One of a five-part series
The flat affect: how a teenage mother’s impassive demeanor left an “indelible” impression at trial
But experts say an accused’s demeanor doesn’t reveal guilt—or innocence
By Erin Bacon, Ethan Cohen, Madeline Fox, Drew Gerber, Anika Henanger, Sophie Hoblit, Marissa Mizroch, Isabel Schwartz and Katie Way
The Medill Justice Project
An 18-year-old mother sits in a cramped room across the table from two investigators with children’s protective services. Her 11-week-old daughter, Alicia, lies in intensive care. They start asking the mother: What happened? Did you violently shake your daughter?
“During our entire interaction, she had a very flat affect; no emotion,” Robert W. Peck Jr., one of the investigators, testified at Tonia Miller’s murder trial about a year and a half later.
In an interview for this article, Miller says she gave off nothing—she wore a blank expression. But behind her emotionless wall, she says she was scared. She just wanted to see her daughter. She felt desperate, trapped.
“I just remember being so completely confused and upset that they wouldn’t let me see her, and then just angry about them accusing me of something,” Miller says. She remembers “feeling completely helpless.”
Daniel Buscher, the lead prosecutor, brought up Miller’s demeanor 23 times during the three-day trial in April 2003. Buscher and witnesses for the prosecution referred to what they called her “flat” affect 10 times.
“Everyone at the hospital was incredulous as to her demeanor. Flat. No emotion,” Buscher said in his opening statement. “Didn’t even want to be near or touch her own child. Is that what you would expect of any parent, let alone a mother whose child had just died or was going to die? This isn’t a matter of being in shock…Now why would a mother whose child is about to die be very guarded in her answers…? One thing. She was guilty, she got caught and the walls were coming down on her.”
Miller was convicted of second-degree murder and sentenced to 20 to 30 years.
Buscher did not respond to multiple requests for comment for this article.
At least two jurors were struck by Miller’s impassive demeanor in the courtroom, according to interviews with The Medill Justice Project. And an alternate juror said the prosecution “made a big deal” of how a nurse testified about the “passive” demeanor of the accused teenager. What’s more, Miller’s defense attorney then, Edwin L. Hettinger, says of her detached demeanor: “It was huge. They were slamming us from the start on that.”
The flat affect of an accused has been used in many other criminal cases to great effect. According to a 1990 Newsday article, a prosecutor questioned whether the demeanor of Marty Tankleff, a 17-year-old accused of murdering his parents, was that of “an intelligent, cold and self-centered person with a convenient loss of memory.” Tankleff was convicted but won his freedom on appeal. Ronald Cotton looked “guiltier and guiltier” as his trial went on, according to a juror interviewed by Frontline in 1997. Cotton, who was convicted of rape and burglary and then exonerated by DNA evidence more than 10 years into his sentence, maintained a flat affect throughout his trial, the juror said. Publications around the world demonized Amanda Knox, the college student who was famously accused of murdering her roommate while studying abroad in Italy, for her apparent lack of remorse. In 2014, before she was found not guilty, the Daily Mirror ran a cover story branding her “The Ice Maiden.”
But interviews with many experts in body language, forensic psychology and law enforcement say uniformly there’s little if any connection between an accused’s demeanor and guilt or innocence.
“Body language and facial expression are very, very poor ways to determine guilt,” says Susan Bandes, Centennial Distinguished professor of law at DePaul University in Chicago.
In Miller’s case, the prosecutor didn’t just mention her demeanor many times, he also asked several witnesses about it.
“She seemed very passive,” testified Debbie Lillibridge, a registered nurse in the emergency department where Alicia was admitted. “I recall her standing two to three feet from the patient’s bed. No emotional outbursts or anything. Just stood there, looked at the baby for a few minutes and then left the bedside area and went back to a family room.”
In his testimony, Russell Bell, the other investigator with children’s protective services who interviewed Miller, said, “I would say that what surprised me [was] that she wasn’t more—there wasn’t crying, there wasn’t the things we generally see when a child was in the condition that this child was in. That’s what surprised me.”
The baby doll
Hettinger, Miller’s attorney at trial, told her he would hand her a baby doll while she was testifying. He says he also told her to hug the doll and say, “I love my baby. I would never harm her.” Miller remembers being told to “grab the doll and just cuddle it.”
But “by the time we got into trial,” Miller says, “I couldn’t remember half of what [Hettinger] said.”
When the moment arrived, Hettinger asked Miller to take the doll and demonstrate how she shook her child after she became unresponsive.
“Do you think you can demonstrate with this doll like this?” he asked her.
“Probably not because I don’t remember exactly how I did it,” she responded.
Hettinger moved on, but the demonstration—which was meant to show that Miller was trying to revive her child, not harm her—had failed.
“I just remember being really frustrated and confused and not knowing what to do,” Miller says.
Over 13 years later, Hettinger says he remains “haunted” by that moment in the trial and says that was when he could feel his case for Miller’s innocence weaken dramatically.
“That’s one of the reasons why I’ve always kind of figured we lost the trial, was with that misplayed strategy,” Hettinger says.
Joan Miller, a juror—no family relation to the accused—says, “I do remember her flat affect” but “I don’t know that it impacted my decision.” Still, while the juror was observing the accused, she was thinking, “She’s not a motherly figure, or not responding how I would think she should respond.” Had the juror been the accused, she says, “Thinking now how I would react and knowing how I am, I probably would have been emotional and more stirred up.”
In an interview for this story, Ginger Kamps, another juror, says her first impression of the case was how young Miller was and that she was “absolutely emotionless.” She says the jury took careful thought and time and focused only on the evidence presented, but she admits Miller’s demeanor left an “indelible impression.”
David Kirby, an alternate juror at Miller’s trial, says he was surprised by how quickly the jury reached a guilty verdict—in a matter of hours. He says he saw Miller’s demeanor not as evidence of her guilt but as a human reaction to the trauma she had just endured.
“I thought she was in shock,” Kirby says. “I know I would’ve been.”
Scholars have been writing about body language since at least 1605, when Sir Francis Bacon mentioned the subject in “The Advancement of Learning.” In 1974, the Reid Technique, developed by a police officer who later wrote a seminal but controversial book on the topic with a Northwestern University law professor, was introduced to help law enforcement officers interrogate suspects. Now, police departments and Fortune 500 companies across the country train their officers and employees in the tactics of body language analysis. In 2013, the Government Accountability Office, Congress’ investigative arm, said the Transportation Security Administration had spent nearly $1 billion to “identify persons who may pose a risk to aviation security through the observation of behavioral indicators.” And the Pentagon confirmed in 2014 it was, at one point, spending $300,000 a year to examine the body language of foreign leaders.
But it’s widely considered a myth that anyone can detect innocence or guilt with certainty just from body language.
“There’s no research that says that a guilty person has a certain look,” says Joe Navarro, a retired FBI agent who specializes in nonverbal communication and consults with companies about how to use body language.
Interviews with many experts echoed the idea that interpretations of body language are fallible. It’s “extremely unreliable,” says Mark Cunningham, a clinical and forensic psychologist in Seattle. “Reports of demeanor at the time of the event are very problematic to make inferences from.”
Bandes, the DePaul professor who specializes in criminal procedure and law and emotion, says people have “ingrained ideas” and “folk knowledge” about how a person should act in a given situation. So when they respond differently than anticipated, people view them as less credible.
“People react in all kinds of different ways to trauma, and there are different influences that make people act very differently than the script might predict,” Bandes says.
James L. Trainum, a criminal case consultant and former Washington, D.C., police detective, says, “People always judge people on how they think they would react to something, and unless you’ve been through something, you don’t know.” He adds, “Suddenly, because these people aren’t acting the way we think they should act, we think they’re guilty.”
When people hear about traumatic events, like the loss of a child, they create a picture in their mind about how they would respond, says Cunningham, the forensic psychologist. The “obvious picture” is one “of grief and tearfulness, bordering on hysteria,” he says. “If that picture is not met, then this is abnormal, and that abnormality must be evidence of culpability or malevolence.”
Navarro, the retired FBI agent, says non-verbal analysis can be useful for law enforcement to feel out leads and investigate further. For example, if suspects are comfortable when answering a question about where they went to high school, but nervous and hesitant when talking about what they did the night of a murder, Navarro would note that difference. But he says even that reaction wouldn’t prove guilt.
In the Miller case, Alicia’s body bore no significant bruises or broken bones, often tell-tale signs of abuse. Tonia Miller had no criminal history and no accusations involving any other children, including her older daughter, according to records. What’s more, there were no witnesses and no physical evidence that directly tied her to the crime.
In a largely circumstantial case that focused on dueling medical opinions, experts say it can be tempting for jurors to gravitate to more intuitive signs of guilt. Such was the case with Joan Miller, the juror, who says she found the medical evidence difficult to understand and wishes it had been explained in “layman’s terms.” Wendy Heath, a psychology professor at Rider University in New Jersey who researches factors that influence juror decisions, says she found demeanor impacts juries when other evidence is lacking. Heath and other researchers conducted a study, published in 2004, in which an actress displayed varying levels of emotions—flat, moderate or high—while delivering identical testimony. She also manipulated the strength of the evidence at the mock trial, which was then presented to a jury for evaluation.
“When the evidence was strong, [the defendant’s] level of emotion didn’t matter,” Heath says. “But when evidence was weak, then emotion played a role in their decisions and the flat affect was not a favorable way to go.”
A flat affect could be one result of a person’s history of trauma, says Cunningham, the forensic psychologist: “We shouldn’t be surprised that their expression is different from somebody who has not suffered.”
In a recent interview at Women’s Huron Valley Correctional Facility in Ypsilanti, Michigan, Miller remembers being about 5 years old on the floor in the fetal position suffering from debilitating migraines—and her father’s wrath.
“He’s kicking me, telling me to suck it up,” Miller says. In her father’s house, she says, “weakness was crying.” Her father died in 2006.
In a report prepared for Miller’s sentencing, the Michigan Department of Corrections interviewed her grandmother, Bettie Hoskin, with whom she lived for a year when Miller was 10. “Whenever Tonia was abused or mistreated she would retreat emotionally, curling up into herself,” according to the report. “Mrs. Hoskins [sic] compared Tonia to abused women.” Hoskin died in September.
By all accounts, Miller was reserved throughout her childhood, even when the father of her older daughter threatened to take the child out of the country. Joyce Wesner, Miller’s neighbor then, recalls Miller speaking calmly, simply saying, “That’s not going to happen… If you want to see her, you can see her here.”
Wesner says, “She was mad about it, she was upset about it, but… she didn’t really yell and scream or flip out.”
Thirteen years in prison have forced Miller to confront her emotions. When she speaks, she cracks jokes about prisoners fighting with broken chicken bones from the cafeteria. When she talks about Alicia’s death, Miller’s brown eyes water and she pauses to compose herself. When she thinks about her release, about seven years away at the earliest, she doesn’t know what she’ll want to do.
Not long ago, Miller went through a program to get inmates to identify and cope with their emotions.
“Growing up, we weren’t allowed to show a lot of emotion, and I think that really damaged me on the stand,” says Miller, now 33.
In 2003, when she was sentenced to prison at the age of 19, she says she was “petrified.” She wouldn’t talk to anybody. But then she opened up to her bunkmate, who had a daughter about Miller’s age. They had long talks about love and life, and it got her thinking about how you’re supposed to be treated by others.
“Out there,” Miller says of her grief, “I didn’t have the tools to show that emotion.”
This investigation was conducted by nine undergraduate students at Northwestern University as part of an investigative journalism course taught by Prof. Alec Klein, director of The Medill Justice Project, which supported the class work. Amanda Westrich, director of operations at MJP, and Allisha Azlan and Rachel Fobar, MJP associates, also contributed to this report.