Faculty Spotlight: Annette D'Onofrio

Linguistics scholar examines how speech defines our perceptions



Annette D'Onofrio
IPR associate Annette D'Onofrio studies how language defines our perceptions.

Growing up in Minnesota, linguistics scholar and IPR associate Annette D’Onofrio was always attuned to the nuances of language. Her mother’s first language was Korean, and her father has a Long Island accent.

She originally planned to major in English, but taking a linguistics class her first year at the University of Pennsylvania made her realize how much she loved the subject—and how it challenges the idea that there is a right and a wrong way to speak.

D’Onofrio’s academic career took her to different parts of the country, introducing her to a “hybrid” of various features of speech. People often do not notice their own dialects until they use certain words. For her, that happened when she said diagonal to a group of fellow graduate students at Stanford University.

“They were like, ‘Whoa, where did that come from?’” D’Onofrio said. “Moving to the East Coast and then West Coast for school very quickly made me aware of the features that I had. And I always say to my students, ‘Don’t lose those features. They’re part of your identity.’”

Identity is key to understanding why people talk the way they do, D’Onofrio said.

“How we speak—not just what we say, but how we speak—affects how other people perceive us,” D’Onofrio said.

The Chicagoland Language Project

In leading the Chicagoland Language Project, D’Onofrio has been examining a typical Chicago accent, otherwise known as the “Northern Cities Vowel Shift.” The ongoing project is designed to understand language and identity in the city, starting with the Beverly and Morgan Park neighborhoods on the South Side. 

One of the distinguishing features of the Chicago accent is the way that the short A is pronounced. Elements of the Northern Cities Vowel Shift like the short A were strengthening over the last 50 years, but now that is starting to change.

“What we're finding is that the younger speakers in these communities—even though they've grown up in Chicago their whole lives, they live down the street from their parents, and they're still in Chicago—they're starting to have less of that accent,” D’Onofrio said.

Understanding why the accent is receding is the big question D’Onofrio and her fellow researchers are examining. She is exploring a few theories: One reason might be that a social stigma became associated with the accent, so people get rid of their accent. Another could be that people have become more mobile, leading to more non-Chicagoans in the community. But D’Onofrio said those explanations do not seem to apply to Beverly and Morgan Park. So, to dig deeper, D’Onofrio asked the study participants, “Who speaks with a Chicago accent?”

“They'll mention police officers, firefighters, and union members,” D’Onofrio said.

Younger people, she explains, might be shifting towards different types of occupations that are less affiliated with the working class, or wanting to orient away from the types of people whom they associate with the accent. 

Another possible explanation—though D’Onofrio added that it needs further research—is that white speakers started using the Northern Cities Vowel Shift during the Great Migration (1916–1970), as a type of “symbolic white flight,” to differentiate themselves from African Americans moving out of the South into Northern, Western, and Midwestern cities. The recent disappearance of the vowel shift in Chicago, then, could be a way for younger generations to distance themselves from those ideologies.

How Politicians Speak

D’Onofrio, with IPR sociologist Beth Redbird, is also examining what meaning we can draw from the way a politician speaks on the campaign trail.

The two researchers have been studying the stump speeches of then-presidential candidates Barack Obama, John McCain, and Mitt Romney in 2008 and 2012 to examine when they each dropped the “g” at the end of a word ending in “-ing,”—for example, “working” versus “workin’.”

For Obama, they found that the overall education level of the surrounding county where the campaign rally was held influenced his use of "-ing." He said “-ing” more in places that had a higher level of education and when discussing specific policy matters. He used “-in” more in counties with lower education levels and when talking about issues related to “ordinary Americans.”

One reason for this difference, D’Onofrio posited, could be that it was a way for Obama to show that he could be simultaneously easily relatable and qualified for the job.

McCain, however, used “-ing” 90 percent of the time, no matter the issue or location.

“There’s something to this idea of, ‘I'm authentic. I speak the same way no matter where I am’ that he [McCain] might have been playing up,” D’Onofrio said.  

How Expectations Change What We Hear  

Our social expectations can also influence the way we hear people speak. In one study, D’Onofrio examined three social personas: the “valley girl,” and the “business professional,” and the “Chicago Bears fan.” She played the same audio clip—for example, a clip that was somewhere between the word “sack” or “sock”—to different people.

Some were told they were hearing a valley girl, while others were told they were hearing a business professional, and a third group was told they were hearing a Chicago Bears fan. That description influenced how they categorized the same audio clip; some heard “sock,” while others heard “sack,” depending on their expectations of how these different types of people would pronounce these vowel sounds.  

What is interesting, D’Onofrio noted, is how a simple vowel sound can sway our perceptions of others. 

“We have ideas about how a valley girl might sound, how a business professional might sound, and that actually shapes what word you're hearing from the same signal,” she said. “[It shows] that we have these representations in our minds of certain types of people and how they speak, and that can actually influence how we take in their language.” 

D’Onofrio has applied this concept to race, too. In a forthcoming paper, D’Onofrio is examining how visual cues can impact how much of an accent Americans detect in a person of Asian descent. Despite hearing the same audio, those who saw a photo of someone who dressed like a stereotypical Korean celebrity, rather than a person dressed in a manner more typical of an American university student, heard more accented speech. Furthermore, this influenced how likely they were to remember what the speaker was saying.

“Just based on this picture, they think [the audio] sounds accented, and therefore they don't encode it as well, which is kind of scary, if you think about it,” D’Onofrio said. 

D’Onofrio’s work is helping to uncover the bias embedded in the response to speech. Policing how people talk is “another form of inequality that doesn’t get talked about as much, because the assumption is that you can learn to not speak this way,” D’Onofrio said. “We think we have more control over that aspect of our identities than other aspects.”

Annette D'Onofrio is assistant professor of linguistics and IPR fellow.

Photo by Christen Gall.