Humanities, Research

Berkeley Voices: A language divided

Rachel Jeantel was on the phone with Trayvon Martin moments before he was killed in 2012. But when she testified at George Zimmerman's trial, the jury deemed her an unreliable witness. Why?

Follow Berkeley Voices, a Berkeley News podcast about the people and research that make UC Berkeley the world-changing place that it is. Review us on Apple Podcasts.

See all Berkeley Voices episodes.

Rachel Jeantel on the stand during her testimony in the trial of George Zimmerman
Rachel Jeantel (pictured) was on the phone with 17-year-old Trayvon Martin moments before he was killed in Sanford, Florida, in 2012. But when she testified at the trial of George Zimmerman, the man who killed Martin, the jury deemed her an unreliable witness. Zimmerman was later found not guilty.

Jacob Langston/AP

There are countless English varieties in the U.S. There’s Boston English and California English and Texas English. There’s Black English and Chicano English. There’s standard academic, or white, English. They’re all the same language, but linguistically, they’re different.

“Standard academic English is most represented by affluent white males from the Midwest, specifically Ohio in the mid-20th century,” says UC Berkeley sociolinguist Justin Davidson. “If you grow up in this country and your English is further away from that variety, then you may encounter instances where the way you speak is judged as less OK, less intelligent, less academically sound.”

And this language bias and divide can have devastating consequences, as it did in the trial of George Zimmerman, who shot and killed 17-year-old Trayvon Martin in 2012. 

This is the second episode of a three-part series with Davidson about language in the U.S. Listen to the first and third episode: “A linguist’s quest to legitimize U.S. Spanish” and “One brain, two languages.”


Read a written version of the podcast episode:

In 1996, the Oakland Unified School District passed a resolution that recognized Black English as a language variety, distinct from what’s widely considered standard English.

“The idea at the time was there were many African American students who didn’t have literacy levels in academic, effectively white, English that was required of them in the school system,” said Justin Davidson, an associate professor and sociolinguist at UC Berkeley.

“So they were trying to find ways to mitigate that. And one of the solutions was to ask, which seems very sound: ‘Why are we expecting people whose home language is African American Vernacular English (AAVE) to be suddenly completely fluent and/or proficient in an English that’s not theirs, that was not used in the home, that they were not exposed to, probably until the school system?’

“So they thought: Well, one of the ways to mitigate that is, Why don’t we have bilingual education with, effectively, Black English and white English?”

Oakland schools piloted a bilingual literacy program called Bridge that acknowledged that Black English is real — and teachers used it to help students connect to and learn standard English.

“So effectively,” Davidson said, “by teaching students in their native variety, that helps to then acquire information, knowledge and then jump to a different variety that society sort of prescribes that they need to have in order to be successful in the school system.

“And it lasted over a little over a year, and it was extremely successful. Their figures of literacy rates advanced, like, by 300%. It was great.”

But there was a general misunderstanding of the program and its intentions, and it was met by widespread criticism and ridicule. The dominant narrative that took shape was that Oakland was trying to dumb down education by accepting Black English as a legitimate language of its own, and even teaching it to students.

In reaction, California Congress passed a bill that mandated that no federal funding could be used for bilingual education programs that involve any variety of language connected to English.

“So it means that a program like this will not receive any federal funding because it doesn’t count as bilingual,” he said. “It’s just two Englishes, and we don’t fund that. So they lost their funding, and the program stopped.

“And the effects of that today are that, again, you cannot get any federal funding or support of any kind if the language in question is connected to quote-unquote ‘English.'”

There are between 5,000 and 7,000 different languages used today in the world, said Davidson. Linguists haven’t studied or documented all of them, so they’re working with a relatively small sample and tend to focus on languages that are less marginalized. So the majority of linguistic theory and linguistic research has been on English varieties.

“In societies where education is a main staple and mandatory, the educational system is where we get all sorts of social information about what’s intelligent, and therefore what’s acceptable,” Davidson said. “And the kinds of English and other languages that we teach in the school system reflect certain Englishes, certain dialects, certain varieties of English, and denigrate others.”

Everyone has linguistic bias, he said, and it’s conditioned based on society and our environment.

In the U.S., for example, standard academic English is a variety most represented by affluent white males from the Midwest, specifically Ohio.

“What comes from that is that if you grow up in this country and your English is further away from that variety, then you may encounter instances where the way you speak is judged as less OK, less intelligent, less academically sound, etc., etc., etc.

“The types of stereotypes that are out there about, I don’t know, Southern English — Texas English, for example, rural, hick English, that sort of thing — that’s the exact same forces at play. It’s that the way that certain people produce language is inferior to others. And all of that isn’t, for linguists, an actual question of inherent goodness. It’s all just social dynamics. It’s who is in power, and who are the people making decisions about what sounds intelligent.”

“A really famous example is, we know the SATs, the vocabulary and, sort of, the language section of the SATs, those were developed by a group of affluent white male speakers to address college-bound students of the time,” said Davidson. “And at the time, college-bound students were just that group: affluent white males.”

The SAT was first administered by the College Board to about 8,000 college applicants in 1926. In the decades that followed, the “scholastic aptitude test,” as it was known, became required to attend most colleges in the country.

And while more than 2,000 colleges across the U.S. have, in recent years, stopped requiring the SAT for college admission, including all UC schools, others have actually reinstated the SAT, including MIT, Yale and the public university systems in Georgia and Florida.

“Of course, these exams are written by and for certain groups in society,” Davidson said. “And so they don’t have to make an academic leap, right? And so they don’t have to learn a new dialect of English or a new dialect of whatever language they speak in order to do well on these exams. And so it’s not surprising then that we see, not just in the United States, but across all countries everywhere, racial and linguistic disparity in standardized exams.

“So there are all sorts of linguistic injustices pervasive to every society, to every country. And sociolinguists are working hard to try and make these issues more known, more part of public discourse, so that ideally, we address these things in at least the school system.”

The California bill that cut off federal funding for bilingual education programs that involved varieties of English is still in effect today.

“For example, if you go to court, you are not afforded an interpreter if you use Black English because it doesn’t count as another language,” Davidson said. “It’s a sociopolitical question. Linguistically, Black English and white English are not the same in the same ways that California English is not the same as Texas English, is not the same as Boston English.

“But the way that we recognize languages politically in this country, African American Vernacular English does not count for funding like that.”

A recent consequence of this happened in the trial of George Zimmerman, who shot and killed 17-year-old Trayvon Martin in 2012 while Martin was visiting family in a gated community in Sanford, Florida. Zimmerman, who was in a neighborhood watch group, was found not guilty. His lawyers argued that he was acting in self defense.

“And there was a witness, Rachel Jeantel,” said Davidson. “She was a witness to Trayvon Martin’s death. And her English is not white academic English. It’s African American Vernacular English. There were no interpreters involved in that court case. So the jury heard her testimony, and the case proceeded.”

Jeantel, who was on the phone with Martin moments before he was killed, testified before the jury for nearly six hours. But her testimony wasn’t brought up a single time during the jurors’ some 16 hours of deliberation.

After the trial, jurors gave interviews on national news programs. And later, linguists from Stanford University gathered these interviews, re-read transcripts and reviewed other investigators’ interviews with jurors to see if the jurors understood Jeantel’s words.

“And they didn’t,” Davidson said. “There was a language barrier. So the jury didn’t understand what she was saying. And she was written off as this uneducated, unintelligible, non-reliable witness.

“And it’s literally a question of language discrimination. If Rachel spoke German, there would have been a translator obligatorily afforded so that the jury could understand them. But not for English and English varieties.

“So that’s a very recent and impactful case where our understanding of what counts as a language really matters.”

Most people in the world, Davidson said, don’t have any exposure to thinking about language in this way.

By understanding — and, ideally, appreciating — linguistic diversity, he said, we can begin to contextualize our own reactions toward language varieties that don’t align with standard academic English. It’s the first step, he said, toward lessening the negative effects of linguistic bias on society.