City Life

Oh, Baby

It was the fall of 1995, and Edmunds had the world at her feet. The striking blonde and her husband, Dave, were the proud parents of two little girls, Carrie and Allison, and were thrilled to be expecting their third in February. Edmunds felt fortunate she could stay home and care for her children in the family's new ranch home in Waunakee. During the week, the three went to the library for story hour, played in the park or frequented Waunakee's newest attraction, a McDonald's PlayPlace. Weekends often meant cookouts with the neighbors; Sunday mornings were reserved for services at Crossroads United Methodist Church.Because she loved kids and wanted to help other neighborhood families, Edmunds started watching a few tots during the week. One of her new charges was six-month-old Natalie Beard, daughter of Waunakee residents Tom and Cindy Beard. From the start, Natalie was a difficult baby. "She fussed or cried all the time," recalls Edmunds' neighbor and friend, Patti Larson, who also babysat. "We'd go for walks together, and any time the stroller went over a sidewalk bump Natalie would cry. If there were any kind of noise, she'd startle. Audrey was constantly holding her and trying to soothe her and make her happy. But she never got frustrated with Natalie. In fact, it was the opposite. She'd say, 'Oh, this poor baby. I feel so bad for her.'"Early in the morning of October 16, Edmunds and her girls got up and ate breakfast; they'd soon be walking Carrie to preschool with some of the neighbors. At 7:35 Cindy Beard dropped off Natalie, telling Edmunds the baby was irritable, that she'd been up twice the night before and had taken less than half her morning bottle. Natalie continued to fuss after Beard left, so Edmunds placed the baby in the master bedroom and propped the bottle of formula in her mouth, hoping the quiet room and bottle would calm her while she got the other kids ready.During the next half hour Edmunds checked on Natalie once, and all was fine. But when she went to dress Natalie for the walk to preschool at 8:35, she quickly realized something was amiss. Natalie made some funny noises, says Edmunds, and was limp when she picked her up. Formula dribbled out of her nose and mouth, and then she became unresponsive. Fearing Natalie was choking, Edmunds sped out of the house and called to her neighbor for help. And that's when the nightmare began. 

Coined in 1972, the term "Whiplash Shaken-Baby Syndrome" described children, typically under three years of age, who were violently shaken but often showed no visible signs of harm. Because their brains and neck muscles weren't fully developed, their internal injuries were severe--the typical "triad" included brain swelling and brain and retinal hemorrhaging, much like being in a serious car accident or falling several stories. The one positive, the theory went, was that a victim immediately becomes unresponsive, making it a snap to figure out who's guilty: the last person with the child. This was still the view held by most medical experts in October 1995, when Audrey Edmunds ran outside her house screaming, clutching an unresponsive infant. 

After the ambulance and police arrived, Med Flight whisked away Natalie. Dave Edmunds, who had just arrived at his new job in the Twin Cities, turned around and drove back home. That afternoon the couple raced to UW Hospital to see how Natalie was doing. "I was so distraught, thinking about what had happened," recalls Edmunds, still assuming Natalie was a choking victim. "I kept thinking, 'Why did I leave her with a bottle?'" She had no idea of the storm clouds quietly gathering around her.The physicians examining Natalie found a severe presentation of the classic Shaken Baby Syndrome symptoms. So when Natalie died that evening, Edmunds was immediately presumed guilty. The fact that Edmunds hadn't watched Natalie for the past four days and Natalie had only been in her care a mere hour before her tiny body shut down was irrelevant. Nor did it matter that Natalie had no outward signs of abuse, such as the rib fractures or bruised arms sometimes found on SBS victims. Or that much older hemorrhages were found inside her brain. Or that everyone who knew Edmunds loved her. Or that no one ever saw her abuse a child. All that mattered was Natalie had the classic SBS symptoms, and when she became unresponsive, Edmunds had her."There was no critical thinking about things that might have gone differently," says Stephen Hurley, her defense attorney and one of the most successful trial lawyers in Dane County. "It was complete tunnel vision."During the investigation that followed, prosecutors would wave off information about Natalie's father, Tom, indicating he was a nervous dad who suffered from migraines and was often irritated by her incessant crying. Also dismissed was Natalie's health history, which included numerous ear infections--she was being treated for one the day she died--and dozens of calls and trips to the doctor. They even ignored the fact that Natalie's parents had taken her into the doctor for lethargy, irritability and vomiting, symptoms that can indicate brain injury, several days before her death.Shelly Rusch, Dane County assistant district attorney, says there was no reason to consider any of those factors because the case was one of simple timing. (Rusch was not involved with the case in 1995, but worked on it recently.) "We prosecutors were lucky," she says. "Natalie's injuries were so serious, any person suffering from them would have been immediately symptomatic. It is what it is." 

John Plunkett is a Minnesota pathologist who has questioned SBS for years. Initially, he was considered part of the lunatic fringe; now he's treated with growing deference. Plunkett first became interested in the subject around 1985, when a defense attorney asked him to look into the death of an eighteen-month-old girl. The girl's mom said she'd been standing on the arm of a sofa and fell, hitting her head on the floor. She was brought to the hospital with retinal hemorrhages, brain swelling and subdural hemorrhages. The State alleged the mother shook her. "I said, 'Well, couldn't this fall have caused the death?'" says Plunkett. "They said, 'We never see it, so short-distance falls don't cause serious injuries in kids.' I said, 'But how do we know that?' And they said, 'Because we never see it.' That's just a circular argument, so I started looking at Shaken Baby Syndrome and realized something was very wrong."But back then, few people agreed. Although Plunkett and other interested experts began studying everything from shaking's biomechanical effects on an infant's head and neck to whether a baby could have a lucid interval between the time shaking occurred and the baby became unresponsive, the medical and scientific communities were still largely united ten years later, when Natalie died. So rather than prosecutors having the burden of proving Edmunds' guilt, in reality Edmunds' defense team had to prove her innocence. In 1995, that was nearly impossible."When I took on the case, I couldn't find any experts who thought [Natalie] wasn't shaken," Hurley says. Further complicating matters, the death of a young child causes intense emotions. Prosecutors feel driven to give voice to the innocent lamb, while the public is hell-bent on finding who's to blame.Hurley found just one medical expert who could help. Like the prosecution's experts, Mary Dominski, a pediatric neurologist with Dean Health System, believed Natalie had been shaken. But Dominski was the only physician to consider relevant her medical history and extreme fussiness the previous day, leading her to conclude Natalie was moderately shaken before she arrived at the Edmunds home. Then, Dominski theorized, Natalie suffered a major seizure at daycare from the prior shaking, culminating in her death.Lacking additional experts, numerous friends testified to Edmunds' stellar character: She had boatloads of patience. She loved kids. They never heard her utter an angry word. They dropped in unannounced all the time and never saw anything amiss. "Audrey has that personality people automatically gravitate toward," says good friend and former neighbor Shelley Murphy. "She'd do anything for anyone." Might Edmunds have been overwrought and momentarily lost it, shaking the ever-fussy Natalie? "I never, ever even considered she might have done it," says Murphy. "I understand she was the last person with the child, but anyone who knew her knew there was just no way."Unfortunately for Edmunds, nearly everyone judging her had never met her. 

The prosecution easily amassed eight medical experts. And while they weren't in total agreement regarding Natalie's injuries and how they combined to cause her death, all believed she was an SBS victim and Edmunds was the one who killed her.Lead prosecutor Gretchen Hayward, now retired, pulled out all the stops during Edmunds' trial, painting her as an outwardly nice person with a violent, hidden side. "Gretchen said very dramatic things, like I slammed Natalie's head into a blunt object," recalls Edmunds, interviewed recently near her home in Minnesota. "But Natalie had no head trauma, no skull fracture. I simply found a choking baby! They also tried to implicate me by saying everyone has a dark side, that my friends didn't know the real me, that I was stressed out because I was pregnant and moving, even though I was happy about both events. They tried to create a scenario that just didn't exist."Despite the overwhelming opinions against her, Edmunds appeared headed for acquittal, Hurley says. And then she took the stand."She looked like the proverbial doe caught in the headlights on cross-examination," Hurley remembers. "She kept looking at me when the jury asked a question, like I was coaching her. She fell apart."But what average American on trial for murder wouldn't be more than a little rattled when grilled by a seasoned prosecutor, questions Dean Strang, who later served as Edmunds' appellate lawyer. Nevertheless, Edmunds' fate was sealed. On November 26, 1996, she was pronounced guilty of first-degree reckless homicide. Her children were just five, two and nine months."I wanted to burst into tears and scream," says Edmunds. "I couldn't imagine being put into prison. I'm still shocked about it." When Edmunds arrived home that night, her conviction was already being blared across the ten o'clock news. She curled up beside her sleeping five-year-old, Carrie, and began to sob. 

In the mid to late nineties, most people prosecuted for SBS were charged with second-degree reckless homicide, says Strang. If they professed innocence and were convicted, it typically meant seven or eight years behind bars. But Edmunds' conviction meant the jury felt she treated Natalie with "utter disregard for human life." Dane County Circuit Judge Daniel Moeser handed down a shocking eighteen-year sentence, barring Edmunds from raising her own girls."The prosecution really pushed the edges here," says Strang.It was Dane County's most notable SBS case to date, and Judge Moeser apparently wanted to make a statement.And so it was that in February 1997, sixteen months after the nightmare began, Edmunds found herself climbing into a van full of female convicts on its way to Taycheedah Correctional Institution, a maximum-security prison in Fond du Lac, seventy miles northeast of Madison. Sitting among seasoned criminals, dressed alike in their drab green prison garb and clanking metal shackles, Edmunds was still numb. Yet she was confident her attorneys would secure her release post haste because, well, she was innocent."I never dreamed I'd be in prison eleven years," she says. "I never even thought I'd be in prison one year." 

Hurley turned the case over to Strang to start the appeals process, and Edmunds began her new life as a prisoner: Sleeping on an uncomfortable iron bed in a tiny cell. Urinating out in the open in a tin toilet. Signing up to use the shower or phone but not always given permission. Lacking freedom to do the tiniest things, like turn on a light or grab a snack from the fridge. Doesn't sound bad at all if you've committed a heinous crime, but if you're wrongly convicted, it suddenly appears downright inhumane. 

It was right around this time the Louise Woodward case hit the news. Woodward was a nineteen-year-old British nanny working for a Massachusetts couple with two boys. After one of their sons was taken to the hospital with skull and brain injuries from which he later died, Woodward was charged with shaking him to death. Although she pled not guilty, she was convicted. The highly publicized trial caused America's medical and forensic experts to start taking a hard look at the science behind SBS.They examined whether retinal hemorrhages can be caused by something other than shaking. They studied whether a child with SBS-type injuries can experience an extended "lucid interval," or a period of hours or days after injury where the child appears normal before becoming noticeably impaired or unconscious. They began to wonder if it was possible to violently shake a baby and cause SBS's famed triad of injuries without leaving some kind of visible neck injury. They explored similar brain injuries and death caused by lesser forces than vigorous shaking, such as short, accidental falls, or even conditions like vitamin deficiencies or immunizations. Little by little, compelling research began emerging that babies with brain swelling, retinal hemorrhages and brain hemorrhages aren't slam-dunk cases of SBS. And one by one, experts began changing their minds.Patrick Barnes was one. A well-respected pediatric neuroradiologist at Stanford University, Barnes was the prosecution's star medical witness in the Woodward case. Today, he regrets that testimony, noting something as mundane as an ear infection can spread to the brain with dire consequences. Unless there's evidence of an impact on a baby such as a fractured skull, Barnes says SBS is more myth than science.Minnesota's Plunkett, who has extensively studied the biomechanics of shaking, says SBS is a myth, plain and simple."There isn't any Shaken Baby Syndrome," he argues. "To cause subdural and retinal hemorrhages by shaking, you'd have to shake a baby twenty to twenty-five times per second to achieve the required force, or load. The most that's humanely achievable in a ten-pound model is three to four cycles per second. It's a simple mathematical equation."Case closed.Except it's not that simple. It never is. 

At some point, Edmunds realized she was in trouble. Perhaps it was after her first appeal by Strang failed in 1999. Or when the next two petitions for her release were denied. Maybe it was after her first mandatory parole hearing in 2001. Despite being a model prisoner with no behavioral problems, plus immense community support, her parole request was summarily dismissed because she was deemed unrepentant. And in denial."Audrey Edmunds loved children. I believe that," says assistant D.A. Rusch. "I believe she properly cared for almost every child she took care of. But sometimes when people do something horrible, they don't want to believe they did it. And after a period of time, they convince themselves they didn't."Meanwhile, the years ticked by. And neither side found peace. The Beards, who didn't want to comment for this story, divorced. So did Dave and Audrey. For nearly five years, Dave Edmunds fought to keep his family together, raising their daughters with help from Edmunds' parents and faithfully making the grueling five-hour drive from the Twin Cities to Taycheedah every other weekend so everyone could be together. But when Edmunds was denied parole in 2001, Dave apparently couldn't take the strain anymore, Edmunds says, and threw in the towel. Edmunds' friends and relatives began bringing the girls to visit, but now only monthly."There were years of tears during the girls' visits," Edmunds says. "It never, ever got easier, and I never got used to it. But hope became my religion. Without hope, you're crushed."Then, in 2003, Edmunds finally got a break. 

The University of Wisconsin Law School's Wisconsin Innocence Project was founded in 1998 to aid prisoners with plausible claims of innocence. Part of the Innocence Network, an umbrella organization of several dozen Innocence Projects worldwide, Wisconsin's has helped secure the release of twelve people. Of the five hundred requests for assistance that pour in annually from across the nation, says co-director and clinical law professor Keith Findley, the group investigates thirty to forty. Priority goes to the strongest Wisconsin-based pleas.Although the Innocence Project hadn't handled an SBS case before, Findley was interested in Edmunds' almost immediately."Her trial and appellate lawyers were convinced of her innocence," he says. "But even more significantly, there was new research which had led one of the State's witnesses against Audrey at the trial to conclude his testimony was in error. Once I heard that, I knew this was something we had to look at."That witness was Robert Huntington III, the pathologist who performed Natalie's autopsy. Finding the classic SBS triad of injuries, he had testified it was "highly probable" Natalie was injured shortly before she became comatose at Edmunds' home. But Huntington's conviction began to fade just three years later, after he performed an autopsy on another infant with injuries similar to Natalie's. When she was brought to UW Hospital, this girl was described as fussy and clingy, but interactive and responsive, much as Natalie had been in the week before her death. Yet it took trained hospital personnel more than fifteen hours to detect signs of brain injury. To Huntington, the elapsed timeframe between the girl's initial symptoms and eventual collapse now meant it was certainly possible Natalie was injured well before she reached Edmunds' home.The Innocence Project prepared a motion for a new trial, based on new medical evidence and the current turmoil about SBS among experts. (Physicians generally favor the old science, while forensic pathologists typically side with the new.) But after an evidentiary hearing in 2007, including testimony on Edmunds' behalf by Huntington (as well as five other doctors), Judge Moeser denied the motion. The Innocence Project appealed the decision, and in January 2008 the District 4 Court of Appeals overturned her 1996 jury conviction, ruling Edmunds was eligible for a new trial. In the meantime, she was free.Ten years and 352 days after Edmunds was led from the City-County Building in shackles (not that she was counting), she was released into the blustery grip of an early February snowstorm and her girlfriends' embrace. The State eventually declined to retry Edmunds and, at last, it was over.

At the time of her release, Edmunds' daughters were sixteen, thirteen and eleven. Natalie would have been twelve. The Beards had both remarried, but neither had other children. Edmunds moved to Minnesota to be near her children, taking a job at a Kwik-Trip and moving in with a friend.She longs for a place of her own to share with her daughters. But that will have to wait until her finances improve. Although the State dismissed all charges against Edmunds, she isn't eligible for any compensation for wrongful imprisonment."I see the struggles she's having and it just upsets me so much," says Edmunds' friend Larson. "Things are certainly better than a year or two ago--she's out of prison. But I'm frustrated she still has all these other things to deal with."Edmunds is determined not to be bitter. She knows there's no way to reclaim her girls' childhoods, so she's focusing on the present and the future. But she hopes her case draws attention to the new SBS medical findings and helps others in her predicament."This was never about me," she says. "There's some bigger purpose to all of this. I just hope it hasn't scarred my girls' lives."Defense attorney Hurley isn't quite as forgiving. For starters, he's angry with physicians. "This case is all about the arrogance of doctors," he says. "Physicians are asked their opinion, and rather than surveying the data and saying there's uncertain science, they divide into camps that say these symptoms mean X or these symptoms mean Y. They think they're correct and there's no way they can be wrong."Hurley's also frustrated with the prosecution's blind faith in the legal process. "Half of the medical profession doesn't believe what was said in court ten years ago, yet … the prosecution is unwilling to admit a person spent eleven years behind bars because they were wrong."Rusch isn't the happiest camper, either. Still convinced Edmunds shook Natalie to death, she says the new science merely confirms the prosecution's initial theory. "Babies just don't die and have all of this bleeding in their head," she says, noting that while the new research now shows retinal hemorrhages, brain swelling and brain hemorrhages can each occur for reasons other than shaking, there have been no documented cases of the triad occurring together unless a baby was shaken.Wrong, says Findley. "There are indeed numerous documented cases of retinal hemorrhages, brain swelling and subdural hematomas coinciding in cases where there was no shaking. In fact, there are no adequately documented cases in which shaking alone caused that triad of signs; it is only a theory, unproven. Unfortunately, this is another one of those cases in which prosecutors cling to discredited theories of guilt, despite the new evidence."Tit, tat. 

Despite Edmunds' release, the prosecution is satisfied she served most of her term (her mandatory release was just one year away). Still, Rusch says, "All the parents wanted was for Audrey to say she was sorry."And all Team Edmunds wants is for the prosecution to apologize. "They just did not want to admit they were wrong, especially Judge Moeser," says Larson. "There's more information now, so just admit it. But there was no apology, no nothing. It's so frustrating. And it makes me wonder how many people are in prison who shouldn't be."Which is precisely where the focus is shifting. Courts in the United Kingdom, Australia and Canada are re-opening old SBS cases; numerous have been reversed. While U.S. courts aren't at that point yet, Findley says American juries are starting to refuse to convict SBS defendants in light of the new medical research. Even more striking, the Kentucky Supreme Court is currently reviewing whether to conduct a discretionary hearing on an SBS case where a lower court judge granted a defense motion to prevent prosecutors from informing the jury the deceased infant had bleeding on the brain and retinal hemorrhaging because there were no outer signs of abuse. As of press time, the court had not ruled."Things are happening slowly, but I think this new understanding will filter into the U.S., and Audrey's case was one step in that direction," says Findley, who was inundated with requests for SBS assistance following Edmunds' release. 

No matter what side of the issue experts are on, all agree on one thing: The term "Shaken Baby Syndrome" should no longer be used. Inflicted Traumatic Brain Injury or Abusive Head Trauma are now favored, says Rusch, because the term SBS suggests a manner of death. "But no two babies are killed the exact same way," she says. "ITBI or AHT move away from saying you know exactly what happened, because you don't unless you witnessed it."Exactly Audrey Edmunds' point.Melanie Radzicki McManus is a contributing writer for Madison Magazine.

At the request of Jill Karofsky, former Dane County assistant district attorney, we are posting the State's Motion to Dismiss Prosecution as the reader may find it helpful in understanding the various positions in this complicated and controversial case.


Madison Magazine Subscription

Get Madison Magazine delivered to your office or home.

Gift subscriptions now available!

Subscribe Now

Dining & Drink

E-Newsletter Registration

This Week's Circulars