Language identification, Accents and Dialects

How and why do we make assumptions based on the way people speak? What is considered “proper,” or “correct” language – and who decides that, anyway?

TRANSCRIPT

Ahmed Ali Akbar: Take a moment and try to hear your own voice in your head. What do you hear? Do you hear, an accent?

[voice clip]: I don’t hear an accent in my own voice, but it’s funny that certain people do.

Ahmed Ali Akbar: What about others? Do they hear an accent? What do they say about it?

[various voices]: I will often have people comment that I don’t sound quite Black in quotes. / I’m Pakistani. Non-Pakistanis were surprised about how clean my accent is. / I’ve heard that the Midwest accent is more pleasant. It’s friendlier, it’s polite.

Ahmed Ali Akbar: Do people ever make fun of you?

[voice clip]: My husband definitely pokes fun at the way I say certain words, like, uh, I’m embarrassed to say it, sauce and dog.

[various voices]: Some people bad and rude. / Yeah, people can be rude.

Ahmed Ali Akbar: The way we speak carries a lot of baggage. It’s the product of where we’re from, of our education, our experience, and it’s super personal and sensitive. And if you people don’t notice your accent, life can be a little easier than if people do notice your accent.

[voice clip]: Growing up, people would always just ask me, you know, where did you grow up? And I’m like, I’m from the same place you’re from. It made me pretty self-conscious about my voice, to be honest.

Ahmed Ali Akbar: Growing up in Michigan, people commented on my parent’s accent, but very rarely on mine. I actually was often validated or rewarded for the way I spoke. I volunteered to do the morning announcements, for example, and I won awards at Model UN. I thought this meant I had “no accent.” So I felt like it was okay to make fun of the way some of my classmates talked – the ones who had more of a Michigander accent. They would say things like bag-el instead of bagel, melk instead of milk. Then when I got older, and moved to the East Coast, I was the one who started getting made fun of. People would comment when I’d say “crayn.” or “pop” for soda so I stopped. [sound of soda can opening] That’s when I realized: I have an accent. Actually, everyone does. There’s nothing neutral about the way anyone speaks. We can’t completely control how we speak. Yet, we tend to have biases against those who speak differently than us. So what is going on here? In this episode, we’re going to explore what our accents and dialects say about us, how we read other people’s accents, and how race, class, and power hang over it all. From Crooked Media and Duolingo, I’m Ahmed Ali Akbar, and this is Radiolingo. Today’s episode, What Accent? [music break] Accents are the unique ways people pronounce things. They can be found in things like vowels, like the bag-el vs bay-gel pronunciation. But they can also be present on consonants and emphasis too. You might say button, buddon, or buh-on. Our accents come from all sorts of places: it might be that we’re American or British, from Spain or Mexico. It may be what city or region we’re from, from Osaka or Kyoto. And it might be our ethnic group, if we’re Black from the South or white from the Midwest.For this episode, I spoke to a lot of different people about their voices, their accents. Many people could hear their accent. Others couldn’t. And so, I wanted a better way to talk about this than just asking people about their accent. And we found just the person to do so.

Dennis Preston: My name is Dennis Preston, and I am a professor of linguistics at the University of Kentucky.

Ahmed Ali Akbar: You prefer Dennis and not Dr. Preston?

Dennis Preston: I don’t like Dr. at all. If you want to go fancy, you may call me Professor Preston. If you don’t want to go fancy, you can call me Dennis.

Ahmed Ali Akbar: Dennis is widely considered the founder of a modern field of linguistic research called Perceptual Dialectology. I know that’s a big phrase, but it helps us understand a bit more about the ways we speak and our dialects. A dialect is made up of three ingredients. We’ve already covered the first, accent. The second is any unique vocabulary found in a certain dialect, a can of soda vs a can of pop. Pop is the Midwestern term. The final ingredient in the dialect pie is grammar. Here’s some very Western Pennsylvanian grammar. The pop needs to be opened, vs the pop needs opened. That second one is from Western PA. It’s super important to understand that regional dialects aren’t some degraded version of a language – they’re genuine ways of speaking with relatively consistent features. They come from a group of people, the Midwestern way of speaking makes sense to Midwesterners.

Dennis Preston: There are regional dialects, but there are also ethnic dialects. There are also gender dialects. There are social dialects which have to do with status and class.

Ahmed Ali Akbar:  Dialects of English are mostly mutually intelligible between each other. You might be confused for a second when someone says, “The pop needs opened” but you’ll figure it out, eventually. But your first thought might be, boy, is something wrong with this guy? But it’s not wrong. It’s a dialect.

Dennis Preston: When we associate dialects, we very often associate those dialects from lower status, local vernacular speakers.

Ahmed Ali Akbar: You hear someone speaking in a different dialect and you begin to make assumptions: where they’re from, how intelligent they are, and maybe what they should sound like. So, I did my little impression of a Midwestern accent. But let’s play an actual clip of someone with a Midwestern accent and dialect.

[voice clip]: So she talked as loud as she could at the top of her lungs. So I laughed and walked across campus to another person’s room to sleep.

Ahmed Ali Akbar: And now let’s compare that to what linguists call standard English. It’s a prestige way of speaking that has features and sounds used by white, upper middle-class speakers. Think about the way a newscaster speaks.

[newscaster clip]: This is a story about a little boy falling in love with a puppy.

Ahmed Ali Akbar: So how do you compare the two speakers? Did you think the newscaster clip sounded educated, whereas maybe that Midwestern sentence sounded kind of like a yokel, less educated? That’s the bias that we hold. This also happens to Southern dialect speakers in the US. Dennis says that he as a Southerner often challenges people’s expectations.

Dennis Preston: So I’m from Louisville. Kentucky, and I sound like I’m from Louisville, Kentucky. But very often when I talk to people, they say, well you don’t sound like you’re from Kentucky at all. And I say, what the hell? You want me to sound like The Dukes of Hazzard? Or for an older generation Little Abner?

Ahmed Ali Akbar:  People assume a professor of linguistics should speak with a so-called standard English dialect. We’re expected to engage with Standard English when we interview for a job or communicate in professional settings, with customers. And when we violate the norms of that prestige dialect, people can get mad.

Dennis Preston: People don’t just believe there’s a good English and believe there’s a bad English. They believe it in spades. I mean, they say, nothing makes me angrier than to hear somebody say X, or I lose it when I hear people say this or I want to puke when I hear. I mean, people have visceral reactions to language which does not suit their standard.

Ahmed Ali Akbar: Standard English was what I was deluded myself into thinking I spoke in high school. It’s what we’re nudged towards in language class. But the truth is basically no one in the U.S. actually speaks this standard English. It’s not very standard at all.

Dennis Preston: The idea that a standard English exists for spoken language is just a kind of absurdity. Some people have nostalgia. Like every parent does, says, oh, no, kids are taking English to hell in a hand basket. Just listen to the way they talk. And of course this happens every generation. Plato, by the way, said that about younger people in Greece.

Ahmed Ali Akbar: So, how does Dennis Preston study dialects and peoples’ perceptions of other ways of speaking? He researches everyday people. Like you and me. And one of the methods is as follows, so imagine Dennis hands you a map of the United States, with regions, borders, and states. He asks you, where do people who speak most differently than you live? You might take your map, and circle New York, the South, Texas, New Orleans, as places where people are speaking differently than you if you’re not from those places. Then, Dennis asks you, how would you describe them? You might say something like, southern people are polite. New Yorkers are rude. By doing this exercise, Dennis finds all sorts of interesting judgments.

Dennis Preston: Hick, hillbilly, talk funny, speak through their nose, South, Yahoos, Yankees. And sometimes they write descriptive phrases. One of my favorites, a South Carolina young man, no young woman. And she wrote, Northern, scratch and claw. So scratching and clawing, of course, is a nice kind of Southern stereotype of the North. Y’all are not laid back. You don’t sit out on your front porch and have iced tea or whatever it is. You don’t say, come around and see me sometime and stuff like that. You’re running and working too hard and you’re not friendly to one another.

Ahmed Ali Akbar: Where do you think we learned these prejudices about the way people speak?

Dennis Preston: Oh, they’re the prejudices of our speech community. They’re the prejudices of the people we grow up with. We grow up in a community and in the same way that we grow up in a religious or non-religious belief or whatever we hold about those things, then we hold the same things about people and those people things extend the language. And there we are. There is no such thing as language prejudice, it doesn’t exist. The only thing there is is people prejudice.

Ahmed Ali Akbar: Let’s take, for example, the contraction ain’t.

Dennis Preston: And it’s always been your experience that people with less education say ain’t, which is not true, but it’s always been your experience and your belief. Then pretty soon you don’t even have to go to the people anymore. You can just say, ain’t. Oh, ain’t is dumb.

Ahmed Ali Akbar: Or think about a posh British accent – you hear it in movies all the time to signify that someone is smart, or authoritative. Now, there’s nothing inherently more authoritative about that accent, it’s just an association Americans have with Brits. We think we can tell a lot about a person’s social standing based on their accent and dialect. Now, what about where I’m from?  It turns out Dennis has done work on dialects in my home state of Michigan. So I asked him about what people thought about their accents and dialect there.

Dennis Preston: They say, no, no, we just speak standard English. We’re good old upper Midwestern speakers. And some of them even go so far as to say they don’t have a dialect.

Ahmed Ali Akbar: This was exactly what my classmates who I made fun of would say. To my ear, they had strong Michigan accents. But they believed strongly that they were speaking Standard English. Dennis has seen this over and over in his studies.

Dennis Preston: So another great thing that we’ve done in perceptual dialectology is to do some experimental work which shows that people don’t hear their own and other voices very well. My good friend and co-author, Nancy Niedzielski at Rice University, says this is when your brain messes with your ear. You have a prejudice, a stereotype about the way a person should talk. And then you play something for them, and if it doesn’t match it, their brain will adjust their ear. The most famous example in Nancy’s early study from years and years ago, that people from Michigan could not hear their own vowels. They couldn’t recognize them as what they said.

Ahmed Ali Akbar: In other words, they mentally couldn’t hear the difference between bag-el and bay-gel.

Dennis Preston: Because people from Michigan think that they are standard English speakers. When they hear these vowels, they automatically adjust them to standard. And therefore the brain is messing with the ear.

Ahmed Ali Akbar:  My classmates in Michigan, mostly the white kids, they weren’t able to hear their own accents. They thought they were speaking in standard English. So me telling them they were speaking funny, triggered a reaction in them, that same reaction Dennis mentioned, no, we speak good English.

Dennis Preston: I think some linguistic insecurity which previously wasn’t there and you’re partly at fault. You making fun of these little local white kids for the way they talk.

Ahmed Ali Akbar: Can you tell me what you mean by linguistic insecurity? Is this a technical term?

Dennis Preston: It is a technical term. And my definition of it is that you have anxiety about not having the linguistic resources to do what you want to do with language. So if you’re around your home folks, you never have linguistic insecurity cause you have the linguistic resources to do all that you want to do. But then you go to work, then you go out in public, then say you give a speech or something like that, and all of a sudden you say, I don’t know if I can handle this. Or I don’t know if I’m going to present myself in the way that I want other people to understand who I am, where I’m coming from, and all that kind of stuff. So if you got those feelings, then you got linguistic insecurity.

Ahmed Ali Akbar: I experienced this linguistic insecurity too. I told you about how I stopped saying pop and crayn. Eliminated my Midwestern elements. Well, I also had some elements of a Pakistani dialect, I think. Growing up, when we were getting ready to sleep, we would say we were putting on our night suit. When I went off to college, I had a very embarrassing moment where I said, hey, you’re wearing your night suit to class? That’s weird. And this classmate turned to me and said, what the hell is a night suit? This is a pajama. So due to linguistic insecurity, I’ve actually stopped saying night suit and started saying pajama.

Dennis Preston: Wait a minute. Wait a minute. You said puh-jamas. Where in hell did you learn how to say puh-jamas instead of Michigan pa-jam-as?

Ahmed Ali Akbar: In a quirk of my accent, Dennis noticed I don’t pronounce the vowels of pajama with a Michigan accent.

Dennis Preston: No self-respecting Michigander would say puh-jamas. They would also be pa-jam-as. Actually, they would say pa-jaam-as. Right? [laughter]

Ahmed Ali Akbar: Before this interview, I had no idea my accent on this word was different from other Michiganders. Even though I don’t think I have much of a Michigan accent, I felt kind of ashamed that I don’t pronounce words like other Michiganders. I chalked it up to insecurity and trying to seem cool on the East Coast and not a Midwest yokel. But my quirky pronunciation of pajama? I think it might be from Urdu. Because the word pajamas came to English from the Kurta pajama. That’s a type of clothing that I’ve worn many, many times in my life. And it’s pronounced like that pa-ja-ma, not pa-jaa-muh. I’m surrounded by Pakistani accents. Honestly, sometimes I don’t even notice them. And I must absorb some of that in addition to the Midwest and East Coast sounds. In Dennis’s work, second generation immigrants like me usually do have unique characteristics about the way they speak. Especially when they live in some density, like in Southeast Michigan.

Dennis Preston: We’ve done work on populations, uh, which were not native speaker populations. We worked on the Hamtramck Polish. We worked on, uh, Dearborn Lebanese Arabic. We also did quite a lot of work on Spanish, which of course a very large population in Michigan. And what we find is that interestingly, even by the third generation, the vowel systems of those speakers are still not completely adjusted to the Michigan form.

Ahmed Ali Akbar: So maybe I say pajama because of Urdu. Or maybe my vowels just aren’t fully Michigan yet. There are all these ways in which I’ve softened my dialect because I’ve been teased but there’s also instances when I’ve teased others. Those kinds of interactions can really cause people to be shy about the way they speak. It goes back to linguistic insecurity. It’s a pretty universal thing when you venture out into the world where people speak differently from you. I feel like I’ve done a lot of work to try to not be ashamed of the Midwest or being Pakistani or the way I speak and then whenever I asked people about their accents for this episode, I was stunned by how much shame people felt about their accents. I also wondered if this happens in other languages, and what that looks like. So I asked someone from the Crooked Office who speaks a lot of languages Ines Sainz de la Maza. She’s the assistant to Radiolingo’s Executive Producer.

Ahmed Ali Akbar: Do you have an accent when you speak Spanish?

Ines Sainz de la Maza: Yes, I do.

Ahmed Ali Akbar: And you’re aware of that?

Ines Sainz de la Maza: Yes. I’m aware of it because people have told me, and it’s people who speak Spanishm more specifically from Spain–, hat I have an accent from Madrid. And it’s a very it’s the be [speaks Spanish] part of Madrid, as we would say, in Spanish. [speaks Spanish]

Ahmed Ali Akbar: Just I’m curious, like, does that accent say anything about the person that you are? Does that make sense? Like–

Ines Sainz de la Maza: Like my Spanish accent? YeIt’s rich people. Yeah, it’s the, like, rich, preppy part of town. That’s what I speak.

Ahmed Ali Akbar: I also asked people outside the office about their accents in other languages.

[voice clip]: I speak Chinese with a Taiwanese accent. So when I speak to Mandarin speakers from outside of Taiwan, they will inevitably say, oh my God, you sound so Taiwanese.

[voice clip]: My Syrian friends make fun of me all the time that I have this very exaggerated Damascus accent when I speak in Arabic.

[voice clip]: I grew up learning Chinese from my parents and speak just like them [speaks Chinese] and they are from Shanghai and Suzhou of China. So they have accents from those respective regions. I never realized what those differences were until I started taking Mandarin Chinese classes and I had to sit down with the teacher. And she explained to me that I was speaking with this accent.

Ahmed Ali Akbar: The subtext of all these stories is that having an accent or a different dialect can make you feel self conscious. When someone comments on your accent it makes you feel like you don’t belong. You sound different. It can create linguistic insecurity. And that sucks. But it’s something that we all experience to some extent.  Even beyond that, though. The lie that there’s some proper way to speak it can be weaponized to perpetuate existing inequalities, and prop up racism and classism. Our next guest has some ideas about that.

Rachel Weissler: Who decides what’s proper and correct? Its just about which one’s given power in a social situation, and which ones are not..

Ahmed Ali Akbar:  That’s after the break. After talking to Dennis, I wanted to get more specific about how our biases impact speakers of “non-standard English.” I spoke with Dr. Rachel Elizabeth Weissler. Rachel is a postdoctoral scholar in psychology, linguistics, and Black Studies at the University of Oregon. Her work focuses on one of the most significant stories of language in American history, what scholars now call African American English or African American Language.

Rachel Weissler: One of the hypotheses of how African American English came to be is that Black people were brought from the African continent against their will to the United States through the South. And so you have this language contact situation that happens between the people in the South in the United States and these enslaved peoples.

Ahmed Ali Akbar: Often, when two groups of people who don’t speak the same language come in contact, a new language emerges so they can communicate. Rachel says African American English emerged out of that history. But it’s not the same African American English that it was when it was born. No language is, including Standard English

Rachel Weissler: So when I think about English, I don’t think about just one English, right? There’s a spectrum of Englishes, world English is, and those changed over time. We don’t sound like Shakespeare and things like this.

Ahmed Ali Akbar: Similarly, African American English or AAE has also evolved from its earliest days. And, it’s evolved in different regions, people speak differently in New Orleans than they do in Detroit or Houston.

Rachel Weissler: People that first were enslaved and brought to the United States sound very different in the slave recordings that we have versus the way that Black people might sound now in the United States. And even within the Black people that are here now, depending on when you came to the United States, depending if you’re a second generation, West Indian American, Haitian American and things like this. All of these different ways of speaking as a Black individual in the US will vary.

Ahmed Ali Akbar: Rachel says scientists historically had a too-narrow view of the world of African American English.

Rachel Weissler: So historically African-American English, when it was being studied, it was being studied in a particular person. So it was Black, young, urban men. That’s who was speaking African-American English, and that’s who was being researched. We weren’t thinking about all genders of different people. We weren’t necessarily considering different age groups. We weren’t considering influences of other languages.

Ahmed Ali Akbar: So there’s a lot of work to be done especially since research shows that speakers of any version of African American typically face discrimination in some form, from the housing market to the legal system. One study about housing discrimination showed that white-sounding callers were more likely to hear back from landlords than Black-sounding callers. In other studies, people who were perceived as Black on the phone were more likely to be denied access to health care. And in the criminal justice system, African American English speakers are at a disadvantage. Rachel told me about one study looking at Black AAE speakers in the courtroom.

Rachel Weissler: People get transcribed differently when they’re on the stand. And if you’re speaking different varieties of English and you’re not well acquainted with those, you might transcribe someone’s testimony incorrectly, which could then affect the outcomes of certain cases, and we’ve seen evidence of this.

Ahmed Ali Akbar: The consequences of language discrimination continues. It’s so baked into society that it’s been programmed into our technology.

Rachel Weissler: There’s research and evidence from the tech world that shows that these assistants that we have, whether it’s through Siri or Alexa, they’re not responding well to non-standardized American Englishes. So we’re seeing these biases happen even in the A.I. that we’re training.

Ahmed Ali Akbar: I wanted to know, what’s going on in our brains when we perceive a dialect different than our own? Rachel studies this very question by actually looking at people’s brains.

Rachel Weissler: I use EEG or electroencephalography. What easier electroencephalography is, is a essentially a cap that goes on somebody’s head with a bunch of electrodes.

Ahmed Ali Akbar: Rachel likes using EEG because it can tell us how the brain reacts almost in real time. She can see how the brain reacts to sounds within 200 to 300 milliseconds.

Rachel Weissler: And so EEG gives us this nice opportunity to say, when we hear something, what does our brain do on sight?

Ahmed Ali Akbar: Rachel talked about two experiments she did using EEG. In the first experiment, she had participants listen to sentence structures that are typically associated with either Mainstream English or AAE.

Rachel Weissler: So I can give you an example. So if I say something like, the clown, he is blowing up balloons at the party. That is, is obligatory. We need it in standardized American English. But in African-American English, you can say, that clown, he is blowing up balloons at the party. But you could also say, that clown, he blowing up balloons at the party. You don’t need that is. It’s not obligatory.

Ahmed Ali Akbar: So Rachel played sentences for participants, and then she looked for spikes in brain activity. Those spikes, to her, would signal that participants were processing the types of speech differently.

Rachel Weissler: So what I found in my work is that broadly we saw different neurophysiological responses on the EEG depending on standardized American English or African American English. So people were distinguishing them.

Ahmed Ali Akbar: Rachel’s experiment concluded that speakers were almost instantaneously perceived differently by participants depending on whether they used Standard English or African American English. This demonstrates the fact that our brains rapidly take what we hear, and form conclusions about the speaker that draw upon our prejudices and experiences. In her second experiment, Rachel played two voices for the study participants. The two voices used different emotional inflections.

Rachel Weissler: So, for example, I’m fine versus I’m fine versus I’m fine, again. The last one, you know that uh oh, Rachel’s not fine. What’s wrong? Talk to me.

Ahmed Ali Akbar: Then she asked listeners to identify whether the speaker was Black or white based on what they heard.

Rachel Weissler: My results from that study were that Black voices were most correctly identified as Black when they were in angry or neutral conditions. White voices are most identified correctly as white when they were in happy conditions.

Ahmed Ali Akbar: There’s this negative cultural stereotype in the US, that Black people tend to be more angry and emotional. Rachel’s studies demonstrate that people believe that intuitively. They’re so deeply ingrained in listeners that their brains make these assumptions about people unconsciously and in seconds. Actually, fractions of seconds. But researchers haven’t come to a consensus about what drives us to make these assumptions. Is it prejudice, or something else?

Rachel Weissler: I often respond to that and say, okay, I hear that, that this is an open question. This also has to be met with all of this other evidence that I described earlier, right. That we’re seeing that people are getting the apartments or they’re not or they’re not getting the jobs because they don’t sound professional and things like this. So these concepts that are factual and have been shown and affect people on a daily basis have to be reckoned with. And I have hypotheses. I do believe it’s based in bias, but those things are all still up for debate and currently in the process of the research right now.

Ahmed Ali Akbar: People who speak AAE experience discrimination. At the same time, AAE is being increasingly adopted by non-Black speakers. Celebrities and kids on TikTok have been called out for using accents and dialects that they think approximate AAE.

Rachel Weissler: People like to use it because it makes you sound cool. It makes people sound inviting.

Ahmed Ali Akbar: The thing is, when non-Black people do this, they don’t experience the same real-world harm that Black people do for using AAE.

Rachel Weissler: And that brings up a whole other kind of warnings of what does it mean? Who gets to use it, and who benefits from using it versus who is discriminated against for using it.

Ahmed Ali Akbar:  nd a lot of times, non-Black speakers aren’t even using AAE correctly. Remember how I said dialects have a consistent grammar that in-group people use? Non-Black people often ignore or are ignorant of these rules.

Rachel Weissler: When I teach students, I’ll often show them tweets and say, look, this person used finna, but they used it in the wrong structured space. How would we correct the sentence to actually be grammatically correct? And things like this, right.

Ahmed Ali Akbar: But is this an example of a natural way language changes? What do you think about the kind of Gen Z argument that this is just the way we speak now, that non-Black folks have natively are incorporating because of the internet, African American English? How do you feel about that? [laughs]

Rachel Weissler: So, I mean, you know, it is not unusual for anything about Black people to be appropriated by the mainstream, whether it’s hairstyles or music or clothing. And of course, language falls right into that. And so to that argument, I say, yeah, you know, people are appropriating and they’ll continue to do so. And Black people will continue innovating and continue being on the forefront of linguistic change.

Ahmed Ali Akbar: Variation and change are the natural state of language. People have strong feelings about the way English ought to sound, what ways of speaking sound funny and what ways sound good. We have this immense capacity to learn about others via language. But we also have biases that we must work to overcome. The best way is to learn, about others’ dialects. About their grammar. About those sounds they make that are uncomfortable for us.There are no good or bad ways of speaking. Maybe, just maybe, I’ll start saying pop, and crayn, and night suit again.

Ahmed Ali Akbar: Radiolingo is an original podcast from Duolingo and Crooked Media I’m Ahmed Ali Akbar, your host, writer and producer. From Crooked Media, executive producers are Sandy Girard and Katie Long. From Duolingo, executive producers are Laura Macomber and Timothy Shey. This episode was produced and co-written by Elizabeth Nakano and story edited by Lacy Roberts. Our associate producer and fact checker is Brian Semel. Our theme and original music is by Carly Bond with mixing sound design and additional music by Hannis Brown. Additional research and production support from Crooked Media’s Ari Schwartz and Duolingo’s Cindy Blanco, Emily Chiu, Alexa Fernandez and Hope Wilson. Special thanks to Crooked Media’s Danielle Jensen and Gabriella Leverette and Duolingo’s, Michaela Kron, Monica Earle and Sam Dulsimer for promotional support.