of the population like, “Oh, yeah, they’re not going to notice it if you’re doing the change really slowly.” And you’re each year reducing the roast time just by little bits. But over time then, the company obviously saves a lot of money and is able to offer something that is very low quality in the end, but people haven’t noticed the shift. It’s evil, isn’t it? But to the extent that companies that continue to do that maybe will catch on, but I also agree, it’s not very catchy.
Martine: Maybe catching has something as well to do with all the continents in the middle.
Bodo: Oh, yeah.
Martine: Skimpflation, I mean it’s pretty hard to– and even shrinkflation is the same. It’s just too many consonants.
Daniel: That’s right. Out of the first, what, like seven phonemes in there, six of them are consonants. Pretty crunchy.
Martine: Kiki.
Bodo: But that makes it sound worse, doesn’t it? I mean, in a way, this fits.
Hedvig: [chuckles]
Daniel: Well, there is one thing that I think is very cool about this. I have never noticed flation as a combining form before, but it was there all these years. There was inflation and deflation. But then since the 70s, there has been stagflation, which was inflation plus stagnant growth or unemployment. There’s also been foodflation, agflation, taxflation, which is bracket creep. And then, slumpflation, which is inflation accompanied by a slump in output and employment. ChrisP on our Discord points out that Stephen Colbert used the term ‘meatflation’ on a recent show, which is probably not the same as inflating meat although it could be, because they do put air in hot dogs. So that could be.
Hedvig: Oh. There seems to be that a lot of people want to make ‘flation’ a productive stem that you can add things to. That’s for sure. People are playing around with it. And they’re describing phenomena that exist. To people who follow economics news, it might be handy to have a shorthand for some of these things, so that can make sense.
Daniel: The important thing about a combining form is that the parts of it, it has to you have to pick a part that’s recognizable.
Hedvig: Yeah, flation.
Daniel: And flation is definitely one of those. There you go. Okay, we’re moving on. This one comes from Pontus on Facebook who says, “Word of the Week has to be bones day and no bones day. If you don’t know what I’m talking about, check user @Jongraz on TikTok. It’s a phenomenon.” Okay, who knows this one?
Hedvig: No idea.
Bodo: Hmm, no.
Martine: No.
Daniel: [laughs] This one comes from Jonathan Graziano, a 30-year-old social media manager from New York. Every day, he makes a video of himself taking his pug dog, Noodle, out of bed putting him on the floor. Sometimes, Noodle stands up. He has bones. It’s a bones day. Get out there and do your thing. But if Noodle doesn’t stand up, bleh, it’s a no bones day. He has no bones. So, on the no bones days, you’re supposed to just take it easy, take care of yourself. It’s a no bones day. Don’t feel like you got to conquer the world all at once. I like it.
Martine: Yeah.
Hedvig: I call those probably headache days. Sometimes I wake up, and at the moment I gained consciousness in the morning, I have a headache already and I’m like, “Okay. Today’s going to be a light day.” Yeah.
Daniel: We’ve got to have those.
Hedvig: Oh, yeah.
Daniel: We’ve got to self-care. I keep talking about self-care, we’ve got to do that.
Hedvig: Yeah, for sure.
Martine: Yeah, that’s nice.
Daniel: So, no bones day.
Hedvig: No bones day.
Daniel: Hedvig, you’ve got a couple?
Hedvig: Yeah, I’ve got one that I heard from the podcast Friends at The Table, which is actual roleplaying podcast. I think it was Allie on Friends at the Table who said, “Youm.” Or Yom. I think she said Youm. Do you want to guess what it means?
Daniel: Okay. I want to guess. I think it’s an answer to whom. So, whom is speaking? Youm are speaking, that’s whom.
Bodo: [laughs]
Hedvig: It’s not that. Any other guesses?
Daniel: Okay.
Hedvig: [laughs]
Daniel: Dang it.
Hedvig: That’s a good guess.
Daniel: Thanks.
Hedvig: Anyone else?
Bodo: No guess here.
[laughter]
Bodo: It looks weird, youm.
Daniel: Y-O-U-M?
Hedvig: I’ll give you a clue. It’s related to y’all.
Bodo: Oh.
Daniel: Okay.
Martine: I was wondering whether or not it’s the plural for you.
Hedvig: It is you some.
Bodo: Oh.
Daniel: Oh, some of you all.
Hedvig: Yeah, because she was saying something, and then she said something like, “Youm, went over there or something.” And everyone started laughing. It was like, “What do you mean?” And she was like, “There’s you all, y’all, and then there’s you some. Youm.”
Martine: I like it.
Hedvig: I like it as well. [laughs]
Daniel: It’s not y’all.
Hedvig: It’s not the same as inclusive, exclusive directly, but sort of, a bit. It’s like some of you, not all the ones I’m addressing.
Martine: It reminds me of– and there are languages that have, so besides having a singular and a plural, and some has a dual for specifically two. Also, some that have a paucal for small groups.
Hedvig: Exactly. It’s very much like a paucal number. Yes. Usually, paucal is three to five-ish. Usually, it’s more than two, but not as many as what the plural is doing. Yeah, it’s a paucal second person– I really thought it was fun.
Martine: Cool.
Daniel: Okay. It is interesting. Although I will make the minor quibble that the all in y’all doesn’t really mean all of you. It’s just plural.
Hedvig: But it is plural.
Daniel: It’s not referring to–
Hedvig: And this is paucal.
Daniel: Yeah. That’s a minor thing. I’m glad to see a paucal pronoun.
Hedvig: Yeah.
Daniel: Like they have in [unintelligible 01:22:21] for example, which they have the two of you, the three of you, which is very cool. Okay. Ooh. We’re almost done. Last one, Hedvig?
Hedvig: My last one is BreadTube.
Daniel: BreadTube. I’m no longer surprised by the fragmentation of communities, but, yes.
Hedvig: [chuckles] So, you assume this is about?
Daniel: I probably know already. So, I guess I should say at last, but I’m going to say first. So, it’s a community of YouTubers who have baking as an interest. Am I right?
Hedvig: No. You’re wrong.
Daniel: What? I’m not.
Hedvig: It is related to YouTube.
Daniel: Fine.
Hedvig: You got that right.
Daniel: Yeah, because tube is a pretty obvious combining form, and that’s kind of new. Okay. I yield the floor.
Hedvig: Does anyone else have a guess? This is really hard. You have to know it.
Martine: It’s not an actual tube that bread comes out of?
Hedvig: No, it is related to YouTube, the video-
[crosstalk]
Hedvig: -web platform. Yeah, and it’s related to YouTube.
Daniel: I want a BreadTube now. Somebody give me a BreadTube that delivers hot, buttery rolls to my face.
Martine: So, it’s nothing to do with food then? What about money?
Hedvig: It is– how to say it? It’s related to Peter Kropotkin’s book, The Conquest of Bread. Does anyone know what that is?
Bodo: No.
Daniel: No.
Hedvig: It is a book explaining how to achieve anarhco-communism, and BreadTube is a term for a series of YouTube channels that primarily deal in socialist, communist, anarchist, or other left-wing perspective video essays
Bodo: Wow.
Daniel: Okay.
Martine: I have to check them out.
[laughter]
Hedvig: BreadTube creators refer to themselves sometimes as BreadTube or LeftTube. If you, for example, know about Lindsay Ellis or Natalie Wynn also known as ContraPoints. They are well-known BreadTubers or LeftTubers. It can be quite a fun place to hang out. Liv Agar, I would also count as BreadTube. They sometimes do what’s called algorithm hijacking. So, they will discuss similar topics that content creators of the right-wing discuss to hack the algorithm to be recommended to right-wing people. Yeah, anyway, I watched some BreadTube channels and I learned about the BreadTube phrase. It’s from this book, The Conquest of Bread. I just thought it was a fun word because it could be, like I think Martine was going for, a tube but you like cook bread in or something.
[laughter]
Daniel: Or the dough to make rolls just to come in this tube that would explode when you unwrapped it, because it was just under such pressure.
Hedvig: Yeah, exactly.
Daniel: But that’s not it.
Hedvig: Yeah.
Daniel: I’m sorry, Hedvig, I’m being dense, but can you connect the dots. The Conquest of Bread is about anarchy/leftism. What is the bread?
Hedvig: It probably is the daily bread, the bread that workers eat, the production of the means. The conquest of bread, I think is a metaphor for the conquest of the means of production, the conquest of the resources in society.
Daniel: Okay, I’m back. Okay, good. All right. Well, that was quite a journey. I feel we’ve gone all over the place, but slowness, Omicron, note the pronunciation, scariant, metaverse, skimpflation, no bones day or bones day, youm, and BreadTube. We’re trying to get them all out before the end of the year.
[laughter]
Daniel: Our Words of the Week. Now let’s get to some comments just quickly. This one’s from Kitty via email, but not the same person as pharoahcat, “Dear Daniel, Hedwig, and Ben, after listening to your latest Episode Number 39, I thought I’d write in and say hi. I’m an undergraduate studying linguistics in the UK, and I’m a big fan of the podcast. My friend introduced it to me sometime last year and she’s a longtime fan, who still refers to you guys as Talk the Talk.”
Hedvig: [crosstalk] Yeah.
Daniel: Yeah, we were there for a long time. “On the latest show, you read some listener comments on the two generativism shows. And following that, I thought I should tell you guys how much I enjoyed them as a student that cares quite a lot about linguistics. I think you guys did something very valuable with those two shows.”
Hedvig: Yes.
Bodo: Aw.
Martine: Aw.
Hedvig: This is my pet project. This fills a hole in me I didn’t know I had.
Daniel: An approval-shaped hole. “My uni department is very generativist. So much so, I’ve been labeled the functionalist in my year group. And it’s very hard to get an unmotivated evaluation of generativism both for and against or to adequately contextualize what we learned about generativism versus functionalism. Your shows provided a clear, comprehensive, and balanced look at generativism that I think students really need. It was particularly nice to hear your guests, Taylor and Adam, since what they work on is so different from what gets taught in my courses. I’d also not seen anywhere else where generativism and nativism are deliberately separated as concepts. Hearing you talk about that has really made me reconsider my stance. Maybe I mainly take issue with the nativism aspect and less so the generative aspect.” Me too, actually, I’ve got to say.
“In any case, Daniel said that he’s proud of those shows and he should be and all of you should be. Thanks for reading my email. I look forward to many more Because Language episodes in times to come.”
Martine: Aww.
Hedvig: That makes me very happy. Thank you so much. It is for people in your situation, Kitty, that we made these shows and I’m so glad you found it valuable.
Daniel: Yay. [unintelligible [01:28:17] on Instagram says, “Just wanted to share my favorite router password. It was prettyflyforaWi-Fi, but given my newfound irrelevance to society, you’ve probably already heard it so I don’t expect anything new. I’ve elected to use fuckthesister instead as it was my favorite piece of toilet graffiti. Anyway, keep up the great work. Kath, the middle-aged lady who’s only a couple of letters away from Karen.” Thanks for that Kath. If anybody else wants to share their clever Wi-Fi passwords or Wi-Fi router names, we are here for it.
Finally, Ozzie via email. “I’m a regular listener though a bit behind who just finished her episode with Sylvia Sierra. Great work as always. I’m excited to see Dr. Sierra making the rounds because I read her original paper on intertextuality a while ago and immediately thought, ‘Wow, this is straight up me and my friend group distilled into a paper.’ I have a gentle correction for you because you seem like the kind of folks would appreciate it. Now I have to give some background we were talking about Mandy Patinkin, the actor and we were saying, ‘Oh, wait, he played Inigo Montoya in the Princess Bride, but wait a minute, he’s not Spanish. He’s Jewish.’” Ozzie says, “Mandy Patinkin being Jewish is not the alternative to him being Spanish.”
Hedvig: This is true.
Daniel: “Despite the best efforts of Ferdinand and Isabella, there are still Jewish Spaniards. The alternative to him being Spanish is him being American.” Thanks, Ozzie, for that correction. We do appreciate it.
Hedvig: That is true. That’s a good point.
Daniel: Yep. All right, Dr. Bodo Winter and Dr. Martine Grice, how can people find out where you are, what you’re doing, and your general awesomeness?
Bodo: [chuckles] General awesomeness is very nice, thank you. That’s actually a good word too. Yeah, I’d say on Twitter. I’m @BodoWinter on Twitter.
Daniel: Cool. Martine?
Martine: Yeah, I don’t have a Twitter account. Oops.
[chuckles]
Daniel: Good for you.
Martine: Follow Bodo.
[laughter]
Hedvig: Fair enough.
Daniel: If you want to follow Martine, just follow Bodo instead.
Martine: [laughs]
Daniel: Dear Martine, hi Bodo.
[laughter]
Daniel: Thanks so much for hanging out with us. This has been a lot of fun and I really appreciate the work that you’re doing.
Martine: Thank you.
Bodo: Thank you. Yeah, it’s been fun.
Martine: Yeah, I liked it
[music]
Hedvig: Wasn’t that a fun show? Now, if you have a question or comment or, just like Kitty, you want to come by and say hi? You can get in touch with us. We are BecauseLangPod on all the things and you can also send us a good old-fashioned email. We are hello@becauselanguage.com, one word. In particular, I thought it was very nice to hear Kitty’s thoughts about the generativism episodes. They were a little bit of a challenge for us to make and a bit of a more deep dive than we usually do. So, we’d love to hear what you thought of that. Besides getting in touch with us, we would also love it if you would maybe tell a friend about us. You can do like Dustin of Summer Stories and just recommend us to anyone who say they’re interested in a new podcast on Twitter, or tell a friend about us in real life, if you have friends who meet in real life or on Zoom, Skype, Google Hangouts, wherever the kids are these days.
Besides that, another way you can help out the show is by leaving us a review in all the places you can leave a review. And fortunately or unfortunately, however you look at it, Apple Podcasts is one of the best places to leave a review and we do have a new review. This one comes to us from [unintelligible [01:31:45].
Daniel: [unintelligible 01:31:48]
Hedvig: I’m Swedish, I’m going to guess this [unintelligible [01:31:50].
Daniel: This person is from Norway.
Hedvig: Well, same thing.
[laughter]
Hedvig: I don’t think they would be offended even. [unintelligible [01:32:00] so it sounds like it’s to do with toddling, maybe, being a toddler. Anyway, they say, “A lovely piece of language. This show is amazing. I followed over from the Talk the Talk days. I also studied linguistics at university and this way, I get to keep following the discussions that are going on. I also love how the show manages to cater both to us nerds and introduces concepts to people who have not yet studied it.” That is exactly what we want to do. We want to cater to the ones who are already into linguistics, that are being linguistics [unintelligible 01:32:29], but also make it approachable and accessible to newcomers. So, that’s great if we can achieve that. Thank you so much, [unintelligible [01:32:37].
Daniel: Thanks, [unintelligible [01:32:41]. And thank you to our patrons because they help us to make episodes and release them for free without annoying ads. Except for that one spectacular vernacular thing. It also helps us to make transcripts, and what that means is that you can read and search the show if that’s what you would rather do. It means they’re accessible and searchable. So, thank you to all of our patrons. Thank you. Thank you. Thank you. Thank you. Thank you. Big shoutout to our top patrons, Dustin, Termy, Chris B, Matt, Whitney, Helen, Udo, Jack, Kitty, Lord Mortis, Elías, Michael, Larry, Kristofer, Andy, James, Nigel, Kate, Nasrin, River, Nikoli, Ayesha, sneakylemur, Moe, Steele, Andrew, Manú, James, Rodger, Rhian, Colleen, glyph, Ignacio, Kevin, Jeff, Dave H, Andy from Logophilius, Samantha, zo, Kathy, Rach, Taylor, Cheyenne, Felicity, Amir, and national treasure Kate B. Thanks to all our patrons for your support.
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’ll catch you next time. Because Language. Because Language. Because Language. Thanks.
Hedvig: You’ve got to work on your vocal fry. You need more vocal fry. You know that, right?
Daniel: You know my big inspiration is, of course, Ken Nordine, who had a vocal fry. He could just fry all the way through the sentence.
Hedvig: It’s so nice when people have a good–
Daniel: [in vocal fry] Because Language.
Hedvig: Because Language.
Daniel: That wasn’t good. There’s good fry and there’s bad fry. That was bad fry.
Hedvig: Ah, yeah, maybe. But it’s nice. Vocal fry is nice.
Daniel: I’ll work on it.
Hedvig: [in vocal fry] Yeah.
Daniel: [in vocal fry] Yeah.
[Transcript provided by SpeechDocs Podcast Transcription]