Rachel Feltman: If you stubbed your toe right now, what sound would you make? According to linguists, the chances are pretty good that your answer would be surprisingly similar to the one given by someone on the other side of the planet—even if you speak totally different languages.
For Scientific American, I’m Rachel Feltman. I’m here today with our friend Allison Parshall, an associate news editor at Scientific American who often covers biology, health, technology and physics. She recently covered a study all about the words humans use to express pain—and the wild similarities between those exclamations.
Allison, thanks for coming on to chat. Hopefully it won’t be too painful.
On supporting science journalism
If you’re enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Allison Parshall: [Laughs] Ouch. That’s all I have to say.
Feltman: So why are linguists talking about pain?
Parshall: Well, linguists are talking about everything always.
Feltman: Talking about talking.
Parshall: But linguists are talking about pain because the words that we use to express it might actually tell us something about our shared biology and the evolution of kind of language in general. It’s a pretty big topic; it’s part of why I’m so interested to talk about it today. But this particular discovery actually started out pretty small, with a French language researcher named Maïa Ponsonnet and what she calls a pretty, quote, “naive observation.”
She was—does a lot of research on language and emotion, and she was in Australia studying Aboriginal languages when she noticed that the words many of those languages used to exclaim their pain, like “ouch” in English, sound similar to exclamations of pain in her native French. So that word in many Aboriginal languages sounds something like yakayi, and in French it’s aïe, and in English it’s “ow,” and—do you know any in any languages?
Feltman: Yeah, I used to study Mandarin, and I remember that there’s an ai in there: ai— ai-yo. And I’m terrible with my tones, so don’t [laughs] …
Parshall: I think it’s first tone for both.
Feltman: Yeah, it’s first tone for both, which I, I remember because it feels so appropriate for an exclamation of pain [laughs].
Parshall: You’re, like, high-pitched yelling, basically.
Feltman: Yeah, exactly. I have studied and failed to master many languages [laughs] in my life, including Italian and Spanish, and I think in both “ai” is just a very common exclamation of pain.
Parshall: Yeah, there’s a lot of “ai.” Basically wherever you look around the world you’ll get a lot of “ai,” you’ll get a lot of “ow,” you’ll get a lot of “ah.” In Japanese it’s itai; I don’t actually speak Japanese at all, so hopefully that’s not totally wrong. But—so you might be seeing a pattern here, and basically Maïa Ponsonnet, when she was doing her research, noticed this pattern, she made this observation, but, like, two languages, what can you really conclude from that? Nothing.
So they decided to do a much broader-scale study. There’s not a lot of research looking into this type of word—it’s called an interjection: so, like, “ouch” or “wow” or “yay.” And basically they surveyed 131 languages from dozens of language families around the world. They gathered these exclamation words for pain, as well as for joy and disgust: so, like, “ouch,” “yay” and “yuck,” respectively.
And then they found that when it comes to pain, her observation was not a superficial similarity at all. These pain interjections—but not joy and disgust interjections; we’ll get to that—are more similar to each other across languages than they are to other words in their language that aren’t pain interjections. So “ouch” and “ai” are more similar to each other than “ouch” is to other words in English and “ai” is to other words in—pick a language that “ai” is in.
Feltman: Italian, Spanish [laughs].
Parshall: Italian. Yeah, Spanish. So basically it showed that it seems likely that this isn’t a coincidence and that there’s some sort of hidden factor that we humans all have in common that has shaped these words similarly.
Feltman: Wow, that’s really cool, and I also love that it started with, you know, what the researcher calls a “naive observation” because it is the kind of thing that was right in front of us all along, and you have to wonder how many people were like, “That’s a silly comparison I just made. There’s obviously no real connection there.” So really cool that she explored it.
So when we say that these are similar sounds, you’ve talked about this a little bit, but exactly what sounds are we discussing?
Parshall: We’re basically talking about this “ah” sound [written as [a] in the International Phonetic Alphabet]. In “ai” and in “ow” you can hear that both of those start the same; they both start with [a]. They’re what we call diphthongs, so it basically is two vowels right next to each other. So either you could have [a], or you could have “ah-ee,” or you could have “ah-oh.” And so in English “ouch” kind of has that [a] sound. Even though we spell it with an o and we think of [a] typically being an a, but phonetically they’re very related.
So it’s possible that [a] is also just a really common sound in language and maybe that was why it was popping up in these pain interjections. But the researchers compared these pain interjections to other words of the same length in each language and found that, no, these [a] sounds actually are just more common in pain interjections than they are in other words across the board, basically.
So I guess my question is, like, if you were the scientist and you’d seen that data—you’re seeing these [a] sounds popping up all over in pain interjections—do you have any ideas of what you might think up as a hypothesis to explain it?
Feltman: Yeah, that is a great question. I feel like I would be tempted to hypothesize that there’s something innate about the experience of pain that relates to these sounds, like maybe something about our physiological reaction to pain. I don’t know. I’m not a linguist, so [laughs].
Parshall: [Laughs] Well, if I went over and pinched you, what sound might you make—just, like, involuntarily?
Feltman: Probably, like, “ah!”
Parshall: Yeah: ah, ah! So the researchers basically had the same idea. They thought, like, “Well, maybe these noises that we make involuntarily, just kind of these primal sounds, maybe for pain it does kind of sound more like an [a] sound across the board.”
Of course we don’t even necessarily know that that’s true. These, like, more primal vocalizations that aren’t language but still communicate something are not super widely studied. So what the researchers did is gathered samples of these nonword vocalizations from five languages and cultures around the world for pain, joy and disgust—so I guess you could imagine [a] for pain, “ee” [written as [i] in the International Phonetic Alphabet] and “uh” [written as [ə] in the International Phonetic Alphabet] for disgust would maybe be what I would do if I was prompted.
And they found that for disgust the most common vowel sounds in these—so kind of like I just said: [ə]—that …
Feltman: Yeah.
Parshall: You can hear it. For joy it was [i], so kind of like what I just said. And for pain it was that familiar [a] sound. So this held true across five different cultures and language groups—which does seem to support their hypothesis. It suggests that it’s possible that these [a]-based pain words from across the world are in some way similar because of a common origin in these, like, more primal noises that we make.
The study can’t really prove that; it’s a very difficult thing to prove. But it suggests that we might say “ouch” or aïe or yakayi because at one point all of our ancestors were just screaming “ahh!”
Feltman: So you mentioned joy and disgust. Are we seeing these kind of phonetic similarities in the linguistic expression of other emotions?
Parshall: Not in the linguistic expression necessarily. So interestingly the study of vocalizations found, for joy and disgust across those five different languages, you had similarities, like joy was [i] and disgust was [ə], but that did not translate to, like, the linguistic words that we use.
So basically, like, if disgust vocalizations were [ə], in English, for a disgust interjection, in the linguistic world, you might say “yuck” …
Feltman: Right, or “ew.”
Parshall: Exactly. And, like, you can see, there’s more cultural layers being added here. There’s a bunch of things you could say to convey your disgust in terms of interjections. And the theory that the researchers had for why joy and disgust don’t seem to have this similarity in the words we use to describe them: so as those layers differ between cultures, the words we use to express them are going to evolve to sound very different, even if they all kind of did start from the same place as that primal sound of [ə].
But pain is a feeling; it’s not really an emotion. Like, I feel like it’s kind of misclassified to say pain is an emotion. It’s, like, more of a physiological response. You could argue that disgust is, too. For sure, there’s some sort of, like, physiological thing we have like the gag reflex. But pain is subjective and complicated, and it does differ across cultures, but our bodies do have, like, literal pain receptors that either fire or they don’t. So it’s possible that might be why it’s the only response they saw here that has similar vocalizations and interjections across languages.
Feltman: So what can that tell us about language?
Parshall: It’s all very hypothetical, but it’s all so interesting to me, and that’s what really gets me about this, is the idea that there are these more, like, primal, mysterious, unseen forces that are shaping the words that we humans develop to describe the world, regardless of where you’re from on the planet, this is pretty new because for a long time there was this really dominant idea in the field of linguistics that language is fundamentally arbitrary.
So what I mean by arbitrary is that, like, the sounds kind of don’t matter; we just kind of tie them to meaning and make them have meaning but that the sounds themselves don’t have any meaning. Like, when I say “bird” to you and you think of a bird that’s just because you’ve learned to associate [the sound of the word “bird”] with a bird but not because there’s anything birdlike about those sounds I just made.
Feltman: In fact, one of my [laughs], my cousin-in-law’s baby, who is German and learning English as a second language, uses the word “bird” to talk about any animal that excites her and, by extension, really anything that excites her [laughs].
Parshall: [Laughs] Really anything. I feel like that, that’s quite common with kids, too.
Feltman: [Laughs]
Parshall: And so the, the idea was that the sounds themselves are arbitrary, but we know that that’s not right because language is full of nonarbitrariness. Like, for example, there’s onomatopoeia. Can you think of any of your favorite onomatopoeia, which are words …
Feltman: “Splat.”
Parshall: “Splat.”
Feltman: Yeah, that’s a great one. “Crunch.” “Crackle.”
Parshall: Mm, those are good ones. I like “boom.”
Feltman: Oh, “boom” is great.
Parshall: “Swish” is fun because it really invokes the …
Feltman: Yeah.
Parshall: Like, “shoo,” you know?
Feltman: Yeah—“hiss.”
Parshall: But so these are words that basically reflect how—we’re basically trying to use a word to just—to replicate a sound.
Feltman: Right.
Parshall: And that can do some interesting things, like putting a b in front of things that we think of as more, like, explosive, like “boom” or “blast” or “bang.” I mean, there is something, “buh” [written as [b] in the International Phonetic Alphabet], that is …
Feltman: Yeah.
Parshall: Literally a plosive consonant is what linguists call it; there’s air exploding from your mouth.
So there’s all of these really subtle associations we can make. And I did want to ask: Have you heard of the “bouba-kiki” effect?
Feltman: I have—you know I have [laughs].
Parshall: [Laughs] Can you recount the bouba-kiki effect for me?
Feltman: Yes, the bouba-kiki effect is the fact that—at least among English speakers and I think some other languages, too—people associate “bouba,” but also, like, b sounds in general, with sort of soft, rounded squishy things and “kiki” and similar sounds to spiky shapes, and there’s, like, the “bouba” and the “kiki” shape that are used in this study, but it’s also sort of more broadly: it’s, like, there are spiky words, and then [laughs] there are squishy words.
Parshall: Yeah, I mean, I, I love your response, too, because I get to say, like, it’s definitely not just English. The cool thing about it is that it’s basically, like, languages and cultures all around the world. It’s been replicated pretty well.
If you give someone two shapes, one is bulbous and one is spiky, and you ask which one is “bouba” and “kiki,” they will tend to put “bouba” with the rounded one and “kiki” with the spiky one. But I did just kind of, like, tip my hand there a second because I called them bulbous and spiky …
Feltman: Right, yeah.
Parshall: Which you can hear—bulba, “bouba”; spiky, “kiki”—like, you can hear it in there.
And this finding has been replicated a lot, and there’s other ones, too. So really recently researchers found an even stronger association: that the trilled r sound, so the [r] sound [as it’s written in the International Phonetic Alphabet], is associated with roughness and the “luh” sound [written as [l] in the International Phonetic Alphabet], or what we would say as “L” in English, is associated with smoothness.
And that’s across the world; that’s even in languages that don’t use these phonemes in their language or that group r and l sounds together as the same phoneme, like Japanese. And I just think that’s so cool, especially because, of course, we’re going to hear the sound [r] and think instinctively to make a metaphor to texture, right? Like, it sounds rough and ragged and ridged—I’m not even really cherry-picking here. Like, I tried to look up synonyms for “rough” …
Feltman: Yeah.
Parshall: That started with l, for example, and I couldn’t find any. This is all cherry-picked, but, like, still.
But l, you can think of, like, lilting, flowing, gliding, silky. There’s all of these …
Feltman: Mm.
Parshall: We, we have these associations, even if they’re kind of flying under our radar perceptually, and that’s what the linguists are all trying to find. And I think they’re so cool; it just—there’s so many of these little associations we don’t even know to make.
For example, like, even within specific languages there’s things called phonesthemes—I hope I’m pronouncing that right. These don’t generalize across languages generally, but, like, in English, for example, the word—the phonestheme “gl-” is often associated with light, so, like, a glow or glimmer. And I just think that’s so interesting. We don’t tend to notice these, but by doing a really—look and—across one language or across all languages, as many languages as you can pull together, you can kind of start to see these patterns that go under the radar, and it raises the questions of why it’s like that.
Feltman: Yeah, totally. Well, speaking of questions, what are the next big ones to answer here?
Parshall: Yeah, I mean, the big question is basically, like, what else is out there? The researchers, to follow up on this study specifically, they are curious if maybe they would see different results if they looked at consonants and not just vowels.
So in some languages, words that evoke or respond to disgust use this consonant that we don’t tend to have a lot in English but that goes like “chuh”—I hope I’m pronouncing that right; it’s represented by an [x] in the International Phonetic Alphabet. Or maybe there’s, like, plosive consonants like b or p—like “buh,” “puh,” “tuh” [written as [b], [p], [t], respectively, in the International Phonetic Alphabet]—maybe those are more common in disgust. I’m just throwing things out there; there’s no necessary reason to think that it’s like that. But, like, what are the other hidden similarities in the ways that we talk?
And Ponsonnet and her colleagues are actually using AI now to try to uncover more of those similarities, which is kind of the thing that AI was always really good at, right? Just ingesting a ton of data and telling researchers where there might be a signal and connections that, like, our human eyes can miss. So I definitely am excited to stay tuned and have my mind blown by future little hidden symbolisms.
Feltman: Yeah, absolutely.
I saw that you recently put together a roundup of, like, this and a few other studies on language from 2024. Are there any other favorites you wanna share from those highlights?
Parshall: This is kind of similar to this finding—at least, I was thinking about this a lot when writing the story—but we had an article about this study that aims to understand why some words in a language survive and other ones go extinct. Kind of, like, what are the rules for survival of the fittest for language? Because we know that our vocabularies expand and contract, and, like, some words go out of existence and some new ones get born.
So in this study the researchers basically had people play this big game of “telephone.” They had a story that they were told—the participants were told—and then the participants had to remember and rewrite that story for other participants. And they found that more concrete words lasted longer than more abstract ones; more emotionally exciting words, whether they were positive or negative, lasted longer; and also words that were acquired at a younger age.
So this paring down doesn’t necessarily reduce language to baby talk over time, but you could think of it as, like, making language more efficient, which I thought was a kind of interesting dynamic. And then, of course, we pull in new, complicated words as we go to suit our needs.
Feltman: Sure, “brain rot.”
Parshall: “Brain rot,” exactly [laughs], which I hope—I don’t hope is dead; I just think that now, once Oxford English Dictionary gets ahold of it, it should be [laughs] …
Feltman: [Laughs] Yeah.
Parshall: I think we might be done saying it.
And this one isn’t really linguistics, but researchers have come a really long way in brain-computer interfaces, which—it might be a little overly simplistic to call them mind-reading devices, but they’re basically mind-reading devices …
Feltman: Right, they’re literally mind reading, not necessarily what people think of when they say “mind reading” [laughs].
Parshall: Yeah, they’re not—but, like, they’re able to look at the brain with—like, take measurements from the brain using electrodes and then translate that into coherent speech. And this has been a slow process of it getting slowly better and better and better, but it seems like this year we hit a really important milestone, which is: there was a man with ALS who was part of a clinical trial, and he regained the ability to talk to his family. Researchers now say that the technology is at the level of quality where it can actually be useful to patients …
Feltman: Mm-hmm.
Parshall: Like, it’s correct enough of the time that it’s the kind of thing people can really start using …
Feltman: Wow.
Parshall: Which is pretty incredible.
Feltman: Yeah.
Parshall: And then also, one of the more fascinating brain language stories I got to read this past year was from SciAm’s very own copy editor Emily Makowski, and she wrote about her experience with something called ticker-tape synesthesia, which I had never heard of but is fascinating. Basically whenever she hears or thinks words, she basically gets, like, mental subtitles …
Feltman: Mm.
Parshall: That kind of scroll in front of her eyes, and it’s not something she can turn off. It’s, at times, I think she describes it as at times helpful, at times annoying, and I—her story of how she came to discover that that was the case for her fairly recently, I thought, was very funny and interesting. So I would really recommend reading her story about it.
Feltman: Wow, yeah, I somehow missed that one, so I will definitely have to check it out. And it seems like there were a lot of cool linguistics findings in 2024, and this “ouch” study is right up there, so thanks so much for coming to talk to us.
Parshall: Thanks for having me.
Feltman: That’s all for today’s episode. We’ll be back on Monday with our usual science news roundup.
Science Quickly is produced by me, Rachel Feltman, along with Fonda Mwangi, Kelso Harper, Madison Goldberg and Jeff DelViscio. Today’s episode was reported and co-hosted by Allison Parshall. Emily Makowski, Shayna Posses and Aaron Shattuck fact-check our show. Our theme music was composed by Dominic Smith. Subscribe to Scientific American for more up-to-date and in-depth science news.
For Scientific American, this is Rachel Feltman. Have a great weekend!
