In her 20 years traveling the world as a freelance writer, Rachel Louise Snyder has covered a hurricane in Honduras, a tsunami in Indonesia, and the forced sterilization of women in Tibet. But no experience abroad scarred her more than reporting on the hidden world of domestic violence in a solidly middle class Massachusetts suburb.

In “A Raised Hand: Can a new approach curb domestic homicide?”, published in The New Yorker on July 22, 2013, Snyder begins with the harrowing story of Dorothy Giunta-Cotter who “knew someday that her husband, William, would kill her.” The traditional methods for helping domestic violence victims didn’t protect Giunta-Cotter, who was killed in March 2002, just two days after her husband was released without bail on charges of assault and violating a protection order.

Snyder posits, in her New Yorker piece, that a relatively new and deceptively simple program called the Domestic Violence High Risk Team might have saved the woman’s life.

As often happens with the best stories, Snyder discovered the Amesbury, Massachusetts, program by accident. It was 2010 and she had recently returned from six years of living and reporting in Cambodia. Feeling adrift, she was visiting friends in the nearby coastal enclave of Newburyport, about an hour north of Boston. It just so happened that these friends, the godparents of her daughter, were the writer Andre Dubus and his wife. Andre’s sister, Suzanne Dubus, is a high profile advocate for victims of domestic violence and the chief executive officer of the Jeanne Geiger Crisis Center, which was using this prevention method.

“I was standing in Andre’s driveway one day and Suzanne drove up,” recalls Snyder, who is now an associate professor of journalism and creative writing at American University. “She told me about this program and I was just stunned that domestic violence was a social ill that we could actually do something about. It seemed so unbelievable to me.”

Snyder spent the next month shadowing Suzanne Dubus and learning about the “Danger Assessment Tool,” which scores risk factors for domestic homicide with the aim of prevention. Compelled by what she found, she started researching the story immediately after that conversation.

As Snyder traveled the long road from idea to publication in The New Yorker, she came to understand how legal and ethical dilemmas deeply influence reporting on domestic violence. She had been pitching the magazine with other story ideas for more than a decade. Although none of her pitches had been accepted, she had developed a good relationship with the magazine’s executive editor, Dorothy Wickenden, who told her she thought this pitch had potential. Editor David Remnick turned it down at first, Snyder recalls, but she kept advocating. She felt domestic violence as a subject of journalistic inquiry had been ignored or underplayed for too long.

She recalls making this argument: “'I know everyone thinks this is a women’s story, but the fact is if women were killing their husbands in these same numbers this would be on the front page of every newspaper across the country.' And David Remnick turned around the very next day and assigned it to me. What we didn’t know was that it was going to take us the next two and half to three years to actually get it to print.”

(From the start, Snyder envisioned expanding the article into a book. “No Visible Bruises: What We Don’t Know About Domestic Violence Can Kill Us” was released today, May 7, 2019.  A part of that book was adapted for an opinion piece, published this past weekend by The New York Times, in which Snyder argues that the legal system needs to take more responsibility for prosecuting domestic violence. Snyder also wrote a piece for The New Yorker in December 2015 titled "No Visible Bruises: Domestic Violence and Traumatic Brain Injury.")

As Snyder reported the 2013 New Yorker article, the first big setback came after she spent a year with a woman whose story she had envisioned as the narrative backbone of the piece. The woman ended up as a secondary character, identified by the pseudonym of Lisa Morrison. The problem: Morrison felt it was too dangerous for Snyder to interview her ex-husband. If he found out she had been talking to a reporter, Morrison feared he would find her and kill her. Snyder’s journalistic ethics, and a well-placed fear of lawsuits, precluded using Morrison’s side of the story without hearing the ex-husband’s version of events. But her humanity wouldn’t allow her to interview the ex-husband without Morrison’s permission.

She was stuck: “Once I lost the narrative, I didn’t know how I was going to tell the story.” She started exploring other options. She tried to tell the story through the viewpoint of men accused of abuse. She tried gathering dozens of police reports to stitch together a narrative, but she couldn’t find a throughline.

Snyder, who has an MFA in fiction, had initially resisted using Giunta-Cotter’s story as the lead anecdote and narrative thread because “if I kill off my main character immediately, what’s going to keep people reading?” She also felt the anecdote wasn’t truly representative of her piece, which essentially is a positive story about a program that works to prevent the kind of horrific death that Giunta-Cotter endured.

But by interviewing the police officer who had answered the call, Snyder was able to recreate the detailed circumstances of Giunta-Cotter’s death. She came to believe that starting with this scene would work because she was able to establish the stakes of domestic violence, and thus keep people reading.

Snyder underestimated one aspect of reporting the story: the toll it would take on her own emotional health. She had covered so many disasters around the world that she thought she was immune. “I didn’t have a self-care plan,” she says.

One day, soon after returning from a reporting trip where she sat in on a meeting to review the details of a domestic homicide, she had a breakdown. She was walking her dog, listening to Pearl Jam, when she fell to her knees and started crying. She called a therapist, who told her it was a textbook case of  “vicarious trauma,” a kind of secondary post-traumatic stress disorder that affects people who work with traumatized populations. To regain her equilibrium, she took almost a year off from reporting the story.

[pq]Snyder underestimated one aspect of reporting the story: the toll it would take on her own emotional health.[/pq]

Snyder says that she has not been the direct victim of domestic violence. But after her article was published, she learned that her stepmother had been abused in a previous marriage. “There’s always resonance around these issues,” Snyder says.

The way Snyder sees it, the women’s movement brought “the battered woman” into the public eye in the 1970s and '80s, but journalists have dropped the ball on sustained coverage. Perhaps because it is viewed as a women’s issue, domestic violence hasn’t received as much journalistic attention as other social ills, such as homelessness and poverty, she says. The subject “needed to be explored from a 21^st century and feminist perspective,” she says.

In Snyder’s view, that means avoiding reporting on domestic homicides as isolated murders and starting to portray them as years-long narratives of abuse and intimidation that could have been stopped using some common-sense solutions.

The annotation: Storyboard’s questions are in red; Snyder's responses in blue. To read the story without annotations, click the ‘Hide all annotations’ button, which you will find just below the social media buttons in the top right-hand menu, or at the top of your mobile screen.

A Raised Hand

Can a new approach curb domestic violence?

Rachel Louise Snyder

The New Yorker ~ July 2013

Dorothy Giunta-Cotter knew that someday her husband, William, would kill her. Why did you decide to let the reader know Guinta-Cotter is dead rather than keep readers in suspense? The simple answer is that her death changed everything. It changed everything in that community about how they categorized domestic violence. It changed what we thought was possible with domestic violence homicide. But I also wanted readers to know the stakes immediately. I know it’s risky to start with a death – deaths, hospital scenes, funerals… they’re often so cliché – but I wanted readers to have Dorothy in the back of their minds the whole time they read. They met in 1982, when he was twenty and she was fifteen: a girl with brown eyes and cascading dark hair. Over the course of twenty years, he had kidnapped her, beaten her, and strangled her with a telephone cord. When she was pregnant with their second child, he pushed her down the stairs. After visits to the emergency room, he withheld her pain medicine and, at one point, forbade her to wear a neck brace.

Dorothy and William had two daughters, Kaitlyn and Kristen. Once, in a rage, William sat on Kristen’s chest until she couldn’t breathe; she was eleven. Another time, angered by what she was wearing, he hit her repeatedly in the head. That day, Dorothy took Kristen from their home, in Amesbury, Massachusetts, and drove to a shelter for victims of domestic violence in Maine. (Kaitlyn, who was seventeen, stayed behind in order to graduate from high school on schedule.) Dorothy feared that William knew the local network of domestic-violence shelters; in Maine, she felt, she would be safe.

There she filed a restraining order, telling the judge that her husband would kill her when he found her. But the judge denied the order, citing a lack of jurisdiction. So Dorothy returned with Kristen to Massachusetts, where she met Kelly Dunne who worked at the Jeanne Geiger Crisis Center, a local domestic-violence agency. Why did you decide not to tell the story through Dunne’s eyes? In a way, she is the hero of the story. If not, why not? Dunne is whip smart, but at the end of the day, the stakes aren’t hers. She orchestrates from behind a desk, which is dull in terms of writing a scene, but the real reason is that the life and death stakes are for victims like Dorothy. You just can’t compete with the emotional pull of that. The center helped Dorothy file a restraining order and found a room for her and her daughters in a longer-term shelter. But Dorothy refused. She told the center’s lawyer, “If I’m going to die, I want to do it in my own house.”

Under the terms of the order, William was required to move out. The crisis center changed the locks and gave cell phones to Dorothy and her daughters. Ten days later, William violated the restraining order. He hid in the garage until Dorothy, who was on her way to a job interview at a local supermarket, came in. He grabbed her and put his hand over her mouth. “Stop screaming or I’ll shoot you,” he told her. Kaitlyn, hearing the struggle, ran downstairs to find her mother being held hostage by her father. “Her mouth was bleeding . . . and she appeared terrified,” Kaitlyn later wrote in an affidavit. “I . . . stood with my mom and dad to make sure nothing was going to happen.” After two and a half hours, William left; the next day, Dorothy went to the police station and filed a report with a detective named Robert Wile. She told Wile, “Every time I talk to him, he scares me.” What are the challenges of writing about the difficult subject of abused women in terms of attracting readers and keeping them engaged? I’ve never met a single victim who thinks of herself as “the typical victim,” which is to say that any of us could be victims. So the key to me is making these people – women AND men – feel fully formed on the page. This is where my background in fiction is useful. We needed to see Dorothy move around, talk, interact. We needed to see her growing desperation. That’s what keeps readers, no matter the subject.

Wile issued a warrant for Cotter’s arrest, and on March 21, 2002, William, accompanied by his lawyer, turned himself in at the Newburyport District Court. His previous record showed only a few traffic violations and bad checks. He had a steady job as a cable installer and coached a local youth sports team. The judge released him on five hundred dollars’ bail.

Five days later, William arrived at Dorothy’s house armed with pepper spray, handcuffs, ammunition belts, and a sawed-off shotgun. Kaitlyn was at a friend’s house; Kristen opened the front door. William pushed past her, broke down the door to Dorothy’s bedroom, and dragged her out. Kristen ran upstairs and called a neighbor, who called 911. The police arrived minutes later. When the dispatcher called Kristen back to confirm their arrival, William picked up the downstairs phone and told her to call off the police or “someone’s gonna get hurt real bad.” Outside, the police could hear Dorothy screaming. When Officer David Noyes kicked down the door, William shot Dorothy at close range; it was as if a grenade had gone off in her body, Noyes later said. Then William reloaded the gun and turned it on himself. Kristen had been hiding under her bed, the phone to her ear; the entire episode was captured by the 911 operator.

This is an amazingly detailed recounting of this case. How did you nail down all the information? Records? Interviews? Both? Did you try to talk to Dorothy’s family? What was their reaction to this graphic description of her death? There was lots of paperwork on this, and people who’d been there — like police offer Noyes — described it to me in detail. The whole thing was captured on the 911 call, so we knew everyone’s location and the dialogue. I had Noyes draw the layout of Dorothy’s house for me: where she ran and where she went down. I do this all the time, have a subject draw a rendering of a scene I didn’t witness. Her family thought it was sensationalized. They weren’t happy and wrote a letter to The New Yorker. I understood what they meant, but, in fact, Noyes went into much more graphic detail that I left out. The New Yorker wrote back to them and they were OK with it in the end. It’s just never not going to be painful for these families.

 

The Jeanne Geiger Crisis Center is situated in a secured red brick building in downtown Amesbury, an hour north of Boston. To insure the safety of clients and employees, no signs mark its presence. The waiting room provides toothbrushes, toys, secondhand clothes, self-help books, and boxes of Kleenex. Behind the reception desk is a large playroom. Writers are often confused about which details to include. I assume these come from your close observation. Can you explain how you came to include them? Were there other details you left out as irrelevant? I look for what I call intellectual details, details that do more than sit on a page. What marks a space as individual? The crisis center is just like any other office; what makes it unique is the security, those items someone who’s fled in a panic might need (like toothbrushes, or clothes). These kinds of things aren’t what you’d have in a usual office. So those details help set the scene, sure, but they also have an emotional underpinning to them. I look for details that will do two things at once narratively.

 

Kelly Dunne, who is forty-two, is the center’s chief operating officer. After graduating from college, in 1997, she became a volunteer at the center, working at the district court as an advocate for victims of domestic violence. On her first day, thinking that she might handle one or two cases in the divorce stage, she brought a book to occupy her in her spare time. When she arrived, five women were waiting to file restraining orders. One had spent the weekend locked in the basement; another had been kicked down the stairs. “I remember thinking, Are you kidding me?” Dunne said. “This is what’s going on in this town over the weekend?” This is such a great quote. How do you decide when to quote someone directly, rather than paraphrase? I have a background in public radio, so my ear is trained for voice inflections to some degree, when someone yells, grows quiet, sobs, whatever. In this moment, Dunne was just aghast. And so I knew it was a good quote. I tend to try to underquote rather than overquote a source. Plus, Newburyport is this charming seaside New England town; domestic violence and homicide are not our primary references for it. Her quote underscores this.

One in every four women is a victim of domestic physical violence at some point in her life, and the Justice Department estimates that three women and one man are killed by their partners every day. (Roughly eighty-five per cent of the victims of domestic violence are women.) Between 2000 and 2006, thirty-two hundred American soldiers were killed; during that period, domestic homicide in the United States claimed ten thousand six hundred lives. Context is key for numbers. How did you come by this comparison? What others did you consider? I had a different figure that came from a researcher, and then when The New Yorker fact checker  went to fact check through the FBI’s homicide reports, she found the figure was wrong, that in fact the number was much higher. So the props go entirely to her for this. The New Yorker fact checkers are literary superheroes. This figure is likely an underestimate, as it was pulled from the F.B.I.’s Supplementary Homicide Reports, which gather data from local police departments, where homicide reporting is voluntary.

Dunne attributes the prevalence of domestic violence, in part, to a deep cultural misunderstanding of how violence operates. We assume that victims incite abuse, or that if the situation at home was truly threatening they would leave. Restraining orders, when filed, are thought to keep perpetrators away. And, if a woman fails to show up in court to renew a restraining order, the assumption is that the problem has somehow been resolved. “We now know that it means exactly the opposite,” Dunne told me. Here, you sum up the problem so clearly and simply. What process do you use when coming up with the nut paragraphs for a major feature? In this case, it was really that I was so ignorant and learning all this myself for the first time that I could just list my own assumptions and myths. I was the sort of the every-person reader. It’s much harder to do this now, because I know so much. But at the time, I literally had all these myths and assumptions up-ended by Dunne and Dubus, so I could just write about the ways in which I learned, the things that surprised me. I learned a trick from working on “This American Life,” that you should be able to fill in the blanks of this sentence: “The thing about XXXXX is XXXXX.” And I use that in everything I write now. It forms the basis of a nut graf for me. There’s the exterior organizing principle or subject of whatever the piece is about, but what is the internal voice of the piece grappling with?

In 2005, Dunne created the Domestic Violence High Risk Team, coordinating the efforts of her agency with those of local police departments, hospitals, state legislatures, and the courts to prevent domestic-violence homicide. The crisis center is funded by federal and state grants, private foundations, and fund-raising. Last autumn, the center received a four-hundred-and-fifty-thousand-dollar grant from the Department of Justice’s Office on Violence Against Women, in Washington, D.C., to help the high-risk team adapt its model to several other communities around the country. Vice-President Joseph Biden has championed the high-risk program; in October of 2010, at an event to mark domestic-violence-awareness month, he said, “We need to replace what we have been doing, and replicate this kind of success.” The high-risk team’s methodology is simple: it strives to prevent domestic-violence homicide by predicting when it might happen.

 

After Giunta-Cotter’s murder, newspaper editorials skewered the local police and the judge who had released William on bail; Bill O’Reilly, on his Fox News show, called for the judge’s resignation. Reporters sometimes shy away from writing about cases where much is already known. How did the notoriety of the case affect your choice of this story and your reporting? In this case, a lot of people had reported on Dorothy’s death, but not what came next, which is even more amazing — the advent of trying to predict domestic violence homicide. So I would probably shy away normally, too, but in this case it was the after story that mattered. Suzanne Dubus, the chief executive officer of the crisis center, convened a meeting between the district attorney and members of the police department, including Detective Robert Wile, who had taken Dorothy’s final police report, in order to analyze why the standard response procedures had failed. Suzanne is the sister of a close friend of yours. What impact did this connection have on your reporting and writing? What advice would you give to journalists whose personal and professional lives overlap? Yes, Suzanne is the sister of the writer Andre Dubus III, who has been a dear friend of mine for 25 years. The first thing I did was let my editor know how I’d come by the story and although I’d never met Suzanne, we decided to pull her role back a lot in the piece just to keep those aisles clear. I’m lucky because Suzanne comes from a family of writers and had zero problem letting Dunne take the lead — and the high risk team was really Dunne’s brainchild anyway. What I tell people is that you can use acquaintances to meet other people or find other sources, but people you know should never be a primary subject. Everyone appeared to have done his or her job correctly. The only real digression from protocol came from Dorothy, when she refused to return to a shelter. “This was our ‘Oh, shit’ moment,” Dunne said. The team had no plan besides offering shelter. “Shelter was our plan.”

Since the nineteen-seventies, shelters have been viewed as the best protection for battered women, but they can be profoundly disruptive. Most shelters in Massachusetts are single-family homes in residential areas, where victims and their children are allotted a room and share kitchens, bathrooms, dining, and living rooms with five to seven other families. Historically, boys older than twelve and pets have not been allowed in shelters, and most contact with friends or family, including a victim’s employer, has been forbidden. Dunne says that shelters are often, in effect, a “ticket to welfare.” Staying in a shelter may mean quitting a job and removing children from school, or being unable to care for elderly parents, or missing a doctor’s appointment. Shelters have saved lives, Dunne said, but the burden of change falls on the victim, not the perpetrator.

In the past decade, shelters and clinical-treatment providers have tried to better accommodate the needs of abuse victims. Many now allow teen-age boys to stay with their mother, and families to bring their pets; others permit contact with friends, family, and employers. But most shelters remain chronically underfunded, and advocates like Dunne are criticized for speaking out against the shelter approach. “It’s not a popular opinion to be putting forth in the domestic-violence world,” she said. This story took years to report, write and edit. How much did the published story reflect the initial pitch? Surprisingly, quite a lot. The primary narrative of Lisa Morrison changed, but otherwise all the high risk team stuff was in the initial pitch. I think I turned in a first draft at 9,000 words, and it ran at about 5,000. I wouldn’t pitch a place like The New Yorker unless I had a very solid sense of what the story was about, who I’d interview, where I expected it to go.

In 2003, Dunne attended a conference on domestic violence in San Diego, where she heard a talk by Jacquelyn Campbell, who teaches at the Johns Hopkins University School of Nursing and is widely recognized as the country’s leading expert on domestic homicide. In the nineteen-eighties, for her doctoral dissertation, at the University of Rochester, Campbell interviewed two thousand victims of domestic abuse in Dayton, Detroit, and Rochester, and sifted through police homicide files, looking for patterns. She found that half the women killed by their partners had sought help from the police or the criminal-justice system at least once, and that the single biggest indicator for domestic homicide was a prior incidence of physical domestic violence. The risk of homicide unfolded on a timeline, spiking when a victim attempted to leave an abuser, or when there was a change in the situation at home—a pregnancy, a new job. The danger remained high for three months after a couple split, dipped slightly for the next nine, and dropped significantly after a year. Campbell identified twenty risk factors for homicide, which she used to develop what she called a Danger Assessment tool. Some risk factors were obvious: substance abuse, gun ownership, a record of violence. Others were more specific: forced sex, threats to kill, choking. The sole demographic factor

Campbell identified was chronic unemployment; poverty alone is not a risk factor. Campbell then devised a weighted scale based on the risk indicators. A score of eighteen or more represented extreme danger; fourteen to seventeen was severe; eight to thirteen indicated increased danger; and anything less than eight signified variable danger. In San Diego, as Dunne listened to Campbell speak she realized that Dorothy Giunta-Cotter would have scored an eighteen. Whom did you interview besides Dunne to recreate this series of events? I spoke to Dunne, Dubus, and of course Campbell, but I also had written records. Dunne had taken copious notes, and Campbell has published widely on this. There are others in San Diego as well that I interviewed who didn’t wind up in the final piece.

Dunne and Dubus began to outline how they might use Campbell’s work to predict which domestic-abuse cases were most likely to end in homicide. During the following year, Dunne and her staff met with police officers in Amesbury and Newburyport; district attorneys; probation and parole officers; batterers-intervention group counsellors; and hospital representatives in order to devise a program that would identify potentially lethal cases. Their first meeting, in 2002, revealed that each department had operated in isolation. Neither the judge nor the hospitals were aware of Dorothy Giunta-Cotter’s history of abuse. The police knew about the restraining order against William, but the judge and the prosecutor handling the hearing didn’t have access to his file, or to Dorothy’s affidavit, which chronicled the two decades of abuse. “It’s in the cracks that murders happen,” Dunne told me. Her goal was to identify high-risk cases and create a plan of action to keep victims safe and out of shelters; the crisis center would serve as the central point of communication. In early 2005, the Domestic Violence High Risk Team began accepting cases.

 

One morning last fall, Dunne met with three staff members from the crisis center: Sara Hammond, a case manager; Kate Johnson, the community-services coördinator; and Connie Martyn, an advocate and a counsellor. It looks like you were there for this meeting? How did you balance your need for detail and access with concerns about confidentiality? I was listening in on a speaker phone, and then one by one I interviewed each of the three of the four folks in the room (Johnson, Martyn, and Dunne). I always, always overreport. But for this story, The New Yorker allowed me to check certain details with Dunne so that they wouldn’t compromise anyone. But before Dunne or anyone would talk to me about any particular case, they always, always have a written release from a victim — this is true even for the high risk team meetings. Remember, I reported on this story for just shy of three years; so I had lots and lots of choices in what to use. The day before, Lisa Morrison had called. She had first come to the center several years earlier, when she was married to a man named Glenn. (These are not the couple’s real names, and the details of the case have been modified slightly in order to protect Lisa’s identity. She was frightened at the prospect of being quoted; the details that were included are common to many of the cases that Dunne and her colleagues see.) Can you walk us through the process of deciding to give Lisa anonymity? Arduous! So, with The New Yorker’s permission (both attorney, head fact checker and my editor), I would read a detail to Dunne and she would tell me if she’d seen that detail in other cases, and then I could use it. This is some of the most careful writing I’ve ever done. Morrison’s story sounds specific on the page, but in fact… I purposely labored over everything. So, for example, we never learn her age; we never learn how many children she has, or how old they are, or where she lives or what her job is. Her ex is a vet, we learn, but I never refer to any particular conflict zone. So if you read her story, you’ll see that it’s both specific and sort of vague. I made them squishy (eg. “twisted her leg"), but in real life I had much more specific details.

 

Over the years, Glenn had pushed Lisa repeatedly, once shoving her into a wall as their children watched; on another occasion, he twisted her leg as she tried to run away from him. After a tour with the military, he was given a diagnosis of post-traumatic stress disorder. Lisa considered a divorce, but she feared his response; he was an alcoholic and had begun monitoring her whereabouts. Did you interview Lisa at length or is most of this through Dunne? Initially, it was all through Dunne. We wanted to see where the case would wind up as it went through court, and I didn’t want Morrison to have to stress about an article at the same time that she was worried about her own and her kids’ safety. But once her case went through the system — it took about nine months — I talked to her. The crisis center provided her with an attorney and a clinical social worker, and, two years ago, helped negotiate an end to the marriage. Lisa got a full-time job and now had a boyfriend, whom I’ll call Thomas. She maintained a cordial relationship with Glenn, who had visitation rights with the children every other weekend. But Lisa and Thomas had decided to move in together, and when Lisa told Glenn he began calling her repeatedly. When she stopped answering, he left messages warning that he would take his own life, that everything would soon be over, that he didn’t know what he was capable of doing. He asked Lisa to send him recent pictures of the children, and told her to take good care of the family. One day, he left more than forty messages.

Although Morrison’s case hadn’t made the high-risk roster during her divorce, Dunne and her colleagues were now concerned. Lisa’s children were due to spend the upcoming weekend, unsupervised, with Glenn. Pickups and drop-offs are particularly dangerous times for victims and their children. Several years earlier, a victim and her abuser, who had been divorced for years, met for a routine child visitation, and the ex-husband locked the children in the car, then bashed their mother’s face into a wall as they watched. Devastating detail — where did it come from? Yup. This is one that stays with me. It came from Dunne in a meeting.

Dunne asked Johnson whether they had grounds to suspend Glenn’s visitation rights. They didn’t, and Johnson reminded her that Lisa felt that the children’s presence would help Glenn control his behavior. Dunne asked if an unmarked police car could be stationed in front of Glenn’s house, but he lived outside the team’s jurisdiction. She asked whether there were guns in the house; no one knew of any. Reviewing the couple’s history, the team noted that Glenn was supposed to be seeing a psychologist but had stopped. Dunne and Johnson conducted a risk assessment using Campbell’s research, and Lisa scored an eight. She wasn’t in the most hazardous bracket, but Dunne knew that situations can change rapidly.

Dunne said, “This is really provocative behavior on his part,” and again argued that the team should find a way to stop the visitation. But Lisa didn’t want to file a restraining order, fearing that it would exacerbate Glenn’s erratic behavior, and there was no other legal recourse. So, on Friday afternoon, Lisa dropped the children off at Glenn’s house. On Saturday, Glenn stopped taking Lisa’s calls.

 

Alarmed, Lisa drove to the police station and filed a restraining order. The police went to Glenn’s house to serve him with papers, and Lisa retrieved the children and went home. Later that weekend, Glenn sent Lisa several e-mails in which he alluded to an impending death. The e-mails were a violation of the restraining order, so the police arrested Glenn, and held him in custody, pending what in Massachusetts is called a “dangerousness hearing.”

The dangerousness hearing is one of the most effective tools available to the high-risk team. A standard hearing determines bail based largely on flight risk. With a dangerousness hearing, even defendants who have clean records can be held until trial if they are deemed to be a sufficient threat to their victims or to their community. At the time of William Cotter’s threats to Dorothy, the statute was seldom used in cases involving domestic violence. Although many states have some version of preventive detention, very few advocates are aware of it, Dunne said.

Historically, such statutes have been used in gang or drug cases, though Massachusetts has seen a marked increase in their application to domestic violence. Viktoria Kristiansson, a legal adviser for AEquitas, a Washington, D.C.-based organization that supports the prosecution of violent crimes against women, noted that a dangerousness hearing “automatically provides a different context for a judge to analyze the evidence.”

Nevertheless, advocates have to contend with the difficult legal issue of preventive detention. “The Constitution tends to frown upon punishing prospective behavior,” Ronald S. Sullivan, Jr., the director of the Harvard Criminal Justice Institute, told me. Randy Gioia, the deputy chief counsel of the Massachusetts Public Defender Division, says that his office tries to fight dangerousness hearings because people who are held don’t benefit from the rights that someone accused of a crime would get at trial. He said, “Our system is set up to decide what happened as best it can; it’s not set up to decide what will happen in the future.”

Holding an abuser before trial provides victims with time to relocate, save some money, and seek counselling and perhaps find a job. Dunne told me, “We know that arrest, in and of itself, is protective. You’re trying to disrupt that escalating cycle of violence.” Before Dorothy Giunta-Cotter’s murder, Dunne said, the statute was employed “maybe five times in three years” at the local district court. She added, “Now we see an average of two a month.”

The day after Glenn’s arrest for violating his restraining order, he appeared at his dangerousness hearing. The judge ordered him held until his pretrial date the following month. Typically, offenders are held in jail, but because Glenn had a history of threatening suicide he was transferred to a psychiatric ward for evaluation. For Lisa, the team, and even Glenn, this offered the one crucial element that was impossible to adjudicate: time. What kind of conversations did you have with your editors about the need to protect Lisa versus the journalistic value of hearing Glenn’s side of the story? That was really stressful. I went up to New York and spoke with David Remnick. I’d never met him in person. He was, shall we say, alarmed that I hadn’t interviewed Glenn, but he sat for an hour while my editor, Alan Burdick, and I went through the past 30 months of reporting, and the particular danger women like Morrison are in. I remember telling him that the difference between, say, covering a war and covering domestic violence is that when a war ends, the sides need to figure out together how to rebuild society and community. War, in fact, has an end date. With domestic violence cases like Morrison’s, there is no end date. It’s like she’ll be forever in that highly charged war zone. He really got this. He really listened.

 

Dunne’s team, through the courts, often requires that, as a condition of probation, abusers attend forty weeks of specialized group counselling. In the past two decades, batterers-intervention groups have proliferated. The first, Emerge, a counselling and education center in Cambridge, was founded in 1977; there are now fifteen hundred nationwide.

David Adams, a co-founder of Emerge, told me that abusers seldom appear to be angry people, because they reserve their anger for the partner or the partner’s immediate family. “The average batterer is more likable than his victim, because domestic violence affects victims a lot more than it affects batterers,” he said. “Batterers don’t lose sleep like victims do. They don’t lose their jobs, they don’t lose their kids.” In contrast, “a lot of victims come across as messed up.”

Often, victims are substance abusers, or they live in extreme poverty. Many have suffered traumatic, abusive childhoods. These cases are the most difficult to prosecute, not least because the victims can be unreliable witnesses. “This is why batterers are so often able to fool the system,” Dunne told me. “They’re so charming, and the victim comes off as very negative.”

Batterers-intervention groups typically provide the court with information about an abuser’s compliance and willingness to change. They file a monthly report with probation officers and are in regular contact with victims about a batterer’s participation in the group. “We can be the eyes and ears of the court,” Adams said. “Victims are trying to make decisions about staying or leaving; if she’s hearing back from us that he’s still blaming her, that’s useful to know.”

In addition to preventing abuse and intimidation, intervention groups try to help an abuser recognize his own dangerous tendencies. One day, I had lunch with a man who had been ordered by the court to complete Adams’s program. He admitted that he had lied to his group the night before about drinking alcohol—a violation of his probation. Yet he told me that the program had helped him. “When you find yourself in a class like that, you can’t lie to yourself about the decisions you made,” he said. “My life has taken me to a point where I can’t tell myself I’m not that bad.” How did you feel having lunch with this guy? Did he ask you for anonymity? Actually, he didn’t. His name was Chris. I was a little nervous, but I did safety planning. I drove my own rental car and met him there. I told my editor and my husband where we were eating. We sat at a table outside. I wasn’t really too nervous. It’s the guys who’ve killed that make me a little nervous, even if they’re in jail.

 

A week after Lisa Morrison’s case came to Dunne’s office, the high-risk team gathered for its monthly meeting at the police headquarters in Newburyport. The meeting brings together Dunne and Johnson, from the Jeanne Geiger Crisis Center; Wile and other police-department representatives from Amesbury, Merrimac, and Newburyport; parole and probation officers; an Essex County batterers-intervention group; and a nurse from the local hospital. The cases are referred largely by the crisis center or by one of the local police departments.

Fewer than five per cent of the cases from Dunne’s center make it onto the high-risk roster, but, once they do, a response strategy is put in place. Because each office has slightly different legal restrictions, one challenge is to maintain a client’s confidentiality. The district attorney’s office can share basic information about a case, such as a pending court date, but nothing more. Probation officers can suggest terms of supervision, which might include measures like drug and alcohol testing or psychiatric counselling, but they cannot disclose anything about an offender’s private life—a job, where he lives, the results of a drug test. Parole officers can provide information only about when an abuser is on or nearing parole. Dunne and the crisis center can discuss cases with the team only after getting written permission from the victims. How did you gain access to these  meetings? What kind of conditions were set in advance? It took so many months! Dunne went to each office and asked their permission and their conditions. The hospital said under no circumstances could I quote their rep. Probation and parole said they wanted to know what quotes I’d use in advance. The police were like, “You can use anything we say at all, no conditions!” I had to keep track of who everyone was in every meeting. I tried to interview everyone separately, too, even if it was just on background. Too many young reporters rely on email to contact people, and phone to do interviews. Huge mistake. You’ll never bring your work to the next level if everything you do is by phone. I go in person whenever and wherever I can. I want them to see I’m not an enemy of the people.

Dunne’s office now sees police reports on the cases that the center has accepted, and they are often chilling. In one report, a woman told officers that her husband “made threats to me in the past about killing me, putting me in the chest freezer, and then taking my body out onto his boat and chumming me into the ocean. He also stated that he could kill me and put my body in his septic tank.”

As team members went down the list of cases, they looked for changes that might indicate trouble: a victim’s attempt to leave, an abuser going off probation or parole, the violation of a restraining order, the loss of a job, an incendiary Facebook post. In one case, a man assaulted his partner on the way to his batterers-intervention meeting, and was arrested again. In another, a man who had tried to stab his wife with a fork and then threatened to kill her was arrested and held without bail; he had a history of violating restraining orders and probably would be monitored by G.P.S. upon his release from prison. (Domestic abusers violate restraining orders forty per cent of the time.) One team member noted that the G.P.S. was not likely to stop the man. Dunne said the crisis center would try to make sure that the charges weren’t dropped, and organize a plan of action when he was released, in eighteen months.

Team members reviewed their options in each case. Police officers can conduct extra drive-bys or home visits to check for signs of unusual behavior. In Massachusetts, as in most states, if the abuser has a gun it can be confiscated when a restraining order is issued. Time can be added to an abuser’s sentence by combining domestic-violence charges with other criminal charges, such as theft or the possession of illegal drugs. Visitation with children can be supervised or suspended, or the judge can refer cases to the crisis center’s attorney in order to craft visitations that take into account the individual risks.

The team also helps victims find transitional housing and free legal assistance. Team members work with clients to improve their safety; this can involve rehearsing emergency situations, erasing their profiles on Facebook, Twitter, and other social-media outlets, and even changing daily habits, such as where they shop or the route they take to work. In rare instances, the team places victims in a kind of state-sponsored identity-protection program, in which the residential address is kept secret and mail is delivered to a post-office box.

All but seventeen states have passed or introduced legislation to allow the use of G.P.S. in cases of domestic violence. If an offender enters certain “exclusion zones”—ranging in size from a few blocks to an entire township—an alert is sent to the local police and an arrest warrant is issued. “We contain the offender so the victim doesn’t have to be contained,” Dunne told me. The clarity of your prose in the article enhanced its power. How much of this style reflects your own voice, the subject matter or the editing and approach of The New Yorker? Alan Burdick was an amazing editor. He did, as he called it, polish the piece, but the arc, the structure, the rhythm, that’s all me. The New Yorker is such a writer’s house; they actually care about what a comma does to the auditory quality of a sentence versus a semicolon. That’s my kind of geekdom, too. I love it.

 

The Morrisons were the final case of the morning. Some troubling facts had emerged. Before Glenn entered the psychiatric ward, he had repeatedly called Lisa’s boyfriend, Thomas; one morning, he drove to Thomas’s house and parked his car out front, where he sat for hours. Moreover, Dunne and her team had learned from Lisa’s counsellor that Glenn had been released by the ward; strict confidentiality laws had barred the hospital from informing them or the court. But, by the time of his pretrial hearing, he had been readmitted and the case was extended to the following month. In the meantime, the police had begun stopping by Lisa’s house once or twice a day to walk around and make sure nothing was amiss. Dunne was frustrated by the gaps in the system, but by now nearly three months had passed and, statistically, at least, Lisa and her children were safer. “Think about where we were originally,” Dunne said. “What increases safety is you go from no containment options to all kinds of people having their eyes on this case regularly, so if there’s any escalation there’s an ability to react.”

Late last fall, just before his upcoming court date, Glenn broke his restraining order again and followed Lisa in her car. He was charged for the second time with violating an order. Finally, eight months after the first call came in to the crisis center, the court gave Glenn eighteen months’ probation and required him to attend psychiatric counseling. Visitations with the children could continue, but only under third-party supervision.

In the Morrison case, Dunne’s team managed to intervene while the situation was still in the misdemeanor phase. Without the high-risk team, Lisa told me, “I honestly don’t know if I would be where I am.” But she and Thomas feel uneasy much of the time: “We say to each other, ‘Always be on the lookout.’ ” Unlike other crimes, in domestic violence the abuser maintains a presence in the life of his victim, and remains a potential threat, especially when children are involved.

I spoke to a woman whose husband had abused her for years. At one point, he threatened to slit her throat with broken CDs. She is divorced now, lives in a secured, secret location, and has a lifetime restraining order against him; he is not allowed to enter the town limits. Finally, she feels safe enough to go jogging—but she would not speak on the record, fearing that he would somehow find her and retaliate. How did you convince victims to talk to you? How did you gain their trust?Time, time, and more time. Also letting them know I understood the complexities of domestic violence. I wasn’t going to judge them for not leaving, or for going back. And I always put my recorder in front of them so they have control over stopping it to go off the record. She said that “the only way to describe what happened to me is like part of me died. It was just about survival.”

 

In the decade before Dorothy Giunta-Cotter’s murder, in 2002, a domestic-violence-related death occurred nearly every year in Amesbury. Since the formation of the high-risk team, in 2005, Dunne has not had a single case end in homicide. As a woman who cares about this issue, and who has now written a book about it, how do you draw the line between journalism and advocacy? I think that’s an easy one, actually. A lot of journalism is simply investigating where systems break down, what’s not working in society. That’s our responsibility as journalists. But we’re not policy makers. I can’t write or pass legislation. I can’t change the way a crisis center works. All I can do is tell stories to improve those systems and policies, or to hold leaders accountable. “When I listen to the stories of the victims who have been involved with the high-risk team, there is no question that many of them would have been killed,” Mary Lauby, the executive director of Jane Doe Inc., a domestic-violence advocacy organization in Boston, told me. Of the offenders now monitored by G.P.S., not one has committed another act of domestic violence; nearly sixty per cent were held before trial using a dangerousness hearing. Dunne also notes that, of the hundred and six high-risk cases documented in the team’s most recent report, only eight women were forced to seek refuge in shelters. She estimated that, before the formation of the high-risk team, ninety per cent of similar cases would have resulted in the women’s going into shelters.

Dunne and Wile have trained more than five thousand people from thirty states, including three thousand in Massachusetts. Groups from California, Louisiana, Florida, Illinois, and more than a dozen other states have contacted them. Framingham, Massachusetts, was the first to replicate the high-risk team based on Dunne’s model. Mary Gianakis, the director of Voices Against Violence, a Framingham crisis center, and a lead member of the area’s high-risk team, told me that, previously, shelters were the primary resource available to victims. “Now we can say, ‘Look, we’re going to bring the full power of all these resources to keep you safe and monitor your partner,’ ” she told me. To Suzanne Dubus, the need to create a model in which victims are protected, rather than isolated, seems obvious. “Here’s the outrage,” she told me. “It’s really cheap to do what we’re doing. It’s a lot cheaper than murder investigations and prosecutions and jail time.”

In terms of structure, you circled back to Dorothy to end where you began. How did you decide what information to put at the beginning and what to hold until the end? Actually, I wanted a different end, in a scene with Kelly Dunne in her office. I fought for it and lost. That scene now ends my book! But with Dorothy, what I wanted was the ripple effect of her death, how it rang like a bell through the police officers who were there that night, and into the advocates and judges and everyone else involved in her case, and eventually someday even me. So my focus was on what her death meant in terms of preventing other deaths, but also how those present that night went on living, but never forgetting her. And I end with her own voice, her own prediction, which is immeasurably haunting in the context because we know now she’ll be dead within days of having said this.

In their training sessions, Wile and Dunne walk through the timeline of Dorothy and William Cotter’s relationship. The violence began within a year of their meeting; each time that Dorothy threatened or tried to leave, William increased the degree of abuse—what experts call “retribution violence.” In 1996, she married him, a fact that often baffles Dunne’s trainees. “It’s counterintuitive,” Dunne says. “He strangled her, held her hostage. Why would she ever marry him?” But, she adds, “William showed her he would never let her go. So she thinks if she marries him he’ll get less violent.” In its way, it was a rational response to a support system that offered her little means of escape; her final attempt to leave failed when her request for a restraining order in Maine was refused. “We give them this message that the system won’t protect them,” Dunne says.

One evening, I drove around Amesbury with Officer David Noyes, who had broken down Dorothy’s door on the night of the murder. In his cruiser, we passed open fields and low-income apartments near the baseball diamonds of Amesbury Town Park, lakeside mansions, and the Amesbury Golf and Country Club. Green Street, where the Cotters lived, is a single block of lower-middle-class homes built so close together that there is barely room for someone to squeeze between them.

Noyes parked in the small lot beside Dorothy’s old house. A tricycle was on the lawn next door. Noyes said that it was so quiet when he and his team first arrived that he walked around the driveway with the other officers trying to figure out if they had the right house. Then he heard Dorothy: “No, he’s gonna kill me!” Noyes ran up the front stairs, and he heard her struggling with the door lock and the sound of William hitting her several times. When Noyes broke in, Cotter fired, and Noyes was blinded for a moment—a sawed-off shotgun emits a dazzling muzzle flash. Then he saw Dorothy fall. “I had trouble sleeping for years after, ” he said. What made you decide to tell this part of the story from the police officer’s point of view? Simple: he was there. He saw the muzzle flash; he saw her body fall. He could talk about that moment and how it haunted him.

Dorothy was thirty-five years old. In the days before her death, she had told Detective Wile that if she and the girls moved to a shelter William would find them and kill them all. She attempted to avert the worse of two terrible outcomes: the loss of her daughters’ lives along with her own. “If I’m over there,” she told Wile, referring to her house, “there’s a better possibility that it’s just going to be me.” ♦

When did you realize that you had the seeds of a book on this topic? Honestly? Right away. This piece covered the formation of the High Risk Team, but there was so much more I began to learn about domestic violence. I talked to my editor about this being a book the whole time I was working on this piece. It just took me a while to figure out how to tell that bigger story, how to structure the book. But I finally figured it out five or six years down the line.

 

Rachel Louise Snyder is the author of the books “What We’ve Lost Is Nothing” and “Fugitive Denim.” She first contributed to the magazine in 2013. 

This article formed the basis of Snyder’s forthcoming book: "No Visible Bruises: How What We Don’t Know About Violence Call Kill Us" (Bloomsbury, May 7, 2019).