How someone talks is complicated. It depends on where they’re from, where they spent their formative years, if they moved around, and whether they’re consciously trying to alter how people perceive them. But beneath it all are the regional roots of a person’s speech—those patterns of pronunciation, shaped by long histories of settlement and migration, that can pin an individual down to a surprisingly small patch of the map.
Despite this, however, for decades the study of American English left a big territory unexplored on the map. Even as researchers divided northern New Jersey from southern New Jersey, or the inland South from the Gulf Coast, linguistics held as conventional wisdom that African American English was a single entity, regardless of place.
But a growing body of research says otherwise. Taylor Jones, a quantitative social scientist who recently got his doctorate in linguistics, wrote a thesis that investigates regional differences in African American English. As an Army kid who moved every couple of years, Jones said, he knew that what he heard in real life contradicted the official story that Black people in America are a linguistic monolith. Where linguists held that African American English always merged the vowel sound in pen with the sound in pin, a whole catalog of New York City hip-hop rhyme schemes—the Notorious B.I.G. saying, “Givin’ ends to my friends, and it feels stupendous”—showed otherwise.
“I went into linguistics with the background knowledge that African American language use varies from place to place,” Jones said, “and did not see that represented much in the scientific literature.” Following the lead of researchers including Sonja Lanehart, Lisa Green, Hiram Smith, Sabriya Fisher, Kelly Wright, John Rickford, and Sharese King, he set out to document the patterns of variation. While white American English pronunciation groups tend to cluster along East-to-West lines of population movement,Jones found African American English groups stretching South-to-North—along the routes of the Great Migration, where Black people left the Jim Crow South for cities in the North.
I spoke with Jones about the complexity of the different ways of speaking African American English, the role language plays in stigma, how vowel enunciation offers insight into someone’s origins, and how accents shift—and I convinced him to guess where I’m originally from. Spoiler: Despite the circumstances that complicated a pure analysis, such as being on the phone during a professional interview, he got really close.
This interview has been edited for length and clarity.
Julia Craven: Tell me a little bit about the work that you do and your thesis.
Taylor Jones: My research is primarily, but not exclusively, on African American language. I also have work on Mandarin, Persian, Zulu, and other languages, but African American English is one of the main areas of focus for me. I’m a white guy, but I’ve always grown up and spent my life in AAE speech communities. So that’s the language that I grew up around, the language that I’m used to hearing. There’s a ton of research left to be done, and also stuff that we just don’t know about AAE.
The other reason I went to grad school in particular is I think there’s a lot of work around AAE that can speak to current events. It can speak to social justice and to civil rights. That’s not necessarily always everybody’s path into linguistics or sociolinguistics or studying African American English, but it’s definitely the way that I’m coming at the language that my friends, my family by choice, and everybody I grew up around speaks. It is highly stigmatized, and there’s a lot that linguists and linguistics can do.
I want to learn more, just for the sake of pursuing knowledge, but also to challenge that stigma, to challenge linguistic discrimination, and to challenge linguistic discrimination as a proxy for racial discrimination.
How do you think this type of research is going to help remove the stigma from AAE?
That’s a harder question to answer than it sounds like it would be, in part because I think linguists have been very hopeful about what role we can play and historically have been pretty consistently surprised by how things play out in the real world and in politics. I have tempered expectations at this point, especially looking back to the Ebonics controversy in the 1990s, the discussions about speech pathology before that, and even going back to articles like “African American English Is Not Standard English With Mistakes,” or “The Logic of Nonstandard English” by [William] Labov.
But I do think that times are a little bit different now. There’s a little bit more grassroots discussions of language that you see on places like Twitter and TikTok and wherever else that didn’t exist before. The more that we can get the word out about the fact that African American language varieties are valid, are structured, are systematic, and that all of the things that people say to diminish them are not linguistically, scientifically grounded, the more we can explain the link between those language ideologies and discrimination. I’m hopeful that that actually helps us make a little bit more progress.
One of the points of your research is that there isn’t just one way that Black people talk. It very much varies depending on where you’re at, where you’re from, and whatnot. How did you go about collecting such a large variation of data?
The way that I approached that is fairly standard. Normally, what sociolinguists will do is: They’ll have a reading passage and they may have a list of words, and they’ll say, “Can you pronounce these and these words? Do these words rhyme with one another? Do they not rhyme?” And then you may do a sociolinguistic interview.
Because these regional variations had not been described at all in the literature systematically, I just wanted to start with the reading passage because even though people are going to speak a little bit more carefully, it’s still a snapshot of how people speak differently in different places and what is below the level of consciousness for careful speech. So there’s certain words that maybe we’ll pronounce one way or the other, but accents in general, more broadly, we don’t give a lot of thought to.
If you look at the existing reading passages for sociolinguistics, they have their strengths, but a lot of them are very strange. They’re very culturally, linguistically, socially white and very academic in a way that is off-putting to a lot of people who aren’t in academia.
I put together a reading passage called “Junebug Goes to the Barber,” and the idea was to have a story that encouraged people to code-switch and to use more, rather than less, African American English. The way that that was accomplished was telling a story where people are talking.
I shared it on social media and personally with people. From there, it was a phonetic analysis where it’s this coding pipeline. I used what’s called “forced alignment and vowel extraction,” FAVE for short. I use it in part because it’s used by a lot of other linguists for a lot of other varieties of English, so it’s directly comparable to other work that people are doing.
I was really interested in the vowels. So, when you’re speaking, your vocal cords vibrate and you produce sound waves from your mouth and nose. What makes a vowel different from another vowel is basically the configuration of your tongue and your lips, so if your tongue is high or low in the mouth, if your tongue is front or back in the mouth, relatively, and if your lips are rounded or unrounded. You hear them in the sound wave, but that’s something that we can do.
For instance, Washington, D.C., people will say goose and goat with vowels that are much—the tongue is much more forward in the mouth than it is for people in, I don’t know, Gulfport, Mississippi. So it’s the difference between:
Can I ask where you’re from?
I was actually going to ask you if you could tell where I was from.
Oh, I would have to ask you a bunch of words.
I can also read “Junebug Goes to the Barber.” I actually read it before this call, and I was like, “This is fun.”
I liked it. I had some help with it, because I had to run it by a bunch of people and make sure it didn’t have regional stuff. I think it says “Dad” in there now—I think I had “Pops” or something, and in some places that was fine, nobody cares, but in other places people are like, “We don’t say that.”
Do you have “Junebug” nearby?
Yeah, I do.
I have a hypothesis from some vowels, like, what they call the strut vowel gives me some information. Did you grow up in one place and was it in the Southeast?
That’s what I got right now from how you said “point” and from how you say “Junebug.” But narrowing it down is going to be a little bit harder.
All right. Just tell me when to stop reading: “So here’s the deal: Marcus is named after his dad, but everyone just calls him Junebug. He’s turning 12 this week. His momma stay watching him like a hawk. She’s always asking him, ‘Where you going?’ But today, she’s letting Junebug go to the barbershop by himself. That’s probably cuz it’s so”—I hate reading out loud—“That’s probably cuz it’s so near she can see him from down the street anyway.”
There’s certain vowels that I’m listening for now, where it’s to narrow it down. Can you keep going? Is that all right?
“He don’t even care. He’s just happy he can take care of getting a haircut. They’re having a birthday cookout on Saturday and everybody’s going to be there. Everybody. That means the boys from school and the girls from school too—so you know he’s got to look fresh. It’s like his old man is always telling him, ‘Always come correct.’ ”
I’m thinking, because you got to some of the vowels that I was curious about. Now, let me add some caveats, all right? Before I ask you some other questions.
One is that in general there’s certain vowels that are more stigmatized and people pay more attention to and try to correct away from, especially if they work in anything having to do with language. So it may be that the clues for a particular place are something that you are aware of trying not to say in general. I’ll tell you what it is after I ask you, but I want to actually try this, because now you’ve got me thinking about this.
They give a keyword for each of these vowels. They’ll talk about a goose vowel, goat vowel, strut vowel, the one that I’m thinking of is the—usually they talk about caught, because they say the cot/caught merger. In the Southeast and then more as you get closer to Mississippi and then up the Mississippi River, that’s more and more pronounced like:
Yours sounded like it was a little closer to that, but not like Alabama or anything. At least not the speakers that I had from Alabama.
The way that you said bug, which is the strut vowel, suggested to me also parts of the Southeast. It could be consistent with the Carolinas, but the way that you were saying the vowels that are normally called the kit vowel and the dress vowel are not as consistent with North Carolina, for instance, as what I would expect from really casual speech from people from those places. You also didn’t have the hawk or caught vowel—did not sound like Florida, and didn’t really sound like a lot of the speakers I have from Georgia. So I’m leaning towards southern Virginia, but I really do not feel confident that that’s the place.
You don’t have any of the stuff like the mergers before “R” or really classic Washington, D.C., and Baltimore stuff, so I want to rule those out. But again, working in language, high level of education, speaking formally in an interview—that kind of thing all affects this stuff.
Am I even in the neighborhood, or will you be like, “I’m from California, obviously”?
You’re definitely in the neighborhood.
I don’t think I’m going to narrow it down any further, because that’s the cues that I have to go on, like how do you pronounce words like hawk, coffee, dog, that kind of stuff. How do you pronounce words like—
Miss Mary [may-REE]? I assume that would have given you the Carolinas.
Yep! South Carolina?
North. I’ve only heard North Carolinians say “Mary.” I never hear it that way anywhere else.
So now I’m interested, because there’s a split in North Carolina. Where in North Carolina? That’s what I want to know.
Lexington. It’s centrally located, in Davidson County.
What I was listening for is the eastern versus central split. They talk about it like it’s a reversal, but it’s not really a reversal. It’s more like what comes first in the vowel. They talk about splitting and they talk about reversal of the vowel nucleus, meaning the first, heavier part of the vowel. So what they call the kit vowel and a lot of Southern white English and in certain places in African American English would be like “kit”:
Similarly, like “bad” or like “head,” that kind of thing would be like … “head” would be like “head.”
That’s what I was listening for, for eastern North Carolina, and didn’t hear that on certain words. Which is why I was like, “I don’t know.” I was ruling out some of the Gulf states because of the vowel in dad. The most extreme example of that would be parts of Alabama and Georgia saying, like, “dad” and “jazz” music so it sounds like “dead and “jezz” to outsiders. I heard people from the South saying, about the musician in the story, they say, “He play jezz.”
Each of those things add up. They all have their own geographic distributions. You can have like a mental map of it and say like, “OK, well, if I hear this sound, then that restricts it to here, and I hear this other sound and that restricts it to this smaller part of here.” You know what I mean?
I wasn’t sure about southern Virginia, but I got close.
When you said the Carolinas, I was like, “Wow.” This leads me perfectly into my next question, because I always felt like I had lost my accent, and I know it comes out at times. But whenever people move and they live in a new place—like I went to college and I was around a whole bunch of different people with a whole bunch of different accents—I’m just wondering how people’s accents can shift.
There’s a lot of research on that. Education plays a part in it. There’s actually some studies—I think Sabriya Fisher was on one of them—but they looked at people going off to college and found that they did a lot of identity construction and a lot of accent work.
People’s accents change with who they’re surrounded by, and what communities they want to be a part of, and how they want to be perceived, and what they’re aware of versus not aware of. I think the really important thing to remember is that everyone has an accent, always. So if you talk about accent reduction, it’s not that you lose an accent—it’s that you develop an ability to speak in a different accent. But it’s all accents. There’s no such thing as nonaccented speech. Some [accents] have more social prestige and some have less, but nobody speaks without an accent.
Then when you talk about losing a Southern accent, that also depends on what communities you’re interacting in, what they even consider a Southern accent. So among African American English speakers in D.C., what’s considered a Southern accent? It’s going to be different than among white Virginians in the suburbs. And that’s going to be different than somewhere else where those suburban white Virginians are considered to have a Southern accent by people in New York. So it’s a whole complicated system.
What I will say is there’s certain things that people really notice. But there’s other things that people don’t really notice. Like the way that you say out, O-U-T, like, “My accent’s coming out.” Different people in different places pronounce that vowel differently, but nobody really pays attention to how everybody else is pronouncing it, other than linguists. They notice that so-and-so sounds different than some other person, but they don’t really notice it’s because of how he says “out.”
In my dissertation defense, I actually played audio of Jay-Z saying “bounce,” because my advisers weren’t really sure on that one, and I just played them Jay Z saying “bounce.” That’s like very much a New York accent—”Can I Get A,” that’s the song I’m thinking of, I played that—then I played audio from the conductor on the 1 train in Harlem and said, “Just listen to how he says, ‘This is a downtown-bound 1 train.’ ” It’s completely different than other places.
Accents are complicated. Everybody has one, but then what you’re aware of and what the people around you are aware of and what is stigmatized and what is not stigmatized—it’s all going to affect how somebody speaks. That’s where you get into code-switching. That’s where you get into having telephone voice or “white voice,” where people feel like they have to navigate making their speech less stigmatized. I recognize the reasons for it, I recognize that it is strategically sound and intelligent to do, but I also don’t think, in an ideal world, that it should be what people have to do.
What happens is in the United States, one, people don’t really know much about linguistics in general unless they study linguistics, and two, all of those things become pathologized when we’re talking about African American English, even though they’re normal and natural. Tons and tons and tons of languages make exactly these kinds of changes. African American English is unique, but all of the individual things that make it unique are cross-linguistically common and attested. It just gets pathologized in the United States because of our history of race and racism.
Like everything from double negatives—people say two negatives make a positive. Well, not in Russian, not in French, not in Arabic, not in Hebrew, not in all of these other languages. But when Black people do it, suddenly it’s a problem. I think that shapes a lot of how language is talked about, where the American mainstream starts from a position of “What Black people do is wrong” and then finds reasons to justify it. If you go out and look at the data, you actually measure how people speak, you look at the syntax, you look at the grammar, you look at the morphology, it is normal and natural and does exactly what other languages do, but it’s pathologized socially in the United States.
I saw your thread on George Floyd and him appearing to say, “I ain’t do any drugs.” It was very interesting to me that people were hearing “I ate too many drugs.” Your thread was very enlightening, I think, for a lot of people, because to me, “I ain’t do any drugs” is a perfectly said, understandable statement. So seeing language being pathologized during the trial for Derek Chauvin, what are your broader thoughts on the way language has been used?
I mean, the strategy for the defense, as far as I can tell, is to portray him as dangerous and that use of force as being required. One of the ways you do that is you play into stereotypes about the criminal Black man. One of the ways that you do that is you play into stereotypes and fears about Black people being stronger. It’s not true, but there’s a ton of research that suggests that people think, incorrectly, that Black people can tolerate more pain or that they’re stronger or whatever. There’s certain drugs that, they’ll argue, make people more violent, that will make people impervious to pain. So the idea is to play into that concept.
I want to be careful here, because I can’t say for certain that they understood it and misrepresented it. They may have misunderstood it themselves. I don’t know which of those is the case. But either way, they thought that the witnesses and the jury would think that “I ate too many drugs” is a plausible interpretation of what he said. I just don’t think it’s a plausible interpretation if you have any familiarity whatsoever with African American English.
There’s other examples in high-profile cases of Black people’s speech being misunderstood, misrepresented, or completely ignored. We wrote a paper in 2019 about court reporters and their transcript of African American English in Philadelphia and found, among other things, that they were not performing at 95 to 98 percent accuracy, which is their certified level.*
In doing that paper, we looked through some other high-profile cases, and you have everything from judges going to Urban Dictionary to find out what finna means. It’s actually the same judge that said it was impossible to determine whether “He’s gonna shoot me” is an excited utterance, and that has bearing on the admissibility of evidence in court. He said it was impossible to determine because there wasn’t a conjugated form of the verb to be.
Any linguist who’s done any work on African American English will tell you that the conjugated form of the verb to be is obligatory in the past tense, and there’s certain contexts in which it can be reduced or deleted in the present in African American English, but also this is consistent with, again, Russian, with Hebrew, with a variety of other languages. “He’s gonna shoot me,” there’s no ambiguity about the time frame there. So that’s one thing.
Then you have, within the same span of years, the guy who asked for a lawyer and was denied legal counsel because he said … I forget the exact quote, but it was like, I want a lawyer, dog. They denied him legal counsel on the fiction that he was requesting a canine lawyer.
Did you not hear about this one? This was 2017, I think. He was like, I want a lawyer, dog. And they say, Uh, we’re denying the motion for counsel on the grounds that a canine lawyer doesn’t exist.
So there’s a range of making decisions on appeal based on going to Urban Dictionary to try and understand what was said, which tells me that there’s a problem there. But at least they’re trying to do their job. And then there’s being denied legal counsel because you said the word dog. There’s a whole range of ways in which this plays out.
What we saw with the defense for Derek Chauvin is consistent with all of that, because we don’t have a level of understanding, of education, in the United States, and respect for African American speech that would prevent these things from continuing to happen.
Correction, May 18, 2021: Due to a transcription error, this piece misstated the level of certification for courtroom transcription in Philadelphia. It’s at 95 to 98 percent, not 90 to 95 percent. This piece has also been updated to clarify the specific pronunciations of the words dad and jazz that proliferate in parts of Alabama and Georgia.