An analysis of police body camera footage has found that officers consistently use less respectful language in interactions with black citizens than when they are talking to white people.
New research, published in the Proceedings of the National Academy of Sciences (PNAS), used a new artificial intelligence technique to analyze language from 981 traffic stops the Oakland Police Department (OPD) in California made in one month.
The study program looked at how frequently the officers used respectful utterances, such as apologies or gratitude like "thank you", as well as less respectful phrases, such as using informal titles like "dude" and "bro". In order to rate how respectful an utterance was, the team at Stanford University had college students read and rate about 400 terms for how much respect they showed, without knowledge of the race of the driver involved in the traffic stop.
Analysis of the transcripts from the interactions between police and citizens found that black members of the community were 61 percent more likely to hear an officer say the least respectful phrases, such as the command "hand on the wheel". Meanwhile, white citizens were 57 percent more likely to hear the most respectful phrases from an officer.
The study looked at half a million words spoken by the officers, controlling for factors that could have affected the officers' language, such as their own race or background, the severity of the infraction, and the outcome of the stop.
“Our findings highlight that, on the whole, police interactions with black community members are more fraught than their interactions with white community members,” Jennifer Eberhardt, co-author of the study and professor of psychology at Stanford explained, though she added that this might not equate to racial bias.
Though they found non-respectful language being used towards black citizens in routine police-citizen interactions, the researchers stressed that the officers were well behaved.
“To be clear: There was no swearing,” Dan Jurafsky, co-author of the study and professor of linguistics and of computer science at Stanford said. “These were well-behaved officers. But the many small differences in how they spoke with community members added up to pervasive racial disparities.”
The team at Stanford say that this study could help with training of officers, and lead to better relationships between police and citizens.
“I’m hopeful that, with the development of computational tools like ours, more law enforcement agencies will approach their body camera footage as data for understanding, rather than as evidence for blaming or exonerating,” Eberhardt said.
“Together, researchers and police departments can use these tools to improve police-community relations.”
The researchers are currently extending their study, to analyze the language used by community members during the stops, as well as tone of voice used by officers and citizens.
