The Hechinger Report is a national nonprofit newsroom that reports on one topic: education. Sign up for our weekly newsletters to get stories like this delivered directly to your inbox.

Justine Cassell
Justine Cassell

Using a computer to help students learn subjects like science and math sounds like it would be far from the controversial thicket of racial politics, but sociolinguist Dr. Justine Cassell manages to stir up emotions anyway.

Cassell, an award-winning researcher who is associate vice-provost for technology strategy and impact at Carnegie Mellon University, believes that educational technology can play a crucial role in helping black children learn academic content like science and math. In fact, Cassell’s research appears to show that black children can learn more math and science using technology if the computer instructs them while speaking in a black vernacular, also known as black English.

While many educators—black and white—still rebel against the idea of any semblance of black English being used in an educational setting, Cassell, who is white, thinks the issue of language and dialect is so important and so crucial to learning that she refuses to back down.

Related: Poor students often lack a home Internet connection. Is this FCC program a solution?

Cassell and her team have even pioneered a technology to help with the instruction. They created a computerized avatar whose race and gender are so thoroughly ambiguous that black students say the avatar is black, white students say it’s white, boys say it’s a male and girls say it’s a female. The avatar’s clever name? Alex—a name that is also racially and sexually ambiguous.

Cassell, who currently serves as chair of the World Economic Forum Global Agenda Council on Robotics and Smart Devices, felt the negative connotations of speaking dialect while she grew up in Brooklyn. Her Brooklyn accent was so strong that her parents thought it would become a hindrance to her. They recorded her speaking and played it back to her so that she could hear how she sounded—and work to erase the accent.

Question: You’ve spent a great deal of time studying the interaction between people and machines. How does your research intersect with questions of educating poor children and children of color?

Answer: I’m going to contrast my work with some really obnoxious work that’s been done making educational technologies that supposedly look like students of color. So there’s been work done in the past where avatars, or agents, were made in multiple varieties. So there’s a white woman, white man, black woman, black man, who are math tutors on the computer and students are allowed to choose the one they want. I read such a paper more than a decade ago and what I noticed was the black tutor was wearing pants down around his knees and gold chains around his neck and a backwards baseball cap. That’s not a black math tutor. That’s some computer scientist’s stereotypical image of what a black person looks like. A very negative stereotype. And what the results of that study showed was that students preferred to work with an agent of their gender and ethnicity, but that they learned the most math from the white male. Nasty, right? Maybe that’s not surprising, if the black male looks like that, because that’s not our vision of a computer tutor. So I started thinking about how culture gets integrated into educational technology. And I think if you asked most technologists, they would tell you that their technology was culture free. But, of course, that’s not the case because most educational technologies that have images of people embedded into them, what’s called pedagogical agents, have images of white men. So what culture free really means is the dominant culture and not culture free. That is what I decided to counteract.

Related: Lack of diversity among tech leadership could stall advances for children of color

Q: How did you address that?

A: What occurred to me was that in many of these technologies that have images of black people, even if they don’t go as far along the stereotyping road as gold chains and backwards caps and pants around your knees, the black agents speak exactly like the white agents do. And yet we know that identity is carried by not just how we dress ourselves but how we choose to speak in different environments. And this is a delicate topic in America. This is something we have a lot of trouble talking about. I grew up in Brooklyn. I had a very strong Brooklyn accent when I was a kid. And my parents, who were really intent on moving up from one social class to another, made me listen to tape recordings of myself and told me the way I was talking was bad. As you can tell, I lost the Brooklyn accent, although when I’m sick or sad I tend to get my Brooklyn accent back again. In that same way, the talk of how we speak for other cultures, what you might call vernacular, has received a lot of negative press. But we know every person in the world speaks differently in different contexts. Children in many black communities may grow up speaking vernacular in their homes. There is now a fair amount of research showing that working class and poor kids continue to speak vernacular more than middle class black kids, who tend to acquire the ability to switch back and forth to what I would call standard American English, code switching. Oprah Winfrey claims any use of the vernacular at all is a sign of poverty and ignorance and should be eradicated. And so she laid herself open to having some sociolinguists do research on her and it turns out that when she has a black guest on the show, she switches to the vernacular. And we’ve heard the president do that same thing over and over again. And one does that because just like in all the other ways we code switch, it’s a way of indexing commonality, building rapport. So the question I wanted to ask was whether educational technologies could play a role in two things: one, in understanding the way young black children use language, and two—since we live in a society where whatever we believe about dialect, power is connected to the mainstream dialect—could educational technologies help children learn to code switch? So that was the question I started out to ask. And while it helped them learn to code switch, could it help them learn school topics as well, like science and math?

Question: What did you find?

A: We showed an avatar to kids and we told them it was an avatar that a kid in another class on the other side of Pittsburgh chose to represent himself or herself. We said, “You guys are going to do a science task like pen pals. He recorded a tape for you, to tell you about himself. Then he recorded the science task. And you’re going to record a tape so he can get to know you and send back your thoughts about the science task.” We did this using three conditions. We had an all-vernacular condition, where the social part at the beginning of the tape was in vernacular and the science part was in vernacular. We had an all standard condition where the whole tape was in standard English. And we had a code-switching condition, where the social part was in vernacular and the science was in standard English. What we found was the condition where the kids did the best science reasoning was the one where it was all vernacular. Science reasoning means talking about hypotheses and evidence using data to support their hypotheses. It’s kind of a standard way of assessing science learning. So they did better science when they did science in the vernacular. Not only that, other results showed when they spoke the vernacular, they spoke at a normal rate. They would say, “My name is Alex, and I like basketball, and I’ve got two sisters, and they do this.” And then they would say, “Okay let’s do the science task.” The kids who worked with the standard agent had this hyper articulation. “Hii, my name is….” They spoke really slow and in a really labored way. And when we did an acoustic analysis on it, we found that that labored style of speech correlates with cognitive load. So they might not have had enough cognitive capacity left to do the science because they were using so much cognitive capacity to force themselves to speak the standard. You can read that paper, it’s on my website.

Related: Wanted: More solutions for solving the homework gap

Q: This issue is incredibly complicated in the black community, particularly among the black middle and upper middle class. Many parents would not be pleased if they heard someone was trying to teach their children using black English.

A: I have been denied access to schools when I explained what I was doing. That’s a kind of respectability politics, if you want to call it that, that I understand. This is a topic that gets me into trouble all over the place. But it seems to be such an important one because it’s just life. We know multilinguilism has cognitive benefits for children. Children do algebra better if they’re multilingual. We know that being able to move in different circles has both social and cognitive benefits for kids.

Q: How are you going to negotiate the difficult politics in order to get your research to the places that would use it most effectively?

A. What I’ve started to do, not something I was initially interested in—I’m more of a sociolinguist than an education person by training—but I’ve become a lot more interested in teacher training through this research. Because teachers are getting very mixed messages. On one hand school systems are telling them that culture is essential and you have to value everyone’s culture equally. And that’s why they celebrate Kwanzaa, these very superficial but visible signs of culture. On the other hand, they’re being told their job is to make sure that those children speak like scientists, speak like learned people. To those teachers and to the school system’s office, that means speaking the standard. And so what we’ve started to do is use those same virtual children for teacher training that we used with the students. Use them to have a conversation with teachers about ways of switching, about context dependent language use. It’s a really interesting thing to do in Pittsburgh because Pittsburghese is a really strong dialect. It’s very much related to Pittsburgh culture, really strong culture, Steelers, steel industry, we’re the underdogs. It’s a really interesting place to be. What I found is if I start the conversation by talking to teachers about how they grew up speaking and about whether their way of speaking changes when they go back home for the holidays, I can broach the topic in terms of their own code switching and then I can show them videos of kids using technology and looking joyful in their science learning in a way that these teachers may not see in their own classrooms. And then we can start to talk about whether it would be okay to allow children to talk however they want to talk when they’re among themselves, but still be mindful of the politics of the US and teach them about contextual use. And teach them about what they’re going to need when they leave school and go into the outside world. That a teacher is one person you need to talk to and kids are other people and you wouldn’t use some of the words with your teacher that you use with your friends. And can that topic be broached using these virtual children as a stepping stone to understanding context sensitive dialogue use?

Related: Theater helps English learners master Common Core: But can it close the achievement gap?

Q: If you were able to promote and advocate for and get the widespread use of this technology, would that be a way of getting beyond unconscious bias against children of color that some researchers have contended occurs with teachers?

A: I hope so. I really, really hope so. And it’s not just white teachers. Bias against dialect happens just as often among black teachers. I wish I worked with more black teachers. I’m working with an African-American charter school and they don’t have a lot of black teachers. I had an amazing moment with a black teacher that keeps me going. The first day we were talking I was giving her an example of what dialect sounded like and I said, “I be walking.” She said, “If I hear a kid say ‘I be walking’ I would say, ‘You be walking right out of my classroom and into the principal’s office.’” That was the first thing she said to me. I thought, this isn’t going to go so well. But we had this long conversation about how people talk and how she talks and how she teaches her kids to talk and how she has taught her kids to eradicate any sign of dialect. I said, “You know there are signs of culture that you don’t put on always but you need sometimes.” So I ran into her recently and she said, “I gotta tell you, you have changed the way I work with my students. What I realized was that the vernacular is celebrated in our poetry, in our singing, in our novels, and that if I badmouth it, aren’t I badmouthing my own culture? So I started talking to the kids about different ways of speaking and I stopped telling them to speak right. I told them to use their school English.” My heart swelled, I was just so thrilled. Wouldn’t it be nice if they all said school English? If that allows them to believe that these kids are gems all the way through; they’re not indicating they’re going to end up on the street. They’re reflecting a rich culture, a rich cultural tradition. I have as many examples of friends who were told by their parents, “If you speak vernacular at home I will beat the crap out of you,” as friends who have gone to visit their grandmothers who say to them, “What happened to you?! As soon as you go to school in the north you come back speaking like that? Don’t you speak like that to me, kid, or I’m going to whack you.” So it goes both ways. And I would love that to be more normal, natural, recognized, a part of existence, so that kids can get the full promise of an education, as opposed to being corrected and never getting the science. Because I hear that all the time: I hear teachers correcting kids, telling them, “Say it right” and then never answering the child’s question.

The Hechinger Report provides in-depth, fact-based, unbiased reporting on education that is free to all readers. But that doesn't mean it's free to produce. Our work keeps educators and the public informed about pressing issues at schools and on campuses throughout the country. We tell the whole story, even when the details are inconvenient. Help us keep doing that.

Join us today.

Letters to the Editor

At The Hechinger Report, we publish thoughtful letters from readers that contribute to the ongoing discussion about the education topics we cover. Please read our guidelines for more information. We will not consider letters that do not contain a full name and valid email address. You may submit news tips or ideas here without a full name, but not letters.

By submitting your name, you grant us permission to publish it with your letter. We will never publish your email address. You must fill out all fields to submit a letter.

Your email address will not be published. Required fields are marked *