Do you sound guilty? How accents influence criminal judgments

The way someone speaks often sparks quick assumptions. Even if those assumptions are wrong, they can shape serious decisions—especially in courtrooms. A new research project looked at how people judge others based on their accents, and the results point to troubling links between speech, social class, and criminal stereotypes.

Accents reflect where you come from, but they don’t reflect who you are. Still, people often use accents as shortcuts to guess someone’s background, personality, or even behavior. These guesses can lead to unfair treatment, especially when it comes to judging guilt or innocence in the legal system.

Researchers in the UK wanted to find out how strong these accent-based assumptions really are. As part of the Economic and Social Research Council-funded project called Improving Voice Identification Procedures, a team of scientists tested how people respond to different British accents. They wanted to know if certain voices seemed more “criminal” just because of the way they sounded.

“We found a strong link between perceived social status and the perceived likelihood of committing crimes,” said Alice Paver of the University of Cambridge, lead author of the study. “This link was more important than how trustworthy, kind or honest someone was perceived to be. This shows that perceived social class, as judged from a speaker’s accent, is an important predictor of UK listeners’ expectations about behavior, and this might have serious implications in the criminal justice system.”

The team included experts from several fields and universities, including Nottingham Trent. They recruited 180 people from across the UK. Each person listened to 30-second recordings of 10 male speakers, each with a different British accent. These included Belfast, Birmingham, Bradford, Bristol, Cardiff, Glasgow, Liverpool, London, Newcastle, and Standard Southern British English (SSBE).

The recordings were carefully controlled. Scientists removed any names or background noise and adjusted the speed, pitch, and volume so that only the accent stood out. Language experts confirmed that the voices represented typical speakers of each accent.

Participants were divided into two groups. One group rated the voices on 10 social traits—things like honesty, kindness, or wealth. The second group rated how likely it seemed that each speaker had committed certain acts. These included five crimes—such as theft or assault—and five non-criminal but morally charged actions, like defending someone from harassment or cheating on a partner.

Once the responses came in, the researchers looked for patterns. They grouped the traits into four clusters. Two were major: “solidarity,” which included traits like warmth and kindness, and “status,” which included wealth and confidence. Two minor clusters appeared as well: “confident” and “working class.” The “working class” cluster represented the opposite of “status.”

SSBE, often considered the standard or ‘neutral’ UK accent, scored highest for status and confidence. It scored lowest for the working-class cluster. In contrast, more regionally distinct accents like Liverpool or Birmingham ranked lower in status but higher in the working-class group.

When it came to crime, things got more complicated. People linked lower-status accents more often with criminal behavior. But this pattern didn’t hold for all types of crime. Most crimes clustered together. However, sexual assault ratings grouped differently—alongside morally negative behaviors that weren’t always illegal.

Interestingly, some accents stood out. Listeners saw London and Liverpool accents as more likely to commit morally bad actions. Belfast and Glasgow accents were seen as less likely to do so. But just because an accent sounded less criminal didn’t mean it sounded more heroic. For example, the SSBE accent scored low for being likely to defend someone from harassment.

“We didn’t see a strong link between how criminal someone’s voice sounded, and how kind or trustworthy they sounded,” Paver explained. “Instead, there was a much more important link between how criminal a voice sounds and how working-class a voice sounds.”

In other words, people assumed that a speaker who sounded working-class was more likely to commit a crime, even if they also thought that speaker might be kind or honest.

Another key finding was how people judged sexual offences. Most crimes followed the same pattern—lower-status accents were seen as more likely to commit them. But sexual assault didn’t fit this mold. Listeners didn't rate working-class-sounding voices as more likely to commit sexual offences than others.

This break from the pattern might suggest a shift in cultural stereotypes. In the past, people may have unfairly linked sexual assault with working-class men. But now, people may be more likely to associate that type of crime with people from higher social classes.

The researchers stressed that this result could have serious legal consequences. If jurors or witnesses bring their accent biases into the courtroom, they may judge a suspect unfairly—even without meaning to.

The study shows that accent-based judgments can lead to real problems in legal settings. If people hear a suspect or witness speak and assume things about them, it could influence court outcomes. This isn’t just a theory. It’s a real risk.

The team behind the study is working on new voice identification guidelines. They want courts to test for voice bias before using voice recordings as evidence. And they hope legal professionals learn to recognize their own accent-based prejudices.

“The team are currently drafting new guidelines for the implementation of voice line-ups,” said Paver. “We support the use of pre-testing to screen for voice bias. We also hope that anyone encountering voice evidence in the criminal justice system is warned against letting voice- or accent-based prejudice influence their decisions. These stereotypes could have real-life legal consequences.”

The researchers also said more work needs to be done. Future studies should include women’s voices and test a wider range of accents. It’s also important to look at how strong an accent sounds and whether that affects how people judge character.

This research doesn’t just open our eyes to how we hear others. It also challenges us to think about fairness. If the sound of someone’s voice can shape how we judge them, even without facts, then justice may not always be blind.

Research findings are available online in the journal Frontiers in Communication.

Note: The article above provided above by The Brighter Side of News.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.