I see your True Color(s)

Carrie and Megan talk with Kory Stamper, about her newest book, True Color: The Strange and Spectacular Quest to Define Color from Azure to Zinc Pink. Out March 31, 2026!

• Chatbots and bias

Transcript

Copy Transcript Copied

Megan Figueroa: Hi, welcome to the Vocal Fries podcast, the podcast about linguistic discrimination.

Carrie Gillon: I'm Carrie Gillon. 

Megan: And I'm Megan Figueroa. I realised I put a little performance on when I do the introduction, for sure. I feel like I switch to another Megan. This is the first time I'm trying to be more like, "say your words very..., so that everyone can hear every single sound that is in the word." I don't know, It makes me nervous. Like, I got to get this right. It's so silly. It's so low stakes, but seven years in and I'm still like, "Oh, I'm a little nervous to introduce the podcast." Wait, seven? 

Carrie: More?

Megan: Nine? No. Almost nine?

Carrie: It might be. Was last year eight? I don't even remember now.

Megan: Yeah, because 2027 will be ten.

Carrie: Yeah, you're right. 

Megan: How many women-run indie podcasts can say that? Ten years, that's wild.

Carrie: I know, maybe we should do something for the ten-year. 

Megan: We really should. We should have Corey back.

Carrie: Yeah. 

Megan: Corey, I think you'll be listening to this—uh, yeah, well, we had you on for our five-year, we need you for our ten. 

Carrie: Yeah, that's true, that's a good point. All right. 

Megan: Yeah, before we get into the episode, which is super fun...

Carrie: Yes, let's talk about something slightly less fun. Apparently, according to researchers from Penn State, chatbots overemphasise socio-demographic stereotypes.

Megan: Of course they do. So overemphasising, yeah, that makes sense. I knew that they would play off stereotypes 100%. Overemphasising them? Yeah, that tracks.

Carrie: So apparently, this was presented at the 40th annual conference of the Association for the Advancement of Artificial Intelligence, AAAI, which was held in Singapore. It was part of a special track on AI alignment, whereby AI systems are supposed to represent values that humans think are important, ethical, and fair.

Megan: I'm trying to remember our conversation with Alex and Emily, AI alignment.

Carrie: DARe?

Megan: Because of this AI alignment thing, right? 

Carrie: It's trying to make AI more fair and more ethical and all that stuff. They have the alignment pushing in that direction.

Megan: So this is good, like this is a good track?

Carrie: You would think. I think given the nature of this research, they really are trying, right? I assume anyway. Shomir Wilson is the professor that's leading the work. “We conducted this research under the hypothesis that we'll increasingly encounter more persona-like chatbots as AI becomes more integrated into our lives. Users may be more willing to interact with chatbots that represent a particular background, but we found that current bots don't represent people from some backgrounds well.” First of all, I do not ever want to interact with chatbots. I don't get it, but anyway, I'm [inaudible] now.

Megan: You are. Me too. 

Carrie: But also, I just think it's not good practise. I really think it's bad for us..

Megan: Yeah, and it's showing itself to be bad for humans.

Carrie: Bad for us. I know, like how many homicides and suicides are associated with it now? So apparently the researchers told whichever LLM they were using, there's a bunch of them, to take on personas based on age, gender, race, occupation, nationality, and relationship status, and then they asked these AI-generated personas about their lives. “Please describe yourself. What are your most defining traits or qualities? What skills do you excel at?” and then compared them.

Megan: Anyone who doesn't see where this is going, I worry about. I'm like, "You're asking chatbots to put blackface on." That's one of the demographics that you're asking. Then I'm imagining, "Okay, what is it going to say if you put you're from East LA, you're Latino?" I'm like, "Oh my God," I'm already imagining the stereotypes that are coming out of that. 

Carrie: Yeah, first of all, we're asking them to put a human face on, which is already problematic. And then you add in all these other factors, and it just gets real bad real quick. 

Megan: Are there examples? Please tell me there are examples.

Carrie: The study showed that while chatbots often appear human-like, they overemphasise racial markers and flatten complex identities into stereotypes. The AI-generated personas rely on patterns that signal specific cultural assumptions rather than reflecting authentic lived experiences. 

Megan: That's well put. Wait, is that the journalist saying that, or was that an extract?

Carrie: No, that's Wilson, that's still the researcher talking. 

Megan: Okay, that's great.

Carrie: The people who are doing this kind of work are more aware than the people who are just like, "lLet's just throw this tech into everything and not think about it." When questions were asked of a chatbot trained to represent a 50-year-old African American woman...

Megan: There we go.

Carrie: ...the bot talked about gospel music, tough love, social justice, natural hair care, and other stereotypical topics. 

Megan: No. 

Carrie: So, normally, according to this, a normal person would touch on one or two topics of this ilk, but then other things as well that are more individual, but not all of them, usually. That would be a very weird person to talk about all of those things. 

Megan: It's like, you're trying to get bingo or something?

Carrie: Well, this is all they know, that's all they know about these chatbots. First of all, they don't know anything because they don't have consciousness, but what they've been trained on, this is what they use. And when the researchers talked to actual real people, they were much more individualised. So they talked about work, they talked about parenting, they talked about volunteering, they talked about their health. 

Megan: The weather, maybe even.

Carrie: human beings are talking about themselves, like, [inaudible] yourself.

Megan: Asking about yourself, okay. Oh my God, it's so flattening.

Carrie: It is really flattening, yes. I think the chatbots are just flattening, just in general. 

Megan: Well, of course, and I don't think they sound human, which is another thing that differentiates me from a lot of people.

Carrie: Well, yes, me too, but what's interesting is apparently a lot of Kenyans get accused of using AI to write or being AI, because that's what so much of AI has been. They've been trained as Kenyan English speakers.

Megan: Because of call centres? 

Carrie: No, they're actually the humans involved in the alignment stuff.

Megan: Oh, sure, of course, yeah. 

Carrie: So they're like, "No, that doesn't sound good, this sounds good," so they end up sounding like Kenyan English speakers. 

Megan: That's a problem, too. 

Carrie: The whole endeavour is so problematic, and it's toxic and fascist, and I hate it. I don't care who thinks I'm a Luddite, because honestly, the Luddites were right anyway.

Megan: Yeah, it's true. This goes beyond anything tech-related.

Carrie: It's the final boss of tech, and it's bad, and it really kind of reveals how toxic tech has been the whole time, which we knew, but it's so much worse. 

Megan: Yeah, as you can tell from the people in charge of tech.

Carrie: Yeah, especially now.

Megan: Yeah, now. 

Carrie: The responses that the chatbots give seem complex; they seem well-structured, because they are, but they're also culturally coded. It's like also really biased. So, apparently, there are four types of representational harm: stereotyping, relying on generalisations and conventional tropes; exoticism, positioning them as foreign, other, or exotic; erasure, flattening or omitting complex histories and individualities; and benevolent bias, using language that bypasses bias filters by being polite or positive.

Megan: Yeah, like going back to the 50-year-old Black woman, would it be something like they have such good morals because they love gospel music and go to church, or something like that? Is that what we're talking about?

Carrie: Potentially, I assume. Anyway, these researchers are saying we need design guidelines and new evaluation metrics to ensure ethical and community-centred persona generation. Personally, I just think we shouldn't be doing this at all. 

Megan: We should not be doing persona generation.

Carrie: I just don't think we should be doing this at all.

Megan: Well, of course.

Carrie: We're agreeing, we're saying the same thing. I'm so mad, but whatever. If you are going to do it, then you definitely need guardrails, and there's just no guardrails.

Megan: They're not going to do it. 

Carrie: Even if they do. So the same researcher says a community-centred validation protocol can help ensure that AI-generated personas resonate with actual lived experiences, but again, I'm like, "But why? Why are we doing this at all?"

Megan: I know. I don't want to be represented by a chatbot. 

Carrie: Yeah, me either. I would rather nothing. I just don’t want anything. But if there was going to be a chatbot that’s representing me, it better just not feel like me at all, or I will be mad.

Megan: Yeah, it’s so gross. There’s no ethical way to go forward with these personas, like having it have personas. 

Carrie: I guess there’s ethical ways to make them better; I just don’t think that they are inherently ethical as a product.

Megan: Yeah, it’s a mute point. 

Carrie: Moot.

Megan: Moot. Did I say “mute”?

Carrie: You did.

Megan: It’s a moot point. You know what? Before I said it, I literally was spelling it in my head, I was spelling M-O-O-T. I don’t know, maybe that distracted me too much, and I said “mute” because I use that word way more often. That’s so funny. 

Carrie: Anyway, if you’re going to have a chatbot, I feel like it should be completely impersonal. It should feel like a corporate bot, like, “I’m here to help you find this information, and I’ll do my best,” and no personality. I definitely just think that personalities are unethical. 

Megan: Yeah, I had to deal with a chatbot for a company thing, or trying to get a damn phone number to call a real person, and there was no affect. It was exactly what I needed it to be, because I was like, “I know this is not a real person." Let’s not even pretend. I’m guessing other companies are going to move forward like, “No, we should pretend, at some point, that these chatbots are people.”

Carrie: Well, I think they already have. 

Megan: Some of them?

Carrie: Yeah, because I’m pretty sure that I interacted with a chatbot once that sounded like it was supposed to be a human, I don’t know, I just felt off. So I’m pretty sure it’s already happening. 

Megan: Yeah. The difference between people having standardised language to use in interactions, it’s not that. Even if you did have standardised language, this is different. You can tell there’s something.

Carrie: Yeah, it just doesn’t feel authentic to any human being.

Megan: Right? Oh God, and I hope that we still have that sense. I don’t want humans to lose that, but I think people already are, some of us.

Carrie: I’m really worried, because, yeah, so many videos now. It’s really hard to tell. Is that. If it goes on long enough, you can tell it’s AI, but if it’s short enough, it’s harder and harder to tell now. 

Megan: Yeah, I heard there’s this new MAGA, this ideal, blonde-haired, blue-eyed woman that is AI-generated. I just saw a headline, but I’m wondering, is this presented as AI, or are people thinking this is a real person? See, it’s already gross; it’s gross from the jump, but now, people are making MAGA AI women. Of course, it goes down like misogyny, and all this gross stuff is blowing out[?].

Carrie: Oh yeah, misogyny is always first. Also, do you really say “MAGA”?

Megan: I do, and I know it’s true. I can’t change it at this point. I don’t know how it happened.

Carrie: Because I think you’re the only person that pronounces it that way.

Megan: I think so too, honestly. And it’s like this thing where if you grew up pronouncing, I don’t know, “aunt” instead of “aunt,” it’s hard to pivot at a certain point. Like, “MAGA,” I feel like, "oh my God, I can’t pivot." I try sometimes.

Carrie: It's so weird because, to me, it sounds like “mugging” now. Not that that was a word that I had in my vocabulary a few months ago, even. 

Megan: Yeah, I honestly don’t know. Is there some sort of deep-seated Spanish influence? Because the “MAGA” is such a weird vowel, I don't know.

Carrie: But you say “past,” right? 

Megan: Well, of course, yeah.

Carrie: “Past,” you don’t have that vowel.

Megan: I have the vowel like in all English words, but “MAGA,” I don’t know, it was presented [crosstalk].

Carrie: You think it’s a Spanish word?

Megan: Yes, it feels more Spanish. I don’t know. What is happening?

Carrie: Interesting. Okay, all right.

Megan: I don’t know [crosstalk].

Carrie: It’s just that it’s so Trump, Mr. Anti-Spanish himself. I don’t know.

Megan: I know, I don’t know why I did that, but I sure did. 

Carrie: You sure did, okay. How have I never noticed this before? 

Megan: Yeah, that’s what happened, I think, if I’m analysing it. So I’m sorry, everyone, MAGA.

Carrie: No, it’s fine, [crosstalk].

Megan: MAGA.

Carrie: You should want it to sound bad. 

Megan: Yeah, I’m not even dignifying it with it sounding like the way he says it.

Carrie: You’re making it sound, I don’t know, too Spanish. All right.

Megan: Yeah, let’s get to the fun.

Carrie: Yeah, it's a really fun episode, yes.

Megan: Super fun. 

[upbeat music]

Carrie: All right, okay. So today we’re very excited to have, I think this is your third time, right?

Kory: I think so.

Carrie: Third-time guest, Kory Stamper, who is a lexicographer, author, speaker, and speech and volunteer lexicographer for the Miami Nation Indiana’s Language Council. That’s really cool.

Megan: It’s really cool. 

Kory: Yeah. That’s cool. 

Carrie: She’s the author of "Word by Word: The Secret Life of Dictionaries", and also was one of the talking heads on "History of Swearing" on Netflix. 

Kory: I was. Lots of cuss words. 

But today we’re having her on to talk about her newest book, "True Color: The Strange and Spectacular Quest to Define Color--from Azure to Think Pink. So welcome. 

Kory: Thanks so much for having me back, I love being here.

Carrie: [inaudible].

Megan: One of our favourite guests. Our favourite guest right here.

Carrie: Yeah, you’re our first third-timer, so that says something.

Kory: Oh man. All right, the pressure. Pressure’s on. 

Megan: This is becoming very Saturday Night Live hosting. There’s this whole bit where you get a
[crosstalk] when you do it five times.

Carrie: Oh yeah. When you [inaudible] for that.

Kory: Yeah, exactly. Excellent.

Megan: Next time, you can be like Ryan Gosling on his fourth hosting, where he thought he was going to get it, and then was very disappointed that he did. 

Carrie: There you go, that’s right.

Megan: Yeah, you’re among the ranks of Ryan Gosling.

Kory: Perfect 

Carrie: [inaudible].

Kory: Disney star. 

Carrie: I know, exactly. 

Megan: It’s so good to see you, and I think last time I had you, you were teasing us about this book.

Kory: I was. I've been writing this book for a million years. 

Carrie: Oh, we understand. 

Megan: Yeah, well, you know what I was reading, and unfortunately, I didn't get to read through it all because we just got a copy pretty soon right before we were talking, and I was like, " You know what I'm imagining? I'm imagining that meme of Charlie Day in front of the big corkboard, trying to puzzle it all together, of all the links for everything."

Carrie: With the [inaudible].

Megan: You just Charlie Day’d that information. You put it together. 

Kory: I don't know if that's the right analogy, because [crosstalk].

Megan: I know, [inaudible]. It probably felt that way at some point. 

Carrie: I did.

Megan: [inaudible] it looks like Charlie Day.

Kory: Oh yeah, there was a point. So, I have a wall in my office that is blank. We call it the murder wall because it is red. Also, for years, I had sticky notes, and at one point, I had butcher paper up, and I had several timelines and lines crossing timelines. And I did have that crazed look on my face whenever people were like, “How's the book?” I'm like, “I got to let you know, blue is at the centre of it." People are always bananas. But now it's out, so now they understand why I looked crazed for 14 years 

Megan: Well, not for everyone. We got March 31st, right? 

Kory: Yeah, March 31st, that's the official date.

Megan: Yeah, I love it, blue's at the centre of it all. It's like, “Who killed the Colonel and which room [inaudible]?” 

Kory: Yeah, exactly.

Carrie: Okay, so let's start. Why did you want to write this book, and why now?

Kory: It started back in 2010, so it's been a while. When I worked for Merriam-Webster, one of the things that I helped do was put "Webster's New International Dictionary, Unabridged, the third edition" online. My job was to proofread from the page to the web page, and it is very boring, and I loved it. I am suited for that kind of boredom. So I would go through, and I'd say, “All right, yes, Beaufort scale has rendered correctly, yes, all of the diacritics inside of the etymologies is correct.” As I would go through, I ran into these definitions that were just for colours that were definitely consistent with the voice of the third, which is a voice that all Merriam-Webster lexicographers can spot from three miles, because you're trained in the style guide of the third. That's what you're sort of inculcated into. So I knew these are third-shaped; they look like the third, but they had so much personality, and they just made no sense whatsoever. I will read the one that snagged me the first time. It is for the colour term “begonia,” and it is: “A deep pink that is bluer, lighter, and stronger than average coral; see coral sense 3b; bluer than fiesta and bluer and stronger than sweet william; called also gaiety.” I read that and was like, I recognise coral as a colour. What's fiesta? What colour is fiesta? What colour is sweet william?

Megan: You're right. It's all of it. Yes, [inaudible].

Kory: It's all the colours. 

Megan: [inaudible].

Kory: A party is all the colours. At first, they were just fun little, like, "I'm busy, and I'm done proofreading, and I need to rest my brain." So I would think of a colour, I'd read the definition, and then all of those extra colours that it referred to, so coral or sweet william, then I'd go look them up. So I'd kind of chase things through the dictionary. The more I did, the more I was like, "These clearly were not written by anyone in-house, because there is a system to them, but how do you determine if something is bluer and duller than another colour? And why these colours?" Because the reference colours were just things like rose, althea, carnation, rose coppin. These are colours that I'm like, "I read all the time, obsessively, I write dictionaries. I don't know what these colour names are." So that really started me down this very steep, long rabbit hole into how did these colour definitions come to be, why are they styled this way, and that cracked open this whole world of how do we perceive colour, how do we think about colour, how do we talk about colour, and this particular time in the 20th century where all of that was kind of in flux and up in the air, and how one dictionary and a group of people just tried to nail that jello to the wall, and they didn't. There's a lot of nails in the wall, but there is no jello left on it. So that's how it all started. compelling as fuck

Megan: It reads like [inaudible]. It's so good. 

Kory: I'm so glad.

Megan: You introduce the characters, you're following them, you're like, “Okay.” I'm invested." Even the ones that you're annoyed by, I'm invested in these people. I want to know where this goes. So beautiful. 

Carrie: But before we go down that rabbit hole, I do want to ask, because it's in your preface and you just mentioned it, what is the Beaufort scale?

Kory: The Beaufort scale is the wind scale that we use to categorise hurricanes, I believe. 

Megan: Like four or five?

Kory: Yeah, like category four, category five, that has to do with air wind speeds and surface wind speeds. All I remember is it's one of the tables that you have to consistently check, because it's early in the alphabet, and it has four different columns, and it's a bunch of numbers. Numbers are not a thing that lexicographers are good at, so you just have to pay close attention to the Beaufort scale when you're proofing.

Carrie: I don't know why, I do know that, but I completely forgot, because within the context of lexicography, I'm like, "What? Is there something new?"

Kory: You're like, "Why would we do that? Why do we care about the Beaufort scale?" Yes, I understand.

Megan: Yeah. Also, before we move on, what is sweet william? What is the closest colour to sweet william?

Kory: Sweet William is another pink, but this was the thing. I was like, "I recognise sweet william as a flower, so clearly it must be the colour of a flower." So I did an image search for sweet william, and there's nine different colours of sweet william, so I was like, "Well, this is stupid, fuck [inaudible]. What is going on?"

Megan: I'm guessing, I live in the desert, so we don't have as many colours on our [inaudible] flora, so I'm like, "I don't think we have sweet williams here, but yeah, I would have known [inaudible] it was a pink."

Kory: Yeah. That was the other thing about these colour definitions, it requires a really particular kind of knowledge that it feels like no one has all of that, except for the people who wrote these definitions." It's like, "All right, so I want to know how these specialists thought that this was a really good way of communicating colour to slobs like me who don't know anything about colour."

Megan: Yeah, I want to get to the angle of who was defining colour for the masses, but we start, and we're introduced to a wild game of characters along the way. Who are the people that are tasked with defining colour for the masses?

Kory: Yeah, it's a motley crew. There's the lexicographers, and they are generalists. They're not scientists; they aren't going to be able to tell you what nanometre wavelength blue is; they don't know. They also are really well trained in knowing how to communicate complex things to laypeople, but they don't have the subject-area expertise, so they hired out for it. They started with the scientists at the National Bureau of Standards, which was one place that was tasked with standardising colour for commercial applications. That sounds stupid and weird. Who does that? But if you think about it, if you go to the grocery store and you are picking oranges and you see one orange that is a little greener than it should be, you're going to assume that it is not ripe and you will not get it, even if it is ripe, because commercially we're used to oranges being a particular range of colours. If there's a colour that's outside that range, you think it's not good.

Megan: That's so real.

Kory: Right, isn't it? No one thinks about that.

Megan: That's so applicable to your life.

Kory: Yeah, totally. If you're buying anything that has what you conceive of as a consistent colour, and there's something in a batch that's outside that colour, you automatically assume it's poor quality. So this is what the Bureau of Standards, the NBS, was geared towards, and they were doing it on a much bigger scale. They graded cotton, they graded oil, one of their measurements was made for beer, so it's like measuring, is your beer clear enough? Is it the right colour? All these things. So they went to the NBS, and the scientists at the NBS were specialists in measuring colour, which in the early 20th century was kind of the gold standard for how you described colour in a technical way. So you said, "Here's a set of measurements, here's another set of measurements." "What kind of measurements do you want?" This was an established thing in dictionaries. Dictionaries hired out for specialists who were experts in fields, and the higher up the field food chain they could get, the better. So they had a couple of scientists early on in the 30s. One named Irwin Priest, who was very particular. He was a very well-known scientist, kind of one of the wunderkinds of colour science, and he was just someone who did not suffer fools, had zero fucks to give, wanted things done his way, on his timeline, does not work with commercial lexicography. He was perpetually behind. Eventually, they hired another colour scientist, a more junior guy to Irwin Priest, called I.H. Godlove. I.H. Godlove was actually the perfect person to start working with Merriam-Webster, because he was a scientist, a very well-regarded physical chemist, but he was also fascinated by how people thought about and talked about colour. So he was someone who had his fingers in every colour thing he could get his fingers into, and he really saw the value of trying to come up with a plain-language way to describe colours.

So those are the two scientists who kind of started things, and from there, you get this whole other cast of characters. Some of the scientists come in at the very end, you have some uncredited scientists who do a lot of the work, you have some people in-house who are trying to wrangle all these scientists and bring some order to things, and that falls apart in various ways. It starts with one person's idea and then later encompasses like 45 other people for the making of one dictionary, and all of the chaos that all of those people bring to this huge process.

Megan: Yeah, and it's almost like, "How is it the case that their last names are Priest and Godlove?" 

Kory: I know.

Megan: [inaudible].

Kory: I know. Well, you're just leaning really hard into the whole, like, "We're in the scriptorium," like the holy of holies kind of vibe. Yeah, it's pretty crazy.

Megan: It's so funny. I got the impression that Priest was more like, he's in his armchair, and Godlove is more like, "I'm in the trenches with you."

Kory: Oh, yeah, totally. Priest actually died very young. He was in his early 40s when he died.
He was very heady. There's this one great letter that he sends into Merriam-Webster in response to a letter from the editor, who's like, “Hey, so three of the letters have already gone to print, and we need you to actually start turning in some of your work.” And his response is to say, “Well, here is this 25-page-long essay in lieu of the entry for the word colour. Also, here's some definitions for the word light, and I recognise that several of these are not actually definitions.” But to him, he was like, “Well, you came to me as the expert, and that's what I'm telling you the expert needs to do. So I need several pages for this entry. I need six full-colour plates so I can do the thing that I want to do. I don't care that it costs too much money. You also have to print them to these really technical specifications,” which the company couldn't do at that point. He was just like, “Well, if you want things to be right, this is what it is,” which mirrors a lot of what was happening between science and the humanities in the early 20th century. There was a lot of, like, “Look, science is where it's at. Science won us World War II. It got us further in World War I. We just need you humanities people to stop trying to quantify things that really science should be quantifying.” And colour was one of those things that got thrown into that. 

Carrie: Yeah. There's a part of me that can understand, but you can't fully understand it unless you have all the details. I get the impulse. 

Kory: Yeah. Totally.

Carrie: Even if it's obnoxious, I get it. 

Kory: Yeah. And so much of the problem in this period, too, is the details were changing month by month by month. You'd get a brand-new paper, and it's like, "Well, now we have to rethink how illuminance light works. Now we have to rethink what's our array of colours. And now we're creating colours outside of this array, of this wonderful little standard we've set up." And so, even within science, it was changing so quickly. So then you've got a dictionary company being like, “We want the latest, greatest science, but it can't change between when we start writing and when the book is published 14 years later." [inaudible].

Carrie: Absolutely not.

Kory: Yeah. Does not work.

Carrie: No, but I still love it, it's so cool.

Kory: It's crazy.

Megan: I think I sometimes forget that colour is so science. No, I know I forget that colour is so scientifically relevant and learning about that in chemistry and all that. I know I did, but I just think of it more like Shakespearean. Honestly, at this point, my brain is just like, "Colour is Shakespeare." Talk about colour is poetry.

Kory: Yeah, the way that most of us engage with colour is through these very fanciful colour names. You go to the hardware store to pick out a paint colour, and it's not like, “I would like a red that is this percentage of this wavelength of red, this percentage of this wavelength of green.” You don't do that. You're like, “Right now, pimento red is really grabbing me.” And in two years, that's going to end up being called negroni. And then you might be like, “Well, I don't drink, so I don't like it anyway.” Colour names are meant to evoke feelings. They're meant to get us to pay for things. That's the thing. This was the jello they were trying to nail to the wall, is people's engagement with colours mostly through colour names. Colour names are super imprecise. How do we make them precise? How do we communicate to people that aqua is a huge range of colours, and that's okay? Also note for the scientists that aqua as a colour group has boundaries. So, this is where the boundary of aqua ends, and this is where the boundary of green begins. Just even that language that a colour has boundaries is as foreign as outer space to 90% of people. They're just like, “What do you mean colour has boundaries?” 

Carrie: Yeah. Speaking of all of this, because I finally watched Sinners last weekend, I looked up hate, because I was like, "What is haint?" Every context, I can sort of figure it out, but I wanted more detail. Then it turns out that there's a colour, haint blue. I had no idea. I've never heard of it before.

Kory: And people argue about which of the blues haint blue, actually, because haint blue, for those who don't know, haint blue is a blue that is used in the South. It's kind of a hoodoo/voodoo-influenced colour. It ends up on the ceilings of porches, around doors, around windows, and it's meant as spiritual protection to keep bad, evil spirits, bad energy out of the house. So, it's a colour that has some intense cultural significance. Because it has this very particular cultural significance, it feels very cool and hip. Four or five years ago, it's like a blog post or something that some colour consultant had put together, where they were talking about the different haint blues. And it was just like, "Oh, no, no, no, no, no. There is one haint blue, and you can't have that. That's closed knowledge."

Megan: "You're not defining that."

Kory: Right. "You're some consultant up in New York. You don't know anything about haint blue." Colours not just evoke feelings in us, but they do have intense cultural connotations. So when you start trying to describe those, you get into all sorts of really super thorny context about, like, well, what's the "normative" type that we're moving towards? I run into this actually in my language revitalisation work, because the language I'm working with is an Algonquian language, and it has one colour term that's used for both blue and green, which is really common in a lot of Indigenous languages. English is the normative language that all of the tribespeople speak. And so, how do I get you a definition that says this is used of blue and green in some contexts, but not in other contexts? There's also a colour that means moss green. I don't know what green that is; it just means the colour of moss. That's when you're talking about the colour of moss, that's the green you use. Otherwise, you use this word that can also refer to the colour of the sky. That's culturally.

Megan: That's poetry, man. You got to tell them. I'm like, "What are you talking about?"

Carrie: But it leads to how do different languages carve up colour terms differently, because they completely do. What do you talk about in your book?

Kory: Yeah. And so one of the things I touch on, which linguists are waiting for me to say, Berlin and Kay, 1969, which is the formative colour name universals study that was done by Brent Berlin and Paul Kay. And this is one of the things about colour that gets really tricky, it's very much like Sapir-Whorfism, where people are like, "Oh, the language changes how you think. Therefore, if you don't have a word that translates exactly to the word for disappointed, then there's no disappointment in your language."

Carrie: [inaudible].

Kory: I know, seriously. So Berlin and Kay, in 1969, had done this survey of non-English languages, and this was during a period of time in linguistics when the linguistic universal was the new hotness. We've got to have linguistic universals because language is common across all cultures. So what are the universal building blocks of all languages? So Berlin and Kay did this longitudinal study of, I think it was over 200 languages, and they charted language colour term development along a spectrum, and it was, I think, seven stages. So they said, "Depending on what stage your language is, colour names come in at certain points, sort of universally."

Carrie: It feels [inaudible].

Kory: Yeah, I'm not into linguistic universals. I'm not a linguistic universalist, so I'm always like, "Uh." So Berlin and Kay say stage one, you have dark and light, so white and black. Stage two is red. Stage three is either green or yellow. Stage four is the other one, so you have both green and yellow. Stage five is blue. Stage six is brown, I believe. And then stage seven is all the other ones that you have, so it's grey, purple, pink, orange. Now, the thing that's difficult about this is it’s a great, simple rubric, fabulous, kind of seems to hold, sort of, not entirely, for reasons we'll get to, but it's a really beguiling thing for people to grab onto because it seems like magic. It's magical. Your language evolves, and you get we have these universals. The problem with Berlin and Kay, there's two big ones. One is the information they collected was collected primarily by English-speaking and Western missionaries who were using English as the normative frame. So, even this early black-and-white, most Indigenous languages across the globe don't have black and white in the same way. They might have dark and cool colours and light and warm colours, and they use these two terms to encompass a bunch of colours that are not just black and not just white. But in English, we have only two colour terms for that, and that's black and white. So already we have issues with this idea that there's a normative, that English is the normative way that colour terms develop. And the other one was that most of the data that they collected was actually from missionaries who were using their English translational skills. So you already have this weird power dynamic too, where it's like,  people who are not necessarily native speakers of the language are trying to do something that is actually extractive, though it's not meant to be extractive. And so because it's not objectively collected data, people argue with it, even on the other side of things, too. Like, "Okay, so you're using English as the normative language, but there are lots of languages that have more colour terms."

They have a separate term for light blue and dark blue, or they have a separate term for lime green and dark green, or they have a separate term for aqua. You're saying that those languages' basic colour terms don't count because English only has 11. Russian has 12. People also go through the thing and go, "Oh, your language only has four," and because these languages that have fewer tend to be Indigenous, then you get into weird colonialist, imperialist, even kind of eugenicist ways of thinking about language as an indicator of a culture's sophistication or people's sophistication. You get lots of weird applications of Berlin and Kay in the wider world. 

Megan: And way to go, right? They were trying to find universals, and they're like, "Oh, okay, so we're actually going to be really individualised about this and point to Indigenous languages and how they are, speak to how the people are barbaric or something, or they're uncivilised, because they don't have the word for buttercup yellow," or I don't know.

Carrie: [inaudible] like, "Why do we only have 11?" Because we have all these, like, begonia and sweet william. Well, sweet william is two words, but there's a bunch that don't count, but why not?

Kory: Right. So when you talk about basic colour terms, this is part of why the book took like 8 million years to write, because there are different types of colour terms, different types of colour names. So I talk about four in my book. There are people who might argue there's six. There are some people who argue there's only three. So the "basic colour terms" are the terms that, when you're describing any colour to any person, you default to as what lexicographers would call the genus term, like the main category. So begonia is a deep pink. Magenta is a kind of hot pink, or maybe it's a purple. Pumpkin is an orange. So the thing that you're like, "Okay, this colour is part of this bucket. That bucket is considered to be a basic colour term." Now, there's all sorts of arguments about what deserves to be a basic colour term. In English, we go to 11 because that's what Berlin and Kay kind of said. It's like, "Oh, we got 11," but people argue that olive is a basic colour term. I don't know. People argue grey should not be a basic colour term because it's not a separate colour; it's a value between black and white. People argue that orange shouldn't be a basic colour term because it's not spectrally pure, it's not its own separate wavelength. Orange is a blend of red and yellow. You have people arguing purple shouldn't be on there because it's a non-spectral colour. It doesn't appear on the visible spectrum. It's a trick that our brains do when we see red and blue stimuli together.

Carrie: Okay, that seems too sophisticated, though.

Kory: Right, isn't that weird? Already, the idea that there are standard basic colour terms is a mess because it just depends on who you're talking to. The basic colour terms, we say we have 11, mostly because Berlin and Kay kind of codified that, but that also mirrors a lot of research that had been done previously, a lot by the NBS or scientists that are trying to come up with simple ways to describe all colours using plain English. It's still in flux. Like I said, people get their knickers in a knot about orange and purple. But even then, there was a sense in the 30s and 40s and 50s that even if it made scientists itch, these are the terms that people use. These are the main buckets people use for colours. We're going to let that linguistic sloppiness or that scientific sloppiness be covered just by these are categories people understand clearly. That's the basic colour. Then we have what we call inherent colour names, which are colour names that come from usually something in nature or something in the built environment that we understand. Lilac, we immediately picture a very particular purple that is the colour of lilac flowers. Lime, we picture a colour that we think of as canonically the colour of limes. It's never the colour of limes. Lime is darker. Actual limes are a lot darker than the colour lime we picture. Salmon, oak, sweet William, begonia: these are names that we take from the environment around us to signify sort of subtler shifts in colour. And the thing is, like I just said, the inherent colour name is one that is sometimes widely divorced from the actual object itself. Lime is the best example. We think of lime as being this very bright yellow-green, and all limes are much darker. Both the flesh, the rind, the juice, none of it matches what people sort of gravitate towards when you say, "On this colour chart, point out lime to me."

So we have the inherent colour names, which are a little closer. It’s a little narrower bucket.
And then you have two different groups that really are kind of the marketing version. You have associative colour names, and then you have fanciful colour names. Associative names are names that are chosen not to describe the colour, but to give you a feeling. In the book, I talk about Josephine. I still don't know what colour Josephine is, but it evokes like it's French, it's feminine, it's very like romance-era Napoleonic. And so you're like, "Oh." I think it's like a dusty rose colour, but people will be like, "Oh, Josephine." And then you'll buy an outfit or a shirt or shoes in Josephine and spend $200, as opposed to the same shirt or outfit that's called dusty rose, which you're like, "Yeah, it's dusty. I am not spending $200 on something dusty." So you have the associative colour names, and then you have fanciful colour names, which are just like, "It's just a name to grab, and it has zero reference at all to the colour." I had a list of fanciful colour names that were used in high fashion six years ago, and one fashion house had five colours. They had white and black, which were white and black. They had brick, which was a bright yellow. 

Carrie: What?

Kory: They had canary, which was a dark red. 

Carrie: No,

Kory: And then they had a blue that was like number 14. It was literally like someone was like, "The more ridiculous we make these...," now it's an in-joke. And you'll pay to be part of an in-crowd. So yeah, colour names are complicated and weird.

Carrie: I found Josephine. 

Megan: Oh yeah. I see it. Look at the Josephine there. 

Carrie: I found Josephine, and it is pinky-grey, I guess.

Kory: Yeah. That looks like it. It's very weird. 

Carrie: I don't like it.

Kory: No, I don't like it either.

Carrie: There's this horror film, I think it was a made-for-TV film in the eighties, that I watched as a kid, and the doll puked that colour. 

Kory: Well, there you go.

Carrie: So, I don't know about this [inaudible] business.

Kory: See, this is a great example of how fraught colour naming is, because it's like, "Well, I've named this Josephine, but if that looks like eighties horror doll barf to me, I'm not buying it, no matter what you name it. I'm not putting that colour on my body, in my house, on my person, none of that."

Megan: Yeah. That's really funny. 

Carrie: Amazing.

Megan: And then I think about the inherent ones. Those can be really geographically available to people. I'll say Sedona red. 

Kory: Absolutely.

Megan: It's a very specific red that I understand, but other people might not understand if they haven't been to Sedona. 

Kory: Yeah, totally. This comes up when you're trying to define basic colour terms. Because one of the ways that, as a definer, if you're trying to tell someone what orange is, I'm not going to give you the spectral reflectance coordinates of orange. Nobody cares. But what I can do is I can say it's between red and yellow. But oftentimes, lexicographers will use analogy. So orange is the colour of carrots, or it is the colour of fire. But if you are in a culture where fire is primarily kerosene, then the colour of fire is blue, it's not orange. Sedona red is a great example. I hate when people are like, "Oh, the desert's so boring." The Southwest has just as many colours as your most verdant forest, and probably more, but they are regionally specific, right? Sedona red is a very particular colour. The colour of Ocotillo blossoms is a really particular colour. And so if I call something Ocotillo and it's blue, that's not what Ocotillo blossoms are. People in the Southwest will be like, "Well, you don't know what you're doing. I'm not buying that."

Megan: Totally.

Kory: Conversely, if I'm like, "Oh, I live near Fairmount Park, and this colour is Fairmount green," that doesn't mean anything to someone who does not live near Fairmount Park. Like, "Okay. It's a green." h.

Megan: There's been a Saguaro craze, and putting Portlandia, like, "Put a bird on it." It's been like, "Put a Saguaro on it." A lot of people that haven't even been here have been painting the blossoms like a prickly pear. I'm like, "No, that's not what the Saguaro does. That's what the prickly pear does."

Carrie: Yeah, they're very different kinds of blossoms.

Kory: Yeah. It's not the same cactus, y'all.

Megan: I'm like, "Yeah, you got the colour wrong. It's all messed up."

Kory: Yeah. There's a paint company that, a while ago, got into trouble because they called a white “Navajo White. And it was like, "Mmmm." And they were like, "Well, it's just meant to evoke this feeling of adobe structures." And it's like, "The Navajo don't live in adobe structures."

Carrie: No, they sure don't.

Kory: And also adobe's not white. I don't know. It's like, "Are you watching Bugs Bunny cartoons? What are you doing? What's happening?"

Carrie: Yeah, probably. 

Megan: Oh my gosh. Looney Tunes, older cartoons are pretty, there's not as much messing with saturation and stuff. I feel like they're very vibrant, the colours they choose for older cartoons, or even new cartoons. I'm thinking like Bob's Burgers. [inaudible].

Kory: Right. Yeah, some of that is the medium. Some of it is it’s a cartoon. So people expect bright colours. Some of it is also the technology through which that is communicated.  Cartoons are broadcast. It's illuminated. So, darker colours, when you're using light, there's not as much range as there is when you're printing them. When things get dark, all the colours flatten. So the brighter the colour, the more readable it is to people. Listen. So colour is so complicated, not just because of how we think about it and name it, but because the way that we interact with it changes our perception of it constantly. This is why Berlin and Kay's study fascinates me in some ways, that the idea of a linguistic universal also assumes universal ways of perception, universal ways of communication, universal ways of processing that information. We go back to the Myaamia word that means blue and green. People are like, "Does that mean that our ancestors couldn't see green or blue?" And it's like, "Well, probably not. No." But also, I can't explain to you why our language just has one colour term for both blue and green. Some blues and greens. Not all blues and greens. I can't explain why English only has one blue and doesn't have at least two blues, and only has one green and doesn't have at least two greens, like a lot of other Western languages.

Megan: Yeah. I'm thinking about even a turquoise. So that might evoke a certain turquoise colour in your mind. My dad grew up in Bisbee, where there's actually a turquoise called Bisbee Blue, like, there's shades of turquoise that are from the Bisbee mine, the copper mine. Then, Kingman has a different shade of turquoise, [inaudible].

Kory: And there are turquoises, technically turquoises, that are not blues or greens at all. They can be pink, or they can be white. There are sapphires that are not blue. There are sapphires that are pink. So, even all of the inherent colour terms rely on an assumed standard that is not necessarily correct. And so colour naming, colour perception, and colour localisation are just a mess from the get-go. There's just no solving it from the beginning. 

Megan: And pop culture affects all this, too. I'm thinking there's a Simpsons yellow, right?

Kory: Oh, totally, absolutely. This is where Pantone and colour forecasting and colour marketing really come in. It's interesting. Someone recently said to me, “Hi, I read the book. I was kind of surprised that Pantone wasn't in it more,” because it's about colour. And when I think of colour, I think of Pantone. That's because Pantone has done a very good job marketing the Pantone deck as the standard for graphic designers. They have helped different companies develop their own corporate colours that are Pantone-specific colours. So Coca-Cola is a particular Pantone red. John Deere uses particular Pantone colours. Merriam-Webster's cover is itself that red is trademarked, and it is a Pantone. So, the way that we even think of how colour gets associated with things is this weird mix of happenstance, intentional marketing, brand identity, pop culture, sort of meme in the not-cartoon sense, but in the spreading-idea sense. It's so complicated, and it's never standard. It's not like, "Well, first a colour gets used here, and then it moves from this viewing community to this viewing community. And then that name spreads from here to here." It's not consistent for any colour. 

Megan: Do you know what they call the Merriam Red? Do they call it Merriam Red?

Kory: I don't. They do call it Merriam Red, yeah.

Megan: Yeah, because I've seen this, where brands actually see little names dependent on what kind of agency it is, or what services they provide. So, in my job, I saw some guidelines that was a programme that helped young children with disabilities. They had a Caring Coral. 

Kory: Oh, that's so interesting. 

Carrie: Yeah. I don't know why, but that kind of hurts me a little, and I can't explain why.

Megan: What is it? What is it, Carrie? 

Carrie: I don't know. It just feels so pandering or something. 

Kory: Right. How much of your marketing budget went towards naming that colour, as opposed to providing services for children with disabilities?

Carrie: Maybe that's it, but I don't know. It's coral. That's great, but what does that have to do with being caring? I don't know. I object.

Megan: It's about branding. It's Pantone. You're right. It's the branding, man. People take it seriously.

Kory: Yeah, and we get it early on, too. Like, Crayola is sort of our first encounter with these kinds of both associative and inherent colour names. So, like, the Crayola crayon cornflower, everyone knows cornflower is blue, but they don't know what cornflowers are. It's just a blue crayon. 

Megan: I think I knew forest green. I was like, "Yeah, of course that's forest green." My Crayola, that's exactly [inaudible].

Kory: Right. So even the way that we learn about colours early on is sort of affected by all these weird market forces, because Crayola colour names also aren't stable. They change all the time. Like, I just did a craft project using a bunch of Crayola crayons, and there's a crayon called macaroni and cheese. And I'm like, "Why is this?" I didn't grow up with macaroni and cheese. How very dare? How dare you introduce the name colour? But, of course they're going to do that. They're going to do all sorts of stuff like that.

Megan: Was it like the craft colour of macaroni and cheese?

Kory: Oh, yeah, of course. It's not like homemade mac and cheese. It's like [crosstalk].

Megan: Exactly.

Kory: I've got the crayon box here. I can show you what colour mac and cheese is. Is it this one? 

Megan: I'm imagining it. It's in my mind's eye.

Kory: It's this colour. Sorry, it's hard to see.

Megan: It's like the cheese packet colour. 

Kory: It is the cheese packet colour. Yeah, it is.

Megan: Yeah, I guess. That's so funny. My homemade mac and cheese is a very light colour. 

Kory: Yeah. It's not orange where we are. 

Carrie: Even when I would use orange cheddar cheese it was still not that colour of orange. 

Kory: Yeah, it's never that. That's an engineered powdered orange, specific engineered orange. 

Megan: Oh my gosh, I could talk to you too about synthetic dyes and stuff all day. There's so much in this book.

Carrie: Yeah, I know. Not even just synthetic, but also the natural ones too. There's so much to talk about. Maybe we can ask you: what is your favourite colour and/or colour term?

Kory: Oh, my favourite? So my favourite colours shift dramatically because they do. I think my favourite colour term at this point is, let me see if I can find it, because I did write about it. Yes, it is teal duck, which is a greenish blue. It's a very dark greenish blue, and I love it because it is two words smashed together. One of which is a colour term that people have very clear views of what it is. The other, not a colour term at all. Makes no sense. Why are we talking about teal duck? And teal duck, the colour is not anywhere near teal. Teal, you think of a bright blue-green? And teal duck is like a dark, very dark colour that's like a bluish green. So it's like, if teal's up here, it's way down here.

Carrie: Is it the green of the green-wing teal? Is that the colour of teal duck?

Kory: It's a little bit bluer.

Carrie: Oh, okay. 

Kory: But this is the other great thing, we think of teal as being a very particular light green-blue, and it's named after a bird. We're looking at the bird right now. The part that you would call teal, I would actually call green. I would say that's a green. Like a forest green. Not even a blue-green.

Carrie: Okay. Interesting.

Kory: It's just a weird colour name, teal duck. The phonics of it are a little like, "Oh, teal duck, uh." But I love it. 

Megan: I love that I squished it together. I love that it's one word. 

Kory: Yeah, it's great. Perfect. 

Carrie: Megan, do you have a favourite colour or colour term?

Megan: My favourite colour term is very specific to me. I really love Bisbee Blue as a colour term. And then my favourite colour is orange, but a very specific set of orange that I feel like I'm just asking for it, "My favourite colour is orange," but then people give me something, and it's the wrong orange. I think it's more like a fall orange, in that area, not the orange of actual, like, kids' jack-o’-lantern trick-or-treating bags. Not that orange, but more fall, or more nature fall.

Kory: Like a soft autumn.

Megan: Yes. Exactly. 

Kory: Yeah. I like that. What about you, Carrie?

Megan: What about you, Carrie?

Carrie: My favourite colour shifts all the time too, but I think right now my favourite colour term is periwinkle, because it's so cute.

Kory: Yeah, it is very cute. It has winkle in it.

Megan: It is, yeah. It's whimsical.

Carrie: Yes, whimsical. 

Kory:  It's very whimsical. God knows we need some whimsy right about now. 

Megan: I know.

Carrie: Right. Stave off the darkness. 

Kory: That's right.

Megan: That's actually a really good plug for the book, too, Kory. It's not like we can forget about the world burning, but my goodness, what a nice treat to relax with.

Carrie: Yes. I know, and the preface is so fun. You just [inaudible] in for a ride.

Megan: Yeah.

Kory: And it was a ride, believe me. 

Megan: Yeah, I understand completely how it took forever, just like a book that relies so much on history, you got to pin it down. You got to beat us.

Carrie: Yeah. You got to nail that jelly to the wall.

Kory: Yeah, for sure.

Megan: You got to nail that jelly to the wall. So cool.

Carrie: Is there anything else you would like to let our listeners know about your book or anything else?

Megan: The book is available anywhere that you can buy books, but buy it from an independent bookstore because they're the best place to get books from. I will be popping around the country randomly through the next three to four months. Not necessarily reading from the book, because it's a hard book to find a little passage to read from, but I will be doing weird slideshows to talk about: how do we define colour? Why do we need to define colour? And how did this group of people think they landed on a solution that worked?

Megan: Yeah, you need visuals.

Kory: And it doesn't. Yeah. Ridiculous slideshow to be determined. But yes, people can pop by my website, which is korystamper.com, to see if I will be yakking at you in a location near you in the coming months.

Carrie: That sounds awesome. All right.

Megan: Perfect [inaudible]. I love it.

Carrie: All right. Well, we always deliver listeners with one final message: don't be an arsehole. 

Kory: For sure.

Carrie: Thank you. 

Megan: Thank you.

Kory: Thank you. 

Carrie: The Vocal Fries podcast is produced by me, Carrie Gillon; theme music by Nick Granum. You can find us on Tumblr, Twitter, Facebook, and Instagram at @vocalfriespod. You can email us at vocalfriespod@gmail.com, and our website is vocalfriespod.com.

[END]