1163. This week, we look at what it’s like to be a "language detective" with Sali Tagliamonte and how she used her own teenagers as a research lab. We look at a 25-year study on how the phrase "be like" became a permanent fixture of English, why the word "very" is suddenly making a comeback with younger generations, and what happens to our language when we spend all day talking to AI.
1163. This week, we look at what it’s like to be a "language detective" with Sali Tagliamonte and how she used her own teenagers as a research lab. We look at a 25-year study on how the phrase "be like" became a permanent fixture of English, why the word "very" is suddenly making a comeback with younger generations, and what happens to our language when we spend all day talking to AI.
Sali Tagliamonte, University of Toronto
🔗 Join the Grammar Girl Patreon.
Thank you to the members of the Order of the Aardvark at Patreon:
🔗 Share your familect recording in Speakpipe or by leaving a voicemail at 833-214-GIRL (833-214-4475)
🔗 Watch my LinkedIn Learning writing courses.
🔗 Subscribe to the newsletter.
🔗 Take our advertising survey.
🔗 Get the edited transcript.
🔗 Get Grammar Girl books.
| HOST: Mignon Fogarty
| Grammar Girl is part of the Quick and Dirty Tips podcast network.
| Theme music by Catherine Rannus.
| Grammar Girl Social Media: YouTube. TikTok. Facebook. Threads. Instagram. LinkedIn. Mastodon. Bluesky.
[Computer-generated transcript]
Mignon Fogarty: Grammar Girl here. I'm Mignon Fogarty, and today I'm here with Sali Tagliamonte from the linguistics department at the University of Toronto, a self-described "language detective." You are just going to love hearing about her work today. Sali, welcome to the Grammar Girl podcast.
Sali Tagliamonte: Thanks! It's great to be here.
Mignon Fogarty: So happy to have you. So, we have so much to talk about, but I want to start with your work on teens and teen conversation and the words they use. First, I am so curious how you even do this research because I keep imagining the Steve Buscemi meme, like, "Hello, fellow kids." So, how do you get past that?
Sali Tagliamonte: That's not how you do it. That's not how you do it. The simple and very straightforward answer is, well, first of all, I'm a language scientist. So language antennae are always, always working. But the great thing about being a language scientist is when you have five children.
So as my kids were becoming preadolescents / adolescents, I found myself in a whole new world of interesting new linguistic phenomena that was right at my breakfast table. I had previous to that, I had done a lot of work on rural dialects and older people, because that's where interesting linguistic phenomena live that are dying, and we want to catch them before they're gone. But when my kids became teenagers, it was like, "Whoa, something cool is going on here."
And so, what you do in order to study that is: you do nothing. You do not say to your children anything that would let them think that you are listening to them. Right? You just say nothing. And so, for a good long time, I said absolutely nothing. And I would volunteer to take kids to soccer practices and music lessons and anything. And I put all the kids in the back—all their friends, anybody who needed a ride. And then I would listen.
And I discovered a lot of interesting phenomena, which I then took to my research lab and said, "Okay, let's go interview adolescents." And that's how I got started with the research that eventually became my book, Teen Talk. I mean, my kids will not let me analyze their language—that would be a little bit too invasive. But I knew what to look for. And I also knew I could go back to my kids and get confirmation of what I was noticing and what I was thinking about, what I thought was going on. And sometimes they said, "Oh, Mom, you're off, you're way, way off," and other times they would go, "Really? That is so cool." And in some of my research papers, I even cite my kids! Some of them have willingly provided me with interesting examples. So, it's a question of using your linguistic antennae to see what's available and to find interesting things right before your very eyes—or ears, as the case may be.
Mignon Fogarty: What were some of the most interesting things that your kids said that was maybe validated by your research, or that you really followed up on in your research?
Sali Tagliamonte: Okay, I have two stories about that. The first story was way back in the early days of instant messaging. And what I heard was at the keyboard, and I thought, "when did my kids learn to type that fast?" I mean, there was more to it; there was more going on there than just, "I'm typing my history paper." It was like rapid-fire stuff, and it would stop and then rapid again. And I was like, "Okay." So, I was the proverbial parent over the shoulder.
And of course, what immediately happened was like, "Get lost, Mom! I'm talking to my friend." So I was like, "Okay, now what's going on there, right?" And so that's when I did the big study on MSN communication. But that is a very dated research project because you can't study MSN anymore because it doesn't exist. Nobody uses it. And the other thing is all the devices have changed. So, can’t really do what I did then anymore. Yeah
Mignon Fogarty: Right, now there's no "tack-tack-tack" on the keyboard. They're doing their thumbs on the phone.
Sali Tagliamonte: Yeah. And it's a different venue, and you can't ever really tell whether someone's on a keyboard or a smartphone or what they're doing. They could even be using audio, talking into their phone. So, the study I did then was sadly of its time, but it was great fun because I found out a lot about what kids were doing on instant messaging.
Of course, the uproar at the time was that they were just bastardizing the language, and they were screwing it up and all this stuff. Well, they were using a lot of funky ways of spelling things and using the keyboard to emphasize things with capitalization and reduplication and all kinds of— whatever they could do to embellish the language—because that's what creative kids do. But they really weren't changing the grammar. They can't because they've got to be able to communicate their message.
So it was kind of an interesting story. So the paper that came out was entitled, "LOL.” You know, this is not happening. LOL for real? No, there's no change in grammar.
Mignon Fogarty: I mean, we still need ways to show laughing online. I think your findings are still completely applicable because you found that texting isn't ruining kids' ability to use language. I thought the part about formal versus informal places to use was very interesting.
Sali Tagliamonte: Yeah, it really is. And you know, social media picked up on that—even television. You'd see this kid, and they have the kid typing all this wacky stuff, and even you have the kids like correcting their own grammar by putting in the informal thing that they think is cool, right? So I'm like, "oooh!" I think they really know the grammar. They're just playing around. Because you can't play around with grammar unless you know what the grammar is, right? So it was fun!
Mignon Fogarty: And they understand. They understand where they need to use formal language and where they don't.
Sali Tagliamonte: Exactly!
Mignon Fogarty: Were there any other big myths that you feel like you dispelled in your research?
Sali Tagliamonte: Well, that big myth that the young people are destroying the language. I mean I cannot tell you how many times someone says to me, "Oh, I just hate it when..." And I'm talking about lots of educated people who will say that to me. "Oh, I just hate it when young people say, you know, ‘like’...
Mignon Fogarty: Mm-hmm.
Sali Tagliamonte: "...It's so stupid. You know, I just like hate it." Right. And I think, "Well, you just used it, man!"
Mignon Fogarty: Yeah. Let's talk about "like," because that's another thing that you've studied, and I do think it's fascinating.
Sali Tagliamonte: It is. I mean, when we first started studying it, it was a huge project. So, I said to my PhD student at the time, Alex D’Arcy, who's written a considerable, unlike herself, I said, "Well, I can't do this. It's too big, but it's a great dissertation project." And so she took it on, and she did a great job. And her book, 800 Years of ‘Like,’ pretty much encapsulates what she found, which is that whether we like it or not, "like" is here to stay. It'll be interesting to see what happens two or three generations from now because everybody's using it. And the only people who don't use it are old people. So, let's see what happens.
Mignon Fogarty: Yeah, it's not really kids anymore.
Sali Tagliamonte: No, it's not kids anymore! Because the people that started bringing it into the grammar, closer and closer to the core of syntax, have done it incrementally.
Mignon Fogarty: One thing I find interesting about "like"—and maybe there's a name for this; I'm not sure—because when someone says, "He was like, 'I'm angry,' and she was like, 'No way.'" People say that's a substitution for "said," but it's really not because I feel like you're sort of implying they said something like "I'm angry" and something like “okay.” It's not the same as "said." Is there a name for using "like" that way, that sort of leads into a quotation, but the implication is that it's not exactly what they said, like a normal quotation would be?
Sali Tagliamonte: Well, that's the use of "like" as a quotative frame.
Mignon Fogarty: Okay
Sali Tagliamonte: And there, too, “like” has taken over so. We've been following a single woman for the past 25 years—it's called the Clara Felipe Project (that's a pseudonym). She was one of the highest users of quotative "be like" when she was a teenager. She's gone on to higher-level education, she's gotten an advanced post-secondary degree, she works in a very established and important profession, and her rate of "be like" as a quotative has not changed since she was 16. So, her rate of "be like" at 16 is the same as her rate of "be like" now as she's entering her 40s. That use of "like"—whether you think it replaces "say" or not—it's the dominant form today. So, it has to be doing more than just replacing "say," right?
Mignon Fogarty: Right.
Sali Tagliamonte: Because it used to be "say" and "think" or something was...and also zero. You could usually say, sometimes say "go" if you wanted to, or like other things.
Mignon Fogarty: Yeah, like, "And then he went, 'I'm angry,' and she went, 'No way.'" Yeah, it's the same.
Sali Tagliamonte: Right. So "be like" has taken over. Now here's the question: what's next? You know, now that "be like" is pervasively by just about everyone, even my specialist doctor or whatever—they're using "be like" as a quotative. So, it's not like it's a "dumb person's thing" as everyone once thought. It's just the way the new generations frame quotatives. That's just the way they do it.
So now, what are the next generations going to do? Are they going to keep using “like,” or are they going to come up with a new one? Who knows?
Mignon Fogarty: Yeah.
Sali Tagliamonte: I keep watching, but so far it's just "be like," which of course drives all the older people crazy because they hear "like, like, like, like, like, like, like." And they think it's some kind of idiotic phrase. But it's not; it's just the way young people talk.
Mignon Fogarty: That's a great lead-in to another piece of work you did on interesting age differences with the intensifiers.
Sali Tagliamonte: Oh yeah!
Mignon Fogarty: So can you talk me through that?
Sali Tagliamonte: Well, intensifiers are so fun because the whole idea about intensifiers is that they are supposed to change rapidly. So, it's a great way to tap into linguistic change in progress and to focus in on generational differences. Because one of the most pervasive comments you get about language is, "Oh, my kids don't sound right; they're terrible." And then the old people say, "Oh, the kids are terrible," and the young people say, "Oh, those old people, they sound awful." So there's real commentary; nobody's happy with the opposing sides of the age continuum, right?
So if you look at intensifiers, this is great because you find all kinds of things that point to differences. Cause no one can tell you, if I said to you or talked to people on the street, “Oh how can you tell if someone's old if you didn’t see them?” Nobody can really tell you. But if you say, "Hey, old people use intensifiers that are old." They say things like, "Oh, that's very nice," or "It's a very nice day." or "I'm very happy today."
Mignon Fogarty: Right. We're talking about words like "so," and "very," and "really" that make something more intense.
Sali Tagliamonte: Exactly. If you want to make something intense, like "It's cold today" or "It's hot today," well, how about if it's "really" cold or "super" hot? You know, something like that. Well, young people will typically not say "very." At least, there's a bit of a new story about that.
But the whole idea when I first did the intensifier study was that it was supposed to recycle quickly, and that the standard intensifier was "very." And I first did this by looking at the Friends television show, which was so much fun! I was so sick of watching Barney the Dinosaur with my kids, you know? And then they started liking Friends, so, hey, yeah! And then I noticed that there was something going on. The guys were using one type of intensifier and the girls were using another.
The guys were saying "really, really pretty," and then the girls were saying "so, so really." And I'm thinking "hmm," “so” and “pretty.” Interesting.
So, we did a study of the Friends interactions over a period of, I think it was eight years. We discovered that, sure enough, the women were using more "so," and as the popularity of the show increased, they were using even more "so" and even more "so," and even more "so." So, I thought, well, that means that the incoming new intensifier must be "so." So then I studied it in my corpus of Toronto English, and I discovered that you could see "very" going down—it was an old person's thing—and "really" going up. But "so" was “so.” You know "so pretty," "so warm," "so wonderful."
Mignon Fogarty: Was that rising too, as was seen in Friends?
Sali Tagliamonte: Yeah, it was rising too. So, I hypothesized at the time—this was back in 2008—"okay "so" is going to keep creeping up there." That was me saying that, and I was wrong. It has not overtaken the main intensifier, which is in North American English is "really."
Mignon Fogarty: Really?
Sali Tagliamonte: Yeah, if you want to say it's really cold, you say, "it's really cold," and most people will say that. Now other intensifiers have come in, like "super," or there are like the odd, creative ones like, "oh, it's crazy hot out," or something like that. You can always tell like a journalist or someone who's you know a grammar queen because they will try to be more flowery about language and use things that other people don't. That's how you can tell someone who's you know a language person.
Mignon Fogarty: Fascinating. Well, but yeah, "very" is coming back. You were talking about what's the "new" thing before, and you found that young people are starting to use "very" again, which was really fascinating.
Sali Tagliamonte: Yeah, so, here is—this is the current dilemma. My students and I teach an undergraduate sociolinguistics class—third year, learn how to be a language detective, training the students. We do it by doing a project: go out and interview a couple of friends, male and female, and lret's see what intensifiers they're using. And we did that for, I think, six years. And we discovered that over the six-year period, I thought "so" was going to keep going up, and I was wrong. What happened was "very" that was going up.
So this paper just came out recently—in the last six months—it's in a journal called Language Variation and Change. And we make the argument that "very" has come back, and we provide some argumentation about why that's true. So now, the next thing I'm doing is we have a corpus where we have people that we interviewed in 2002 and people we interviewed in 2018. We're checking those materials to see if it's really true that over the course of the two timeframes "very" has actually increased. We haven't got all the data processed on that, but hopefully we will soon.
Mignon Fogarty: Hmm, that's interesting. Well, when you send your students out to do field research like that, what are some of the big things that you tell them to make sure they do or watch out for?
Sali Tagliamonte: That's a big question. I mean, the capacity that young people have to talk to another person and get them to tell them stories is a skill. And you have to learn how to do it. It's even harder when you have to sit down and get a stranger to talk to you. Because most kids these days never talk to strangers—I mean, "stranger danger." Don’t go talking to strangers, right?
So, part of what I teach students to do is to do what you do: interview someone effectively so that what you eke out of them is their—whatever you’re after, their "oeuvre," their main findings. But to do sociolinguistic work, what you really want is just that person to tell you stories. What is their experience in life? What are the things they think about? What do they care about?
And so, what I tell students to do is be an honest listener. Sit down and say, "Hey, tell me a story about your life." I mean, don't say that right away! Say, "Where were you born? Where did you grow up? What kind of games did you use to play when you were a kid? What was your street like? Who was your best friend? Did you ever get blamed for something you never did?"
Everybody has, right? There's always a story about what you got blamed for that you never did. And so, what comes out of those kinds of interviews or informal conversations is people tell you stories. And when they do, they tend to use intensifiers and quotatives—and all the apparatus of language that language scientists like to study.
Mignon Fogarty: Yeah, so you're not asking them about their language; you're getting them to talk and then observing their language, recording it.
Sali Tagliamonte: Yeah, because if you ask someone about their language, they'll just not tell you about their language, right? They will like make up stuff.
Mignon Fogarty: Right.
Sali Tagliamonte: "Oh, I never use 'like.’ ‘So' as an intensifier? No, I'd never do that."
Mignon Fogarty: But you get them to tell an emotional story about being falsely accused, you get some intensifiers!
Sali Tagliamonte: Yes! Exactly, which is exactly why in a real detective interrogation, they just keep interviewing someone and asking them the same questions over and over and over again, right? So that's not what we do. We have a pleasant conversation with people.
I say to people, "What is a sociolinguistic conversation like?" It's like a trip down memory lane. It's usually very cathartic for people because they've never sat in front of another person and had that person intently listen to them. And my students tell me, many years after having studied with me, that that is one of the things; it was one of the courses that has held them in the best stead for the rest of their lives. Because everywhere they go, they just pull out their skills for doing a sociolinguistic interview. And they can interact with anyone, anywhere. Family supper? Great-Aunt Susan is beside you. "What the heck do I say to her?" Oh, "Where were you born, Aunt Susie?" You go to a cocktail party and your boss turns up and you're like, "Oh yeah, hey, where are you from anyway?" So students tell me it's the best skill they ever learned how to do.
Mignon Fogarty: That's wonderful.
Sali Tagliamonte: Yeah.
Mignon Fogarty: It's like a course not only in linguistics, but just in good conversation.
Sali Tagliamonte: It's simple, but it also opens up this huge, amazing, and wonderful world of linguistic variation. It just comes out. It's there. You can't stop it. You know we all have it.
Mignon Fogarty: Mhmm
Sali Tagliamonte: You just have to have someone in front of you that's willing to listen—like you're listening to me, even though you have a plan, right?
Mignon Fogarty: Yeah, yeah. No, it reminds me... I always, before the holidays, I try to tell people you know the things they can talk about with their relatives. I say, "Ask your grandparents why they call the couch a 'Chesterfield,' for example."
Sali Tagliamonte: Exactly.
Mignon Fogarty: You know, we do "familect" at the end of the Tuesday show—the family dialects. And it's like it's a great time to ask your parents and grandparents, "Why do we say this weird thing?" And you're likely to get not only an answer to the language question—I guess that's a little bit more of a direct language question.
Sali Tagliamonte: Yes
Mignon Fogarty: But also a wonderful family story!
Sali Tagliamonte: That is absolutely true! So, you're right onto like one of the best ways of tapping into well language, yes, but human nature too. And the stories that make up the best parts of being human, right?
Mignon Fogarty: Yeah.
Sali Tagliamonte: And it could be something simple.
Mignon Fogarty: Yeah, no. Speaking of humans, now I want to change gears a little bit because you've also done some work on essentially non-human language, because you've looked at ChatGPT and to see if it can talk like a human. So, can you tell us a little bit about that research and what you found?
Sali Tagliamonte: Well, it's very difficult to talk about that research because it's only an idea about research, right? Because I haven't done it yet.
Mignon Fogarty: Oh
Sali Tagliamonte: I think it would be a great new vista for studying language variation. Because of course ChatGPT and Claude and all of those big language learning models are being scraped from—we have no idea where.
Mignon Fogarty: Yeah.
Sali Tagliamonte: And so, what does it represent? I know that language actually is keyed intimately and importantly to time and place, and social relationship, and social standing, and everything else that makes us a human in the social world. But these big language models—they don't do that, to my knowledge.
But that's an empirical question we have to find out. I mean, I know you can play with the latest systems. And you can say, "Oh, I want it to be more formal. I want it to be more this, I want it to be more that." But according to which time and place? According to which society? You know, there's a very big difference between how someone writes a letter of reference in the United States versus how someone writes a letter of reference in Great Britain. They're very different; the styles are completely different.
Mignon Fogarty: I didn't know that, but it makes sense.
Sali Tagliamonte: Exactly. And that's just a small contrast. So, when we ask ChatGPT or Claude, "I want this letter to be more formal. I want it to be more friendly, I want it to be more this, I want it to be more that" we don't know how anyone who receives that is going to interpret it, because it's not keyed in any way to something real in the universe—like in the secular world.
Now, I don't know what those big LLMs are doing, and really no one does, because there's no accountability yet. But it's easy to find out. You just keep asking them the same questions that I you know ask other people, and let's see what we can find out—without the possibility that we're fooling ourselves, and we're just in a big circular wheel of LLM-produced language.
Mignon Fogarty: Yeah
Sali Tagliamonte: So, it's going to be tough to be a language detective to try to figure out what they're doing. But I'm interested.
Mignon Fogarty: Mhmm. I would imagine... just in the last month or so, I've heard people start saying that they feel like they are starting to naturally write the way LLMs write, because they're spending so much time reading LLM output.
Sali Tagliamonte: Yeah
Mignon Fogarty: And I would imagine as a sociolinguist you would say that that is plausible, right?
Sali Tagliamonte: Oh, yeah! I can't tell you how many people say to me, "I know the meeting's being recorded, so I have to be very careful about how I formulate my questions so that the AI will pick up on the right things to summarize for the meeting." Right?
Mignon Fogarty: Oh wow. I haven't heard that.
Sali Tagliamonte: Kids are starting to know that if they are interacting with AI, it's very important how they phrase their questions, in order to get the right answer, or the answer they're seeking, or the results they're looking for. So, yeah, it's going to change a lot about how we do things. Everybody's using AI, but the question is: how? And I have no answers yet. But maybe in like a year or two, I might have some answers.
Mignon Fogarty: Yeah. Well, it's going to be an interesting area of future research.
Sali Tagliamonte: Yes, absolutely!
Mignon Fogarty: Yeah. And you've done such fascinating research already—just more exciting stuff ahead. Sali Tagliamonte, thank you so much for being here. For our Grammarpaloozians, we're going to have a whole additional segment. Sali has done so much amazing research; we have another whole show we're going to do about her work. So, look for that in your feed, if you're one of our Grammarpaloozians, one of our supporters—and thank you. If you're not, you can sign up now at patreon.com/grammargirl. Or, if not, we appreciate you listening here today. Sali, where can people find you if they want to know more about your work?
Sali Tagliamonte: Just Google me.
Mignon Fogarty: That's T-A-G-L-I-A-M-O-N-T-E. Sali—S-A-L-I, with an "i." Sali Tagliamonte. Thank you so much for being here!
Sali Tagliamonte: Oh, this was so much fun, Mignon! Thank you!