Valerie Fridland writes in her new book, Why We Talk Funny: the Real Story Behind Our Accents, that humans instinctively to use accents to categorize those around us. “We learn to recognize other people as being like us through the way that they sound,” Fridland says. It happens early: studies suggest small children, when choosing friends, favor those who share their accent.

In one study, for instance, five- and six-year-olds were shown pairs of kids on a computer screen, one with a local Canadian accent and one with a British accent. Asked who they wanted to be friends with, they picked the kid with the local accent – even though they lived in Toronto and are exposed to a huge range of accents every day.

Our accent-based judgments lead to serious problems, fueling stereotypes about class, ethnicity and regional background. That can take a toll in a range of high-stakes scenarios, including job interviews, when someone with a posher accent might be deemed more capable than someone with a more working-class one. It can lead to assumptions about how someone thinks, as in a study that found subjects assume politicians with southern accents are making conservative arguments. It can even affect the way juries react to witnesses, as Fridland believes happened in the trial of George Zimmerman.

Fridland, a linguistics professor at the University of Nevada, Reno, is herself a case study in how accents develop. She grew up in the American south – Memphis, Tennessee – with parents who had French accents. “Being surrounded by people who were very aware of outsider opinions of their accents primed me from a young age to be curious about why they were such markers of identity,” she says.

Accent discrimination has been around since ancient times, Fridland writes; it even comes up in the Old Testament, when one Semitic tribe, the Gileadites, identified enemy Ephraimites by demanding they pronounce the word “shibboleth” with ruinous consequences if they got it wrong.

Then said they unto him, Say now Shibboleth: and he said Sibboleth: for he could not frame to pronounce it right. Then they took him, and slew him at the passages of Jordan

We often have kneejerk reactions to accents “because we don’t understand their value”, Fridland says. Developing this understanding – and simply being aware of the potential for bias – can go a long way toward fighting our own prejudices.

Our awareness of accents begins almost as soon as we arrive on the planet. “By about a year old, babies have already figured out the sounds that are prevalent in the languages around them,” Fridland says. Research shows one-year-olds respond to sounds that exist in their languages and not to those that are absent. Over the next few years, they pick up language from their parents and others around them. But it’s not until about age five that their own accents really emerge. At that age, Fridland says, children lose interest in their parents’ speech and decide: “‘Let’s go talk to these cool peers at school, because they’re way, way more fun.’ And that is when the accent starts to really ramp up.”

This explains why an American kid whose parents have French accents, like Fridland, ends up sounding like her peers rather than her parents – and when the reverse occurs, it stands out. Thus it was striking to her childhood classmates when, during show and tell, she described her teddy bear as “yuge” rather than “huge”.

“I spent the rest of that year so aggressively pronouncing my ‘H’ wherever I thought it should go that I became known as the girl with the spitting habit,” she writes.

How these accents sound is, of course, rooted in history. Take, for instance, the fact that American English tends to be rhotic – most of us pronounce the “R”s before consonants and at the ends of words such as “hard” and “car” – while many British speakers do not (there are, of course, exceptions on both sides, for instance in a Boston, Massachusetts, accent or a West Country accent). That’s because the dropping of the “R” became fashionable in London only after the first American colonies were founded, as rapid social change in the city in 19th century fueled changes in pronunciation.

Once we have our accents, and especially once we reach adulthood, the way we talk is extremely difficult to change. Learning a new language provides particular challenges as it requires creating novel sounds – sounds we might not even be able to hear, let alone recreate. For instance, English speakers often aren’t familiar with Swahili’s “mb” sound, so they may hear it with a vowel in the middle and incorrectly add that vowel when speaking. Spanish never has an “st” sound at the beginning of a word, so native Spanish speakers sometimes report hearing a vowel at the beginning of words like “student” and add it in speech.

On top of that, to sound like native speakers, we have to imitate the prosody of a language: its rhythms, tones and stresses. Americans, for instance, “typically increase the duration and loudness of whatever word they may want to highlight in a sentence” and lower their pitch on the last word of a sentence to indicate they’re done speaking, Fridland writes. In Mandarin, on the other hand, is a “syllable-timed language”, meaning “every syllable is said with roughly the same duration and intensity”, while tone is essential to the meaning of a word.

Even in our own languages, changing our accents is challenging. An American who moves to Surrey is unlikely to ever sound truly English. Instead, they might develop what Fridland calls a “blended dialect”. “When people are long-term residents of a dialect area that’s not their own, we do find something called speech accommodation that happens, which is where they move closer to the norms of that region without actually replicating the norms very well,” Fridland says. They don’t sound quite like a local in their new home town, but they don’t sound like a local from their birthplace, either.

The effect intensifies the more we make friends and build community in a new home; listen, for instance, to the writer Bill Bryson, who grew up in the US but has spent most of his adulthood in the UK. (This phenomenon can even occur ever so slightly in the span of a conversation: “You and I talking, if you measured us at the beginning of our conversation from an acoustic standpoint, and then someone measured us at the end of our conversation, our speech would have started to converge” in pitch, accent and vocabulary, Fridland says.)

Given the deep roots of our accents, it’s all the more unfair to judge people according to them. Unfortunately, we do, consciously or unconsciously.

In the trial of George Zimmerman, who was accused of murdering 17-year-old Trayvon Martin in 2012, the prosecution’s key witness, who had a strong African American Vernacular English (AAVE) accent, was “largely dismissed as incomprehensible and not credible”, Fridland writes.

As the linguists John R Rickford and Sharese King explained in 2016, Rachel Jeantel, one of Martin’s friends, testified for six hours – longer than anyone else in the trial. She had been on the phone with him moments before Zimmerman allegedly killed him. But Jeantel was reportedly not mentioned in jury deliberations. “In a sense, Jeantel’s dialect was found guilty as a prelude to and contributing element in Zimmerman’s acquittal,” Rickford and King write.

There have been similar cases in the UK, and a study last year found that people with working-class accents were more likely to be suspected of crimes. “There’s a really good, solid body of literature that suggests that having a non-standard accented speaker or heavily regionally accented speaker can influence credibility ratings of jurors and, in fact, increase attribution of guilt,” Fridland says.

But there are simple ways to minimize the harms these unfair reactions can cause. Often, simply recognizing our biases can go a long way. Studies have shown benefits when an employer is reminded to focus on a person’s innate capabilities rather than how they talk. A simple desire not to appear biased – regardless of the purity of the employer’s intentions – can be enough to mitigate prejudice. To that end, linguists have helped develop jury instructions to reduce bias.

Ultimately, Fridland says, the way we talk is rooted in a universal experience. We start with the same equipment – our mouths and our minds – and are driven by the same thing: the need for social belonging. In Fridland’s experience as a linguist, she has found “most people are genuine in wanting to be better listeners”, she says.

“There are some assholes out there, but the majority of people, I think, given the right tools, want to do better.”