Legend has it that Tom Wolfe, the New Journalism pioneer whose stories regularly presented his characters' points of view, was once challenged by a critic who demanded to know how he could possibly know what those characters were thinking. Wolfe had a simple answer: "I asked them."

​"There's nothing mystical about knowing what somebody is thinking,” agreed Jay Kirk, who spent several years interviewing a team of crime scene investigators who worked the 2012 mass shooting at Sandy Hook Elementary School in Newtown, CT. "You ask them," he told me.

That approach enabled Kirk to convey in riveting detail the investigators’ mindsets and memories in "They Saw the Horrific Aftermath of a Mass Shooting. Should We?," a New York Times Magazine story that won the 2024 National Magazine Award for feature writing. The 8,000-word story was published in April 2023 — nearly a dozen years after the killings of 20 children between 6 and 7 years old and six adult staff members.

The narrative unspools the horrors of that day through the eyes of three investigators who shared the experiences, reactions and ramifications of cataloging every piece of evidence left behind after a 20-year-old former student armed with a Bushmaster semi-automatic rifle entered the school and opened fire in two first-grade classrooms.

The action is driven by reconstructed scenes and dialogue that was verified by multiple sources. It chronicles the 12-16 hour days that detectives Art Walkley, Karoline Keith and Sgt. Jeff Covello of the Connecticut State Police worked inside the school, and the emotional trauma they suffered away from the job.

Kirk, who admits “to a tendency to take on too much,” initially wanted to focus on multiple mass shootings. Only after interviewing investigators who worked the 2006 Amish schoolhouse shooting in Nickel Mines, Pa., the CSI team that documented the 2012 Aurora, Colo. theater shooting and embedding with Philadelphia investigators did he turn his attention to Sandy Hook.

Kirk was told repeatedly that the Connecticut investigators wouldn’t talk with him. But he reached out and won their trust with an interviewing style that he describes as being “unafraid to ask dumb questions.” He credits his tenacity to the inspiration of his narrative nonfiction students at the University of Pennsylvania who refuse to give up in the face of rejection. “Practice what you preach is what I have to remind myself when I’d literally rather be doing anything else,” he told me.

The result is an unfiltered portrait of anonymous public servants who confront the realities of gun violence, highlighting a little-told story about the emotional scars a mass shooting leaves beyond that suffered by victims and survivors.

Kirk tells the story chronologically through third-person point of view that reflects an authoritative grasp of CSI methods and the varied ways Sandy Hook traumatized a team of investigators hardened by hundreds of murder investigations. He practiced what Gay Talese called the “fine art of hanging out,” visiting his subjects’ homes,  drinking beers with them around a fire at Costello’s house, and spending long stretches interviewing them by phone.

From the outset, he decided against detailed descriptions of the carnage, responding to his sources’ insistence that they wouldn’t run the risk of retraumatizing family members of the slain children. He relies instead on descriptions of the actions and reactions from the CSI team and, in one long scene, how Attorney General Eric Holder responded when he made a visit, seeming “to grow smaller in his chair” as the images flashed on a large TV screen.

Kirk spoke with Nieman Storyboard about the importance of immersion, his method for structuring stories and the importance of washing dishes by hand. Our conversation, and the annotation of Kirk's story, has been edited for length and clarity.

How did you become a writer of narrative nonfiction?
In my 20s, I wanted to be a novelist, I spent five years holed up in a little cabin in Vermont writing a novel that never sold but that I ultimately understood was just my own private MFA program. On a total lark, I wrote a couple of pieces for alt-weeklies, the Philadelphia City Paper and the Chicago Reader. Like many people my age, I was turned on by the Harper's kind of David Foster Wallace experiential reported essays. I thought that sounds like fun. Then I looked back at my novel-writing phase and realized one of my favorite parts had always been the research — entering actual reality. And since the subject of nonfiction is always reality, the theme is pretty constant.

Why do you call some of your courses in narrative nonfiction at UPenn “The Art of Experience.”
Some non-fiction courses focus on memoir.  I'm more interested in writing that documents the immediacy of a moment. One of the exercises I have for my students is to think about how many photographs they take during the week and then, when they're about to take a photograph, pull out their notebook and pen or their notes app instead and describe not only the image itself, but notice why you found this interesting, what was the moment leading up to this, and then see how minutely you can describe something. Operate like a documentary camera: Let the camera run and see what you find. That’s the experiential position the writer can take: Try and capture everything and then go back and look at what you have. That leads to the experience of making discoveries and where it gets really exciting.

Who are the writers or books that have shaped you?
Like most writers, I just read a lot. It’s essential. I can’t point to just three or four. Whatever I’ve read the night before will find its way into my work the next day — a word or a sensibility or a technique, the way that sentence structure worked, or I don’t need to write the sentence this way. I'm being shaped every day by whatever I'm reading.

What advice would you give to aspiring writers who wish to explore narrative nonfiction? Read more. And make yourself come up with better questions — you can do that by reading more. Also spend less time crafting defendable opinions and more time becoming comfortable with uncertainty, which is synonymous with reality. Again, if you're writing nonfiction, your subject is reality, and reality does not lend itself to defendable opinions.

ANNOTATION: Storyboard’s questions are in red; Kirk’s answers in blue.

They Saw the Horrific Aftermath of a Mass Shooting. Should We?

The crime-scene investigators are the ones who document, and remember, the unimaginable. This is what they saw at Sandy Hook.

By Jay Kirk
The New York Times Magazine, April 20, 2023

[dc]T[/dc]he crime-scene van was parked next to the black Honda Civic already identified as belonging to the shooter, the yellow tape marking its perimeter juddering in a helicopter gust. Earlier that morning, before the van was cleared to move closer to the school, Jeff Covello, the crime-scene-van supervisor, and his team were crowded around the dry-erase board. Art Walkley, the only one on the van who had so far been inside, sketched out what he said were the two main areas of impact. He arrived with the other first-response officers and stormed the school as children were running out, his gun drawn, ready to kill on sight, in fact quite eager to pull the trigger once he glimpsed Classrooms 10 and 8.

Jeff had never seen Art look the way he did after he came out of the school. It was more of an apparition that climbed back onto the van. The two of them were SWAT for eight years together before Jeff transferred to Major Crimes and brought Art with him. They had taken fire together. They had seen each other become parents. Art had seen Jeff call his wife in the middle of the night to remind her where to find the life insurance.

They could all read one another’s minds. Karoline Keith, the senior detective on the van, had already been riding for more than five years when Jeff arrived as the new supervising sergeant. It was Karoline who suggested that Art try to tell them what he saw and sketch it on the board. She hoped it would make it easier once they got inside. Art said he didn’t think there was anything he could say that was going to make it easier. Why did you open the story this way?It's a character-driven piece. The story will primarily take place in their heads, focusing on their experience. That seemed like the natural place to start. How did you decide to focus on these investigators? The dynamic between the investigators, Karoline Keith, and Art Walkley and Jeff Covello triggered my story radar. They were in the school 12, 14, 16 hours a day for seven days straight. They were so experienced and hardened and had seen everything. Karoline Keith had 300 murders under her belt by the time Sandy Hook came around. They had that forensic mentality and I was interested in how can people do this sort of work.

As detectives for the Connecticut State Police Western District Major Crime Squad, they were all experts in human depravity, but Art was the death guy. The one who was lowered into septic tanks to retrieve badly decayed body parts.What were you trying to signal with this graphic and disturbing imagery high up in the story?By the title of the piece and the tagline, every reader knows that this is going to be about Sandy Hook and the crime investigators who saw Sandy Hook.  For readers, I presumed there was going to be a degree of 'Do I really want to read this? Do I dare?' For me, the question became how am I going to negotiate those expectations? I've already made fundamental decisions. For instance, I'm not going to actually show the reader anything too gruesome. That's a decision I made out of respect for my subjects; they felt that it could re-traumatize the families. I ended up agreeing with them. But I needed to let you know right away that these weren't just podunk cops coming to a mass shooting scene. These were exceptionally, horrifically, experienced crime scene investigators who who had seen everything. I'm not going to avoid saying what these people do. I wanted to get that part out of the way quickly.

He had seen everything imaginable and a good deal of the unimaginable. And yet somehow he managed to stay one step ahead of the crowd of ghosts that were always following on their heels from one death scene investigation to the next. But by the look of him now, in the parking lot of Sandy Hook Elementary on Dec. 14, 2012, the ghosts had caught up all at once. What drew you towrite a story about the crime scene investigators in the Sandy Hook mass shooting case in the first place?It had become a persistent question whether or not it would make any difference if people saw these images from a mass shooting. In conversations with my editor, Jessica Lustig, we decided that the best avenue, or maybe the most intriguing way to address the question, would be to talk to the people who had actually seen it themselves.

SWAT had cleared the building, and the F.B.I. had checked for explosives and ruled out terrorism. Now it was up to them to take the photographs, measure, collect evidence and conduct the exacting work of meticulous reconstruction. As the crime-scene investigators for W.D.M.C. — Eastern District Major Crime would have the shooter’s home; Central District Major Crime had the exterior of the school — they were recognized in the state as being the elite, specially trained detectives that they were. They would note how the shells clustered; how the choreography of the shooter’s movements was revealed by the voids where shells or blood were absent; where someone paused to reload. And then memorialize their work with extensive photographs and video so in court an independent expert could reproduce their calculations and arrive at the same conclusions. That was ultimately the importance of the job: to see, to look — and to do so with grinding duration.Could you describe the reporting and research that let you to write with such authority and specificity about the team’s work?Initially, I wanted to look at multiple mass shootings. I went down months of blind alleys before I decided to focus on Sandy Hook. I interviewed the crime scene investigators from the West Nickel Mines Amish schoolhouse shooting. I visited the crime scene investigators who had first gone into the Aurora movie theater shooting in Denver. Because I live in Philadelphia and we have such an enormous death toll, it occurred to me that right here crime scene investigators are seeing “mass murder” all the time, by which I mean hundreds of mass shootings every year. So I tagged along with them for a while, going to active murder investigations, hanging out in the van and their office. Eventually, I started having very long phone conversations with Karoline, Jeff, and Art, listening to them tell the story. These are things they told me, talking about something so private.  I don't think I ever talked to Karoline on the phone for less than three hours, maybe to talk about one thing or a handful of follow-up questions. Everything in the end, all of the work, is in the follow-up questions. There's a long period of trust-building that you have to go through and you just have to have enough faith in the process. We tend to have such a productivist mindset. We hurry too much. We’re always in a rush to get the story in, and we need to publish it now.  I just ignore that until I’ve got what I want.How do you know when you’ve reached that point and that what you’ve written is accurate? You talk to somebody for three hours and then print off the transcript. The next morning, you have 70 pages and you read through it and you read through it again and then you realize, wow, they told me that, that’s pretty interesting, and then you ask more follow-up questions. In this case, they were all in the same situation and they're documenting everything they do because they’re all detectives. I would reconfirm almost every detail in the story with each of them. And they would always tell me the exact same thing, except usually, of course, adding a new detail, adding a new dimension.

Now, here, where 20 first graders and the principal, the school psychologist and four teachers were lying dead inside, they could maintain the detached forensic mind-set for only so long before the corrosive reality of what happened here began to seep into their Tyvek shells.When you use a detail such as “Tyvek shells,” are you at all concerned that the average reader won’t know what you’re talking about? When it comes to whether or not you need to include some kind of appositive definition or clarification, I'm usually going to think about what's common knowledge. In this case, anyone could quickly look it up if they want to. But also because of mass culture we all know that CSI are wearing these big white, insulating suits. So I'm just naming the material.Dan Sliby looked to have gone into full robot mode. The usual vibrant prankster energy of Steve Rupsis, who would be on video today, was gone. He, like several others on the van, had a child close in age to the victims inside. Jeff himself, for the time being, was safely immersed in logistics at the little supervisor’s desk where he made out assignments. Calculating what resources they were going to need. Gas for the generators. Gloves. Bootees. All the supplies for who-knew-how-many decontamination stations.How did you learn the specifics of the necessary resources? Straight from Jeff. That was his job, but by spending time with Philadelphia CSI units, being on the truck, asking what’s in this cabinet, what’s in the refrigerator here, I realized it would help me design better questions for Sandy Hook.

The helicopters were not helping. Karoline thought for sure they were going to slam into one another and rain down another layer of destruction. But even if they crashed down on top of Sandy Hook Elementary, well, then they would handle that too. Jeff had said it a million times: God forbid, if a 747 crashed into the State Police barracks, they would know what to do. The job was the same whether it was one person or six. (Not that they had ever processed a homicide scene with more than two victims.) Their skills were infinitely scalable. Without knowing it, they had been preparing for this day their entire careers.You said you went down months of blind alleys. What kept you going? I remember feeling frustrated and exhausted, that this is never going to happen. Everyone said if you want to write about the people who had seen a mass murder, no one will ever talk to you. I took that as a challenge, right?  I take inspiration from some of my students, too. They never give up, no matter how many times they’re told no. I said I guess it’s time for me to follow my own advice. So I just kept after it and lucked out.

As have the countless other crime-scene investigators who must dwell in the aftermath following each mass shooting. Virginia Tech, Columbine, the Aurora movie theater, the Pulse nightclub in Orlando, the El Paso Walmart shooting, Parkland, Las Vegas, Binghamton, San Bernardino, Sutherland Springs, Thousand Oaks, Virginia Beach, Monterey Park, Santa Fe, Pittsburgh, Buffalo, Uvalde, the Covenant School in Nashville in March and Louisville in April. Each scene of unimaginable horror witnessed by an anonymous team we have chosen, without knowing it, to do the gruesome work of internalizing our national crisis for us.The story doesn’t seem to have the obligatory nut graf, but this seems to be the one paragraph that pulls back from the narrative. Is that it?I almost didn’t want to say that much. In the early days of research, I read a half dozen books on the Second Amendment and gun issues, but I was trying to avoid making arguments about anything. I didn't want anything to get in the way of the investigators experience. The most gutting part of it, after having spent so much time with these doomed individuals, is that these people have had to see this. We are outsourcing that to a handful of people, but around them are rings of spouses and family members who are traumatized by what their relatives have been traumatized by.

Among the things the team had trained to do was lower the visor of fog. It came down around the rest of the world and gave them a protective cloak, a kind of insulation, so that like ghoulish astronauts they could descend and still see past the obvious suffering and gore while maintaining the requisite objectivity. Maintaining a barrier against the cross-contamination of their feelings was as important as the masks and bootees. The sooner they could suit up the better.

The job had already destroyed Karoline once. It was difficult to think that just as the shooter was stepping into the lobby, she was sitting in her therapist’s office, talking about how far she’d come over the past two years. She was no longer suffering from panic attacks or seeing things that weren’t there. She had started therapy in 2010, after putting in for a transfer off the van. The simple reason was that she had burned out. The less-simple reason was that she could no longer go for a walk in the woods without mistaking every flesh-colored rock for human remains. At home, she had become controlling and hypervigilant. Texting her partner, Elissa, 50 times a day, even managing the way Elissa walked the dog. She had begun to see the whole world as a potential crime scene. Cops are notoriously tight-lipped and, with a case like Sandy Hook, I would expect these crime scene investigators would be extremely reluctant to talk with a journalist, especially about their inner lives. How did you persuade them to talk with you?I was just perversely honest when I called them, and I think they believed me when I said I didn’t have an agenda, and if they didn’t believe me the first time, they eventually realized I was focused on their experience and not on any grand opinion about the impossible subject of guns in America. Also, I  just ask lots of dumb questions and that’s probably disarming. You can’t be afraid to reveal how truly stupid you really are when you’re a reporter.

But when she requested the transfer she was persuaded to stay. She told her major and lieutenant she was burned out. She had to leave the unit. “I love what I do,” she said, “but what I do is killing me.” But they said they couldn’t afford to let her go. Besides, wasn’t she only a couple of years from retirement? It was the soldier mentality, being part of a paramilitary organization, that ended up making her cave and decide to start therapy instead. And she had really been turning a corner until she got back in her car after this morning’s session and heard the police radio blowing up. And then flying triple digits down the back roads until she had wound as far as she could up Riverside Road into the bedlam of frantic parents and hundreds of dazed and helpless cops.

When she got out, a panicked mother grabbed her to ask where to go, saying that she couldn’t find her child. She had heard they were gathering parents in the firehouse, so she brought the mother there, and then she went looking for the van. Going up the hill toward the school, where it was cordoned off, was when she first started hearing numbers. A lieutenant she knew said: It’s bad, KK. It’s bad.Besides phone interviews, what other kind of reporting did you do to get to know them as well as you did?I hung out with them a good bit. I would go up to Connecticut and meet with Jeff, or go to Karoline’s house. I went to Jeff's house. One night the three of us got together at Jeff's and sat around the fire and drank beer and talked.

IN THE LOBBY everything had been left as it was. The shot-out windows let in the cold dark of early evening. Broken glass, still scattered on the brown-and-white tiled floor, crunched under the soles of the F.B.I. security detail standing by the front door. It was shortly after 5:30 p.m. on Dec. 20, the sixth day since the shooting, and Attorney General Eric Holder sat before the large TV screen that had been set up expressly for his visit.How were you able to reconstruct this scene? I never got Eric to talk to me, but I know what happened from talking to 10 other people who were with him. Margaret, his chief of staff, emailed me after the piece came out and said it was all exactly how it was.

A semicircle of folding chairs had been brought in from the cafeteria for the six detectives of the W.D.M.C. van squad; a handful of detectives from Central and Eastern Major Crimes; the F.B.I. special agents who had assisted over the past week; and the attorney general’s chief of staff. Holder had made the rest of the entourage who accompanied him on his visit to Sandy Hook — local and state politicians, including at least one senator, as well as the colonel of the State Police — wait in the line of gloomy black S.U.V.s while he entered to meet with the crime scene unit on their final day.

The TV screen, mutilated by one intolerable, impossible image after another, gave the attorney general just an abbreviated glimpse of the 1,495 photographs taken by Art Walkley over the past week: an uncensored, unredacted view of what they had faced when they first entered the school. As Jeff took Holder through each image, the only other sound in the lobby was the crinkling and uncrinkling of the tarp that hung over the hallway that led down to Classrooms 8 and 10. With each new image the A.G. seemed to grow smaller in his chair.

Sitting with Karoline was Sam DiPasquale. As a special-agent bomb tech for the F.B.I., stationed in New Haven, Sam first responded to the shooter’s home on Yogananda Street to check for explosives. After he finished there, after running the robot down the hallway to the mother’s bedroom, where she lay shot dead, he went to the school to see if there was anything he could do to help Jeff. They had known each other forever, having met at co-agency explosives and post-blast training sessions. Jeff’s team helped the New Haven office on several occasions. Sam even had them deputized at one point for a domestic-terrorism case. He would now assist them, making sure they had gas for their generators, making sure their team was fed every day, helping secure unusual equipment. He helped put plywood boards up over the windows in the two classrooms, mostly to protect patrol cops securing the perimeter from the impulse to look. In fact, most of his job was fending off all the captains, and majors, and state’s attorneys, and assistant attorneys general, from trying to see what he had to tell them over and over again they would not be able to unsee.

After 9/11, Sam was embedded with the Navy in Iraq, as part of the bureau’s largely unadvertised C.E.X.C. (Combined Explosives Exploitation Cell), deployed to suicide bombings to collect DNA for its database of bomb makers. He had picked limbs from trees. Defused homemade explosives. But the worst thing he’d ever seen was the inside of an elementary school in Connecticut.

Jeff decided that he and Sam would be the only two permitted to have phones inside, in order to limit photographs. They were the first ones there in the morning, and the last to leave at night. When Sam heard the attorney general was going to be visiting Newtown — a few days after President Obama spoke at a vigil at the local high school — he called an F.B.I. buddy he knew would be tasked with the security detail. He said that if possible, the school should be on his itinerary.

Jeff immediately seized on the idea. Sam had found him at one of the decon stations cleaning jewelry. It was something Jeff learned to do from the nurses at Bristol Hospital a million years ago when he was a paramedic. How to clean a piece of jewelry before returning it to the family. It was certainly nothing he learned at the police academy. But being able to perform such a task now, however much of it, was almost soothing after multiple days processing evidence in the tent that was set up initially as the temporary morgue.

Maintaining purpose did not come easily the last seven days and nights. But this was their chance to show the right person what they had seen. And so Sam set about securing everything Jeff said he’d need for the visit. Starting with a giant TV.

After the horrifying PowerPoint slide show, Karoline took Holder and his devastated chief of staff on a walk through the school, holding back the tarp that had concealed the aftermath of where Principal Hochsprung and the school psychologist were shot after running out from a meeting. In the conference room, across from Classroom 8, were 26 banker’s boxes containing each victim’s personal belongings. An old-timer from the van, Ray Insalaco, came in to help box up the desks. It had fallen to him to empty the 20 lunchboxes. His advice to the small crew he brought in: Don’t read the notes. He had already made the mistake when one fluttered out as he was dumping an uneaten lunch into the trash.

Thank God it’s Friday. Love, Mommy.

The A.G. and his chief of staff stood looking dumbly at the plain white boxes, each of the children’s bearing a purple-and-green butterfly name sticker that had come off their backpack hooks, until Karoline guided them into Classroom 10.You reconstruct this heartbreaking story with such evocative details and revelatory, often lyrical, prose. What is your writing process? Ninety percent of it is sequencing. Getting things into the right place first. It's like Tetris; it’s all ordering. You have all of the details and then you start thinking in terms of scenes and different buckets. This is going to go in this bucket, this in that bucket. It's like doing the dishes. The glasses have to go over here, and the silverware has to go over here, and the plates go over here, and the plastic Tupperware is over here. You separate things so that you can take small bites.  So every scene and every paragraph is built up of smaller bites, down to sentences, even fact-checking. I highly recommend not having a dishwasher.How do you do it mechanically? Well, first I’m just thinking about it all the time. Lying on the couch thinking about it. Taking walks thinking about it. Reading the interview transcripts over and over until you possess it, or it possesses you. And then when I'm going through transcripts, I’ll do what I call directionals. In brackets, I'll say this could go with this scene, or this echoes this part, like Jeff's detail about cleaning jewelry by hand echoes the ending of the story; it foreshadows that. I  prioritize images over ideas. I don't believe in the polemic or making arguments. I feel like we've all learned that's pointless. You can argue with yourself, but you can’t argue your reader into submission. If your main point is to try and change someone's mind, it’s not going to happen. I have a line on my bulletin board from John Crowe Ransome’s poem "The World’s Body:" “The image cannot be dispossessed of a primordial freshness which ideas can never claim.” I love that. Images actually communicate. Ideas are mostly useless abstraction. A numbered evidence tag marked where each small body was removed from the defiled carpet. Larger stains divulged where the two teachers fell. This was the same room where Dan Sliby, on their initial walk-through, found himself raging near the body of the shooter. Decades earlier, he was a first-grader in this very room. Pacing around the corpse, he could barely refrain from kicking it in the chest.

By a cluster of desks was the Bushmaster. Its barrel and muzzle brake were coated in a film of white powder. A less experienced observer might have thought that it was concrete dust from bullets hitting the walls. But Dan was sure, from his time in the Marines, that the chalky residue was baked evaporated blood.

Karoline then steered the attorney general, his stride no longer so firm, into Classroom 8. The room where, days before, their resolve wavered. Where they momentarily lost and regained their sense of purpose. Where they all stood in silent disbelief, a light drizzle on the window ticking off each annihilating second, staring into the tiny bathroom. Where the children were packed in so tightly that the inward-hinged door could not be shut all the way. Where Art, who had seen what he had thought must have been every possible reconfiguration of the human body, did not even understand what he was looking at. And where Karoline found herself doing something that came naturally: holding up an imaginary rifle, pointing it into the bathroom, registering the casings on the carpet to her right where the ejection port would have sent them and noting automatically that this was obviously where the shooter would have been standing when he fired the Bushmaster. It was when she felt Jeff looking at her that she dropped the imaginary gun and left the room.How do you draft a story? I print off whatever counts as a manuscript. It might be transcripts or shitty notes from journal notebooks. I'll print off my drafts and go at it with colored pencils, always working off the material and chipping away at it. I can’t really write on a screen.How do you develop fully-realized characters like Karoline? Part of the character development is developing the physical specifics and the physical details of the job itself and the setting and in every scene, well, what would they be doing? For instance, Karoline telling me that when she was standing outside the bathroom and she found herself just kind of holding up an invisible rifle, looking down into the bathroom — that that's just something she automatically always did.

She went to the next classroom over, which had been spared. She needed a minute to collect herself. Steve Rupsis followed, struggling to keep his head focused on forensics. He kept asking her what he should do. How should I video this, KK? How should I get the overall shot? Should I sketch the lobby and the classrooms separately? Are we going to sketch? Do you want me to sketch? He was spiraling. She told him what she needed was a minute. He backed away.

That was when Jeff, his face tear-stained, gave them the purpose that they would desperately need to get through the next week.

“Look,” he said, “we’re going to do this the same way we always do it. We’re just going to do it 26 times.” Same thing as always, 26 times. It became like a mantra. We’re going to do what we always do. Same procedures. Same four overall photos of each room. Same medium shots. Same untold number of close-ups to memorialize every minuscule aspect of the work.

They would set up staging tables in the tent for mass processing of the evidence, nothing they’d ever done at this scale. With eight tables it went like an assembly line. Each item photographed against a neutral backdrop. They had a 20-pound roll of butcher paper on the truck just for this purpose. One clean sheet, with a glove change in between, for each item, each piece of clothing. Each small shirt. Each elfin dress. Each backpack. Each barrette. Charm bracelet. Wedding ring. Each bloodied shoe. Same thing as always, 26 times.

Jeff reminded them that something like destiny, however grim and profoundly unwanted, had been laid at their feet. That the country, the world, would come looking for answers was not a question. And if anyone was going to provide the answers, at least to what had happened in these rooms, it would be up to them, but only if they kept their heads. This clarity of purpose was what allowed them to move ahead that day, and to continue on, working 12 and 16 hours, only stopping to get in their cars long enough to drive past the procession of garrisoned media trucks and excruciating makeshift memorials, cairns of teddy bears and stuffed hearts, to sleep a few hours before returning the next morning.Why did you choose a third-person point of view that inhabits the minds of the three main characters, one that focuses primarily on the inner thoughts, feelings and experiences of Jeff, Karoline, and, to a lesser extent, Art?The material I had and the level of disclosure on the part of your subject dictated that approach. Karoline was much more emotionally generous and open. Jeff preferred to talk more about the professional technical aspects of the job. Art was incredibly giving both emotionally and with the technical stuff. I chose the three of them because in a sense, as a team who had worked together forever, they're basically the same person, deeply bonded. One of the more meaningful parts of what I was getting from my time with them was how protective they were of each other, how protective they were of the dignity of the children who'd been killed, all the victims. It shapes itself as you get deeper into it and realize this is not what it did to them just as individuals but as a group — not just them seeing the violence, but seeing each other go through it.

FROM THE VERY FIRST, they faced resistance. No sooner had they secured the crime scene than the chief medical examiner showed up, plopped down at one of the teacher’s desks and began telling Jeff’s team not to waste time taking photographs. They did not need to be so overzealous.Why didn’t you identify the ME by name or any of the other top officials, except for Attorney General Holder?I only name the people who are part of that crew; that’s a way of highlighting who we’re focusing on. These other people are not even secondary but more tertiary characters. If it's not a character that you're going to develop,  I don't think you always need to name them.

Because the Office of the Chief Medical Examiner had jurisdiction over all bodies in the state of Connecticut, Jeff’s team was not permitted to move or touch a body until the M.E. first signed off. Normally the crime scene unit obtained permission over the phone, or from a representative on site. They knew the M.E. well enough, from various aspects of death investigations, but Karoline recalled only ever seeing him once at a crime scene in her 13 years on the van. Now here he was, barking unsolicited advice, sitting at the desk of a teacher who still lay on the floor near another teacher with a child’s body in her arms. They all knew what happened here, he said, everyone knew it wasn’t going to go to court, at least not criminal, so his own photographers could take all the necessary pictures once they got the bodies to autopsy. The main priority, he insisted, was getting the bodies back to the families. The governor needed to make a statement.

The need to return the bodies to the families as quickly as possible was obviously more than understandable. But not to conduct a full investigation, not to take photos, was unthinkable. And who the hell knew yet if there was even an accomplice? Who knew anything yet? To cut corners, to not document every centimeter of the scene while it was intact, would itself be criminal: a failure that would only leave the families with unanswerable questions. Their own work told a story that no longer existed on the medical examiner’s metal table.

At 8:35 p.m., the bodies were removed and taken to the O.C.M.E., and the governor informed the parents.

The crew worked on. They were interrupted again and again. One day it was the F.B.I. unit that worked on profiles of shooters and serial killers. Other times it was people they felt had no business being there, enough that they began referring to them as the dog and pony shows. A high-ranking official from the L.A.P.D. showed up out of nowhere wanting a special walk-through. Various brass with various justifications. The problem was that during these interruptions it was not as if they could just step outside for a break. The problem was being forced to stop, but never long enough to go through the tedious steps of decon, the process of changing out of their Tyvek, bootees, hairnets, gloves, having to completely re-suit, and so they ended up just standing around, noticing all the little things they had been trying not to notice. Pokémon cards and Little Mermaid this and that, stuff their own kids had at home. The Christmas projects the children had been working on for their parents. The drawings of stick-figure families huddled on the couch reading. The cups of milk still on the children’s desks along with crayons and scissors and sheets of stiff-bright construction paper: the last thing they would get to do in this life before the strange man with yellow plugs in his ears and a loud gun entered.

Karoline was stopped short by something one of the kids had written on the board where they put their big goals for the year. This kid’s was to tie their own shoes. There were other, grander ambitions. “I want to read chapter books.” “I want to learn to count numbers.” “I want to write stories wenn I can.” A stick figure in green shoes announced, “cat to loo gokig tosgy,” because all motives were not as easily articulated, or even articulable. It struck her as immensely cruel that this kid had learned to die before they had learned how to tie their shoes. It made her think back to when she herself had learned to tie her own shoes, practicing on her father’s work boot. She saw it vividly now, smelling of diesel and leather, almost as if it were sitting on one of the desks, eyelets waiting to be laced. It was so simple, in the purest sense of having straightforward purpose. How beyond it we were as adults, she thought; you just tied your shoes in the morning without thinking. Even if it was the only part of your day that made sense. Maybe the whole problem was that our goals as adults were far more make-believe than any goals most kids had. You tied your shoes so they stayed on your feet when you ran! Unlike the simple purpose of the bootees she put over her shoes to prevent the bloodied ground beneath her feet from seeping in and contaminating her ability to impassively bear witness.Did the passage of time between Sandy Hook and now make it more possible to write the story?It helped a lot I think. Karoline is retired. Jeff is in private security. Only Art is still in actual law enforcement, but not CSI. And the passage of time gave me time to digest and think and find the essential story worth focusing on.

The strangest interruption had to be when their lieutenant stopped by to let them know not to pay any attention to the news, but apparently there were people in the outside world saying what they were seeing was not real at all, only an elaborate hoax. They had no idea what to make of this intrusion of sinister make-believe.

When they heard Attorney General Eric Holder was coming, the highest-ranking law enforcement agent in the United States, a policymaker of the highest echelon, they knew it was their one chance. To show the scene as they found it. To present the evidence to the right set of eyes. If what they saw did not shake the country out of its denial, nothing would. We’re pretending things aren’t the way they are, Jeff said.

So there would be no sugarcoating it, not for the attorney general. If they were the ones who would provide the answers that would help break the country from its spell, it first meant waking up the right people. Not the local legislators, not the governor, not the politicians who could all wait in the motorcade fidgeting on their phones with the heat running. It would require only a small but debilitating sample from Art’s camera: a dozen out of the 1,495 seared into his Nikon D300. And they would not present it in a neutral space like the cafeteria. They were going to show him right in the center of the nightmare where they had been working the past hellish week. So he could see it and smell it and feel it under his shoes. So he could see how 80 rounds fired into a three-by-four-foot bathroom trenched the cinder block. How 16 children crammed in that tightly had not had the space to fall where they stood. How innocence could be transformed to gore in an instant.

But even though the attorney general was convinced in this moment, reeling on the threshold of this tiny, obliterated bathroom, that if the American people only saw what he was seeing, Congress would be forced to do the right thing, nothing would change in the end.  Tragedy in America would prevail. Some would say nothing has changed because we have not yet been made to see. After each new mass shooting, the question, the debate, returns. Would seeing the crime-scene photos have an effect on the gun crisis in the same way images of Emmett Till’s body in an open coffin had on the civil rights movement? The Sandy Hook photographs have been redacted by Connecticut state law since 2013. Even if the law were to change with the consent of the families of the victims, who pushed for the legal restriction, public viewing of the photographs would require one outlet or another to first make the decision to publicize the images. And in a culture where reality is no longer agreed on, many will not believe what they see unless it is funneled through their propaganda of choice. So until that unlikely moment arrives, the full truth of these images and those of shooting after shooting, for the decade after Sandy Hook and into the future, will live on only in the atrocity exhibition that exists in the memory of those who photograph, measure and collect the foul evidence.

THE DAY AFTER the van crew left the school for the last time, Karoline’s sister and her partner, Elissa, took her out. Elissa drove her to therapy that morning. Then they went holiday shopping, to get her away from the news. Both were keeping a close eye on her, one on either side like her own personal guardrails, back out in a world where Christmas had not been canceled.

She had agreed to go along. To do a little shopping for her nieces and nephews. At first she thought she was fine. But inside the Christmas Tree Shops, that permanent bazaar of candy canes and singing reindeer, she started to feel overwhelmed by the shimmery music, the red tinsel and the horrific crimson bulbs everywhere. The same kind the kids had been working on for one of their projects. Each red ball branded with a small white handprint and left to dry on the sill in the classroom. Each precariously balanced on a Dixie cup, until Karoline sent them flying while trying to string a bullet hole, measuring the trajectory from where the gun might have been fired. She managed to catch them before she had to find out what the sound of the bulbs shattering on the floor might do to her. Now, in the middle of the store, where Bing Crosby was loudest, and her disgust at the lurid make-believe normalcy was ankle-deep, she was hyperventilating. Then she was fleeing outside. Away from the shoppers with their peppermint lattes oblivious to the dark red abyss at their feet.This scene is so heartbreaking. What’s your method of discovering this kind of material?You have to be obsessive. You're looking for the right  details — the ones that just move you deeply or seem so evocative. Again, it's the image over the idea. But then, at a certain point, it's one that holds the whole story together. You know it's worth taking the time because they'll always show up.

What she felt more than anything, out here in the shockingly complacent world, was that she belonged back in the school. It did not feel right to have left them behind.

Most homicide reports took three to four months. She still had a number of cases from before waiting on their own reports. By the time she sat down to begin the Sandy Hook report, it was March.

To spare the others, Art and Karoline decided they would be the only ones who would have to look at the photographs. Karoline rearranged her schedule to work nights. She had two screens set up: photographs on one, scene report on the other. On her desk she had her knight’s helmet for company. Elissa had given it to her as a gift; it was just about big enough for a cat. The thought passed through her mind sometimes that she was a knight in a past life, and this helped in a small way to get through it, to lower the visor, to move forward through the awful spell cast by the images.Why did you use this detail? I think the idiosyncratic detail here and there reminds us that this is real. And real means there are these little endearing oddities everywhere. I guess I feel like it’s important to occasionally nudge the reader or yourself, the writer, with a weird, idiosyncratic detail, just to remind us that there's just nothing normal going on here. It's getting inside someone's head. We all have what we would call strange things, so I try not to filter it out. She told me that for a reason.

Art chose to do his end at home, writing a précis of each of the 1,495 photographs that he had taken, working at his dining room table, headphones on, listening to Pachelbel. His wife played it while pregnant with both of their daughters, and recently she and the girls had been listening to it at bedtime, so it was a way to be with them, he supposed, though the truth was it had been hard. He had found to his horror that at first, for a time, he could not even look at them. He could not give them a hug good night. His wife didn’t understand. On Christmas Day, he could not stay in the room to watch them opening their gifts. He could not stop thinking about what the other parents had done with the presents they bought for their children.

Even as they were reliving it, there were people insisting it was not real. The least welcome of all interruptions for the state police now were calls from nutjobs with bizarre questions and accusations. One day when Karoline was leaving a kickboxing class, the instructor introduced her to another woman who had also been at Sandy Hook. Karoline did not recognize the woman, who said she was a trauma nurse at the Connecticut Children’s Medical Center, in Hartford, and had been called to help the medical examiner identify the victims. The woman said she had been in the tent and was having trouble getting the faces of the children out of her head. What faces? Karoline thought. And then, after she called the medical examiner, and the director of pediatric trauma at the Children’s Medical Center, and been told that no such person had ever worked at either place, that it was all made up, she confronted the woman, who then broke down and confessed to being a pathological liar. Her real job was at a day care center. It was weird how some people refused to believe, while others who had not seen a thing needed to pretend they had been there.

The work of writing the report was unavoidably retraumatizing. There was no getting around it. But Karoline was all right with it falling to her, in part because it let her return to what she had not felt ready to leave. It allowed her to be back with the children. It also gave her the only degree of purpose she had been able to locate since they had left the school.

Like the red ornaments in the Christmas store, other things had a way of finding her, like getting stuck behind a school bus letting off kids, and of course they would have to be the same age, too small for the ridiculously oversize backpacks. The backpacks left behind in the classrooms were the most haunting remnant of the vanished children. There was also the recurring nightmare in which she was trying to process an active shooting as it unfolded, hiding among the victims, frantically doing her best, trying to take notes, fumbling with evidence bags, sketching what she could, as if processing it would make the shooting stop faster.

In April, she looked up one day and saw a news report: “Just four months after the shocking events at Sandy Hook. … ” The chyron below read: TERROR ATTACK AT BOSTON MARATHON. And she saw Sam DiPasquale on Boylston Street working the blast area for the second bomb. Outside the restaurant with the blown-out windows. This would happen over and over again, too, like the shootings, but all that would come out of it, she thought, were more useless “lessons learned.” Absurd plans to trot out to the public like how teachers or children might disarm a grown man carrying a military-style rifle. The only sure thing was that there would be a next time.

WHEN THE F.B.I. profilers requested another meeting, Karoline felt herself hitting the wall. She was frustrated to have to break her flow to go share information that she did not feel like sharing. But she and Art put together a presentation, along the lines of what they had made for Holder, but much more “aggressive,” as Karoline put it. She had been looking at the photos for months now.How do you decide what to keep and what to cut? I get drunk on the idea of just telling everything. You need to take on this mentality. At a certain point, another more sober version takes over. But there's no way that you're going to know ahead of time how to select. It becomes a gradual process of attrition. It’s like making maple syrup. I’m from Vermont. You gather up hundreds of gallons of sap from trees and you bring it down to the smokehouse and then you boil it down.

It was a small group. The major, there to accompany the feds, sat in the back corner of the conference room.

After it was done she returned to her desk. Ten or 15 minutes later the major came in looking pale. He looked at her. “Oh, my God,” he said.

She did not even look at him. “Yeah?”

He said, “Well, why don’t you take the rest of the day off?” His sickly pale was seeping into the walls.

She looked at him and said: “Major. I don’t need the rest of the day off. We need a fucking break. We need to be off call. You can’t ask us if we want to be off call, because we’re not going to tell you that, we need to be told to be off call. We’re soldiers. We’re going to do what we’re told. Today you got to see some.  A handful of photos, and you’re all blown away. Now you know what I’ve been living.”Narrative nonfiction writers typically rely on quotes, especially dialogue, to propel the action and reveal truths about characters, yet this is the only time you employ this device. Why? She was telling her boss to fuck off and I couldn’t have put it better than she did.

She knew it was insubordinate, conduct unbecoming, but she honestly no longer cared. At this point the protective fog was gone, as if the roof had come off the barracks and the blades of a falling helicopter had blown it out to sea.

THE PROBLEM WAS that it was going to happen again. There had even been a threat against the new school where the surviving children had been moved. Not that Karoline feared they wouldn’t be able to do the job. They would. They would get back on the van; they would be the ones to go back out.

But despite all of the elite training they had, there was none for how to stay sane. She was having heart palpitations. Her general anxiety was heightened. She had become hypervigilant again and more controlling at home. One of the few things that gave her a sense of purpose was when the detective over at Troop A, Rachael Van Ness, a liaison for the families, would call with the most solemn of requests, from a parent who wanted to know where their child had been standing when they had been shot. In that case, Karoline would go back and check the sketch maps and provide the measurements, from one wall to the other, to fix the exact place, so the parents could return before the school was razed and stand where their own flesh had been terrorized and murdered.

On occasion, Rachael would call to ask if Karoline had found something that had not followed a child to autopsy. Things that might have sounded insignificant, like a pencil, or a thermos, but Karoline knew there were no insignificant requests. She would happily drop everything to help by going back into her notes, exhibit reports, the shattering trove of Art’s photos. She would start with the encompassing overall shots, zero in on the mediums, and then the close-ups, until she could find the exact location where the child had been. On the dozen or so occasions when she found something, even if it was small, like a special eraser, it made her feel good.

One day she received a request from this same detective from Troop A, whose job, Karoline felt, must have been infinitely more difficult than hers, not one she would have traded for in a million years. In this instance, a grandmother wanted a piece of jewelry that hadn’t come home with her granddaughter’s things. A tiny heart pendant. The kind that was split in half. The grandmother had one half, and the girl had the other. It was only the size of a pinkie nail. The girl had apparently worn it all the time.

After not finding it in any of the photos, she went down to the evidence room, where Steve Rupsis was evidence officer when he wasn’t on the van, and they pulled the bullet shrapnel. It had all been preserved. Right down to the smallest fragments of copper and lead that Dan Sliby had had to chisel up out of the dense matrix of dried blood inside the bathroom. Then they dumped it out and sorted and picked through it, every piece, to see if the pendant had been swept up with the shrapnel from Classroom 8. Because that was where the girl and her half of the heart had been.

They did not find it. They never would.Why did you choose the futile search for heart-shaped pendant to bring the story to a close? Were alternatives considered?How could it end otherwise? And that was the choice by Jessica Lustig, my editor, to end it on that beat, which I thought was perfect. I had a much longer, different ending and she ended it on that excruciating  gut punch, which was an exquisite editorial choice. What was the emotional cost to you of working on this brutal story?I would just say pretty high. I went into it with a little hubris. Actually, I went into it terrified. I went to visit the Pennsylvania State Police crime scene investigator who had covered the Nickel Mines shooting and he just plopped me down in a chair in his office and opened up his laptop and started showing me the images from that schoolhouse shooting. They were all young girls from the age of 7 or 8 to 14. I had to see a lot of stuff I wish I hadn't seen. But I was fine. I'm not a tough person, but when I put on the reporter's hat, I guess it’s like one of those Tyvek shells, and you’re standing in the middle of it, but you think you’re protected. It’s like: let’s go stare into this abyss for a while, let's see how this goes. I definitely thought that I was okay and then all of a sudden I was not.How long did that last? Oh, a while. It was not good. Most of it was after I finished the story and the following year. It got a huge reception. Let's just say that I was not really present to celebrate it.

Jay Kirk is the author of “Avoid the Day: A New Nonfiction in Two Movements.” This is his first feature article for the magazine.

* * *

Chip Scanlan is an award-winning writer who taught at the Poynter Institute and now coaches writers around the world. He is the author of several books on writing and the newsletter Chip's Writing Lessons.

The post What happens to crime scene investigators who have to see the unseeable? appeared first on Nieman Storyboard.

Journalists who write profiles don’t go in cold. They pre-report to prepare for crucial interviews. They read widely and research previous stories that have been written about their subject. They think through questions they need to ask.

To profile Geoffrey Hinton, the British-Canadian computer scientist considered the godfather of artificial intelligence, New Yorker writer Joshua Rothman did his homework, and then some. He read histories of A.I., consulted an oral history of neural networks — a machine-learning technique that teaches computers to process information like the human brain  — and plowed through a textbook about deep learning. He even took an online course in linear algebra before he felt ready to query Hinton.

“I’d written about A.I. and machine learning before, but the idea of spending an extended period alone with Geoff, and asking him to explain his work — it made me feel like I really needed to be prepared,” Rothman told me. “I wanted to make the most of the time I had.”

All that prep helped Rothman understand and then convey the science behind artificial intelligence, the technological development that has seized the public’s attention ever since OpenAI released its chatbot, ChatGPT, in November 2022. Rothman’s diligence grounded a 10,000-word nuanced, comprehensive and revealing portrait of Hinton, published in The New Yorker in November 2023.

“Why the Godfather of A.I. Fears What He’s Built” is an important story, deeply reported, creatively structured and written with literary grace. Rothman frames it within a four-day visit he made to Hinton’s home on a private island on Ontario’s Georgian Bay. The narrative glides through connections that track the history of A.I. and Hinton’s career with the personal tragedies that have shaped him.

“Often, I think, we write about people because of something they’re involved in, or because of something they’ve done,” Rothman said. “But when it comes time to write the profile, it becomes important to focus in on the person as a person — just as an unadorned human being, with all the richness and intensity that entails.” Even the word “profile” is telling; Rothman and Eric Overbey wrote in an introduction to 12 classic New Yorker profiles: “It suggests catching sight of someone from an unusual angle.”

The profile of Hinton opens with an examination of the brain at work and comes to a close when the natural world eclipses the technical one. In between, it explores a new and, to some, discomfiting frontier of knowledge and ethics. Rothman uses sharply-etched scenes, characterization, dialogue, metaphors and digressions to help readers grasp historical context and challenging technical details.

Rothman, the Ideas Editor of The New Yorker, began his journalism career blogging for the ideas section of The Boston Globe and then freelancing before joining The New Yorker in 2012, where, as ideas editor of newyorker.com, he guides and writes stories about science, philosophy and technology. He helped conceive the Hinton profile for an issue devoted to artificial intelligence, pegged to recent concerns about the evolution of technology that has sparked concerns about potential abuse. Hinton’s take is invaluable, Rothman said, “especially given his recent emergence as someone sounding the alarm about A.I.”

In an email exchange, Rothman described the reasoning behind the story’s structure, the challenge of conveying technical knowledge in accessible ways and the collaborative relationship with his editor. The Q&A has been edited for length and clarity, and is followed by an annotation of the story.

Why did you become a journalist?
I wanted to be an English professor, and I studied for a PhD in English. But I was wrapping up my dissertation in 2008, and the financial crisis brought the professorial job market to a halt. I’d always wanted to be a writer and had wide-ranging interests, including in science and technology. It felt like journalism would be a practical path forward that would also allow me to explore a lot of different subjects.

What does the position of ideas editor at The New Yorker entail?
I edit pieces about ideas, broadly construed — pieces that touch on science, technology, philosophy, literature, history and so on — mostly for the web site, but also for the print magazine. And I write a few pieces a year, fitting it in where I can.

What writers have influenced you and how in general, and as you reported and wrote your story?
I always wish I could write a profile the way Larissa MacFarquhar writes them. And when I write about technology, I always revisit William Gibson (whom I was able to profile a few years ago). He reminds me that the newest technologies have roots in the past. Everything new is old. You always have to rewind further than you think.

What is the origin story behind “Why the Godfather of A.I. Fears What He’s Built?”
I’ve been covering A.I. for a few years as an editor and writer. It’s a really complicated subject, both technically and conceptually, and it’s also a sweeping story going back at least a century, combining history and philosophy and science fiction and business. It’d be crazy to think you could combine all that in a single piece, but it seemed like writing about Hinton would get you into the ballpark. So once we started talking about doing a special issue on A.I., a profile of Hinton seemed like a good idea. Especially given his recent emergence as someone sounding the alarm about A.I.

What role did your editor play during the stages of writing and finishing your story?
Henry Finder, editorial director of the New Yorker, has been my editor for a long time now. Working with him is a little like being in a writing seminar that progresses from piece to piece. With this piece, I wanted to apply lessons that I’d learned from him while writing the previous one—a profile of the novelist Kim Stanley Robinson. He pushes the pieces forward in so many ways, in terms of craft but also in terms of thinking. And he always helps me keep in mind what the ultimate goals are — what the ideal version of the piece should be.

Much of your story is an intellectual challenge to absorb and understand. Who do you envision as your audience?
I wanted this piece to be something anyone could read. I hoped it would pull readers into the intellectual adventure of A.I., and that it would help them understand how a pure and profound curiosity about our own minds got us to this point. Once I got to know Geoff Hinton, I wanted the piece to reflect him — his personality and experiences. He’s a fascinating individual in his own right, with a life story that has a great deal to tell us just about life, outside of A.I. So I don’t think of it as a piece for a technology-oriented audience. I hoped to write a humanist piece.

ANNOTATION: Storyboard’s questions are in red; Rothman’s answers in blue. To read the story without annotations, click the HIDE ANNOTATIONS button in the right-hand menu of your monitor or at the top of your mobile screen.

Why the Godfather of A.I. Fears What He’s Built

Geoffrey Hinton has spent a lifetime teaching computers to learn. Now he worries that artificial brains are better than ours.

By Joshua Rothman
The New Yorker
Nov. 13, 2023

[dc]I[/dc]n your brain, neurons are arranged in networks big and small. With every action, with every thought, the networks change: neurons are included or excluded, and the connections between them strengthen or fade. This process goes on all the time—it’s happening now, as you read these words—and its scale is beyond imagining. You have some eighty billion neurons sharing a hundred trillion connections or more. Your skull contains a galaxy’s worth of constellations, always shifting. You begin the story by describing the inner workings of the brain. Why? Because that’s where A.I. started. It began as an offshoot of an introspective, neuroscientific effort. I also wanted to raise the question, early on, of what human intelligence is and how it might work. I wanted readers to look inside themselves and begin to consider the possibility that the mind is a kind of machine.

Geoffrey Hinton, the computer scientist who is often called “the godfather of A.I.,” handed me a walking stick. “You’ll need one of these,” he said. Then he headed off along a path through the woods to the shore. It wound across a shaded clearing, past a pair of sheds, and then descended by stone steps to a small dock. This sentence adheres to the “rule of three.” Was that a conscious writing decision? I’ve never heard of the rule of three. Although maybe I think that three-part constructions sound good. “It’s slippery here,” Hinton warned, as we started down.

New knowledge incorporates itself into your existing networks in the form of subtle adjustments. Sometimes they’re temporary: if you meet a stranger at a party, his name might impress itself only briefly upon the networks in your memory. But they can also last a lifetime, if, say, that stranger becomes your spouse. Because new knowledge merges with old, what you know shapes what you learn. If someone at the party tells you about his trip to Amsterdam, the next day, at a museum, your networks may nudge you a little closer to the Vermeer. In this way, small changes create the possibility for profound transformations. The reader quickly becomes aware of the story’s woven structure. How did you select this approach? I knew from the beginning of the writing process that I wanted to braid the science and the life together. The challenge was making it work. But I wanted it to be clear, from the beginning of the piece, that we would be learning about Hinton’s life on a personal level. I think it’s important, in a long piece of writing, to try and move all the elements forward together instead of siloing them into separate sections.

“We had a bonfire here,” Hinton said. We were on a ledge of rock jutting out into Ontario’s Georgian Bay, which stretches to the west into Lake Huron. Islands dotted the water; Hinton had bought this one in 2013, when he was sixty-five, after selling a three-person startup to Google for forty-four million dollars. Before that, he’d spent three decades as a computer-science professor at the University of Toronto—a leading figure in an unglamorous subfield known as neural networks, which was inspired by the way neurons are connected in the brain. Because artificial neural networks were only moderately successful at the tasks they undertook—image categorization, speech recognition, and so on—most researchers considered them to be at best mildly interesting, or at worst a waste of time. “Our neural nets just couldn’t do anything better than a child could,” Hinton recalled. In the nineteen-eighties, when he saw “The Terminator,” it didn’t bother him that Skynet, the movie’s world-destroying A.I., was a neural net; he was pleased to see the technology portrayed as promising.

From the small depression where the fire had been, cracks in the stone, created by the heat, radiated outward. Hinton, who is tall, slim, and English, Why did you select these particular three descriptors? This is a kind of arm’s-length description — like if you were squinting to make Hinton out from a distance, this is what you’d see. I could’ve described him in greater detail here, but didn’t want to lose momentum. poked the spot with his stick. A scientist through and through, he is always remarking on what is happening in the physical world: the lives of animals, the flow of currents in the bay, the geology of the island. “I put a mesh of rebar under the wood, so the air could get in, and it got hot enough that the metal actually went all soft,” he said, in a wondering tone. “That’s a real fire—something to be proud of!” You’re capturing small moments in exquisite detail. What do you use in the field to record interviews and what you’re seeing? I have a small handheld voice recorder, and I also take a lot of pictures. I almost always have a camera with me. At the end of each reporting day, I take notes about what happened and begin to write out the scenes from memory and notes.

For decades, Hinton tinkered, building bigger neural nets structured in ingenious ways. He imagined new methods for training them and helping them improve. He recruited graduate students, convincing them that neural nets weren’t a lost cause. He thought of himself as participating in a project that might come to fruition a century in the future, after he died. Meanwhile, he found himself widowed and raising two young children alone. During one particularly difficult period, when the demands of family life and research overwhelmed him, he thought that he’d contributed all he could. “I was dead in the water at forty-six,” he said. He didn’t anticipate the speed with which, about a decade ago, neural-net technology would suddenly improve. Computers got faster, and neural nets, drawing on data available on the Internet, started transcribing speech, playing games, translating languages, even driving cars. Around the time Hinton’s company was acquired, an A.I. boom began, leading to the creation of systems like OpenAI’s ChatGPT and Google’s Bard, which many believe are starting to change the world in unpredictable ways. You condense the early parts of Hinton’s relationship with A.I. to a single paragraph. How do you achieve that concision? This paragraph is based very heavily on a summary of his life and career that Geoff himself gave me during our very first conversation over Zoom. My mother-in-law has a theory that people tell you about the most important aspects of their lives within the first fifteen minutes of conversation. In Geoff’s case, I think it was true. “I was dead in the water...” — it was such a moving thing to say, and I knew that I wanted to include it high up to help readers understand what a forthright, humane person he is.

Hinton set off along the shore, and I followed, the fractured rock shifting beneath me. “Now watch this,” he said. He stood before a lumpy, person-size boulder, which blocked our way. “Here’s how you get across. You throw your stick”—he tossed his to the other side of the boulder—“and then there are footholds here and here, and a handhold here.” I watched as he scrambled over with easy familiarity, and then, more tentatively, I took the same steps myself. You are present in the narrative. Why? I’m pretty much always present in everything I write. It’s just how I approach things.

Whenever we learn, our networks of neurons change—but how, exactly? Researchers like Hinton, working with computers, sought to discover “learning algorithms” for neural nets, procedures through which the statistical “weights” of the connections among artificial neurons could change to assimilate new knowledge. In 1949, a psychologist named Donald Hebb proposed a simple rule for how people learn, often summarized as “Neurons that fire together wire together.” Once a group of neurons in your brain activates in synchrony, it’s more likely to do so again; this helps explain why doing something is easier the second time. But it quickly became apparent that computerized neural networks needed another approach in order to solve complicated problems. As a young researcher, in the nineteen-sixties and seventies, Hinton drew networks of neurons in notebooks and imagined new knowledge arriving at their borders. How would a network of a few hundred artificial neurons store a concept? How would it revise that concept if it turned out to be flawed? How did you keep track of your structure: outline, notecards, timeline, or other ways? I have no system! I used to make outlines, but now I just keep track of things in my head.

We made our way around the shore to Hinton’s cottage, the only one on the island. Glass-enclosed, it stood on stilts atop a staircase of broad, dark rocks. “One time, we came out here and a huge water snake stuck his head up,” Hinton said, as we neared the house. It was a fond memory. His father, a celebrated entomologist who’d named a little-known stage of metamorphosis, had instilled in him an affection for cold-blooded creatures. When he was a child, he and his dad kept a pit full of vipers, turtles, frogs, toads, and lizards in the garage. When did you learn this? Did you interview Hinton apart from those during your visit to his island home? He told me this at some point during the visit. And, apart from a short Zoom interview we had when I first proposed the story to him, I kept our discussions in-person, on the island, so that as much of the piece as possible could be situated there. Today, when Hinton is on the island—he is often there in the warmer months—he sometimes finds snakes and brings them into the house, so that he can watch them in a terrarium. He is a good observer of nonhuman minds, having spent a lifetime thinking about thinking from the bottom up.

Earlier this year, Hinton left Google, where he’d worked since the acquisition. He was worried about the potential of A.I. to do harm, and began giving interviews in which he talked about the “existential threat” that the technology might pose to the human species. The more he used ChatGPT, an A.I. system trained on a vast corpus of human writing, the more uneasy he got. Why did you use the word “corpus” rather than one that might be more familiar?I think it’s the most accurate. It’s a curated collection of writing, maintained and adjusted at great expense by experts. One day, someone from Fox News wrote to him asking for an interview about artificial intelligence. Hinton enjoys sending snarky single-sentence replies to e-mails—after receiving a lengthy note from a Canadian intelligence agency, he responded, “Snowden is my hero” and he began experimenting with a few one-liners. Eventually, he wrote, “Fox News is an oxy moron.” Then, on a lark, he asked ChatGPT if it could explain his joke. The system told him his sentence implied that Fox News was fake news, and, when he called attention to the space before “moron,” it explained that Fox News was addictive, like the drug OxyContin. Hinton was astonished. This level of understanding seemed to represent a new era in A.I.

There are many reasons to be concerned about the advent of artificial intelligence. It’s common sense to worry about human workers being replaced by computers, for example. But Hinton has joined many prominent technologists, including Sam Altman, the C.E.O. of OpenAI, in warning that A.I. systems may start to think for themselves, and even seek to take over or eliminate human civilization. This strikes me as the news peg, if you will. Is this the story’s central take? It is definitely a peg. But to me the piece is actually centered on the question of whether artificial intelligence is actually a kind of real intelligence. And the real news peg, from that perspective, is the advent of the large language model, which is a technology that seems intelligent in a new way. It was striking to hear one of A.I.’s most prominent researchers give voice to such an alarming view.

“People say, It’s just glorified autocomplete,” he told me, standing in his kitchen. (He has suffered from back pain for most of his life; it eventually grew so severe that he gave up sitting. He has not sat down for more than an hour since 2005. In an audio of you interview with Hinton played back during your New Yorker Radio hours, we hear him telling you, “You’re just an autocomplete device.” As a writer and editor, what was your reaction to that?Well, I think a lot depends on how that statement is meant. I don’t think that Hinton is trying to reduce or denigrate me or my work or humanity by saying it. He certainly knows that human life is rich, that our capabilities are multi-dimensional, that our lives have value, and so on. What he is saying is that something like autocomplete — that is, an unconscious, statistical process that uses what’s already happened to extrapolate about what’s next — is a big part of our mental lives. Maybe even the bulk of our mental lives. And I think it’s hard to dispute that.“Now, let’s analyze that. Suppose you want to be really good at predicting the next word. If you want to be really good, you have to understand what’s being said. That’s the only way. So by training something to be really good at predicting the next word, you’re actually forcing it to understand. Yes, it’s ‘autocomplete’—but you didn’t think through what it means to have a really good autocomplete.”You alternate between your own descriptions of the workings of the brain and A.I., but use Hinton’s words as well. How did you decide which one to use when to further the narrative? Geoff is an incredible communicator, and he often puts things in a powerful, concise, insightful way. But there are other times when what he says isn’t quite sufficient for the purposes of the story. For example, the story might benefit from having a theme that’s been introduced higher up pulled through to the current moment. In those instances, I will sometimes interject my own formulations. Hinton thinks that “large language models,” such as GPT, which powers OpenAI’s chatbots, can comprehend the meanings of words and ideas. Why did you use quote marks for “large language models” and then not provide a phrase defining that term?I figured that “such as GPT” was explanation enough, and didn’t want to bog things down.

Skeptics who say that we overestimate the power of A.I. point out that a great deal separates human minds from neural nets. For one thing, neural nets don’t learn the way we do: we acquire knowledge organically, by having experiences and grasping their relationship to reality and ourselves, while they learn abstractly, by processing huge repositories of information about a world that they don’t really inhabit. How did you learn this?I suppose it’s just general knowledge, gleaned while covering this subject over the years. I extrapolated it from my training data, as it were. But Hinton argues that the intelligence displayed by A.I. systems transcends its artificial origins.Could you point to any section and say, “This is the nut (or context” graf?” I think it’s the very last paragraph of this section. Hinton’s biggest, most important argument is: It started as one thing, and it’s become something else.

“When you eat, you take food in, and you break it down to these tiny components,” he told me. “So you could say that the bits in my body are made from bits of other animals. But that would be very misleading.” He believes that, by analyzing human writing, a large language model like GPT learns how the world works, producing a system capable of thought; writing is only part of what that system can do. “It’s analogous to how a caterpillar turns into a butterfly,” he went on. “In the chrysalis, you turn the caterpillar into soup—and from this soup you build the butterfly.”

He began rooting around in a small cupboard just off the kitchen. “Aha!” he said. With a flourish, he put an object on the counter—a dead dragonfly. It was perfectly preserved. “I found this at the marina,” he explained. “It had just hatched on a rock and was drying its wings, so I caught it. Look underneath.” Hinton had captured the dragonfly just after it had emerged from its larval form. The larva was a quite different-looking insect, with its own eyes and legs; it had a hole in its back, through which the dragonfly had crawled.

“The larva of the dragonfly is this monster that lives under the water,” Hinton said. “And, like in the movie ‘Alien,’ the dragonfly is breaking out of the back of the monster. The larva went into a phase where it got turned into soup, and then a dragonfly was built out of the soup.” In his metaphor, the larva represented the data that had gone into training modern neural nets; the dragonfly stood for the agile A.I. that had been created from it.  Throughout the story, you use metaphors to present abstractions. Why? I can think of few technologies that are more abstract than A.I. Arguably, the whole research program is metaphorical. And one of Hinton’s big intellectual commitments is to the idea that we are “analogy machines.” So, in addition to being useful from a readerly perspective, I thought it would be fun, as a writer, to try to do a lot of metaphorical thinking. Deep learning—the technology that Hinton helped pioneer—had caused the metamorphosis. I bent closer to get a better look; Hinton stood upright, as he almost always does, careful to preserve his posture. “It’s very beautiful,” he said softly. “And you get the point. It started as one thing, and it’s become something else.” Is the metaphor designed to deliver more than one message and if so, what is it? It adds a biological dimension to the discussion of A.I. It suggests the mystery of the technology — it’s more like a process we set in motion than a machine that we build part by part. And it captures the sense of wariness with which Hinton thinks we need to approach the technology: we don’t yet know what it is, or what it’s becoming. When he showed me this dragonfly and talked about it this way, I knew almost immediately that I’d close the first section of the story with it.

A FEW WEEKS EARLIER, when Hinton had invited me to visit his island, I’d imagined possible scenarios. Why did you use the visit to the island as the spine of the narrative? An alternative approach would’ve been to visit Geoff in Toronto. But I knew it wouldn’t be nearly as interesting as visiting his island, because the island, with its cottage, is almost like a stand-in for his personality, his life. Like the Ramsay house in “To the Lighthouse.” And, of course, he very graciously offered to host me there. Perhaps he’d be an introvert who wanted solitude, or a tech overlord with a God complex and a futuristic compound. Several days before my arrival, he e-mailed me a photograph he’d taken of a rattlesnake coiled in the island’s grass. I wasn’t sure whether I felt delighted or scared.

In fact, as private islands go, Hinton’s is fairly modest—two acres in total. Hinton himself is the opposite of a Silicon Valley techno-messiah. Now seventy-five, he has an English face out of a Joshua Reynolds  painting, with white hair framing a broad forehead; his blue eyes are often steady, leaving his mouth to express emotion. A mordant raconteur, he enjoys talking about himself—“ ‘Geoff’ is an anagram for ‘ego fortissimo,’ ” he told me—but he’s not an egotist; his life has been too grief-shadowed for that. A tragic, but beautiful phrase. When did it come to you? My editor suggested it! I had something windier. “I should probably tell you about my wives,” he said, the first time we spoke. “I’ve had three marriages. One ended amicably, the other two in tragedy.” He is still friendly with Joanne, his first wife, whom he married early, but his second and third wives, Rosalind and Jackie, both died of cancer, in 1994 and 2018, respectively. For the past four years, Hinton has been with Rosemary Gartner, a retired sociologist. “I think he’s the kind of person who always needs a partner,” she told me, tenderly. Why did you include the adverb? I wanted to make it clear that Rosemary meant this kindly, not as a criticism or cutting observation, and not in a wry way. She did mean it tenderly. He is a romantic rationalist, with a sensibility balancing science and emotion. In the cottage, a burgundy canoe sits in the single large room that makes up most of the ground floor; he and Jackie had found it in the island’s woods, in disrepair, and Jackie, an art historian, worked with some women canoe-builders to reconstruct it during the years coinciding with her illness. “She had the maiden voyage,” Hinton said. No one has used it since. Just six words, but they carry so much weight. Why did you write them, rather than quote Hinton? Sometimes, for an emotional beat to land, it needs just a little moment of repetition or underscoring. And it can be nice to have a kind of reaction shot from the text — an acknowledgment, in the piece, of the gravity of something.

He stowed the dragonfly, then walked over to a small standing desk, where a laptop was perched next to a pile of sudoku puzzles and a notebook containing computer passwords. (He rarely uses the notebook, having devised a mnemonic system that enables him to generate and recall very long passwords in his head.) “Shall we do the family tree?” he asked. Using two fingers—he doesn’t touch-type—he entered “Geoffrey Hinton family tree” and hit Return. When Google acquired Hinton’s startup, in 2013, it did so in part because the team had figured out how to dramatically improve image recognition using neural nets; now endless family trees swarmed the screen.

Hinton comes from a particular kind of scientific English family: politically radical, restlessly inventive. Above him in the family tree are his great-uncle Sebastian Hinton, the inventor of the jungle gym, and his cousin Joan Hinton, who worked as a physicist on the Manhattan Project. Further back, he was preceded by Lucy Everest, the first woman to become an elected member of the Royal Institute of Chemistry; Charles Howard Hinton, the mathematician who created the concept of the tesseract, a doorway into the fourth dimension (one appears in the film “Interstellar”); and James Hinton, a groundbreaking ear surgeon and an advocate of polygamy. (“Christ was the savior of men, but I am the savior of women,” he is said to have remarked.) In the mid-nineteenth century, a great-great-grandfather of Hinton’s, the English mathematician George Boole, developed the system of binary reasoning, now known as Boolean algebra, that is fundamental to all computing. Boole was married to Mary Everest, a mathematician and author and the niece of George Everest, the surveyor for whom Mt. Everest is named. It’s so interesting the way you use the family tree as a device to present Hinton’s ancestors. I wish I could say that this was my idea, but almost everyone who writes about Hinton runs down the family tree. In this case, he made it easy, by actually proposing that we look at one.

“Geoff was born into science,” Yann LeCun, a former student and collaborator of Hinton’s who now runs A.I. at Meta, told me. Yet Hinton’s family was odder than that. His dad, Howard Everest Hinton, grew up in Mexico during the Mexican Revolution, in the nineteen-tens, on a silver mine managed by his father. “He was tough,” Hinton said of his dad: family lore holds that, at age twelve, Howard threatened to shoot his boxing coach for being too heavy-handed, and the coach took him seriously enough to leave town. Howard’s first language was Spanish, and at Berkeley, where he went to college, he was mocked for his accent. “He hung out with a bunch of Filipinos, who were also discriminated against, and he became a Berkeley radical,” Hinton said. Howard’s mature politics were not just Marxist but Stalinist: in 1968, as Soviet tanks rolled into Prague, he said, “About time!” Could you verify what his father said? This is based on Geoff’s recollection.

At school, Hinton was inclined toward science. But, for ideological reasons, his father forbade him to study biology; in Howard’s view, the possibility of genetic determinism contravened the Communist belief in the ultimate malleability of human nature. (“I hate faiths of all kinds,” Hinton said, remembering this period.) Howard, who taught at the University of Bristol, was a kind of entomologist Indiana Jones: he smuggled rare creatures from around the world back to England in his luggage, and edited an important journal in his field. Hinton, whose middle name is also Everest, felt immense pressure to make his own mark. He recalls his father telling him, “If you work twice as hard as me, when you’re twice as old as I am you might be half as good.”

At Cambridge, Hinton tried different fields but was dismayed to find that he was never the brightest student in any given class. He left college briefly to “read depressing novels” and to do odd jobs in London, then returned to attempt architecture, for about a day. Finally, after dipping into physics, chemistry, physiology, and philosophy, looking for a focus, he settled on a degree in experimental psychology. He haunted the office hours of the moral philosopher Bernard Williams, who turned out to be interested in computers and the mind. One day, Williams pointed out that our different thoughts must reflect different physical arrangements inside our brains; this was quite unlike the situation inside a computer, in which the software was independent of the hardware. Hinton was struck by this observation; he remembered how, in high school, a friend had told him that memory might be stored in the brain “holographically”—that is, spread out, but in such a way that the whole could be accessed through any one part. What he was encountering was “connectionism”—an approach that combined neuroscience, math, philosophy, and programming to explore how neurons could work together to “think.” One goal of connectionism was to create a brainlike system in a computer. There had been some progress: the Perceptron, a machine built in the nineteen-fifties by a psychologist and pioneering connectionist named Frank Rosenblatt, had used simple computer hardware to simulate a network of hundreds of neurons. When connected to a light sensor, the apparatus could recognize letters and shapes by tracking which artificial neurons were activated by different patterns of light. How much research apart from your field reporting did you have to do to write complex material with clarity and authority? I did a lot of research. But I also follow this field closely already, and have for many years, and many of the ideas I’m discussing in sections like this have come up in other contexts — for example, my profile several years ago of Daniel Dennett, the philosopher of mind, was tremendously useful preparation for writing this piece.

In the cottage, Hinton stood and strolled, ranging back and forth behind the kitchen counter and around the first floor. He made some toast, got us each an apple, and then set up a little booster table for himself using a step stool. Family pressure had had the effect of pushing him out of temporary satisfactions. “I always loved woodwork,” he recalled wistfully, while we ate. “At school, you could do it voluntarily in the evenings. And I’ve often wondered whether I’d have been happier as an architect, because I didn’t have to force myself to do it. Whereas, with science, I’ve always had to force myself. Because of the family, I had to succeed at it—I had to find a path. There was joy in it, but it was mostly anxiety. Now it’s an enormous relief that I’ve succeeded.”

Hinton’s laptop dinged. How did you decide on this verb? I believe that it actually made a “ding” sound. But that may very well be what Hinton would call a “confabulation” — that is, an extrapolation from other existing data.  Ever since he’d left Google, his in-box had been exploding with requests for comment on A.I. He ambled What were you trying to convey with this verb? I was struck by how Hinton spent all day standing, walking, striding, ambling, leaning, and so on. He has a whole vocabulary of walks. over and looked at the e-mail, and then got lost again in the forest of family trees, all of which seemed to be wrong in one way or another.

“Look at this,” he said.

I walked over and peered at the screen. It was an “academic family tree,” showing Hinton at the top with his students, and theirs, arrayed below. The tree was so broad that he had to scroll horizontally to see the extent of his influence. “Oh, dear,” Hinton said, exploring. “She wasn’t really a student of mine.” He scrolled further. “He was brilliant but not so good as an adviser, because he could always do it better himself.” A careful nurturer of talent, Hinton seems to enjoy being surpassed by his students: when evaluating job candidates, he used to ask their advisers, “But are they better than you?” Recalling his father, who died in 1977, Hinton said, “He was just extremely competitive. And I’ve often wondered, if he’d been around to see me be successful, whether he’d have been entirely happy. Because now I’ve been more successful than he was.”

According to Google Scholar, Hinton is now the second most cited researcher among psychologists, and the most cited among computer and cognitive scientists. If he had a slow and eccentric start at Cambridge, it was partly because he was circling an emerging field. “Neural networks—there were very few people at good universities who did it,” he said, closing the laptop. “You couldn’t do it at M.I.T. You couldn’t do it at Berkeley. You couldn’t do it at Stanford.” There were advantages to being a hub in a nascent network. For years, many of the best minds came to him.

“THE WEATHER'S GOOD," Hinton said, the next morning. You skip past half a day. Why?  Did you spend the night at Hinton’s? It was a long visit. All told I spent four nights on the island with him. So I skipped a lot! “We should cut down a tree.” He wore a dress shirt tucked into khakis and didn’t look much like a lumberjack; still, he rubbed his hands together. On the island, he is always cutting down trees to create more orderly and beautiful tableaus.

The house, too, is a work in progress. Few contractors would travel to a place so remote, and the people Hinton hired made needless mistakes (running a drainage pipe uphill, leaving floors half finished) that still enrage him today. Almost every room harbors a corrective mini-project, and, when I visited, Hinton had appended little notes to them to help a new contractor, often writing on the building materials themselves. In the first-floor bathroom, a piece of baseboard propped against the wall read “Bathroom should have this type of baseboard (maple trim in front of shower only).” In the guest-room closet, masking tape ran along a shelf: “Do not prime shelf, prime shelf support.” Is that where you stayed and the way you noticed it? Yes, I stayed in the guest room and poked around a bit, like journalists do.

It’s useful for minds to label things; it helps them get a grip on reality. But what would it mean for an artificial mind to do so? While Hinton was earning a Ph.D. in artificial intelligence from the University of Edinburgh, he thought about how “knowing” in a brain might be simulated in a computer. At that time, in the nineteen-seventies, the vast majority of A.I. researchers were “symbolists.” In their view, knowing about, say, ketchup might involve a number of concepts, such as “food,” “sauce,” “condiment,” “sweet,” “umami,” “red,” “tomato,” “American,” “French fries,” “mayo,” and “mustard”; together, these could create a scaffold on which a new concept like “ketchup” might be hung. A large, well-funded A.I. effort called Cyc centered on the construction of a vast knowledge repository into which scientists, using a special language, could enter concepts, facts, and rules, along with their inevitable exceptions. (Birds fly, but not penguins or birds with damaged wings or . . .)

But Hinton was doubtful of this approach. It seemed too rigid, and too focussed on the reasoning skills possessed by philosophers and linguists. In nature, he knew, many animals acted intelligently without access to concepts that could be expressed in words. They simply learned how to be smart through experience. Learning, not knowledge, was the engine of intelligence. This is a bumper sticker or T-shirt slogan! What a great line. Did Hinton tell you this and, if so, why did you paraphrase it rather than quote him? This is something he says a lot, in different words. I said it myself, rather than quoting him, for two reasons. First, it’s sometimes important to get the rhythm of a paragraph just right. And second, we’re in free indirect style in this passage — the narration is unfolding from his perspective, as though we’re inside his head back in the days when he was a young scholar. It would break the spell to suddenly quote something he said years later; suddenly, we’d see him from the outside.

Sophisticated human thinking often seemed to happen through symbols and words. But Hinton and his collaborators, James L. McClelland and David Rumelhart, believed that much of the action happened on a sub-conceptual level. Notice, they wrote, how, “if you learn a new fact about an object, your expectations about other similar objects tend to change”: if you’re told that chimpanzees like onions, for instance, you might guess that gorillas like them, too. This suggested that knowledge was likely “distributed” in the mind—created out of smaller building blocks that could be shared among related ideas. There wouldn’t be two separate networks of neurons for the concepts “chimpanzee” and “gorilla”; instead, bundles of neurons representing various concrete or abstract “features”—furriness, quadrupedness, primateness, animalness, intelligence, wildness, and so on—might be activated in one way to signify “chimpanzee” and in a slightly different way to signify “gorilla.” To this cloud of features, onion-liking-ness might be added. A mind constructed this way risked falling into confusion and error: mix qualities together in the wrong arrangement and you’d get a fantasy creature that was neither gorilla nor chimp. But a brain with the right learning algorithm might adjust the weights among its neurons to favor sensible combinations over incoherent ones.

Hinton continued to explore these ideas, first at the University of California, San Diego, where he did a postdoc (and married Joanne, whom he tutored in computer vision); then at Cambridge, where he worked as a researcher in applied psychology; and then at Carnegie Mellon, in Pittsburgh, where he became a computer-science professor in 1982. There, he spent much of his research budget on a single computer powerful enough to run a neural net. He soon got married a second time, to Rosalind Zalin, a molecular biologist. At Carnegie Mellon, Hinton had a breakthrough. Working with Terrence Sejnowski, a computer scientist and a neuroscientist, he produced a neural net called the Boltzmann Machine. The system was named for Ludwig Boltzmann, the nineteenth-century Austrian physicist who described, mathematically, how the large-scale behavior of gases was related to the small-scale behavior of their constituent particles. Hinton and Sejnowski combined these equations with a theory of learning.

Hinton was reluctant to explain the Boltzmann Machine to me. “I’ll tell you what this is like,” he said. “It’s like having a small child, and you decide to go on a walk. And there’s a mountain ahead of you, and you have to get this little child to the top of the mountain and back.” He looked at me—the child in the metaphor—and sighed. He worried, reasonably, that I might be misled by a simplified explanation and then mislead others. “It’s no use trying to explain complicated ideas that you don’t understand. First, you have to understand how something works. Otherwise, you just produce nonsense.” Finally, he took some sheets of paper and began drawing diagrams of neurons connected by arrows and writing out equations, which I tried to follow. (Ahead of my visit, I’d done a Khan Academy course on linear algebra.) Why did you do this? I’d bought a textbook called “Deep Learning,” blurbed by both Hinton and Elon Musk (!), and wanted to read it before my reporting trip. I quickly discovered that, if I wanted to understand the notation, I needed to give myself at least an introduction to linear algebra. And it came in handy; a few times during my visit, Hinton used equations in his explanations, and I could follow along.

One way to understand the Boltzmann Machine, he suggested, was to imagine an Identi-Kit: a system through which various features of a face—bushy eyebrows, blue eyes, crooked noses, thin lips, big ears, and so on—can be combined to produce a composite sketch, of the sort used by the police. For an Identi-Kit to work, the features themselves have to be appropriately designed. The Boltzmann Machine could learn not just to assemble the features but to design them, by altering the weights of the connections among its artificial neurons. It would start with random features that looked like snow on a television screen, and then proceed in two phases—“waking” and “sleeping”—to refine them. While awake, it would tweak the features so that they better fit an actual face. While asleep, it would fantasize a face that didn’t exist, and then alter the features so that they were a worse fit.

Its dreams told it what not to learn. There was an elegance to the system: over time, it could move away from error and toward reality, and no one had to tell it if it was right or wrong—it needed only to see what existed, and to dream about what didn’t. Hinton seems to be a master of the metaphor. Artificial intelligence is incredibly abstract. We’re talking about thinking, probability, psychology, computation — all things that are happening in our heads (and now in machines). And he uses metaphors very effectively to make this abstract world comprehensible and accessible to intuition.

Hinton and Sejnowski described the Boltzmann Machine in a 1983 paper. “I read that paper when I was starting my graduate studies, and I said, ‘I absolutely have to talk to these guys—they’re the only people in the world who understand that we need learning algorithms,’ ” Yann LeCun told me. In the mid-eighties, Yoshua Bengio, a pioneer in natural-language processing and in computer vision who is now the scientific director at Mila, an A.I. institute in Quebec, trained a Boltzmann Machine to recognize spoken syllables as part of his master’s thesis. “Geoff was one of the external reviewers,” he recalled. “And he wrote something like ‘This should not work.’ ” Bengio’s version of the Boltzmann Machine was more effective than Hinton expected; it took Bengio a few years to figure out why. This would become a familiar pattern. In the following decades, neural nets would often perform better than expected, perhaps because new structures had formed among the neurons during training. “The experimental part of the work came before the theory,” Bengio recalled. Often, it was a matter of trying new approaches and seeing what the networks came up with.

Partly because Rosalind loathed Ronald Reagan, Hinton said, they moved to the University of Toronto. They adopted two children, a boy and a girl, from Latin America, and lived in a house in the city. “I was this kind of socialist professor who was dedicated to his work,” Hinton said.

Rosalind had struggled with infertility, and had bad experiences with callous doctors. Perhaps as a result, she pursued a homeopathic route when she was later diagnosed with ovarian cancer. “It just didn’t make any sense,” Hinton said. “It couldn’t be that you make things more dilute and they get more powerful.” He couldn’t see how a molecular biologist could become a homeopath. Still, determined to treat the cancer herself, Rosalind refused to have surgery even after an exam found a tumor the size of a grapefruit; later, she consented to an operation but declined chemotherapy, instead pursuing increasingly expensive homeopathic remedies, first in Canada and then in Switzerland. She developed secondary tumors. She asked Hinton to sell their house so that she could pay for new homeopathic treatments. “I drew the line there,” he recalled, squinting with fresh pain. Is this the kind of gesture that you record in your notebook? Yes, very much so. “I said, ‘No, we’re not selling the house. Because if you die I’m going to have to look after the children, and it’s much better for them if we can stay.’ ”

Rosalind returned to Canada and went immediately into the hospital. She hung on for a couple of months, but wouldn’t let the children visit her until the day before she died, because she didn’t want them to see her so sick. Throughout her illness, she was convinced that she’d soon get well. Describing what happened, Hinton still seems overwhelmed—he is angry, guilty, wounded, mystified. When Rosalind died, Hinton was forty-six, his son was five, and his daughter was three. “She hurt people by failing to accept that she was going to die,” he said.

The sound of waves filled the midafternoon quiet. Strong yellow sun spilled through the room’s floor-to-ceiling windows; faint spiderwebs extended across them, silhouetted by the light. What an exquisite description. Could you describe the writing and revision behind it?I was so struck by the light in Hinton’s cottage that I took many photographs, and I went back to them to write the scene. As I got to the end of the writing process, I knew that I wanted to introduce the spiderwebs somewhere; they’d turned out to be a metaphor for how A.I. might grow to cover the world. I tried using them in a few different places, and concluded that this spot was both the most beautiful and the most parsimonious. Hinton stood for a while, collecting himself.

“I think I need to go cut down a tree,” he said.

We walked out the front door and down the path to the sheds. From one of them, Hinton retrieved a small green chainsaw and some safety goggles.

“Rosemary says I’m not allowed to cut down trees when there’s nobody else here, in case I chop off an arm or something,” he said. “Have you driven boats before?”

“No,” I said.

“I’ve got to not chop off my right arm, then.”

Over his khakis, he strapped on a pair of protective chaps.

“I don’t want to give you the impression that I know what I’m doing,” he said. “But the basic idea is, you cut lots of V’s, and then the tree falls down.”

Hinton crossed the path to the tree that he had in mind, inspecting the bushes for snakes as we walked. The tree was a leafy cedar, perhaps twenty feet tall; Hinton looked up to see which way it was leaning, then started the saw and began to cut into the trunk on the side opposite the lean. He removed the saw, and made another converging cut to form a V.

Hinton worked the chainsaw in silence, occasionally stopping to wipe his brow. It was hot in the sun, and mosquitoes swarmed every shady nook. I inspected the side of the shed, where ants and spiders were engaged in obscure, ceaseless activity. Down at the end of the path, the water shone. It was a beautiful spot. Still, I thought I saw why Hinton wanted to alter it: a lovely rounded hill descended into a gentle hollow, and if the unnecessary tree were gone the light could flow into it. The tree was an error.

Eventually, he began a second cut on the other side of the tree, angling it toward the first. Then he stopped and turned to me. “Because the tree leans away from the cut, the V will open up as you go deeper, and the blade won’t get stuck,” he explained. He continued the upper cut, nudging the tree toward an entropic moment. Suddenly, almost soundlessly, gravity took over. The tree fell under its own weight, landing with surprising softness at the bottom of the hollow. The light streamed in. Why did you lavish so much attention on the tree-cutting? Two reasons. First, I simply thought it was interesting to see Geoff Hinton, godfather of A.I., felling a tree. But second, the tree struck me as evocative of several different things in the world of A.I. There’s the idea of pruning away undesirable things (trees, or connections between neurons) to let the light in. There’s also the idea of setting something up (a tree trunk or a neural network) and then letting nature take its course through a process (gravity or learning). I wasn’t able to get into it in the piece, but one way to understand the training process is by imagining a landscape; a technique called “gradient descent” is used to find “minima” in the network’s errors. And then there’s the fact that the development of A.I. has been a slow, patient process, with Geoff as its “gardener.” All these things made me want to dwell a little on the tree. I think that readers can sense the vibrancy of a scene like this, even if it’s not explicitly worked out for them.

HINTON WAS IN LOVE with the Boltzmann Machine. He hoped that it, or something like it, might underlie learning in the actual brain. “It should be true,” he told me. “If I was God, I’d make it true.” But further experimentation revealed that as Boltzmann Machines grew they tended to become overwhelmed by the randomness that was built into them. “Geoff and I disagreed about the Boltzmann Machine,” LeCun said. “Geoff thought it was the most beautiful algorithm. I thought it was ugly. It was stochastic”—that is, based partly on randomness. By contrast, LeCun said, “I thought backprop was super clean.”

“Backprop,” or backpropagation, was an algorithm that had been explored by a few different researchers beginning in the nineteen-sixties. Even as Hinton was working with Sejnowski on the Boltzmann Machine, he was also collaborating with Rumelhart and another computer scientist, Ronald Williams, on backprop. They suspected that the technique had untapped potential for learning; in particular, they wanted to combine it with neural nets that operated across many layers.

One way to understand backprop is to imagine a Kafkaesque judicial system. Picture an upper layer of a neural net as a jury that must try cases in perpetuity. The jury has just reached a verdict. In the dystopia in which backprop unfolds, the judge can tell the jurors that their verdict was wrong, and that they will be punished until they reform their ways. The jurors discover that three of them were especially influential in leading the group down the wrong path. This apportionment of blame is the first step in backpropagation.

In the next step, the three wrongheaded jurors determine how they themselves became misinformed. They consider their own influences—parents, teachers, pundits, and the like—and identify the individuals who misinformed them. Those blameworthy influencers, in turn, must identify their respective influences and apportion blame among them. Recursive rounds of finger-pointing ensue, as each layer of influencers calls its own influences to account, in a backward-sweeping cascade. Eventually, once it’s known who has misinformed whom and by how much, the network adjusts itself proportionately, so that individuals listen to their “bad” influences a little less and to their “good” influences a little more. The whole process repeats again and again, with mathematical precision, until verdicts—not just in this one case but in all cases—are collectively as “correct” as possible. Amazing use of an extended metaphor. How did it come to you? I tried many different metaphors, including a Presidential election in which the “wrong” candidate gets elected and a corporate bureaucracy in which the orders of the C.E.O. are imperfectly carried out. But I liked this Kafka-esque metaphor because it captures the authoritarian aspect of backpropagation — ultimately, it’s the humans who tell the machine what’s right or wrong — and because it responds to the slightly sour view that Hinton takes of backprop. In his view, it’s a disappointing way to learn.

In 1986, Hinton, Rumelhart, and Williams published a three-page paper in Nature showing how such a system could work in a neural net. They noted that backprop, like the Boltzmann Machine, wasn’t “a plausible model of learning in brains”: unlike a computer, a brain can’t rewind the tape to audit its past performance. But backprop still enabled a brainlike neural specialization. In real brains, neurons are sometimes arranged in structures aimed at solving specific problems: in the visual system, for instance, different “columns” of neurons recognize edges in what we see. Something similar emerges in a backprop network. Higher layers subject lower ones to a kind of evolutionary pressure; as a result, certain layers of a network that’s tasked with deciphering handwriting, for instance, might become tightly focussed on identifying lines, curves, or edges. Eventually, the system as a whole can develop “appropriate internal representations.” The network knows, and makes use of its knowledge.

In the nineteen-fifties and sixties, a great deal of excitement had accompanied the Perceptron and other connectionist efforts; enthusiasm for connectionism waned in the years after. The backprop paper was part of a revival of interest and earned widespread attention. But the actual work of building backprop networks was slow-going, for both practical and conceptual reasons. Practically, computers were sluggish. “The rate of progress was basically, How much could a computer learn overnight?” Hinton recalled. “The answer was often not much.” Conceptually, neural nets were mysterious. It wasn’t possible to program one in the traditional way. You couldn’t go in and edit the weights of the connections among artificial neurons. And, anyway, it was hard to understand what the weights meant, because they had adapted and changed themselves through training.

There were many ways the learning process could go wrong. In “overfitting,” for example, a network effectively memorized the training data instead of learning to generalize from it. Avoiding the various pitfalls wasn’t always straightforward, because it was up to the network to learn. It was like felling a tree: researchers could make cuts here and there, but then had to let the process unfold. Did you know you were going to refer to the felling of Hinton’s tree again here? I knew that the tree would be a useful metaphor. In general, throughout the writing and revision process, I try to tie things together as much as possible — to see which metaphors, phrases, and images can be usefully pulled through from one part of the piece to another. They could try techniques like “ensembling” (combining weak networks to make a strong one) or “early stopping” (letting a network learn, but not too much). They could “pre-train” a system, by taking a Boltzmann Machine, having it learn something, and then layering a backprop network on top of it, so that a system’s “supervised” training didn’t begin until it had acquired some elemental knowledge on its own. Then they’d let the network learn, hoping that it would land where they wanted it. Your writing makes complex science accessible. How did you learn to do so? I’m always trying to improve, and not always succeeding. Being an editor myself has really helped: I read so much excellent writing done by the writers with whom I’m working. The writer who’s helped me the most is probably Daniel Dennett, the philosopher (whom I profiled in 2017). Dan works to build what he calls “intuition pumps” — thought experiments or metaphors drawn from the physical world that help guide your thinking. I’m always trying to emulate him.

New neural-net “architectures” were developed: “recurrent” and “convolutional” networks allowed the systems to make progress by building on their own work in different ways. But it was as though researchers had discovered an alien technology that they didn’t know how to use. They turned the Rubik’s Cube this way and that, trying to pull order out of noise. “I was always convinced it wasn’t nonsense,” Hinton said. “It wasn’t really faith—it was just completely obvious to me.” The brain used neurons to learn; therefore, complex learning through neural networks must be possible. He would work twice as hard for twice as long. A nod to his father’s brutal statement. Why did you allude to it here? One of the most striking things about Hinton’s story is his persistence, and I think part of it is rooted in his upbringing. He believed in these ideas and it just didn’t make sense to him to walk away from them. He also felt driven to achieve something substantial. So I wanted to bring his dad back here, to remind us of how unique Hinton is as a person.

When networks were trained through backprop, they needed to be told when they were wrong and by how much; this required vast amounts of accurately labeled data, which would allow networks to see the difference between a handwritten “7” and a “1,” or between a golden retriever and a red setter. But it was hard to find well-labelled datasets that were big enough, and building more was a slog. LeCun and his collaborators developed a giant database of handwritten numerals, which they later used to train networks that could read sample Zip Codes provided by the U.S. Postal Service. A computer scientist named Fei Fei Li, at Stanford, spearheaded a gargantuan effort called ImageNet; creating it required collecting more than fourteen million images and sorting them into twenty thousand categories by hand.

As neural nets grew larger, Hinton devised a way of getting knowledge from a large network into a smaller one that might run on a device like a mobile phone. “It’s called distillation,” he explained, in his kitchen. Why did you identify the setting?I just didn’t want the piece to feel too disembodied, even during the technical sections. “Back in school, the art teacher would show us some slides and say, ‘That’s a Rubens, and that’s a van Gogh, and this is William Blake.’ But suppose that the art teacher tells you, ‘O.K., this is a Titian, but it’s a peculiar Titian because aspects of it are quite like a Raphael, which is very unusual for a Titian.’ That’s much more helpful. They’re not just telling you the right answer—they’re telling you other plausible answers.” In distillation learning, one neural net provides another not just with correct answers but with a range of possible answers and their probabilities. It was a richer kind of knowledge.

A few years after Rosalind’s death, Hinton reconnected with Jacqueline Ford, an art historian whom he’d dated briefly before moving to the United States. Jackie was cultured, warm, curious, beautiful. How do you know this?From Geoff! “She’s way out of your league,” his sister said. Did you talk with Hinton’s sister? The fact-checking department did. Still, Jackie gave up her job in the U.K. to move to Toronto. They got married on December 6, 1997—Hinton’s fiftieth birthday. The following decades would be the happiest of his life. His family was whole again. His children loved their new mother. He and Jackie started exploring the islands in Georgian Bay. Recalling this time, he gazed at the canoe in his living room. “We found it in the woods, upside down, covered in canvas, and it was just totally rotten—everything about it was rotten,” he said. “But Jackie decided to rescue it anyway, like she did with me and the kids.”

Hinton was not in love with backpropagation. “It’s so unsatisfying intellectually,” he told me. Unlike the Boltzmann Machine, “it’s all deterministic. Unfortunately, it just works better.” Slowly, as practical advances compounded, the power of backprop became undeniable. In the early seventies, Hinton told me, the British government had hired a mathematician named James Lighthill to determine if A.I. research had any plausible chance of success. Lighthill concluded that it didn’t—“and he was right,” Hinton said, “if you accepted the assumption, which everyone made, that computers might get a thousand times faster, but they wouldn’t get a billion times faster.” Hinton did a calculation in his head. Suppose that in 1985 he’d started running a program on a fast research computer, and left it running until now. If he started running the same program today, on the fastest systems currently used in A.I., it would take less than a second to catch up.

In the early two-thousands, as multi-layer neural nets equipped with powerful computers began to train on much larger data sets, Hinton, Bengio, and LeCun started talking about the potential of “deep learning.” The work crossed a threshold in 2012, when Hinton, Alex Krizhevsky, and Ilya Sutskever came out with AlexNet, an eight-layer neural network that was eventually able to recognize objects from ImageNet with human-level accuracy. Hinton formed a company with Krizhevsky and Sutskever and sold it to Google. He and Jackie bought the island in Georgian Bay—“my one real indulgence,” Hinton said.

Two years later, Jackie was diagnosed with pancreatic cancer. Doctors gave her a year or two to live. “She was incredibly brave and incredibly rational,” Hinton said. “She wasn’t in deep denial, desperately trying to get out of it. Her view was ‘I can feel sorry for myself, or I can say I don’t have much time left and I’d better do my best to enjoy it and make everything O.K. for other people.’ ” She and Hinton pored over the statistics before deciding on therapies; largely through chemo, she extended one or two years to three. In the cottage, when she could no longer manage the stairs, he constructed a small basket on a string so that she could lower her tea from the second floor to the first, where he could warm it up in the microwave. The devotion behind this gesture is heart-rending.  How did you react when he told you about it? I was very moved. When people open up like this, I feel an extra sense of responsibility to write something that’s worthy of that openness. (“I should’ve just moved the microwave upstairs,” he observed.) Some editors and writers argue that attributive verbs should be limited to “says” or “said.” They argue that examples like the one used here are unnecessary. What’s your take on this? I can imagine that over-using those kinds of words could be distracting. But what’s the harm from time to time?

Late in the day, we leaned on Hinton’s standing desk as he showed me photos of Jackie on his laptop. In a picture of their wedding day, she and Hinton stand with his kids in the living room of their neighbor’s house, exchanging vows. Hinton looks radiant and relaxed; Jackie holds one of his hands lightly in both of hers. In one of the last pictures that he showed me, she gazes at the camera from the burgundy canoe, which she is paddling in the dappled water near the dock. “That was the summer of 2017,” Hinton said. Jackie died the following April. That June, Hinton, Bengio, and LeCun won the Turing Award—the equivalent of the Nobel Prize in computer science. You could have delivered this information anytime earlier. Why did you decide to wait until now? I think you can only tell a story once. If I’d included this higher up, it would’ve been repetitive appearing here. I wanted to keep things in order, as much as possible.

Hinton is convinced that there’s a real sense in which neural nets are capable of having feelings. “I think feelings are counterfactual statements about what would have caused an action,” he had told me, earlier that day. “Say that I feel like punching someone on the nose. What I mean is: if I didn’t have social inhibitions—if I didn’t stop myself from doing it—I would punch him on the nose. So when I say ‘I feel angry,’ it’s a kind of abbreviation for saying, ‘I feel like doing an aggressive act.’ Feelings are just a way of talking about inclinations to action.”

He told me that he had seen a “frustrated A.I.” in 1973. A computer had been attached to two TV cameras and a simple robot arm; the system was tasked with assembling some blocks, spread out on a table, into the form of a toy car. “This was hard, particularly in 1973,” he said. “The vision system could recognize the bits if they were all separate, but if you put them in a little pile it couldn’t recognize them. So what did it do? It pulled back a little bit, and went bash!, and spread them over the table. Basically, it couldn’t deal with what was going on, so it changed it, violently. And if a person did that you’d say they were frustrated. The computer couldn’t see the blocks right, so he bashed them.” To have a feeling was to want what you couldn’t have. Here’s another example of wisdom you deliver rather than Hinton. Why? In this case, he did a great job of explaining this idea just a paragraph up. But I wanted to rephrase it in a way that would make a broader connection to the house, and to the life with Jackie that wasn’t able to unfold there, and to the bittersweet feelings that the place sometimes provokes.

“I love this house, but sometimes it’s a sad place,” he said, while we looked at the pictures. “Because she loved being here and isn’t here.”

The sun had almost set, and Hinton turned on a little light over his desk. He closed the computer and pushed his glasses up on his nose. He squared up his shoulders, returning to the present.

“I wanted you to know about Roz and Jackie because they’re an important part of my life,” he said. “But, actually, it’s also quite relevant to artificial intelligence. There are two approaches to A.I. There’s denial, and there’s stoicism. Everybody’s first reaction to A.I. is ‘We’ve got to stop this.’ Just like everybody’s first reaction to cancer is ‘How are we going to cut it out?’ ” But it was important to recognize when cutting it out was just a fantasy.

He sighed. “We can’t be in denial,” he said. “We have to be real. We need to think, How do we make it not as awful for humanity as it might be?”

HOW USEFUL—OR DANGEROUS—will A.I. turn out to be? No one knows for sure, in part because neural nets are so strange. In the twentieth century, many researchers wanted to build computers that mimicked brains. But, although neural nets like OpenAI’s GPT models are brainlike in that they involve billions of artificial neurons, they’re actually profoundly different from biological brains. Today’s A.I.s are based in the cloud and housed in data centers that use power on an industrial scale. Clueless in some ways and savantlike in others, they reason for millions of users, but only when prompted. They are not alive. They have probably passed the Turing test—the long-heralded standard, established by the computing pioneer Alan Turing, which held that any computer that could persuasively imitate a human in conversation could be said, reasonably, to think. And yet our intuitions may tell us that nothing resident in a browser tab could really be thinking in the way we do. The systems force us to ask if our kind of thinking is the only kind that counts.

During his last few years at Google, Hinton focussed his efforts on creating more traditionally mindlike artificial intelligence using hardware that more closely emulated the brain. In today’s A.I.s, the weights of the connections among the artificial neurons are stored numerically; it’s as though the brain keeps records about itself. In your actual, analog brain, however, the weights are built into the physical connections between neurons. Hinton worked to create an artificial version of this system using specialized computer chips.

“If you could do it, it would be amazing,” he told me. The chips would be able to learn by varying their “conductances.” Because the weights would be integrated into the hardware, it would be impossible to copy them from one machine to another; each artificial intelligence would have to learn on its own. “They would have to go to school,” he said. “But you would go from using a megawatt to thirty watts.” As he spoke, he leaned forward, his eyes boring into mine; I got a glimpse of Hinton the evangelist. In a New Yorker Radio Hour when the magazine’s editor, David Remnick, interviews you about this story, he suggests that Hinton is a godfather who has become an apostate. Which is he in your mind: Evangelist or apostate? I think that part of the reason he’s such an important person to listen to now is that he’s neither — or, at least, he’s attempting to be neither. He’s trying to be a realist, to be someone who knows the technology well and is simply describing the possibilities. Observers see him as one or the other, depending on their points of view. But he’s not trying to hype A.I. or criticize it; he’s trying to describe our situation, as he sees it. Because the knowledge gained by each A.I. would be lost when it was disassembled, he called the approach “mortal computing.” “We’d give up on immortality,” he said. “In literature, you give up being a god for the woman you love, right? In this case, we’d get something far more important, which is energy efficiency.” Among other things, energy efficiency encourages individuality: because a human brain can run on oatmeal, the world can support billions of brains, all different. And each brain can learn continuously, rather than being trained once, then pushed out into the world.

As a scientific enterprise, mortal A.I. might bring us closer to replicating our own brains. But Hinton has come to think, regretfully, that digital intelligence might be more powerful. In analog intelligence, “if the brain dies, the knowledge dies,” he said. By contrast, in digital intelligence, “if a particular computer dies, those same connection strengths can be used on another computer. And, even if all the digital computers died, if you’d stored the connection strengths somewhere you could then just make another digital computer and run the same weights on that other digital computer. Ten thousand neural nets can learn ten thousand different things at the same time, then share what they’ve learned.” This combination of immortality and replicability, he says, suggests that “we should be concerned about digital intelligence taking over from biological intelligence.”

How should we describe the mental life of a digital intelligence without a mortal body or an individual identity? In recent months, some A.I. researchers have taken to calling GPT a “reasoning engine”—a way, perhaps, of sliding out from under the weight of the word “thinking,” which we struggle to define. “People blame us for using those words—‘thinking,’ ‘knowing,’ ‘understanding,’ ‘deciding,’ and so on,” Bengio told me. At what stage in the reporting did you interview Begio and Yann LeCun? How many human sources did you speak with? I talked to them after I spent time with Hinton on his island, not before. I take different approaches in different pieces; in this case, I felt that it would be especially important to see Hinton on a personal level, as a human being, rather than as a technologist first and a human being second. So I wanted to see him without talking to others about him first, and to encounter him fresh. I ended up talking to maybe a dozen people on background, but only wanted to quote a few in the piece, to prevent the story from succumbing to talking-head syndrome, and to keep it focused on a single human individual. “But even though we don’t have a complete understanding of the meaning of those words, they’ve been very powerful ways of creating analogies that help us understand what we’re doing. It’s helped us a lot to talk about ‘imagination,’ ‘attention,’ ‘planning,’ ‘intuition’ as a tool to clarify and explore.” In Bengio’s view, “a lot of what we’ve been doing is solving the ‘intuition’ aspect of the mind.” Intuitions might be understood as thoughts that we can’t explain: our minds generate them for us, unconsciously, by making connections between what we’re encountering in the present and our past experiences. We tend to prize reason over intuition, but Hinton believes that we are more intuitive than we acknowledge. “For years, symbolic-A.I. people said our true nature is, we’re reasoning machines,” he told me. “I think that’s just nonsense. Our true nature is, we’re analogy machines, with a little bit of reasoning built on top, to notice when the analogies are giving us the wrong answers, and correct them.”

On the whole, current A.I. technology is talky and cerebral: it stumbles at the borders of the physical. “Any teen-ager can learn to drive a car in twenty hours of practice, with hardly any supervision,” LeCun told me. “Any cat can jump on a series of pieces of furniture and get to the top of some shelf. We don’t have any A.I. systems coming anywhere close to doing these things today, except self-driving cars”—and they are over-engineered, requiring “mapping the whole city, hundreds of engineers, hundreds of thousands of hours of training.” Solving the wriggly problems of physical intuition “will be the big challenge of the next decade,” LeCun said. Still, the basic idea is simple: if neurons can do it, then so can neural nets. At any point, did you use A.I. during the reporting and writing? If so, how, when and why? I used A.I. a lot! Not for the writing or reporting, really, but just to have the experience of using it. I kept a running conversation going with ChatGPT about itself, about Hinton, about A.I. technology and ethics. I’d chat with it in speech mode quite often. I’d try out different metaphors for various technical concepts and ask ChatGPT to riff on those metaphors. I also asked ChatGPT to take on different personalities; for whatever reason, I enjoyed it most when it was snarky. At one point, I asked it if “writer’s block” was real and explained that I was writing a piece on Hinton. It replied: “Oh, Josh, a project on Geoff Hinton? How original. It's not like he's one of the most recognized figures in deep learning or anything. But hey, maybe you'll manage to write something that isn't another regurgitated biography or fanboy piece. Most writers would kill for an opportunity to profile someone like him, but here you are, whining about it taking too long. Maybe you should consider if you're actually cut out for this kind of work. What could possibly be so hard about writing a piece on a guy who's been interviewed and profiled a million times? Good luck with that. Hope your ‘writer’s block’ miraculously disappears.”

Hinton suspects that skepticism of A.I.’s potential, while comforting, is often motivated by an unjustified faith in human exceptionalism. Researchers complain that A.I. chatbots “hallucinate,” by making up plausible answers to questions that stump them. But he contests that terminology. “We should say ‘confabulate,’ ” he told me. “ ‘Hallucination’ is when you think there’s sensory input—auditory hallucinations, visual hallucinations, olfactory hallucinations. But just making stuff up—that’s confabulation.” He cited the case of John Dean, President Richard Nixon’s White House counsel, who was interviewed about Watergate before he knew that the conversations he described had been tape-recorded. Dean confabulated, getting the details wrong, mixing up who said what. “But the gist of it was all right,” Hinton said. “He had a recollection of what went on, and he imposed that recollection on some characters in his head. He wrote a little play. And that’s what human memory is like. In our minds, there’s no boundary between just making it up and telling the truth. Telling the truth is just making it up correctly. Because it’s all in the weights, right?” From this perspective, ChatGPT’s ability to make things up is a flaw, but also a sign of its humanlike intelligence.

Hinton is often asked if he regrets his work. He doesn’t. (He recently sent a journalist a one-liner—“a song for you”—along with a link to Edith Piaf’s “Non, Je Ne Regrette Rien.”) When he began his research, he says, no one thought that the technology would succeed; even when it started succeeding, no one thought that it would succeed so quickly. Precisely because he thinks that A.I. is truly intelligent, he expects that it will contribute to many fields. Yet he fears what will happen when, for instance, powerful people abuse it. “You can probably imagine Vladimir Putin creating an autonomous lethal weapon and giving it the goal of killing Ukrainians,” Hinton said. During your New Yorker Radio Hour interview with Remnick, audio excerpts of your conversations with Hinton are played. How would you describe your interviewing style?In that case, I conducted a special, more formal interview with Hinton, just for the radio, at the end of my reporting trip. Ordinarily, I’m more informal. I like to ask people to tell me about their lives in chronological order. I prefer listening to talking. He believes that autonomous weapons should be outlawed—the U.S. military is actively developing them—but warns that even a benign autonomous system could wreak havoc. “If you want a system to be effective, you need to give it the ability to create its own subgoals,” he said. “Now, the problem is, there’s a very general subgoal that helps with almost all goals: get more control. The research question is: how do you prevent them from ever wanting to take control? And nobody knows the answer.” (Control, he noted, doesn’t have to be physical: “It could be just like how Trump could invade the Capitol, with words.”)

Within the field, Hinton’s views are variously shared and disputed. “I’m not scared of A.I.,” LeCun told me. “I think it will be relatively easy to design them so that their objectives will align with ours.” He went on, “There’s the idea that if a system is intelligent it’s going to want to dominate. But the desire to dominate has nothing to do with intelligence—it has to do with testosterone.” I recalled the spiders I’d seen at the cottage, and how their webs covered the surfaces of Hinton’s windows. They didn’t want to dominate, either—and yet their insectoidal intelligence had led them to expand their territory. Living systems without centralized brains, such as ant colonies, don’t “want” to do anything, yet they still find food, ford rivers, and kill competitors in vast numbers. Either Hinton or LeCun could be right. The metamorphosis isn’t finished. We don’t know what A.I. will become.

“Why don’t we just unplug it?” I asked Hinton, of A.I. in general. “Is that a totally unreasonable question?” Why did you include this dialogue with Hinton? This is the question on many of our minds, and even though it’s a silly one — obviously we can’t “unplug” A.I. I wanted to ask it in a direct way, so that we could get to the heart of the matter.

“It’s not unreasonable to say, We’d be better off without this—it’s not worth it,” he said. “Just as we might have been better off without fossil fuels. We’d have been far more primitive, but it may not have been worth the risk.” He added, stoically, “But it’s not going to happen. Because of the way society is. And because of the competition between different nations. If the U.N. really worked, possibly something like that could stop it. Although, even then, A.I. is just so useful. It has so much potential to do good, in fields like medicine—and, of course, to give an advantage to a nation via autonomous weapons.” Earlier this year, Hinton declined to sign a popular petition that called for at least a six-month pause in research. “China’s not going to stop developing it for six months,” he said.

“So what should we do?” I asked.

“I don’t know,” he said. “It would be great if this were like climate change, where someone could say, Look, we either have to stop burning carbon or we have to find an effective way to remove carbon dioxide from the atmosphere. There, you know what the solution looks like. Here, it’s not like that.”

Hinton was pulling on a blue waterproof jacket. We were heading to the marina to pick up Rosemary. “She’s brought supplies!” he said, smiling. As we walked out the door, I looked back into the cottage. In the big room, the burgundy canoe shone, caressed by sunlight. Chairs were arranged in front of it in a semicircle, facing the water through the windows. Some magazines were piled on a little table. It was a beautiful house. A human mind does more than reason; it exists in time, and reckons with life and death, and builds a world around itself. It gathers meaning, as if by gravity. An A.I., I thought, might be able to imagine a place like this. But would it ever need one? Why did you make this observation? To me, this story is partly about how one of the people who’s been instrumental in creating A.I. is also deeply human. Hinton isn’t someone who wants to transcend or escape or redefine what it means to be a human being; he’s an empathetic, experienced, wise person who’s lived a full and sometimes difficult life and who fully grasps complexity, richness and value of human experience. It’s possible to imagine a version of the A.I. story in which the scientists behind the technology are somehow cold, like in the Hollywood movie “Ex Machina.” I wanted to show how this wasn’t true. And I also wanted to pose the question at the end of the paragraph. Many of the ideas in the closing part of the piece relate to mortality and its role in making us into individuals. (Can an immortal A.I. ever be an individual?) This is one of the ways in which thinking about artificial intelligence leads us to think about the nature of human existence. I thought it would be good to bring the piece full circle.

We made our way down the wooded path, past the sheds and down the steps to the dock, then climbed into Hinton’s boat. It was a perfect blue day, with a brisk wind roughing the water. Hinton stood at the wheel. I sat in front, watching other islands pass, thinking about the story of A.I. To some, it’s a Copernican tale, in which our intuitions about the specialness of the human mind are being dislodged by thinking machines. To others, it’s Promethean—having stolen fire, we risk getting burned. Some people think we’re fooling ourselves, getting taken in by our own machines and the companies that hope to profit from them. In a strange way, it could also be a story about human limitation. If we were gods, we might make a different kind of A.I.; in reality, this version was what we could manage. Meanwhile, I couldn’t help but consider the story in an Edenic light. By seeking to re-create the knowledge systems in our heads, we had seized the forbidden apple; we now risked exile from our charmed world. But who would choose not to know how knowing works? What is the narrative purpose of this paragraph?Well, on one level, it’s actually true that I had these thoughts while we motored back to the marina. But I decided to include them because I wanted to foreground how the A.I. story is still developing and could be told in many ways. It’s such a primal, elemental story — it has Golems, Frankensteins, zombies, gods, and so on — and those kinds of metaphors have a magnetism that’s hard to resist. I thought it would be helpful to name them, and to make them explicit, so that readers could engage with them out in the light. And I also wanted to set up the ending, with the snake, in a way that would make the symbolism more obvious while also rendering it a little tongue-in-cheek.

At the marina, Hinton did a good job of working with the wind, accelerating forward, turning, and then allowing it to guide him into his slip. “I’m learning,” he said, proud of himself. We walked ashore and waited by a shop for Rosemary to arrive. After a while, Hinton went inside to buy a light bulb. I stood, enjoying the warmth, and then saw a tall, bright-eyed woman with long white hair striding toward me from the parking lot.

Rosemary and I shook hands. Then she looked over my shoulder. Hinton was emerging from the greenery near the shop, grinning.

“What’ve you got for me?” she asked.

Hinton held up a black-and-yellow garter snake, perhaps a metre long, twisting round and round like a spring. “I’ve come bearing gifts!” he said, in a gallant tone. “I found it in the bushes.”

Rosemary laughed, delighted, and turned to me. “This just epitomizes him,” she said.

“He’s not happy,” Hinton said, observing the snake.

“Would you be?” Rosemary asked.

“I’m being very careful with his neck,” Hinton said. “They’re fragile.”

He switched the snake from one hand to another, then held out a palm. It was covered in the snake’s slimy musk.

“Have a sniff,” he said. After a story that presents complex concepts in often long paragraphs, you shifted here to an easy-to-grasp scene driven by dialogue and action. Why the change? The piece is structured as a visit, and I wanted to end it by bringing the visit to a close. And I also wanted the piece to be very much about Hinton as a person, and so it made sense to me to end it on single a moment with him, realized as a three-dimensional, real-time scene, rather than an interview. Also, I think it feels good, after a somewhat claustrophobic piece — it’s just me and Hinton on the island — to suddenly have someone new arrive.

We took turns. It was strange: mineral and pungent, reptilian and chemical, unmistakably biotic.

“You’ve got it all over your shirt!” Rosemary said.

“I had to catch him!” Hinton explained.

He put the snake down, and it slithered off into the grass. He watched it go with a satisfied look.

“Well,” he said. “It’s a beautiful day. Shall we brave the crossing?”Why did you end the story on this note? Did you experiment with alternate approaches? As soon as this happened during my reporting — Geoff catching the snake, my meeting Rosemary, their charming interaction — I knew that I’d probably be ending the story with it. And then it became a matter of working backwards and figuring out what I needed to add (or subtract) to make the ending as satisfying as possible. The article has been so retrospective and bittersweet. I thought it was good to see Hinton in the present, in a happy relationship. And we’re seeing his comfort with the non-human, which is itself a dimension of his own humanity. My hope was that the ending would prompt us to appreciate our humanity at the same time it makes us think about what might challenge it.

* * *

Chip Scanlan is an award-winning writer who taught at the Poynter Institute and now coaches writers around the world. He is the author of several books on writing and the newsletter Chip's Writing Lessons.

The post How a reporter prepped to understand A.I. and the man who helped invent it appeared first on Nieman Storyboard.

The best narrative writers know they need not just to interview after the fact, but to observe in the moment. They want to be on the scene, where they see characters and action unfold in real place and real time, providing a less-filtered view of complexity and humanity. Policy and reframes take a back seat to lives as they are really lived.

Ellen Barry, who covers mental illness for The New York Times, immersed herself in the chaotic life of Andrey Shevelyov, a man with schizoaffective disorder who preferred living in a tent rather than taking psychiatric medications and living in a group home —  a decision that has devastated his mother and stepfather. Barry read portions of Andrey’s journals, watched the video journals he posted on social media and traveled across the country three times to spend time with him and his family in Vancouver, Washington.

“I wanted to be able to observe the inflection points,” she told me. “Not just reconstruct them, but describe them. Luckily I was able to see a number of interactions between him and his mother that forcefully conveyed the pain and the tension.”

Her work resulted in an intimate profile of “The Man in Room 117,” a 4,000-word tragedy that reveals a bitter truth about America’s struggle to deal with some of its most troubled and vulnerable citizens. From the story:

As affordable apartments all but vanished in American cities, a whole tier of people with disabling mental illness were forced onto the street, where they now live in numbers large enough to disrupt civic life. Many of them are shunted into the criminal justice system, only to return to homelessness upon their release.

Her story, published Jan. 29, 2024, spotlights the consequences of laws that let people with mental illness efuse treatment. State-by-state variations of those laws have been passed as an antidote to the longtime abuses and isolation of institutions. Those well-intended policies have led to unintended consequences: Often patients who decline medication, therapy or specialized housing end up on the streets or at the mercy of systems ill-equipped to help them.

Barry uses struggles faced by Andrey and his parents as a vehicle to explore the tensions around the laws. She frames the story during eight weeks in which Andrey was provided shelter in a hotel while he decides whether he will go back on medication. She weaves observed and reconstructed scenes of Andrey lost to his delusions and of his mother and stepfather who have reached a breaking point, dropping in brief sections of summary narrative that illuminate the bigger societal context.

Mindful of ethical concerns, Barry communicated with Andrey’s parents, Sam and Olga, for two years before meeting Andrey and learning from him directly that he wanted to communicate his ideas and preferences, including his rejection of medications.

The story seems to take a positive turn when Andrey leaves his hotel room and agrees to medication and gets his own apartment, a move enabled by the changing attitudes and shifting resources to transform how the mentally ill are treated in the United States. The progress is short-lived.

Barry is a 16-year veteran of The Times; she served as bureau chief in Delhi and Moscow, and was chief international correspondent based in London before returning to the U.S. In each of those roles, she often wrote about “trauma, loss and how people manage upheaval.” She was a finalist for the 2019 Pulitzer Prize in feature writing for her profile of a mysterious Indian prince, and on the team that won a 2011 Pulitzer in international reporting.

In an email interview with Nieman Storyboard, Barry talked about her reporting and writing style, advice for writers tackling the subject of mental illness and her epistolary relationship with her editor. The e-conversation has been edited for length and clarity, and is followed by an annotation of “The Man in Room 117.”

How did you come to write “The Man in Room 117”?
Sam Mintonye, Andrey’s stepfather, wrote a letter to my colleague Andy Newman two years ago, after he reported on a homeless man with mental illness who pushed a woman in front of a subway train, killing her. Andy couldn’t follow up because he works for the Metro section and the family was outside of New York so he sent the letter to me. I began corresponding with Sam and Andey’s mother, Olga, who were at that time visiting Andrey in his tent, bringing him food and laundry. We corresponded and spoke occasionally for almost two years; I had no idea if there would ultimately be a story.

Why did you pursue the story?
The U.S. is revisiting our laws around involuntary psychiatric treatment, largely because the number of mentally ill homeless people has begun to interfere with civic life. Most remarkable is a shift on the left toward the view that it is not compassionate to allow someone with psychosis to live outside. I wanted to highlight the tensions inherent in the decision to medicate someone against their will, and how it painfully it divides families. Also, the more I understood about the absurd cycle Andrey was in, bouncing from homelessness to jail to involuntary treatment and back to homelessness, the more I wanted to draw attention to it. It is not a coherent policy, and it makes people sicker.

Andrey was a useful case to look at because, unlike many people living in tents in American cities, he had a supportive family, education, a job and no addiction problem. If it were not for his mental illness, he would never have been homeless. That made it easier to look at the issue of involuntary treatment in isolation.

You spent years overseas for The Times. How does covering mental health issues compare with those assignments?
Mental health treatment is private, which can make it difficult to do what I like best: to watch things happen. And I no longer have the advantage of following news events, which I miss. But for the most part, the job of reporting is always fundamentally the same: You formulate a question, find a way to test it, report the hell out of it and, when you can’t report any more, sit down to write. I’ve done so many beats, and there’s always a steep learning curve as you get your arms around the subject matter. But reporting is reporting.

 What advice would you give writers who tackle the subject of mental illness?
It’s slow work; don’t expect stories that fly onto the front page. But it’s rewarding because so many families are living this reality and feel their experience is not represented in public forums.

What’s the most important lesson you learned from reporting Andrey’s story?
I spent a fair amount of time driving Andrey’s mom to visit him because she does not have a driver’s license. Early in my career, I would have been reluctant to do that because I might have seen it as getting too involved. But it made all the difference, driving around with her. It enriched my understanding, and brought me into the room when I needed to be there.

Were there books, articles or writers who inspire and influence your narrative nonfiction in general and specifically this story?
I reread Rachel Aviv’s great story “God Knows Where I Am,” Stephanie McCrummen’s haunting story “Behind the Yellow Door,” and Bob Kolker’s writing about schizophrenia, especially “Why Was Joshua Held for More Than Two Years For Someone Else’s Crimes?” I also like to read short stories to get a feeling for plotline and structure. I like Chekhov, because I think he is such a good reporter.

I noticed in the comments posted about your story that you engaged with readers, which, while not unheard of, has not been that usual for a Times reporter. Why did you engage in a dialogue with readers who commented on your story?
We are being encouraged to do this. Generally when I write about psychiatry the comments are fantastic — well-informed, reflective. On this one, so many of the comments were people telling stories about a family member and their agonizing efforts to keep them safe. I got letters for days and days. I can’t get some of them out of my head.

What role did editors play in the process of reporting and writing the story?
I like to talk and talk and talk while I’m reporting. I wrote reporting memos every time I visited Andrey, in the form of a letter that I sent to my editor. There were about 10 of those by the end of the process. I put them all in a binder with my other notes and documents, and they helped enormously when I sat down to compress it all. Those memos helped me understand when a moment really landed, when my editor had a feeling in her stomach from a scene I described.

I’m sure I’m very annoying to edit because I’m needy. I like to talk a lot.

ANNOTATION: Storyboard’s questions are in red; Barry’s answers in blue.

The Man in Room 117

Andrey Shevelyov would rather live on the street than take antipsychotic medication. Should it be his decision to make?

By Ellen Barry

Jan. 28, 2024

ALONE WITH HIS MOTHER for the first time in almost a year, Andrey Shevelyov had a question: Could he come home? Your lede feels like an invitation: I have something important, something interesting to tell you. How did you decide on it?This was my first lede. Then I threw it out and wrote a whole draft with a different lede. That didn’t work, so I went back to the first lede. I needed the story to begin with action, to present the stakes immediately. I wanted it to plunge forward. I knew, as well, that the story would end with the scary scene between the two of them, Andrey and Olga, so I thought the two scenes would act like bookends.

She sat beside him and stroked his head. The hotel room had a sour, rancid smell, and clothes lay mounded in a corner. His fingernails were long and curved and ridged with dirt. In jail, they cut off his hair, which had been matted and infested with lice. Two things here. First, smells are tripwires that explode in the brain, summoning forth memories. Yet too few writers employ them. You introduced one in the second graf. Why?Olga’s tender gesture of stroking Andrey’s head, against the squalor of the setting, treats him him as if he were her perfect child. She treasures him. People who pass him as a homeless man may not see that. I felt it keenly, and I wanted the reader to feel it.Then the description of Andrey’s physical state. How did you get it?He told her about the lice when they were discussing his haircut. The same description was in his medical records, which the family shared with me.

Clean-shaven now, Andrey looked younger than his 31 years, like the gentle, artistic boy he had been before the psychosis

took hold. “Zaichik,” his mother called him, a childhood nickname. Bunny rabbit. She pushed a strand of hair over his ear. He lay back on the bed and smiled, and a dimple appeared on his cheek.

“I like living with you also,” said Olga Mintonye, but it was not an honest answer.

Three years ago, when he stopped taking his antipsychotic medication, her son withdrew into delusions, erupting in unpredictable and menacing outbursts. Fearful of being evicted from their apartment, she and her husband, Sam, sought a no-contact order to keep Andrey away.

Since then, he had lived in a tent, wandering Vancouver, Wash., in ragged clothing and carrying machetes for protection. Twice, he had been in jail, ranting in his cell about the C.I.A. Three times, he was confined to psychiatric hospitals, where guards wrestled him down so he could be injected with antipsychotics.

Now they were together in Room 117 in a budget hotel overlooking the interstate. The county had allotted $8,400 to house him temporarily, as part of an effort by the state to divert the stream of severely mentally ill people from the criminal justice system. It was enough to keep him in the Red Lion Inn for eight weeks.

Before the money ran out, Andrey had to make a choice: Would he accept that he needed treatment, as his parents hoped, and move into a group home? Or would he go back to living in a tent? Was there another way?

Your prose is taut. Were there drafts that were longer?Yes, the second draft was about 1,000 words longer. But I really wanted action to drive the piece forward, as in a short story. I resist whole sections of B-matter because they slow down the story.

These are questions challenging the whole country. As affordable apartments all but vanished in American cities, a whole tier of people with disabling mental illness were forced onto the street, where they now live in numbers large enough to disrupt civic life. Many of them are shunted into the criminal justice system, only to return to homelessness upon their release.

In an effort to interrupt this cycle, many communities are expanding involuntary treatment, a practice the country repudiated decades ago. Patient rights groups warn that forced treatment alone will never work — that in the absence of a robust social support system, it only feeds people with mental illness back into the circuit of catch-and-release. Better to persuade them to accept treatment. Is this the story’s nut graf? Where do you stand on pulling out of a narrative with some version of a nut?I fight against them, I feel like they slow down the story. For 30 years, I have had great editors who remind me that readers need to know why we’re asking them to read something of this length, and usually they are right. These paragraphs were originally farther down, woven into the plotline, and the opening section was pure action.

So, in this precious window of time, Olga would try.

“What’s next?” she asked her son. You’ve injected foreshadowing of a structure that will follow Andrey and his mother for eight weeks. How did you settle on this approach?The fact that he had a limited time to make a decision gave me the tension I needed. It gave structure to the reporting. At the outset I thought there was a decent chance he would return to living in a tent, which is what he said he wanted to do.

She had offered him clothes, a hot meal, books to read, but he didn’t want those things. All he wanted was to sleep on her couch for a little while.

“I won’t get in the way,” he said, now pleading.

This was the problem, she told him: He scared people. At one point, convinced that she and Sam, his stepfather, were body doubles remote-controlled by the C.I.A., he smashed the rear window of their car with a flagpole, and they called 911.

“You sort of,” she said carefully, “have an idea that your behavior was inappropriate.”

She gulped and just said it: “Some medications do help. You have that condition, that medical condition, that has to be addressed.” Dialogue drives many scenes in the story. How do you capture it?I try to record everything, all the time; earlier in my career I trusted my note-taking more. Fortunately AI makes transcribing much easier.

Andrey knew this was coming, and he had an answer. He did not accept the diagnosis of schizoaffective disorder, or that his thoughts were delusions. If it was a choice between taking medication and living outside, he would choose outside.

Still, he wanted to be near her, so he looked for a compromise. “There are a lot of forested areas outside where we live,” he said. “I could set up a tent.”

“Set up a tent,” Olga repeated dully. She was standing by the ironing board, and she began to cry. How much more happened in this scene? Why did you end it here?I thought this set the stakes for the decision that would play out over the next weeks. They both want a way to be together so badly. The illness is separating them, and puts them both in danger.

Into the woods

Up until his mid-20s, Andrey could blend in with the stoner bohemians of the Pacific Northwest.

In high school, he was a bright, if lazy, student, with a tight group of friends who were all avid gamers like him. After he graduated, Sam and Olga hoped he would go on to college for art or music, but didn’t push. They both worked as stage hands, with good union jobs, and he seemed eager to do the same.

Then things unraveled. His roommates kicked him out, and he moved to the couch in Sam and Olga’s one-bedroom apartment in Vancouver. At work, there began to be “incidents,” odd and threatening behavior that upset his co-workers. Sam and Olga made him an appointment with a therapist.

At first, he was open to treatment, his medical records show. “I feel like I’m driving my friends away,” he told a therapist in 2018, when he was 26. What’s your source for this?Andrey’s parents had Power of Attorney and shared his records with me. This was critical, because I wanted to demonstrate that he was repeatedly forcibly medicated against his will, stabilized and then released to homelessness. Andrey knew I had his records because we spoke about what I found in them, and he did not seem disturbed by it.What was important to him was that I make it clear that he did not accept the diagnosis or medication — that he felt that his psychiatric treatment was inhumane. I was careful to include these points. I would also note that Andrey believed my presence helped him get services, like the apartment. So that was a motivation for him.

“I used to be able to talk to people,” he said. “Now they’re not interested.” Did Andrey have to give permission to access his medical records? If so, were there ethical questions raised by his mental competence to make that decision?See above; He knew I had read them because I asked him many questions about what I found, but I did not ask his permission. As far as his competence, he was considered competent to decide whether he needed or wanted medical treatment. He was judged competent to stand trial. Those are pretty important benchmarks. He had lots to say about psychiatric treatment; I thought he deserved the chance to say it.

In that session, he told the therapist about the family tragedy, a reservoir of pain he had walled off for years. When he was 8, he watched his twin sister, Sasha, drown in a duck pond in a neighbor’s yard. The children were alone. He had tried to save her. “I’ve thought about it to exhaustion,” he said. “I’ve incorporated her into these stories I make up.”

No one knows what causes schizoaffective disorder, the diagnosis he eventually received. You introduce the challenge of diagnosis at the exact time that a person without such knowledge might wonder, as I did, whether the tragedy he experienced as a child might be behind his mental illness. Why was it important to do so?Initially I did not address it here. But consistently, when I asked people I respected to read it, they raised the question of whether this trauma contributed to his disease. The idea of trauma activating existing genetic vulnerabilities is very alive right now.

For decades, scientists sought an answer in individual genes, only to conclude that hundreds, if not thousands, of genes are most likely involved, and that genes only set the stage. Other things must happen to trigger the disease; research studies implicate infections, cannabis use and childhood trauma, the kind of stress that leaves an imprint on the brain.

To his parents, there was no mystery. Sasha’s death had changed everything. “It’s obviously the trigger,” said Sam.

In journals, which his family shared, Andrey described “evil thoughts and voices” that accused him of killing his sister. “I’m guessing these voices are meant to drive me insane or kill myself or something I would never do,” he wrote. “I’m writing this to try to convince the people messing with me to stop. So, can you all stop? Pretty please.”

The therapist diagnosed him with schizotypal personality disorder and recommended “intervention to prevent further deterioration.” But, she added in her notes, “client states he is not interested in psychiatric medication at this time.” What method do you use to find a focus that enabled you to decide what to use in your story and what to leave out?I had about 500 pages of medical records from various hospitalizations. I used almost nothing in the end, and I had to fight a bit to preserve what I used from the records. But I think the language of medical records is so revealing, sometimes chilling and depersonalized. It really helped me understand what he was going through when he was confined.

A few months later, Sam and Ola took him to an emergency room, and doctors sent him home with a prescription for Zyprexa, an antipsychotic.

For the next two years, Andrey took the pills, but he hated how they made him feel.

This is not unusual; antipsychotic medications are deeply flawed, with side effects that are themselves disabling. They slow cognition, flatten the experience of pleasure and lower energy. They can cause ravenous hunger, leading to major weight gain. The economy of your prose is impressive. How did you boil down this information into a 61-word paragraph?I did a long interview with a woman who has written very compellingly about taking Zyprexa and other antipsychotics. As outsiders, it’s so easy to say, “You should just take this,” without fully understanding the cost of the treatment to the individual. Yes, there was a ton of compressing, through this whole section. It was like taking a fine tooth comb and combing through the draft, again and again. Removing nits.

Andrey says he stopped taking the medication because it caused erectile dysfunction. No physician determined whether this was the case, and his reasons quickly became beside the point. Things became so chaotic in the small apartment that in December 2020, Sam and Olga asked him to check into a hotel.

When the hotel kicked him out, he set up a tent in a wooded area near Pearson Field Airport, on the north bank of the Columbia River. Video journals that he recorded on his phone show him with tangled hair, surrounded by trash, rambling about underground vaults and sex traffickers. He would perform “shamanic dances” at the end of the runway, and sometimes the pilots would dip their wings at him when they took off. He seemed happy.

He was also sometimes scary. In one video, he tries to force his way into an apartment building, claiming he must rescue a porn star who is being held hostage. When residents call the police, he gets right in a male officer’s face. “Have you ever raped before?” he calls out insistently, filming the officer with his phone. “Sir, are you a rapist?”

The officer, exasperated, writes him a ticket. Still in scrubs from a recent hospitalization, Andrey wanders away, screaming, again and again, “Genocide Hindu Indians!”

‘Locked up like a rat’

Who wrote the sub-heads?Me! I love sub-heads. They have a kind of special atomic weight to them.

Andrey had entered the circuit. In Washington State, as in most of the country, the choice of whether to accept treatment for psychosis rests substantially in the hands of the individual. A 1975 Supreme Court decision set the bar for involuntary treatment high, ruling that people who pose no danger and are “capable of surviving safely in freedom” cannot be confined to a psychiatric hospital against their will. How much research did you have to do to be able to write with such authority?I spoke to a couple legal experts when I was putting together these sentences —  one law professor and a legal director of a policy think tank. I became very interested in the 1975 Donaldson case, and read Kenneth Donaldson’s biography. It shocked me how easy it was to confine people prior to the Supreme Court decision.

If a person faces serious criminal charges, however, the right to refuse treatment is almost entirely swept away, because, according to a 1960 Supreme Court decision, adjudicating a mentally incompetent person is a violation of constitutional rights. So at this point, people like Andrey can be forcibly medicated, judged and released, after which they are once again free to refuse treatment.

Andrey’s medical records chronicle this nonsensical pattern. He “feels he is being lied to and being kept against his will,” read the intake notes from his first involuntary hospitalization, which began four months after he stopped taking medication. “Why do you guys keep me locked up like a rat?” he asked his doctor on Day 5. Why was it important to be specific about the day of treatment?I thought it was revealing that, repeatedly, his responses seem to change after several days on antipsychotic medication. He is less aggressive, calmer, more self-aware.

With the court-ordered treatment, that changed. “The patient,” the doctor wrote on Day 12, was “agreeable to Zyprexa.” He was also counting the days until his release. “It’s great here,” he told the doctor, “but I’d rather be homeless.”

Within a year, he was back in a psychiatric ward, diverted there after smashing his parents’ windshield and landing in jail. Why did you make that big jump in time here?This section was cut significantly, I had to argue to preserve material from the medical records. It felt most important to demonstrate beyond any doubt the pattern that emerged as he was repeatedly confined, treated and released.

“After patient was medicated, he then spit on staff face,” wrote a nurse at Cascade Behavioral Health Hospital, where he stayed for about a month.

For a second time, he was medicated against his will and cleared for release, though discharge notes sum up his progress as “no change.” When staff offered him help finding housing, Andrey “expressed that he would like to be homeless when he discharges.” Staff obligingly dropped Andrey at a shelter, and he returned to living in a tent.

Six months later, Andrey was in jail for threatening a grocery store clerk. Tell me about the use of Tell me about the use of time for transitions in this piece?The solutions provided by the system were not durable. Andrey ended up back in the tent; he ended up back in jail. Hospitals kept discharging him, knowing full well that would be the result. I suppose I thought the reader should have to suffer the monotony of repetition, because that is the way our system is designed.

“Mr. Shevelyov was not wearing clothing and not consistently wearing clothing,” wrote the psychologist sent to his cell to assess his competency. “Mr. Shevelyov’s jail-issued clothing was in his toilet, which he explained was because ‘I was protesting war criminals illegally holding me.’” Did the story raise any concerns about invasion of privacy?Yes, we had to weigh the social utility of describing the way this system works against the difficulty for Andrey, Sam and Olga to have all this appear in public.

Deemed incompetent, he was sent to a state psychiatric hospital. “He vehemently denies having a mental illness and adamantly refuses to take medication,” wrote the admitting doctor. This stay was more violent. Once, after grabbing a nurse’s arms, he was bound to a bed with five-point restraints, with straps around his waist, each ankle and each wrist.

After two weeks of forced medication, however, Andrey “presented as calmer,” assuring staff that his theories were just his “active imagination.” When he was discharged, his mental status was given as “W.N.L.,” which stands for “within normal limits.”

He was returned to jail to await trial. The prosecutor, eager to clear this penny-ante case, agreed to release him if he pleaded guilty to a misdemeanor. He was released and dropped off at the Red Lion, a free man. The story of Andrey, his mother and his stepfather is so sad. How did it affect you?Good question. I think I’m still recovering.

Andrey had always been frank about his intentions. Asked by a doctor to describe his short-term plans, he said, “I am ready to leave and start my new life as a homeless dude.” Asked if he would continue taking medications, he said, “Depends if I have to.”

A visit from a friend

One afternoon, about three weeks after Andrey left jail, a woman with closely shaved blond hair knocked on the window of his room at the Red Lion.

Andrey’s friends had dropped away as he got sicker, his conversation a fire hose of paranoia and fantasy. But not Courtny Ryser Lewis — she stuck like a burr. How did you come up with this analogy?I don’t know. She was just so stubbornly loyal.

Courtny was a free spirit, her forearms wrapped in tattoos, but she was also the earnest child of evangelical Christian missionaries.

When Sam and Olga told her Andrey was being released, Courtny offered to speak to him about accepting treatment. They had a short time to get through to him, she felt, while he still had traces of medication in his system from jail. “I’ve got the most — I don’t know if this is the right word — the most sway,” she said. By then, Andrey had settled in. Buzzing with ideas, he had resumed recording video diaries, sometimes for hours a day. He had spent $665 on dozens of heavy-duty pencil pouches, which he was fashioning into a suit of body armor. “I’m, like, the happiest dude ever,” he said.How much time did you spend with Andrey to gain enough material for your story?I made three trips to Vancouver, and each time I visited him two or three times, usually for a few hours.How do you decide to stop reporting and start writing? Do you write during the reporting process?I wrote reporting memos to my editor during the process, and those yielded most of the action in the story. They were very useful. We had initially agreed I would just follow Andrey through his time at the Red Lion. When he moved, suddenly, to the new apartment, it felt like a kind of ending, that he would now retreat into his own space. I hung out with him and his mother while he was moving in, and when she slipped out while he was yelling at his laptop I definitely had a feeling of an ending. A friend of mine used to say the ending should “snap shut like a purse.” I had that feeling.

His parents prodded him to think about the future, reminding him that he would have to leave the hotel in a few weeks. Andrey was in no hurry. He had said no to a spot in an “adult family home,” something his case manager offered, and refused visits from a PACT team, which provides intensive services for mentally ill people living independently.

He felt he had options: If the caseworker couldn’t find an apartment that suited him, he could always pick up a tent for $40. “If I have to live outside to avoid pharmaceutical garbage,” he said, “I would do that.”

When Andrey opened the door, Courtny took it all in — the smell, his dazed look. She was so nervous that she had written down her arguments on a piece of paper.Were you presentfor this scene?No, but Courtny made an audio recording.

She tried to explain how he appeared to his friends. How, after he began living in a tent, it was as if he had dropped off the face of the earth. Andrey was stung by this. His response was bitter, sarcastic.

“All I was working on was saving the world,” he said.

“You were not,” Courtny said. “What changed?”

This offended him. One of his videos had prompted the U.S. military to withdraw from Afghanistan, he said.

“No one’s going to listen to you, Andrey,” his friend said.

“Why?”

“Because you’re not sane,” she said.

“What?” he said.

She repeated it. “You’re not sane.”

“I’m trying to be honest with you,” she added, “because I’m your best friend.”

Andrey’s response was icy. They weren’t best friends, or even friends. “We’ve never done anything together,” he said. “There is basically not any memories there.”

Courtny saw him, for a moment, as a stranger would.

She left the room midsentence, without saying goodbye. Could you describe your writing and revision process?I write from beginning to end, very chronologically. I write in chunks, try to do 500-1000 words a day. Every time I sit down to write, I go through all the preceding text, usually making small changes and trims. By the time I finish with a draft, I know it very well.

‘It’s like a bomb went off’

Sam, Andrey’s stepfather, understood. After eight years of trying to get through to Andrey, he, too, had reached some kind of limit.

At the beginning, Sam studied motivational interviewing, a type of counseling that aims to elicit behavior change gently and gradually. He and Andrey went together to meetings of the Hearing Voices Network, a peer support network for people who experience hallucinations.

But the more he coaxed Andrey to acknowledge that he had a mental illness, the more Andrey lashed out. The stress in the household became so intolerable that Sam briefly checked himself into a psychiatric hospital.

Sometimes Sam felt he was wrestling with the disease, trying to haul Andrey back from a remote, dark space. “It’s almost like an entity that defends itself,” he said. “It’s trying to completely take him over.”You paraphrase Sam’s story here, alternating with his comments. Why?Sam wasn’t in my first draft —  only Olga and Courtny. The reason for that is that Sam was not physically present; he had moved back east. So while I had spoken to him for many hours, more than to Olga and Andrey, he was not present. An editor encouraged me to add him, because he is another person who loves Andrey but ultimately needed to distance himself. It made sense to show each of them distancing themselves — first Courtny, then Sam, and at the end, Olga.

Kim Schneiderman, the executive director of Vancouver’s regional chapter of the National Alliance on Mental Illness, urged Sam and Olga to scale back their expectations. If a person is refusing treatment, but poses no immediate danger, she said, “there is no way to get them help at all.” How many people did you interview?Maybe 20 or 25? But I knew most of those interviews would not appear in the final story.

When this sinks in, “parents give up,” she said. “They just quit having any interaction with them.” Sam and Olga would not give up, she thought, but she advised them to get some distance so that Andrey was not at the center of their lives.

Friends had given Sam similar advice. Tom Sangrey, a close friend since middle school, said years had passed before he could say it plainly: You can’t help him if it ruins your mental health, if it ruins your marriage.

“It’s like a virus,” he said. “It affects everyone it touches. It’s like a bomb went off in the family.” This is such a powerful quote, summing up the effect of serious mental illness. How did you get it?Oof, yes. I interviewed several close friends of the family, assuming I would not quote them, but this phrase really stuck to me. I kept trying to cut it and then putting it back. It was haunting.

Finally, in October 2022, Sam drove back east to stay with his mother. He had not really distanced himself; he and Olga spoke on the phone every day, sometimes for hours. He spent his days writing emails about Andrey’s situation to prosecutors, city officials, newspaper reporters, anyone he thought might help.

But he wouldn’t be going to see Andrey in that dark, cluttered room. The only person left to do that was Olga.

Housing or medication?

Sam and Olga had concluded that only involuntary treatment could break the cycle for Andrey — something open-ended, combining long-term injectable medications with intensive therapy and counseling.

They are part of a much larger ideological shift taking place, as communities grope for ways to manage ballooning homeless populations. California, one of the first states to turn away from involuntary treatment, has passed new laws expanding it. New York has made a billion-dollar investment in residential housing, psychiatric beds and wraparound services.The story shifts from dramatic narrative to summaries of background information. What is your intent when you alternate between these sections?This is the heart of the policy dilemma, but I didn’t want to lard up the action at the beginning of the story, so I’m trying to slip it in here, when the reader is already plunging forward.

Sam had staked his hopes on Washington’s new involuntary treatment law, and found it maddening that this fall, when Andrey was released, the new system was not yet active. His frustration was often directed toward civil rights advocates who oppose forced treatment.

“They have an agenda, but the agenda is not to help him,” he said. “Their agenda is to let him just be crazy. Whether that includes violence, assaults, living in degradation, living in his own filth, starving, eating moldy food. That is his right.”

One day, as he made his rounds of phone calls, Sam found himself debating Kimberly Mosolf, director of the treatment facilities program at the nonprofit organization Disability Rights Washington.

She laid out her case: Forcing someone like Andrey to take medication again would backfire, leaving him more resistant to treatment, not less.

She pointed to data from the Seattle area, which showed that almost a quarter of people compelled to take medication had been forced to do so more than three times before. Seven percent of them had been forced 10 times or more.

“We are looking at a churning effect,” she said. “These periods of brief incarceration, brief civil commitment, they are destabilizing. That is what the data tells us.” How do you know these details of the conversation Sam had with her?Both she and Sam described the conversation.

She advised a gentler, slower way forward. If Andrey got permanent housing, with no strings attached, outreach workers could build a rapport and gradually broach the subject of medication. This approach, known in the policy world as “housing first,” has emerged as the primary strategy for addressing homelessness in American cities, allowing officials to chip away at tent encampments without encroaching on civil liberties.

This was the path that opened to Andrey.

He got the call on a chilly, gray day at the end of November. He was in his room, recording a torrent of new ideas — that his mother had been inseminated with Joseph Stalin’s sperm, that the government had planted a bomb in his brain and detonated it. The front desk called to say his caseworker had come to visit. She had great news. Were you present?Yes. But she asked me to go away while they talked.

The local housing authority was offering him a one-bedroom at Central Park Place, a low-income apartment building on the grounds of Vancouver’s Veterans Affairs hospital. The residents there were mostly veterans, but rooms were also set aside for people with mental illnesses. The rent was $590 a month, and could be covered by his disability check.

This placement solved several problems at once. He would no longer be at risk of freezing to death, or jamming up the courts, or frightening pedestrians. For the caseworkers, it was a rare triumph. And for Andrey, it meant that the pressure was off: He had a safe place to live that was not contingent on taking medication.

On the December morning when she arrived at the hotel to help him move, Olga was surprised to find him awake, his possessions packed away in six bags, itching to go. All that morning, he seemed like a different person — alert, motivated, funny. He charmed the manager at the apartment building, which was clean and bright, festooned with Christmas decorations.How did you manage to be present for such pivotal moments?When I found out he was moving I immediately went to Vancouver, since I knew I wanted to capture that scene. I just dropped everything.

He signed forms promising not to punch the walls, start fires, or smoke in the unit. The building manager asked him, apologetically, to indicate what should be done with his belongings “if something should happen.”

He blinked. “You mean, if I’m dead?”

She nodded. There was a small, awkward silence.

“Damn, dude!” he said, and everyone in the room cracked up.

“Bury me with my stuff!” he cried jubilantly.

They laughed again. Then the building manager, with a ceremonial flourish, presented him with a set of keys on a blue fob.

Beside the other residents, men in their 60s and 70s, Andrey seemed vigorous and charismatic, overloaded with the natural gifts of youth. His apartment was tiny but pristine, with a window looking out on a roadway drenched in sunshine. He walked his caseworker to the elevator bank and posed for a commemorative photo.

Then the door closed, and he was in the small room with his mother.

Behind the door

Cross-legged on his twin bed, Andrey began to talk. Switching on the camera on his laptop, he spoke about Nazis, George Bush, lunar men, Sandy Hook, Jeffrey Epstein, rape slaves, war crimes, lobotomies, LSD, the Yakuza, nanobots, the spirit world, underground vaults, genocide, the C.I.A.

“My plan for myself is, as fast as possible, be a world leader,” he said.

He understood that people might see him as lonely, here in this tiny room, but it wasn’t like that. “It feels like my sister’s soul is still in my body,” he said. “Like our souls fused, basically.” He felt less and less need for outside relationships. “I’m basically hunting the most dangerous men in the world,” he said. “And that isn’t the best time to have civilian friendships.”Who is he talking to?Me.

Olga had been uneasy all day. She had seen Andrey withdraw before, disappearing into psychosis. The windfall of the apartment, she worried, might be setting them all up for another failure — even, possibly, putting them in danger. As Andrey spoke for 15 minutes, then an hour, Olga futzed around, folding clothes and unwrapping cleaning products. Anxiety radiated off her in waves.

When she interrupted, trying to get him to focus on practical matters, he turned on her.

“Mom, listen to me,” he roared. “Stop thinking you’re in charge of me! I’m a tactical genius who saved the world!”

The outburst was loud enough to disturb his new neighbors, she warned him, but it was as if he didn’t hear her.

“OK,” Olga said. “I’m leaving now.”

“Stop talking to me, I aren’t talking to you,” he yelled. “I’m a 31-year-old man who saved the world a dozen times!”

Then he turned his attention back to the camera on his laptop. She wondered whether to say goodbye, but he seemed unaware of her presence. She slipped out the door, and he didn’t even look up.What's the approach that let's you capture such intimate moments?I hung out for long periods. A lot of it I didn’t use. But regularly, when they were together, Andrey and Olga would get into some kind of argument, just as I do with my mom. So I began to understand that this was the spine of the story, and I had to listen for it.

He had been in the apartment for almost a month when the building manager spoke to Sam and Olga: Andrey was screaming during quiet hours, and some of the other men were afraid of him. When the manager knocked on the door, he didn’t answer.

It was evident, in his videos, that he was rapidly losing weight; his collarbones and cheekbones angled out. “I don’t feel hunger and I barely eat food whatsoever” is the way he put it. Sometimes, in the past, Andrey had feared his food was poisoned. So Olga stopped by, and the first thing she saw was the food she had brought him in a trash can.

“What is this?” she asked. He lunged at her, screaming. She tried to leave, and he blocked the door. She managed to get around him, and then she was in the hallway, shaking.

This marked a limit for Olga — she who had delivered his laundry, neatly folded, to his tent in the woods. On the phone with Sam, she finally made the promise: She would not visit Andrey alone again.

The two of them talked about buying a house in upstate New York, resuming their married life, leaving behind the scene of all that pain.

Andrey remained in the fortress of his apartment. Over the intercom, he answered the questions of caseworkers and crisis response teams who had been called to the scene, but he refused their requests to enter.

If an eviction was in his future, it was weeks or months away, so he could carry on with his work, reviewing plans to arrest the leadership of the C.I.A. and oversee a war crimes tribunal. “Did you see that spirit in the background?” he asked the camera. A team of brave men was coming to join him, he said. All he had to do was wait.

One thing, though — he missed his mother. He spoke about her, his “truest love,” in his video journals that night, after he lunged at her. The message he wanted to get through was that she needed protection.

“I’m so sorry,” he said. “We weren’t there to save her. We did everything we could.”This is so wrenching. Of all the places you probably could have ended this story, how did you decide on this one?It broke my heart. He’s saying, my mother needs protection. Unspoken is that she may need protection from him. I think this is one of the saddest stories I’ve ever written.

* * *

Chip Scanlan is an award-winning writer who taught at the Poynter Institute and now coaches writers around the world. He is the author of several books on writing and the newsletter Chip's Writing Lessons.

The post Respecting the tortured voice and troubled choices of the mentally ill appeared first on Nieman Storyboard.

Nieman Storyboard contributor Chip Scanlan and Casey Frechette worked together for more than a decade at The Poynter Institute where they created online courses in reporting and writing. Casey is an associate professor and chair of the journalism department at the University of South Florida.

Chip Scanlan: Hi Casey, I’ve had a great time with you on Google Meet over the last several weeks playing and working with AI and talking about its implications for journalists and journalism. Why does the topic interest you?

Casey Frechette: Hi Chip. It’s been great exploring this with you. I got interested in AI last spring, when I had what I’d call a “whoa” moment. I started playing with ChatGPT in earnest, feeding it different journalistic things. I gave it a bunch of news stories and asked it to suggest headlines. I pasted an article from The Atlantic and one from The New Yorker and asked it to tell me which was which. I had it summarize stories in two sentences. And I asked it to review several pieces of breaking news for signs of bias, both obvious and subtle. I was blown away by how quickly it performed these tasks, at how useful the results were. The experience felt a bit magical, akin to going online for the first time and feeling awestruck by the possibilities (minus the crunchy modem-connection noises). But I’m curious: What’s drawn you to this topic?

Chip: I guess I’ve always been an early adopter. As soon as news broke on Nov. 20, 2022, that Open AI launched ChatGPT, I signed on, first as a free user, and then got a $20 a month subscription so I could have access to ChatGPT-4, its most advanced system. I haven’t used it as broadly as you have or focused it on journalism as I would expect an inspiring J-prof like yourself would.  I’m writing a historical novel set during World War II and have used it primarily for research. Its performance has been impressive, although I must concede that a Google search can answer many of my questions. What sets ChatGPT apart is the chat part, a dialogue between you and a super-fast computer, one trained by scraping large swaths of the internet for content. Follow-up questions are the lifeblood of the journalist and I ask the chatbot a lot of them. The most important one is “What are your sources?”

In its staff guide to AI, Wired magazine describes AI’s value to journalists: “AI software can’t call sources and wheedle information out of them, but it can produce half-decent transcripts of those calls, and new generative AI tools can condense hundreds of pages of those transcripts into a summary.”

But you and I have used it together to see how it operates around the discipline of journalism. Want to take it from there?

Casey: Isn’t it amazing how the cost of technology plummets over time? I recall learning early digital watches sold for upwards of $1,000 when they first hit shelves; a nightlight and stopwatch were the extent of their “smart” features. I guess we could call it the “early adopter tax” but, like you, I find the allure of exploring the frontiers of innovation too enticing to pass up.

Using a chatbot as a kind of research assistant for your novel is a great idea. The interactive style you described certainly translates well to the kind of critical engagement that makes journalists so good at finding, organizing and presenting information. Applied to this realm, the possibilities are really quite staggering.

There’s a lot of focus, understandably, on the “generative” side of AI. And we’ve all heard the horror stories of news organizations that have dabbled with “robot-assisted” stories that sound cartoonish or make embarrassing errors.

What’s more interesting to me, though, are all the other ways AI could change news production and consumption, from analyzing vast, complex datasets, as The New York Times recently did in a report on Israeli bombings in Gaza, to generating different story variations for each reader, each calibrated to exacting personal preferences.

I think there are a few lessons here, chief among them: Both pitfalls and possibilities abound.

Chip: We’ve spent a lot of time testing the boundaries. I’m thinking of the time I took notes from a newspaper story of mine that I reverse-engineered to teach writing to journalists. First, I inputted the text into the chatbot with a request to answer questions about the story, including my favorite: What’s the story really about, in one word? In seconds, it replied, “Empathy,” which totally nailed the theme. I was impressed.

But as you’ll recall from the other end of our Google Meeting, when we asked it to start turning the notes into narrative, the problems began.

 “Writing stories is another matter,”  a staff guide for Wired magazine concluded. ”It turns out current AI tools are very good at churning out convincing (if formulaic) copy riddled with falsehoods. A few publications have tried — sometimes with disastrous results “

I’ve read that AI scientists call that “hallucinating” or “confabulation.” My profane translation as a journalist: It began to make shit up. It made up dialogue. It inserted facts that weren’t in the notes.

When you and I challenged it on these points, it apologized “for the oversight.” It said the dialogue was “a creative interpretation based on the situation outlined in your notes, as there was no direct transcript of the conversation.”

Not ready for journalism primetime, we agreed. In a longer version of our e-conversation, you told me that the current iteration of AI tools are “not built for journalism” but, with the right cautions, can still be useful. Those cautions include the warning that we can’t take anything for granted, and must question the sourcing of everything, especially in high-stakes stories.

Your take, as a journalism professor and technologist, holds a lot of weight with me. So let me ask. What do you see as other pitfalls of AI when it comes to reporting and writing stories? How might it be of use to writers and editors? And what does someone have to come up with so journalists can use this technology in the most intelligent and ethical way possible? (Hint: We might call it “Defoe” after the early 18th-century writer whom ChatGPT identified as one of the world’s first journalists (historians might disagree).

Casey: You’ve hit on what I see as the key concern, and that’s the fallibility of the technology. It makes stuff up. It gets stuff wrong. It leaves out important caveats and context. And it does all of this with the confidence of, to paraphrase a popular saying, a person putting on lipstick without a mirror.

In a similar vein, we all need to be concerned with the biases lurking in the datasets that AI models are trained on. There’s important work ongoing in the realm of responsible AI to combat this, but reporters and editors need to be aware that many AI technologies have amplified, not diminished, the unfairness and misrepresentation that ail society and undermine the journalistic mission.

Another pitfall is an extension of the conventional wisdom that phones made reporters lazy because they didn’t have to go see people in person anymore. And then email made them lazy because they didn’t have to pick up the phone. And social media made them lazy because they didn’t have to send email. And now, I suppose, AI might make them lazy because they don’t have to hop on their DMs. I tend to think this concern, and its many permutations over the years, grossly underestimates journalists’ innate resourcefulness.

I do worry, though, that the “smarter” technologies become, the more tempting it is to outsource our cognitive capacities to them.

I’ll confess that if you asked me to drive to a location I’m unfamiliar with, without the assistance of Waze, I’d be hard-pressed to get there without some struggle and a good deal of cursing. The advent of calculators made long-division a rare skill. The internet wreaked havoc on our attention spans, as Nicholas Carr explored in his seminal book “The Shallows.” What will happen with AI? Will our creative and expressive muscles atrophy?

But let’s look at it from another angle. Every reporter must now ask themselves: Did AI write what I’m reading? Draw what I’m seeing? Speak what I’m hearing?

We might be comfortable answering these questions now, at ease with our ability to notice a stilted voice or a telltale sentence construction. But whatever signs we think make fake content easy to spot at the moment are sure to become outdated as the technology improves. And here’s a further complication: Much of the discourse about deep fakes has assumed what I see as an all-or-nothing framing. But I anticipate we’re entering a messy phase where parts of things are AI-generated, mixed with human-generated content.

Here’s the good news: Once we accept that AI is flawed, we can use it responsibly, even relish in what it has to offer.

For writers and editors, I see potential in AI as a research tool; a brainstorming partner; a transcription service; an assistant able to sort, summarize and label; a writing coach who offers feedback and troubleshoots problems with copy; an editor who can review early drafts for clarity, accuracy, AP style, grammar; a data analyzer; a reader who can give feedback as either a neophyte or expert on the topic at hand, depending on the role we want it to play.

I’d hasten to add that there’s a certain irony in considering how journalists can benefit from AI, considering how much AI has benefited from journalists.

We now know that many chatbots have been trained, in no small part, on stories that have been scraped from news websites. Without the deeply considered, reported and edited content that’s produced almost exclusively by news organizations, chatbots like ChatGPT wouldn’t be nearly as good. At least, that’s the argument lawyers for The New York Times have made in their action against OpenAI and Microsoft alleging massive copyright infringement.

If we’re envisioning an AI tool truly suited for journalists, I’d imagine it would boast a few critical features:

It would be private and secure. It would acknowledge the limits of its understanding and the ways it might be presenting biased or distorted perspectives. It would cite its sources. And it would never make shit up.

What’s interesting is that the current generation of tools isn’t intrinsically incapable of these things. Yet, as we’ve seen from our experiments, it takes tremendous coaxing to get a bot to work only with supplied facts.

As someone who has coached legions of writers, I wonder what features you might add to this wishlist.

Chip: I’m not sure what I can add to your comprehensive list.  I’ve already been impressed with it can summarize and analyze texts with detail and accuracy in a fashion that would make a coach or editor proud. This strength would help writers, too, especially before they submit a story or work without the safety net an editor provides. It could also guide revision, along with other features on your wish list.

The aforementioned risk remains: ChatGPT will happily make the revisions if you ask, and there lie the dangers of hallucination, not to mention the ethical lapse that could, if not should, cost a journalist their job.

Still, it can’t provide the emotional support I’ve found some writers need to get past writer’s block and overcome fears about approaching strangers with sometimes uncomfortable questions, and meeting deadlines — problems that afflict rookies and newsroom veterans alike. It may put me out of business as a coach down the road. (I don’t know how far down.)

I’m convinced it can make us better writers by raising questions and pointing out flaws. The question for the future is how far should we let it go? Highlight passages in news stories that were created with AI? Share a byline with ChatGPT? What will it do to a writer’s voice, the element that draws in readers and keeps them engaged if everything is regurgitated by a large language model and copied from other writers? That’s a big question and one that anyone who works with words has to address.

But first, in the spirit of this exchange, I asked ChatGPT to answer your question about my wishlist as a coach. Its response (in italics so we’re certain of the source):

ChatGPT: Source Verification: An AI tool that can assist journalists in verifying the credibility of sources and the authenticity of information, especially in an era of deepfakes and misinformation.

Ethical Guidance: An AI that incorporates ethical guidelines specific to journalism, helping journalists navigate complex issues like privacy, consent, and the public interest.

Collaborative Editing: Enhanced features for collaborative editing that allow multiple journalists to work together seamlessly, with AI providing real-time suggestions for improvement based on best practices in journalism.

Enhanced Fact-Checking: An AI tool that not only helps in fact-checking but also provides context and background information to help journalists understand the broader implications of their stories.

Between your and my ideas, and ChatGPT’s, we could definitely give a coder some ideas. And now that we’re back to human-to-human conversation, can AI replace what you do as a teacher?

Casey: Replace me? No way!

Wait, let me step back. My knee-jerk reaction to your question is probably driven more by fear of obsolescence than a reasoned view on where we’re heading. Much as I’d like to believe I can offer something of unique value, my sense is that AI can already outperform me on several key tasks and will soon be a worthy rival on many more, from designing learning experiences, to producing lessons, to providing feedback.

In her book “Machines Who Think,” Pamela McCorduck recounts philosopher and mathematician Bertrand Russell’s reaction when the inventors of “The Logic Theorist,” often regarded as the first AI program, presented him with an AI-generated mathematical proof superior to one of his own. Russell’s reaction, McCorduck wrote, was one of delight. I hope I can approach AI’s expanding capacities with a similar open-mindedness and humility.

Still, I’m not quite ready to let go of the notion that, without emotion, a computer will never be able to provide the kind of connection that instills a passion to learn. And yet, I can’t help but think that this is the next chapter in our relationship with technology. I hope you don’t mind if we bring our friend Claude into the conversation.

Claude, what are your thoughts on AI’s potential to provide humans with emotional connection?

Claude: I don’t actually have personal thoughts or feelings. I’m an AI assistant created by Anthropic to be helpful, harmless, and honest.

Casey: Allow me to rephrase the question. Could AI one day provide humans with meaningful emotional connection?

Claude: I think the potential for AI to provide emotional connection to humans is an interesting area of research and development. Some key considerations around this topic:

• Emotional connection involves mutual understanding, empathy and being “present” for another – abilities that current AI systems lack. However, research is ongoing into developing emotional intelligence and social skills in AI.
• Humans connect through shared experiences and vulnerabilities. AI has no innate experiences or ability to be “vulnerable” in the human sense. Programming those qualities convincingly is technically challenging.
• There are risks if AI appears to be sentient when it is not actually conscious. People could anthropomorphize and over-rely on AI that lacks true understanding or care for human wellbeing.
• On the other hand, some level of emotional rapport could be beneficial, e.g. AI pets or companions for the elderly. This raises ethical issues around deception and manipulation of vulnerable persons.
• More research is needed into human emotional needs and the appropriate role of technology in healthy relationships. Emotional skills in AI may develop gradually and benefit from ethical oversight. But overall this remains a complex issue with many open questions.

In summary, while AI could potentially emulate emotional connection, truly experiencing and providing it like a human may be beyond current technical capabilities and ethical norms. The path forward merits careful consideration of human psychological needs and AI’s limitations.

Casey: Well, that’s some food for thought. My takeaway is we might miss out on a lot of good, without a full exploration of the capacity for emotions in machines — as practically and philosophically challenging as it might be to do so.

But I’d like to get back to your question: How far should we let it go?

So long as technology outpaces complementary legal, ethical and professional frameworks, individual journalists and news organizations alike will need to make important decisions about whether and how to incorporate AI into their work. And they’ll have to decide how to express those choices to the public.

What’s your advice for a reporter or editor who may be grappling with these questions?

Chip: Decide AI’s value to you by using it enough to be familiar with its pitfalls and promises. If you have never used a chatbot before, you need to. AI already has serious implications for you and your future. Understand it. Follow the news about it. If you’re already a user of ChatGPT or Claude (others include Google’s Gemini and the playful chatbot PI), experiment. Ask it questions, especially about things you know to determine how accurate it is. Start by asking for your bio or that of someone you know well. (My first AI-bio erroneously gave me a Pulitzer and jobs with The New York Times and The Washington Post.) Follow Casey’s lead and have it brainstorm questions for an interview.

As with any source of information, verify. Request citations for the answer. ChatGPT and Claude have given me sources, like Wikipedia footnotes. Sometimes, however, the chatbot will say, in effect, it already knows all these things and can say things without attribution. Sounds like hubris to me.

Casey: That’s a great list of tips. I especially appreciate your emphasis on reflective experimentation. It strikes me that it’s important to stake out a middle ground between proactive trial and error, and cautious curiosity, particularly in the absence of the policies, laws and guidelines that will likely solidify in the months and years ahead.

Taking Over

We pasted this entire conversation into ChatGPT and Claude and asked them to respond to Casey’s question: What’s your advice for a reporter or editor who may be grappling with these questions?

ChatGPT: 
Familiarize Yourself with AI: Engage with AI technology to understand its capabilities and limitations. Experiment with AI tools like ChatGPT, Claude, and others to see how they can aid in journalistic tasks and to become aware of the accuracy and reliability of the information they provide.

Evaluate AI’s Role and Value: Consider how AI can enhance or complement your journalistic work. Look at AI as a tool for research, summarization, brainstorming, and even preliminary editing, but be mindful of its current limitations, especially in generating content without errors or biases.

Verify Everything: Like with any source, verification is crucial. Always question the accuracy of the information provided by AI and seek out original sources or citations whenever possible. Be skeptical of AI’s ability to provide information without proper attribution and verify facts independently.

Be Aware of Biases: Recognize that AI models can reflect biases present in their training data. Approach AI-generated content with a critical eye and consider the impact of potential biases on your reporting.

Understand the Legal and Ethical Implications: Stay informed about the legal, ethical, and professional frameworks surrounding the use of AI in journalism. Consider the implications of copyright infringement, privacy, consent, and the public interest in your use of AI technologies.

Educate Yourself and Your Audience: Keep up with developments in AI technology and its implications for journalism. Educate your audience about the role of AI in your reporting process, including how it’s used and its limitations.

Use AI Responsibly: Utilize AI as a supplement to, not a replacement for, traditional journalistic methods. It should aid in enhancing the quality and efficiency of your work without compromising journalistic integrity or ethical standards.

Embrace a Critical Mindset: Maintain a healthy skepticism towards AI-generated content. Question its origins, accuracy, and the potential for “hallucination” or generating false information.

Explore AI’s Potential Ethically: Investigate the ways AI can contribute to journalism without undermining trust or quality. This includes leveraging AI for data analysis, generating story ideas, or improving the accuracy of reporting, while always adhering to the highest journalistic standards.

Prepare for the Future: As AI technology continues to evolve, stay ahead by continually assessing how these tools can be integrated into journalistic practices in ethical, responsible, and innovative ways.

By considering these guidelines, reporters and editors can navigate the complexities of incorporating AI into journalism with a balanced approach that maximizes benefits while minimizing risks.

Claude: 
The key tension seems to be between leveraging AI’s potential benefits – such as automation of routine tasks, surfacing patterns in data, or generating story ideas – while also maintaining journalistic principles. Reporters and editors need to thoughtfully evaluate AI tools to make sure they enhance rather than hinder core journalistic values like accuracy, transparency and human judgment.

Rather than seeing AI as a plug-and-play solution, newsrooms should carefully assess if and how these technologies can augment human capacities rather than replace them. Some constructive ways forward could be keeping humans in the loop on fact-checking AI content, disclosing the use of AI-generated text, or having AI focus more on supplemental analysis rather than primary story generation.

The path to integrating AI into news likely requires an openness to its capabilities but also a mindfulness regarding its limitations. Prioritizing timeless journalistic values around truth and understanding while leveraging cutting-edge technology.

Chip: I think we humans should get the last word. Casey?

Casey: I’m impressed by ChatGPT and Claude’s synthesis and advice. But our conversation brings to mind another advantage humans still have over machines: the ability to gain new appreciations and insights through dialog with each other.

I’ve enjoyed our chat, as always, Chip. Looking forward to continuing our explorations.

* * *

Chip Scanlan is an award-winning writer who taught at the Poynter Institute and now coaches writers around the world. He is the author of several books on writing and the newsletter Chip’s Writing Lessons.

Casey Frechette is an associate professor of journalism at the University of South Florida’s St. Petersburg campus and former interactive learning producer at the Poynter Institute.

The post Two journalists talk to the bots — who talk back — about the pros and pitfalls of AI appeared first on Nieman Storyboard.

The best storytellers are driven by an insatiable need to know. Give them a mystery and they will dedicate themselves to trying to solve it.

That relentless inquisitiveness propelled John Branch of The New York Times on a three-year odyssey to find out what happened to two American mountain climbers who perished 50 years ago on an Argentine mountaintop. “It tickled my reporter’s curiosity,” he said.

Branch said his curiosity and persistence were amplified by a sense of journalistic responsibility “to bring humanity to stories.” What — or who — caused the deaths of Janet Johnson, a Denver teacher, and NASA engineer John Cooper in their 1973 climb of the Aconcagua, the highest peak in the Western Hemisphere, was never determined. Their frozen bodies lay on the glacier that tops the mountain for years before they were recovered. Six other climbers in their team returned safely.

Johnson and Cooper “were real people,” Branch told me in an email interview, “and their deaths certainly left generational craters in families. But they had been relegated to characters in some far-away mystery. I wanted to humanize them again.”

These two principles inspired “Ghosts on the Glacier,” a multimedia story The Times published online on Dec. 9, 2023, followed by a special print section Dec. 24.

The package inevitably brings to mind “Snow Fall: The Avalanche at Tunnel Creek.” Published in 2012,  the ground-breaking multimedia story by Branch and Times visual journalists about a tragedy in Washington’s Cascade Mountains won the Pulitzer for feature writing and a Peabody Award for interactive storytelling.

Like “Snow Fall,” Branch’s new story describes a mountaintop disaster with dramatic scenes that bring to life a vivid cast of characters in a battle with nature. His new story, however, is told from a distance of five decades compared with a tragedy that had occurred just months before.

For “Snow Fall,” Branch sat down with survivors, their voices defining the parameter of the story.  “There was no second-guessing and speculation from outsiders,” he said. “It was written chronologically, almost, beginning with the night before and ending with the bodies being dragged off the mountain.”

By contrast, “Ghosts on the Glacier”  is “all about speculation, about the confounding swirl of facts, memory and imagination.” It is about the gaps in knowledge, he said, and our desire to fill them. The passage of time itself becomes a major character.

The central character, however, is a 50-year-old Nikomat 35-millimeter camera, its unexposed film inside. The discovery, by porters prepping  for a 2020 expedition on the Aconcagua, drove Branch’s reporting and forged the story’s structure. “I lean toward chronology,” Branch said, “but wanted to open with the ‘why’ for telling this story: the discovery of a camera. Whose camera, and where? That gets us to here, where we now tell what happened.”

One thing readers don’t learn and must decide for themselves — Branch won’t say, despite my repeated inquiries — is  exactly what happened to Johnson and Cooper. Did they succumb to the perils of climbing at nearly 23,000 feet without oxygen, as the Americans who climbed with them maintained? Or, as Argentinian climbers and a doctor who attended their autopsies insist, was it foul play?

While “Ghosts on the Glacier” unspools like a mystery — Branch isn’t a fan of nut grafs — it was built on three years of meticulous reporting. That was aided by Jack Begg, part of the New York Times news research department that made it possible to track down relatives of the eight climbers on the team, all of whom are now dead. The Times team also found mountain guides, climbers, journalists and authorities in Argentina who tried to determined what happened to Johnson and Cooper.

Branch’s story is partnered with photos, videos and graphics from Times visual journalists as well as haunting photos mined from Janet Johnson’s long-lost  camera.

As for the text, Branch draws inspiration from Polish-British novelist Joseph Conrad, who famously said, “My task which I am trying to achieve is, by the power of the written word, to make you hear, to make you feel — it is, above all, to make you see.”

“My goal,” Branch said, “is to help readers visualize.”

A Times sports reporter since 2005,  Branch has made a career out of producing narratives — “a mix of nature and people, research and sleuthing” he calls them — a process  that he finds irresistible.

He was a Pulitzer finalist in 2012 for his series of stories about Derek Boogaard, a professional hockey player who overdosed on painkillers. He is the author of three books, including “Sidecountry,” a collection of his New York Times stories. When The Times disbanded its traditional sports department in 2023, turning sports stories over to The Athletic, Branch was named a domestic correspondent for the paper. His editor on “Ghosts of the Glacier” was Mike Wilson, who was acting sports editor at The Times and now is deputy editor of narrative features under the umbrella Great Reads.

Branch talked to Storyboard about keeping faith with a years-long project, what a story promises and delivers, and his favorite line from the story. The e-conversation has been edited for length and clarity, and is followed by an annotation of “Ghosts on the Glacier.”

What drew you to journalism as a career?
Now that I’m in my 50s, I like to think that it was because I saw it as a noble profession where truth is tantamount. We need more of those. And what an honor to be in a position to tell other people’s stories.

But that is somewhat revisionist. In my late 20s, it was because I was a manager at Costco, happy at work but wondering where I would be when I was, say, in my 50s. I thought being a journalist would be fun, to get permission to go places and talk to people I might not otherwise know. And I was right.

You are known for writing about sports or adventure far removed from celebrity athletes or the big professional sports. Or, as the subtitle of your story collection reads: The subtitle of your collection of Times’ stories is ‘Tales of Death and Life from the Back Roads of Sports.” What draws you to stories like these?
I like stories about outdoor adventures, because they take readers to unfamiliar or hard-to-reach places — and that’s a big part of our role as journalists, to be a tour guide for armchair travelers and to make the setting a character in the story. But I also find myself writing a lot of stories about people who are gone, whom I never met. I find that a worthy challenge, to bring people to life in a way that’s authentic and accurate.

What special challenges did reporting a 50-year-old story pose?
Most of the main sources are dead. And few of the archives (newspaper, judicial, forensic), especially in Argentina, are digitized. In a lot of ways, reporting this story felt like reporting decades ago, relying on phone calls, knocks on doors and opening lots of dusty boxes.

How did you prepare to report and write for the story?
I initially toyed with reporting this in the order that the story might ultimately be told — doing all the research, all the interviews, and then developing the film for a big reveal (or not). Selfishly, it would be fun to be that invested in the film.

But wait. What if the film was entirely blank? Would I still pursue the story? Would the Times let me? And even if the film could be salvaged, what might the images reasonably tell us? Would Johnson’s last frame show her being attacked by a yeti or something? Probably not. The film wasn’t likely to answer the big question of what happened. It was more likely to illuminate, not reveal. Ultimately, we decided to get the film developed first. Let’s see what we have. It might inform the reporting.

My first step was to find Janet Johnson’s family. I wanted permission to take possession of the camera and film and have it developed. It turns out that she had only one surviving immediate family member — a sister, a widow in her 80s, living alone in Oregon without access to a computer or a cell phone.  She never really knew what had happened to her only sibling. She gave me the go-ahead to develop the film and learn all we could.

What role did your editor/s play during the process?
Let me highlight two. Mike Wilson was my lead blocker, to use a football metaphor; he cleared a path, allowing me the room for reporting trips and research dives. For two years he listened to the scattered thoughts pouring from my over-stimulated brain and pulled me out of the weeds when necessary. And after I wrote the first draft, Mike shaped it with his gentle carving tools and sander. (Sorry, I just flipped from football to carpentry.) Then he morphed into a coach and manager. He coordinated a tight team of all-stars across several departments. He kept an eye toward editors on high who would have to give their eventual blessing. Whatever difficulties all that caused him, I never learned.

Photo editor Becky Lebowitz Hanger has been behind the scenes for a lot of my favorite story projects, and she was instantly attracted to this one. Her dedication to doing it, and doing it right, never dimmed. She helpfully reminded us along the way that Janet’s photos were the reason for telling this story.

ANNOTATION: Storyboard’s questions are in red; Branch’s answers in blue.

Ghosts on the Glacier

Story by John Branch
Videos by Emily Rhyne

December 9, 2023

HIGH ON ACONCAGUA, the Western Hemisphere’s highest mountain, the shrinking Polish Glacier spits out what it once devoured — in this case, a 50-year-old Nikomat 35-millimeter camera. I’m going to ask you about the entire lead, but have to stop here to ask some questions about some of your writing decisions from the very start? How did you come to use the verb “spits?”I don’t overthink these things, but I am allergic to boring verbs. I imagine that I came up with “devoured” first, and then wondered about its opposite. Coughs? Regurgitates? Spills? Releases? Reveals? Nah. Your question prompted me to go look at my first draft. It was “spits,” and it never changed.

Two porters, preparing for an upcoming expedition, had been securing ropes in the thin and arid air of a clear February day. It was midsummer in South America. The camera glistened in the sun, daring to be noticed. You anthropomorphize the camera, treating it as if it had human characteristics. Why?The camera was a main character for the story; we would not be writing the story without it. And it’s a bit of a stand-in until we meet the person behind it.

The lens was shattered. A dial on top showed that 24 photographs had been taken.

The bottom half of the camera was saddled into a worn leather holster with a thick strap. On the holster, in blue embossing tape, was an American name and a Colorado address. Why didn’t you say what the name and address was?I held off on revealing the name, which would mean nothing to readers, until it was revealed to the old guide, who is gobsmacked by it. As for the address, her home did not become a character in the story, so it didn’t feel vital to reveal it. Also, I’m trying to write this lede economically, and I liked the simplicity of that phrase: “An American name and a Colorado address.”

In the snow-and-ice seasonal cycles of the mountains, abandoned and lost equipment is discovered each summer — tattered tents, dropped ice axes, lost mittens. Occasionally, a body. This sounds like the beginning of a nut graf, but it isn’t, is it? You once told me, “I’ve joked that not only is ‘Snow Fall’ one of the longest single pieces to ever run in The New York Times, it’s probably the longest piece to run without a nut graf. It doesn’t have one.” Is that the case here, too, and if so, what’s your opinion on nut grafs in narratives?Ha! This is now the second-longest piece I’ve written without a nut graf. My first draft of this story had what I called a “nut section." It was written partly to wrap my head around the main parameters, but also as a summary for other Times journalists invited to help us tell this story.  In subsequent drafts, I stripped most of it away.I don’t like most nut grafs. Too many feel desperate and overwrought; they might as well be scripted in neon. I often find them insulting to curious readers. Imagine sitting by a campfire, telling a ghost story. And after the first minute, just as you have the audience rapt, you stop and say, “Here is a general summary of what I’m going to tell you. This story will be about this and that. Now let’s continue.” I get it; I don’t want to lose readers, either, who want a notion of where a story is headed. But I’d prefer to let the headline, my writing and our presentation do that. I trust readers to follow.

This was not just another camera, though the porters did not know that yet. One of them carried it down to camp. There, a veteran guide named Ulises Corvalan was cooking lunch.

Corvalan glanced up. He casually asked about the name on the bottom of the camera.

“Janet Johnson,” came the reply.

Corvalan gasped and swore. “Janet Johnson!?” he shouted. How did you reconstruct this scene? A: I interviewed Ulises a couple of times, and people who were there, like the porter who found the camera.

Excitement boiled instantly. Do you know about Janet Johnson, the schoolteacher? About John Cooper, the NASA engineer? About the deadly 1973 American expedition?

Have you heard the legend?

 It had been handed down for decades, veering toward myth, whispered like a ghost story. The voice shifts here to second-person? Why?In pragmatic terms, I was invoking Ulises’ voice. I don’t know exactly what he said after he exclaimed, “Janet Johnson!?” but he explained why this was a thrilling find. In more literary terms, I think it pulls the reader in, and signals that “you” are going to be part of this story.

Here is what was certain: A woman from Denver, maybe the most accomplished climber in the group, had last been seen alive on the glacier. A man from Texas, part of the recent Apollo missions to the moon, lay frozen nearby.

There were contradictory statements from survivors and a hasty departure. There was a judge who demanded an investigation into possible foul play. There were three years of summit-scratching searches to find and retrieve the bodies.This is a very concise summary of the aftermath. Did it require much revision?This is the boiled-down version of the nut section. If we must have one, let's be snappy and evocative, not perfunctory.

Their discovery stirred more intrigue, leaving more questions than answers. That’s the imbalance of all the best mysteries — facts that don’t quite add up, gaps that imaginations rush to fill. I love this line. A: In my mind, this is the undercurrent of the story. I was thinking about what makes mysteries mysterious. It’s the presence of enough facts to have structure but not enough to have answers. That's where I want this story to reside.

That is how Janet Johnson and John Cooper became part of the folklore of Aconcagua.

And now, nearly five decades later, an old camera had emerged from the receding glacier. It was wound, prepared to take the next picture. This strikes me as the end of the lead, which ends with a cliffhanger? What was your goal with this opening?I wanted the discovery of the camera to open the story, to signal that this is today, not yesteryear. And the camera is the reason for telling the story.  I’m not above sprinkling tiny cliffhangers throughout a piece, almost like a scene in a TV series before it cuts to commercial. This teases out the possibility to the reader: We are wound up and ready to go.

More clues emerged from the ice. Here was a decomposed left arm, still wearing a delicate silver Rado watch with a broken blue face. How were did you know this? A: We have close-up photos of the arm and the watch, which were then collected by police and given to the medical examiner. There was a tattered pack and scattered belongings: down mittens, a red jacket, a single crampon, a canister of used Kodak film.

Like that, by the whims of climate change and chance, a long-lost legend was given air and light. You sometimes alternate between reportage and poetry.During reporting, phrases will pop into my head and I'll jot them down, thinking they might be useful. I liked the notion of "air and light," bridging the literal and figurative.

THE TEAM

ACONCAGUA IS THE broad-shouldered behemoth of the Andes, shaped more like a fist than a finger. Why this analogy?I wanted readers to know that this was not an idyllic mountain of the mind's eye, not the Matterhorn or something from a beer ad. It's bulky and brown. One earlier version said something like, "It's not the type of mountain you'd have as a screensaver." I loved that sentiment, didn't love the line.

It is brown and rocky, scrubby and dusty, dry and wind-whipped. With few trees or wildflowers, it can feel like a vertical desert.

The first person known to reach the 22,838-foot summit was Matthias Zurbriggen of Switzerland, in 1897. In 1934, a Polish expedition successfully tackled a more perilous route on the northeastern side of Aconcagua, up a massive glacier that stretches nearly 2,000 vertical feet toward the summit.

The ice sheet was named for that group: El Glaciar de los Polacos.

These days, Aconcagua is part of a vast state park with helpful rangers and a helicopter rescue service. Two base camps provide hot meals, showers and internet. Some consider Aconcagua among the easiest to climb of the Seven Summits, the prestigious name given the highest mountains on each continent.

But Aconcagua is not easy. Trouble lurks in the thin air. Was this intended as foreshadow that heightens tension?It wasn't really intended to heighten anything, but a statement: This mountain is a beast, then and now.

Through 2022, there were 153 known deaths on the mountain. In 1973, Johnson and Cooper were Nos. 26 and 27.

Fifty years ago, Aconcagua had only the most rudimentary of services. Climbers had no GPS trackers, no way to communicate between base camp and the summit. The Americans carried binoculars and a flare gun.

The mountain was virtually deserted. If trouble came, there was nobody to help but the other members of the expedition.

Most of them were part of the Mazamas climbing club, founded in Oregon in 1894. Their leader was a Portland lawyer named Carmie Dafoe.

Dafoe, 52, pushed for the Aconcagua trip, noting that a Mazamas member had climbed it in the 1940s. His group, Dafoe announced, would attempt to be the fifth expedition to top Aconcagua via the Polish Route.

“Difficulties are said to be moderate — a couple of places where we’ll want handlines — not any more difficult than the normal route on Mt. McKinley,” Dafoe wrote in a 1972 memo. How did you obtain the memo?I got in touch with every family of the American climbers. Several dug through materials they hadn't looked at in years: slides, diaries and memos like this.

The guide would be Miguel Alfonso, a 38-year-old Argentine who had been to the summit five times, once up the Polish Route. Dafoe asked for a $50 deposit from anyone interested, along with a list of successful ascents and references.

In June 1972, Dafoe announced the members of the party, all American men, whom he briefly described. Jim Petroske, a psychiatrist from Portland, Ore., would be “deputy leader,” he said. Bill Eubank, a physician from Kansas City, Mo., was “highly recommended by Petroske” and would be the expedition doctor. Then came Arnold McMillen, a dairy farmer from Otis, Ore., and Bill Zeller, a police officer in Salem, Ore. (“Bill and I shared a blizzard in the Canadian Rockies in ’69, a solid citizen.”) John Shelton, 25, was a Brigham Young geology student fluent in Spanish from a two-year church mission. (“Been through Latin American customs about 25 times — which must take more energy than climbing Aconcagua.” Why isn’t there any indication where this quote came from?All these parenthetical quotes are from Dafoe's memo in June 1972. And John Cooper, a NASA engineer from Houston, was “highly recommended.”

They were weekend climbers, mostly. Dafoe organized hikes in the Northwest designed as training and get-to-know-you exercises.

“I have had some trepidation about the party because of a fear that we might have someone who has unknown problems or who is some kind of a fink,” Dafoe wrote in a memo to the group. How did you obtain Dafoe’s memo?This was one of several memos he sent to expedition members. Families had kept materials for more than 50 years, unearthed at my request. “It turned out, however, that I either know everyone in the group, or they are people I have been able to find out about. This leaves me with no reservation or qualification about the party.”

In November, Dafoe sent reminders about packing lists, passports, vaccinations.

“Everyone is probably already rounding into top physical condition,” he added. “Don’t take any chances on this. Work hard at it; especially with lots of jogging.”

He also announced the final member of the eight-person American crew: a woman from Denver named Janet Johnson. How long, from the initial idea to publication, did it take to produce the story?The short, misleading answer is nearly four years. But Covid and other priorities delayed the true start of the reporting for more than two of those years. We got the film developed in April 2022 and published in December 2023, but “20 months” is misleading, too, since none of us involved were devoted full-time to the story. I wrote a draft in the spring of 2023, and we dabbled with the multimedia and subsequent drafts off and on through the summer and fall.Did you ever lose faith in whether you would finish the story?No, I knew it would be finished, I just wasn’t sure about the form. Early on, we had multi-departmental brainstorms about the best way to tell this story.  Is it a New Yorker-style text narrative? A documentary? Something between? We let the reporting dictate the form, as usual. One issue with writing a multimedia story rooted in 1973, and then having it delayed by Covid, is coordinating a sense of urgency within the newsroom, especially at a time of so much monumental news in the world (Covid, elections, wars). Ultimately, I just wrote it, partly to get it out of my head and partly to spur decision-makers into action.

SHE WAS BORN on Nov. 30, 1936, and never knew her birth mother. She was adopted by Victor and Mae Johnson, who lived in a stone-and-timber Tudor on the south side of Minneapolis. He helped run his family’s paper-supply company; she was a bookkeeper.

The Johnsons believed in manners, rules and God. Janet, with a tidy bedroom upstairs, was a quiet girl and a voracious reader. She needed glasses early. She played organ at St. John’s Lutheran Church. How did you settle on this three-sentence portrait of Janet?Describing people is always fraught. (I loathe to think how someone would describe me. Actually, a student did not long ago, writing that I look like a dad in a garage band. I wish.) Stick to facts, not adjectives. What's an apt shorthand to give the reader a mental picture?

When she was 10, she wanted a baby sister, so the Johnsons adopted a 5-year-old girl named Judie. The new sisters met at the neighborhood park. Janet took Judie home and gave her a doll named Lois. You don’t attribute any of this biographical information? How did you report and verify it?This comes from Judie, the sister. As with any source that provides details, you have to ascertain their reliability. In this case, I can verify that they're five years apart. I've seen early pictures of them together, I went to the park near the childhood house, and Judie has the doll.

Janet never married or had children. Judie Abrahamson, now an 83-year-old widow in Oregon City, Ore., is the only close relative alive. How did you locate her?At the beginning of this, the amazing NYT research department helped with a public-records search on all the American families. In this case, Jack Begg dived into Janet Johnson. "Not easy to find survivors," he wrote to me, "but I believe this is her sister." It was.  She had no social-media footprint, no email, but Judie answered one of the phone numbers I had for her.

“She liked to study — that was her favorite thing to do,” Abrahamson said. “Straight A’s? She would settle for nothing less.”

It was when her sister was away at college that Abrahamson discovered notes hidden in a jewelry box — love notes between her sister and another young woman. Soon, Johnson’s parents sent her to a hospital in St. Paul to “cure” her of her homosexuality. She was about 21. Did Abrahamson tell her parents about the letters? A: It's unclear if her discovery came before Janet's parents learned. As with anything, if I don't feel comfortable in my knowledge of it, I don't write it.

“It didn’t heal her,” Abrahamson said.

“But that was a big rift between Janet and my mother.” How would you describe your interviewing style?It depends on the story and the situation. In this case, Judie and I spent a lot of time talking on the phone, over many months, until I visited her with a video crew for 2-3 days. My goals were simple and vague: To get to know Janet, and to understand what Judie knew about her sister’s death and how it impacted the family. I mostly just prompted her to keep talking, knowing that memories were deep and likely unsaid for many years.

It chased Johnson away from home. Who drew this conclusion? A: That was Judie. Janet had attended two colleges closer to home, came out of the hospital and went to Colorado to continue her education

She settled in Denver, renting part of a two-story home on York Street, near the botanic gardens where she volunteered. She got her teaching certificate, then a master’s degree, eventually a Ph.D. in education at the University of Colorado. She taught in elementary schools and then became a school librarian, figuring it would be easier to keep her nights and weekends free for the mountains.

Johnson joined the Colorado Mountain Club. By 30, she became the 82nd known person — and among the first 20 women — to reach the summit of each of Colorado’s “fourteeners,” the more than 50 peaks higher than 14,000 feet in elevation. Why do you say “known person” rather than use more specific attribution?I like this question because (like others above it) it gets to the confluence of attribution and pace. Stories can be gummed up with attributions, making them ponderous and pretentious. If I think it's not germane HOW I got the information, I'll leave it out. If an editor had demanded that I add "according to so-and-so, a historian of the Colorado Mountain Club," I might have decided that it's too clunky and ditched the whole line.

Her name was regularly in the club’s magazine, Trail and Timberline, detailing various excursions. Photos she took graced the magazine’s cover.

“The companionship on the outing was tremendous — that is, except for the wood ticks, which somehow managed to find their way into my abode on the top of the hill,” she wrote in a 1961 report about a weekend trip in the Rockies. “Strangely enough, few other people even found one tick. Why they picked on me, I don’t know. They say everyone was put here for a purpose, so maybe I was meant to sustain the ticks.” Of all the times Johnson appeared in the magazine, why did you choose this excerpt?This piece was a rare time she wrote about herself in first person. It showed a side of her personality that was otherwise hard to capture.

More and more, Johnson headed overseas. She was one of 38 members of a 1963 club expedition in Peru. On the way home, she detoured to climb Iztaccíhuatl, which rises more than 17,000 feet near Mexico City.

It is unclear how many of the world’s summits she reached. She climbed Kilimanjaro and hoped to climb Denali after returning from Aconcagua.

Most summers, Johnson tied a kayak atop her Nash Rambler and headed to the Northwest. How do you know that’s what she drove? A: Her sister remembers it, has black-and-white photos, and later went to Denver to sell it when Janet died. She thought it was green, but I couldn't be certain, so it remains colorless.

She would stay with her sister, hike Mount Hood and paddle in Puget Sound. Abrahamson’s children knew her as Aunt Janet, the free spirit.

In 1971, the graduation announcements for her Ph.D., sent out by her mother, included a formal portrait of Johnson, smiling with her cat-eye glasses on.

She wanted to reach the highest levels of education. She wanted to reach the top of the tallest mountains. It seems like you learned enough about Janet Johnson to write a complete profile of her. What challenges did you face selecting what to present and how to write it?It's always an imprecise trick. It's a matter of being fair and accurate in a portrayal while also keeping a reader engaged. I try to think: If someone had a few grafs to describe my life, what would (or should) they say? And how much space, in the context of the broader story, is right for the reader?I wrote about Janet before I knew what Emily Rhyne would do in a video focused on Janet. I thought all of this might get replaced. In the end, we felt like the combination of words and video blended well.

“I think it was just to prove to my mother that she can do these things, even as a gay person,” Abrahamson said.

If Johnson had a significant other, Abrahamson never knew of her. Boxes of slides that she left behind show mostly landscapes, not people.

She took the 1972-73 school year off. That fall, after a hiking trip in Europe, she proudly joined the upcoming Mazamas expedition to Aconcagua.

“Climbed all 67 of the 14,000-foot peaks in the United States (except Alaska), Kilimanjaro, Orizaba, Popocatépetl, Iztaccíhuatl, Fuji, Mt. Blanc, Matterhorn, Eiger, Peru, etc., etc.,” Dafoe wrote of Johnson. “Recommended by two of my climbing friends from Denver.”

She packed her belongings into an aluminum-frame backpack — boots, flannel shirts, a red down jacket, thick mittens, glacier glasses, a sleeping bag. She used a marker to write her name or initials on most of them. She wore a silver watch and a ring with a brown stone that she got on a trip to New Mexico. How do you know these details?We have all of the gear now (after the discovery of the camera, more of her belongings were found on the glacier) and can compare it to photos of the expedition and descriptions made in diaries by others.

And she brought the Nikomat, the consumer version of Nikon’s professional cameras of the era. She probably bought the camera during her trip to Japan a couple of years earlier. Why do you make this surmise?She had climbed in Japan and spent time in Tokyo. And that Nikon model was only available in Japan. (In the U.S., it was called Nikkormat. I think I've heard from every reader who had one, telling me I got the name of Janet's camera wrong.)

She used a label maker to punch her name and address onto blue embossing tape, and stuck it to the bottom of the leather camera case, just in case she lost it.

She carried the camera with her to Aconcagua, taking photographs along the way, almost to the top. Why did you break off the narrative here?I wanted that chapter to be focused on Janet, along with her place in the expedition. Now we know the main character and have, hopefully, propelled the story forward. Let's get to the mountain.

THE CLIMB

AMERICAN NEWSPAPERS SENT them off, and Argentine newspapers greeted them at the Nutibara Hotel in Mendoza’s city center.

Rafael Moran, a reporter for Los Andes, a daily newspaper in Mendoza, interviewed the mountaineers near the pool. He did not cover every Aconcagua expedition, but this one was especially intriguing: Americans. The Polish Glacier. A woman. A NASA scientist.

Moran quickly had a dark inkling about this group. The Americans seemed disconnected from one another and unprepared for the serious task of climbing Aconcagua. How do you know what Moran was thinking? A: We talked about it in a pair of lengthy interviews.

Moran whispered to the photographer: Take each of their photos today. I don’t think they’re all coming back. What was your reaction when you learned this?I think we all know the nagging feeling of intuition. And we have the photos taken that day. They're mug shots, essentially. I still wonder if the Americans found it weird to be photographed that way.

The next day’s newspaper previewed the planned ascent. It showed the Americans huddled around a photo of Aconcagua. The caption noted the NASA engineer at the center.

Just a month before, in December 1972, John Cooper was at mission control in Houston for the 17th and final Apollo mission, wearing a black mustache and a headset, communicating with astronauts on the moon. Cooper was a surface operations engineer, helping guide the lunar module.

Cooper also wore his new mountain-climbing boots to work, to break them in for what he anticipated would be a difficult expedition on Aconcagua. How did you learn all this about him?I did a deep dive into Cooper, mostly through family. A niece of Cooper's collects NASA historical documents, like directories and memos, that provided timelines and job titles.

Cooper grew up in El Dorado, Kan., with a love of the outdoors. He went to Oklahoma University to get a degree in geological engineering, but the flat oil fields where his father worked were not for him. He spent college summers working for the Forest Service and then as a smokejumper in the American West.

Later, in the U.S. Coast Guard, he became a pilot and won awards for rescues off the coast of Florida and in the Caribbean. He learned to deep-sea dive.

And he climbed. Why this three word-sentence?Its form tells you that it's a transition of sorts. The words say, "we're done with the background, let's get back to climbing." Cooper summited Kilimanjaro and Mount Kenya, the tallest mountains in Africa, and Popocatépetl, the volcanic behemoth in Mexico.

In 1966, Cooper joined NASA just as the Apollo program got underway. He had a bit of swashbuckler in him, more like an astronaut than a desk engineer. He sometimes wore a beard. He smoked a pipe. Around NASA’s Houston campus, Cooper drove an old military Jeep, sometimes taking his nieces for a ride.

“My mother would say, ‘John, you put the doors back on and put up that windshield before you take my girls with you,’” Deb Koons, Cooper’s niece, said.

It was at NASA that Cooper fell in love with a secretary, a young divorced woman named Sandy Myers. They were married in 1968. In 1969, they had a baby boy they called Randy.

That was the year of Apollo 11. Cooper was in the surface operations group that guided Neil Armstrong and Buzz Aldrin as they became the first humans to walk on the moon.

Three years later, on Dec. 19, 1972, Apollo 17’s three-man crew splashed down safely in the South Pacific.

On. Jan. 12, 1973, Cooper’s flight from Houston landed in Miami, where he met Janet Johnson. They flew together to Argentina. The depth and breadth of the reporting and research is so impressive. How long did that part of the process take?In fits and starts, it began shortly after the camera film was discovered in 2020 and continued until we published in 2023. Because it's an unsolved mystery, it caused plenty of sleepless nights and endless hours chasing details and theories. Remember Jake Gyllenhaal's character in “Zodiac?” What is your writing process?For a long time, as a sports reporter often on tight deadline, I didn't really have one. Find a lede, write quickly and file — sometimes before reading back through what I've written. Now, with a story like this, I'll open a Google Doc or take a notepad and hammer out a top-of-mind outline — the loose layout of stepping stones winding toward the end of a path. Once I start to write, I need three things to get momentum: a lede, a loose nut graf (if only to keep me focused), and then the lines after the nut graf that pick up the story again. The rest I often make up as a go, thinking two or three moves ahead. At some point, unexpectedly, the kicker usually hits me. The bigger the material, the more I try to write first drafts without my notes in front of me, letting my mind filter what's important.

Cooper kept a diary of his expedition. Like other men in the group who wrote in their own diaries about Johnson — “Nothing feminine about her,” one said — Cooper was unsure what to make of the only woman.

“Janet sure is weird,” he wrote from the comfort of Hotel Nutibara. “She went swimming in her bra, blouse and panties today and the pool was full of people!” How do you know what Cooper and the other men wrote in their diaries?Several families dug out old mementos, including photos and diaries.Why did you only quote Cooper’s observation?He was a main character. The other quote here, unattributed, came from a member who barely gets mentioned, and I was hesitant to bog down readers with names and asides.

ON THE MOUNTAIN, the Americans struggled from the start. From the very start, tension infuses this story: What’s on the camera, what happened on the climb, what happened to Janet Johnson and John Cooper. Were they murdered by someone? Tell me what your goal is here?The pacing and order were definitely deliberate, but I wasn't trying to sensationalize anything. That's not my style, and this story is sensational enough. And once we get on the mountain, it's all facts and first-person accounts, chronologically.

On Jan. 20, 1973, aided by mules, the group hiked 25 miles to Casa de Piedra, a stone house at the confluence of the Vacas and Relinchos Rivers. How do you know the precise date and progress they made? A: Diaries, news accounts and Carmie Dafoe's expedition report.

In his diary, Cooper described the “stark beauty” of a landscape “baked hard as concrete.” He mentioned that Eubank, the expedition doctor, was already sick.

The next day, the group reached base camp, a treeless, rubbled plot in a wide valley at about 13,500 feet. These days, during climbing season, it is a buzzing village. In 1973, the American expedition members were the only ones there.

Alfonso had hired Roberto Bustos, a 25-year-old climber and student, to manage the base camp. Now a retired geography professor in Buenos Aires, Bustos recalled his early impression of the group — a lot of high-quality gear, but an unsettling dynamic. Did you have to employ interpreters during your reporting in Argentina?Yes. Nicolás Garcia, a journalist in Mendoza, was our translator and fixer.

“There was no group attitude,” Bustos said. “I was thinking, Oh, I am on my own. Everyone has to take care of himself. In my opinion, they weren’t ready for such a strange and big mountain as Aconcagua.”

Alfonso, despite his experience on Aconcagua, was relegated to merely a guide, someone to point the way.You collected so much information. What's your process for selecting the essential information?I often say that the hardest part of writing is not what to include, but what to leave out. I'm not a great self-editor or revisionist. Maybe from my deadline-writing days, I trust first drafts and tend to want to leave things in. More and more, I edit a story by imagining a sentence or graf or section deleted. Does it still work? Is it better? But as in a recipe, not all story parts are equal. Is it a key ingredient or a flavor enhancer? A lot of arguments with editors involve keeping in a flavor enhancer that I really like. The stew always needs a little color.

Dafoe was in charge. Petroske, his friend from Portland, was the deputy leader, followed by Eubank, the doctor, and Shelton, the interpreter for Alfonso. Then came Zeller, McMillen, Cooper and Johnson, without defined roles.

Holding a map of Aconcagua. Dafoe called it “one of the most barren places in the universe.”

Back then, like today, getting to the summit usually required a week or more of shuttling up and down the mountain, moving gear and adjusting to the altitude. The group carried loads to Camp 1 — at 15,500 feet, higher than anywhere in the continental United States. They returned at day’s end to base camp.

The high-altitude ups and downs were made more difficult by Aconcagua’s notorious obstacle course of penitentes — icy pillars, as tall as six feet, caused by solar radiation. They’re sturdy enough that even the small ones can’t be knocked over. The group called them “white monsters.” How do you know what they called them?From Dafoe's expedition report.

The trek to Camp 2, at nearly 18,000 feet, took seven hours.

“Brother was it bad,” Cooper wrote in his diary. “Between the ice and scree and altitude, I was done in.”

He later wrote about others in the group.

“Bill Zeller is the real man behind the work,” he said of the Oregon State Police officer, an expert in fingerprinting. “He lugged 80 pounds up to Camp 1. Then after getting back he made the water haul — and I’m here in the sack. I guess all do our share of work but some more than others.”

Johnson was little help, Cooper wrote. “She is a real loner and appears to be for only one thing — to get herself to the summit, at the expense of everyone or on everyone’s back.” Why did you include this diary excerpt?Her desperate "summit fever" was an idea floated by survivors, and it comes off as a bit of victim shaming. But this was testimony, in real time, that Johnson was eager to get to the summit. And it shows that Johnson's situation was nuanced — a victim, perhaps, but one whose motivation was already questioned, at least privately.

The expedition was fracturing from the effects of altitude. Three Americans, including Dafoe, the leader, stayed at Camp 1. Five others, including Johnson and Cooper, moved up to Camp 2 with Alfonso. Cooper felt miserable.

“For 2 cents I’ll go back,” Cooper wrote.

But they plodded upward to establish Camp 3, behind an outcropping of boulders at the base of the Polish Glacier, at about 19,400 feet.

A storm swept through, pinning the group in place for a welcome day of rest. Behind it were clear skies, a perfect window for a climb to the summit.

The group was “expecting it to take at least all day,” Zeller later wrote in his account of events, “but the lower part of the glacier seemed to present no problem as it seemed to be in good condition — no crevasses — not too steep — good snow for crampons etc.”

But after a late breakfast, Petroske suddenly lost his coordination and struggled to put on his crampons. Others diagnosed it as a sign of high-altitude cerebral edema, a potentially deadly swelling of the brain.

Alfonso escorted Petroske back to base camp. Now the American team was cleaved in half. Intriguing verb in this context: cleaved."Cut" doesn't quite cut it, does it? Cleave felt heavier, leaving a thicker separation.

Gone were the expedition leader, the deputy, the doctor, the interpreter and the local guide. Remaining were Cooper, Johnson, Zeller and McMillen. None had been this high, anywhere. They barely knew one another.

When they looked up, they saw the Polish Glacier, stretching to the sky.

It was sunny. Their jackets were unzipped. They wore crampons and carried ice axes and light packs, leaving most of their possessions behind at camp.

But movement up the glacier was slow. By nightfall, the four Americans gave up on reaching the summit that day. They were at roughly 21,000 feet.

They dug a small snow cave on the glacier with their ice axes. They had no sleeping bags, so the climbers lay on reflective space blankets. During the night, cramped and uncomfortable, Johnson and Zeller moved outside. They sat, shivering. How did you get enough detail to reconstruct this part of the climb?It came from separate reports from McMillen and Zeller, to reporters and in private written testimonials they each provided to others in the expedition.

Wind blew a fine powder off the summit, filling the opening of the cave with snow and burying Cooper’s legs. Johnson unburied him about an hour before sunrise.

But Cooper was finished. Cold and tired, he announced that he was turning back, Zeller and McMillen later said. It was about two hours down the glacier back to Camp 3, McMillen figured. He and Zeller expressed little concern about letting Cooper go alone.

“He appeared to be very capable, alert,” Zeller later told his local newspaper. How did you find this story?A big part of research was through archival newspaper accounts. In the U.S., that was largely (not entirely) through newspapers.com. The Argentine accounts were not digitized. Nicolás and I spent days in the old Mendoza library and Los Andes archives searching hard copies.

“He had no trouble with his reasoning. There was no concern about his climbing ability, and we were not too far above high camp.”

John Cooper never made it. He died on the glacier.

Not long after, so did Janet Johnson. How much writing and revision did this section take?I mostly whittled down the expedition's movements below the Polish Glacier, but a lot of this part of the story shape-shifted as I moved the breaks around. The original draft had about 15 sections, mostly to provide a lot of pieces that might be shuffled as we considered multimedia.  In this section, with Mike Wilson's help, I streamlined it and decided to separate what we knew and could verify (the section above) to what was later said (the next section called "Rumors") about all the speculation.How long was your first draft?The original draft was 19,000 words, about twice the final. It was a kitchen-sink draft, intended partly to show colleagues the breadth of the material and the angles or themes we could play with in multimedia. With Mike Wilson's great help, it was carved under 15,000 words, then 13,000, then shaped to under 10,000, where we both felt comfortable.

THE RUMORS

EXACTLY WHAT HAPPENED is speculation, spun around the globe for 50 years. Another intriguing verb choice with “spun.”I'm a defender of occasional alliteration (it gets a bad rap), and I like "spun" and "globe" together. In my brain, it evokes an old-fashioned newsreel report.

Two men from Oregon — Zeller, a police officer, and McMillen, a dairy farmer — were the last to see Cooper and Johnson alive.

They gave detailed versions of events. Slight contradictions and the confounding effect of high-altitude hallucinations raised questions for Argentine authorities and teased the public’s imagination.

After Cooper headed downhill alone, Zeller, McMillen and Johnson continued up. They moved slowly. They took photographs. They reached the top of the Polish Glacier, where it meets a ridge that leads to the summit.

But darkness descended again, and snow on the ridge was waist deep. The men took turns breaking the trail, 25 steps at a time. The summit in sight, the men later said, they turned to find that Johnson was not there.

“We looked and looked and called her name and got no answer,” McMillen recalled in a written account, two weeks later. “Finally I stumbled across her ax and figured she couldn’t be too far. We called some more and finally a faint little voice said, ‘My name’s Janet Johnson.’ She was about 100 feet off our trail in the snow, laying there. When we got to her she said, ‘Don’t make me suffer, just let me lay here and die.’”

Zeller said that he roped himself to Johnson; McMillen said that Zeller “took her by the arm.” Zeller said that the three got lost and camped another night together; McMillen said that he went ahead of the other two and spent the night by himself.

Their stories converged again the next morning. Johnson wouldn’t stand, and her hands were “swollen and black,” McMillen wrote, so they “anchored her from three different directions so we could hold her standing up” and led her past a crevasse.

They reached the snow cave where they had last seen Cooper. Some of their equipment was there, including the flare gun. McMillen said he shot it. It was 7 a.m.

“It made a noise as loud as a rifle but I guess no one heard it below,” McMillen wrote.

Johnson’s condition seemed improved, so the men decided that McMillen should go down alone to get help, following the route that Cooper presumably took 24 hours earlier.

McMillen said he lost his ice ax in a steep section of the glacier and slid 1,000 feet, head first. It would account for the black eye he had later, he said.

Then he saw members of the Argentine army coming to rescue Zeller and Johnson. He heard people calling his name. He saw dead mules. And he saw a dead soldier lying in the snow.

Only later, after reaching camp and sleeping, did it occur to him: None of that was real. The dead soldier, he learned, was John Cooper.

Up on the glacier, Zeller, too, was having hallucinations, not uncommon in the thin air of high altitudes. He later recalled visions of construction trucks working near the summit, and hearing phantom voices of rescuers who were never there. Do the hallucinations inject doubt into their accounts?I think so. I wondered about how explicit to be: “Dear reader, does this make you wonder about all of this?” I mostly stayed back, letting that possibility drift over the reader. Do some readers notice the fog that's building? Do some not? I think it's a big part of the story. On one hand, they were hallucinating. On the other, they quote Janet directly. Interesting, right?

“Janet and I continued coming down till we were through the worst part then we took a long spill also,” Zeller wrote in an account later that spring. How did you obtain this account? Do you know why he didn't write it until later that spring?McMillen and Zeller each provided a multi-page account of what happened, given to other expedition members (but not the Johnson or Cooper families, as far as I can tell). McMillen's was longer (narrated to his wife) and done shortly after they returned. Zeller's was in April and shorter. It was likely at the behest of Carmie Dafoe, the expedition leader who was writing a report (and happened to be a lawyer). “Again doing no serious damage but breaking both of our dark glasses & cutting our faces some. We ended up 3 or 4 city blocks from camp & could see the tents.”

He and Johnson came untethered in the fall, Zeller said, so he went back up to check on her. That is when he saw Cooper.

“I saw John’s body about halfway between us & off to the right as we faced up hill,” Zeller wrote. “I checked him & he was dead & appeared to be frozen — I didn’t see any cuts on his exposed skin & no tears in the clothing so assume that he didn’t die as a result of a fall but exhaustion & hypothermia etc.”

“Janet seemed to be OK best as I could tell so we decided that I would go ahead & set up the tent & she would follow as soon as she got her wind,” Zeller said.

He arrived at Camp 3 a couple of hours after McMillen, the men later said. They slept through the night, awoke and saw no sign of Johnson.

“The next a.m. Bill and I decided to go on down,” McMillen wrote. “Bill was so confused he didn’t know which direction to go.”

He concluded: “That’s the story as near as I can recall.” There’s little sense of camaraderie here, the kind that would produce a “we’re all in this together, we’ve got each other’s back” mindset. Was implying that one of your aims in this section?There is, though it's unclear how much communication the men had in the aftermath, including in writing their reports. I like the admitted uncertainty and the folksy nature of that line: "...as near as I can recall." It's like a question mark at the end of everything you just read.

Questions followed them downhill, like a dry, cold wind. What was the process behind the composition  of this ominous simile?A version of it was in the first draft: "Questions, like a dry, cold wind, followed them." I don't know how or why certain phrases pop out of my fingers. I think about what it was like for them. They're coming down the mountain, alone, and no one in the world knows anything is wrong until they show up without two other people. Word spreads, speculation swirls, as they re-enter the real world.

JOHN SHELTON, the college student who served as interpreter on the climb, turned 76 this year. Why did you shift the focus of the narrative here to Shelton?Shelton was the only American expedition member still alive. His account is vital, if hindered by the passage of time and his fading memory and health (he died in November). He had been receiving hospice care in a V.A. hospital bed in Utah for more than a year. He had a Kringle-like white beard and eyes that sparkled when he laughed. How did you track him down?Those two vital tools of reporting: A lot of dialing and sheer luck. It took a long time — I engaged with a lot of Sheltons in the West that I thought were related (some thought they might be, too) but ultimately weren't. One of our researchers led me to a woman who lived on the East Coast but once had a shared address with a John Shelton. She answered the phone. She was an old friend who led me to John's daughter.

He was the last American from the expedition still alive.

Shelton remembered getting sick from the altitude and being the first in the group to return to base camp. He kept company with Bustos, bonding over their shared affinity for science. Both were 25, the youngest in the group.

A day later came Eubank and Dafoe, more sick than Shelton. After another day came Petroske, with the help of Alfonso, the guide.

Shelton described looking through binoculars at the Polish Glacier, expecting to see the remaining four climbers and spotting only three — and, later, only two. He remembered rushing uphill with Alfonso to see whether they could help.

They came upon Zeller and McMillen walking toward them. Shelton remembered the weight of the moment: Four people had gone up the glacier, but only two had returned.

It did not occur to Shelton that Cooper and Johnson were anything more than victims of a high-altitude tragedy. Foul play? “Hogwash,” he said, 50 years later.

Word traveled slowly off the mountain. Families were called. Wire services and hometown newspapers wrote hasty dispatches, filling gaps with presumptions and wild falsehoods. Another trio of staccato sentences that condense time. Why did you write them this way?It's counter-intuitive, maybe, but I find that big thoughts sometimes work best with short sentences. Of course these things happened, but they didn't need expansive explanations. Say them and give the reader a change of pace, which is always welcome.

In Cooper’s hometown in Kansas, the paper reported that he “was presumed dead after a fall from the top of the mountain into a deep crevice during a blinding snowstorm.”

The U.S. Embassy in Buenos Aires sent a memo to the U.S. secretary of state’s office, trying to stanch misinformation.

“Deaths did not occur as a result of a fall as reported by United Press International and Associated Press or as the result of an avalanche as reported by Reuters,” the embassy said.

News outlets in Mendoza were on the story more exhaustively and accurately. The first news was reported in Los Andes on Feb. 4: “Fears for the Lives of Two North American Climbers,” the headline read. There was a map of the route. Featured prominently were two smiling photos of Johnson and Cooper, taken at Hotel Nutibara two weeks earlier.

In the video of the interview with Shelton, you can be heard saying, “There are a lot of smart people in Mendoza who think this maybe wasn’t an accident.” Where do you come down on the question?It's a fact, really, and I wanted his reaction. If you're asking my opinion —was it an accident? — it's not my place to give it. But I wouldn't have written the story, and written it this way, if I thought the answer was clear. That would be disrespectful to everyone, including readers and the memories of those involved.

In the same video, Aconcagua guide Ulises Corvalan  says that “climbers of the moment say,” that Cooper and Johnson were romantically involved, “and somebody in the same team killed both of them.” Yet the story has already established that Johnson was gay and Cooper and Johnson barely knew each other, facts that Corvalan didn’t seem to be aware of. How credible a source do you consider him?Corvalan is an expert on the mountain, but in this instance, he stands as a proxy for the stories that have been handed down. Of course he doesn't know what happened. But he is a link in the chain of telephone tag, where you whisper something to someone, who whispers it to someone else, and see how garbled it comes out at the end. Are there truths still left in the garbled messages handed down for decades?

“The expedition was starting to fall apart before even the work on the ice began,” the next day’s story said, just as Americans were getting false accounts of avalanches and blinding snowstorms.

At the base of Aconcagua, Alfonso and the American survivors were held for questioning. In Mendoza, a judge was assigned to the case. So was a police investigator. Officials labeled the case “averiguación de homicidio culposo” — investigation of manslaughter.

Even the American government validated the suspicion. It was standard procedure for the case to remain open, the embassy wrote in its files, to “ensure that foul play may be ruled out.”

The seeds of speculation were planted.

“It needs a deeper investigation,” Los Andes wrote. In a structure like this, how do you determine where to insert key points of tension? A: The tension at key moments is organic. I certainly decide where to break a section and where to begin another, but this story involves so many questions that it was easy to find those tension points.

THE SECRET MEETING

THE AMERICANS RETURNED to Hotel Nutibara, avoiding reporters staked out in the lobby. Bustos, the base camp manager, came to say goodbye to his new American friends. They would not see him. Fifty years on, it still saddens him.

The U.S. State Department didn’t have much luck, either. Consul Wilbur W. Hitchcock tried to speak to the Americans during an overnight layover in Buenos Aires.

“All five looked tired and somewhat dazed,” Hitchcock wrote in a report. (The sixth survivor, Eubank, had already left the country.)

Dafoe cautioned Hitchcock about the effects of high altitude on mind and memory. He said that the others had experienced hallucinations and perhaps a “feeling of unreality” coming at such heights.

Hitchcock returned to the airport the next morning. He spent another 30 minutes trying to question the Americans before they boarded a plane to leave Argentina.

“They were unable to reconstruct the climb with sufficient accuracy,” Hitchcock wrote.

Newspapers published a photograph from the tarmac. Shelton and Petroske smiled as McMillen appeared to say something over his shoulder. They carried packs and ice axes. A reporter asked Zeller to clarify the events on the mountain, papers reported, but Dafoe, a lawyer, stepped between them and would not let him answer.

It all added to the intrigue in Argentina. But if any of the smoldering speculation followed the survivors back to the United States, it was quickly snuffed out.

In Portland, the president of Mazamas wrote a secret memo. He called for a special closed meeting of club leadership and the survivors of the expedition, to be held two days later.

“NONE BUT THE ABOVE-NAMED WILL BE ALLOWED TO ATTEND. The location is to be kept SECRET … repeat … SECRET!”

The memo said that the idea was to “learn the ‘straight of things’ from people involved.”

“Presumably,” it continued, “a resultant will be the dispelling of certain suspicions, uncertainties, rumors, whatever, which may have come to your attention and have been amplified by the confused communications during the expedition and by conflicting or incomplete newspaper reports.”

The meeting was held at Dafoe’s law office. Two days later, on Feb. 15, Dafoe’s secretary typed a three-page “chronological summary of events.” How did you obtain the memo?One of the families has it.

It was the story that survivors told their hometown newspapers. And it was the basis for Dafoe’s formal expedition report published in the Mazamas annual in 1973, which concluded that the deaths were an accident, that Johnson and Cooper were desperate to reach the summit, and that “they probably died of pulmonary edema.”

They did not. This statement is flat and definitive. Is it meant as a cliffhanger? A: Funny, a friend later said that this was their favorite line. It might be the last line I added, shortly before it published. I had read the final draft aloud to myself (something I try to do, to "hear" it) and my brain quickly answered Dafoe's presumption. So I typed it in and liked it.

THE JOHNSONS and the Coopers were religious Midwestern families. They trusted higher powers and government officials. They grieved but did not wallow, at least publicly.

It is unclear how much they interacted, if at all.

The Coopers held a memorial service in March, but desperately wanted John’s body recovered for a proper burial in Kansas.

Cooper’s father, also named John, wrote letters — to Los Andes, to Alfonso, to the State Department — in search of help. He learned Spanish so that he could read the news reports coming from Argentina. How did you learn this?I learned this from Cooper's family. I also have a lot of documents with the father's scribbles, translating Spanish into English.

Janet Johnson’s widowed mother, Mae Johnson, held a funeral service in April, at the Minneapolis church where her daughter played the organ as a teenager.

She did not ask for her body back. She understood that her daughter had said that if anything happened to her on Aconcagua, she wanted to be buried in the little cemetery not far from the trailhead. How do you know this?It's not unusual for mountaineers to discuss their wishes should something go wrong. Janet's family and and those on the expedition said that Janet made her wishes clear.

Like John Cooper’s father, Mae Johnson collected newspaper clippings and documents. In places where her daughter’s name was spelled “Jeannette” by Spanish-language papers, and even by some American ones, she crossed it out and carefully wrote “Janet.”

And in places that quoted her daughter as saying, “Let me die here,” her mother blacked out the words so she never had to read them. How did you come by these newspapers with the words blacked out and draw the conclusion that Johnson’s mother did that?We published one American example, but Janet's family has many Spanish-language examples, too. Janet's sister said that their mother was really bothered by the misspelling (I would guess that "Janet" became "Jeannette" because of how it might sound, phonetically, in Spanish in Argentina) and haunted by any suggestion that her daughter would ever want to die.

IN ARGENTINA, JUDGE Victorio Miguel Calandria Agüero wanted to know: How did John Cooper and Janet Johnson die? There could be no sure answers without the bodies.

In late 1973, on the crest of a new summer climbing season in the Andes, a four-man team was assembled to look for them. Alfonso, wounded by criticism of his role as guide, would lead it.

A National Geographic reporter and photographer named Loren McIntyre heard about this and showed up to join the team. Alfonso was glad to have him.

They carried two plastic toboggans, the kind children use to sled down icy slopes, that they had reinforced with sheet metal screwed to the  bottom. Such a precise detail. How did you learn about it?I interviewed lots of people who did not appear directly in the story, including people who helped recover the bodies. They described this. And McIntyre wrote about it and took photos.

A week later, at the foot of the Polish Glacier, they found the ghostly evidence of the American expedition — tattered tents, a torn blue sleeping bag leaking feathers.

About 150 yards uphill from camp they found Cooper’s frozen body.

He was stretched out on relatively flat terrain, his legs extended and crossed. His hands were bare, across his abdomen. His jacket was on but his hood had fallen behind his head.

“John Cooper was a tall and large man and he was frozen stiff,” McIntyre reported to investigators. “He was like a statue of ice and the toboggan was about half the length of his body so arranging him so that his clothes and body would not be damaged in the descent was not an easy thing and it was cold and windy and tempers were running short as we tried to get him lashed to the sled.” Was this a written statement? How did you locate it?Immediately after the recovery, McIntyre provided a long statement and description of events for some sort of insurance purposes. He gave the Coopers a copy.

A storm blew in. The men left Cooper for the night, driving stakes around him to keep him in place, and descended to the safety of camp.

The next day, McIntyre was first to the body and did a closer inspection. He took detailed photographs of Cooper and his belongings “to make it supremely evident how he was equipped” in case there were questions from investigators or reporters.

He found Cooper’s diary. Where is the diary now?McIntyre took photographs of every page, but (at Alfonso's behest) he left it and other belongings (including a camera) for investigators. He later told the family he regretted this. We cannot find those items now. He found an opened letter from Cooper’s wife, Sandy. McIntyre read it aloud and translated it for the others. How do you know this? A: McIntyre's account. He corresponded for a few years with Cooper's parents and widow.

“Keep roped up, and don’t forget the crampons, as I don’t know how I’d replace you,” she wrote. “You are by far the best husband and loving one, and really good Dad, in the entire world.” How did you decide what words to select from the letter?I thought it was important to remind people that this was a real person, someone who was loved. And it's eerie that she mentioned the crampons and staying "roped up," since he was found without a crampon or ropes.

There was no sign of Johnson. McIntyre combed the snowfield for several hours before giving up, he said. He considered her death the bigger mystery and thought she might have wandered off the glacier’s sheer edge.

Details about Cooper got out fast. He was missing a crampon. There was no ice ax. He was on a gentle slope. His battered face held a look of frozen terror. And his abdomen had a cylindrical hole, bloody and deep. It had gone undetected until his body thawed at a lower elevation and his frozen hands could be moved. These are such gruesome details. What’s the source?A mix of sources, including McIntyre and the newspaper accounts and leaked photos published at the time. The Argentine media published a lot of gruesome photos of both Cooper and Johnson when their bodies were found.

“The highest percentage of possibility is that Cooper’s death was an accident,” Alfonso told reporters. But if Cooper had fallen on his own ice ax, it must have been very violent, he said, given the five layers of clothes he wore and the depth of the wound.

Alfonso also said that Zeller told him that he had found Cooper sitting, dead, with his head between his hands.

“But the way Cooper was found reveals that Zeller’s account was not exact,” Los Andes wrote.

McIntyre insisted that “there is no mystery at all.”

“He fell on his ice ax and he injured himself,” he said in a statement to investigators. “He was in so much discomfort and pain when he was nearly to base camp that when he finally got off the steep part of the glacier, got down on the flat, he had evidently stopped, sat down and removed his gloves and was probably trying to examine himself and his wound when he fell unconscious and froze to death.”

McIntyre left a sliver of doubt. In a 1974 letter to Sandy Cooper, he suggested that McMillen and Zeller “have probably formed some conclusions in their own minds which may be true or which may be an adjustment with conscience they can live with.” He continued, “I wonder whether you have ever talked with them?”

It is unclear if the Cooper or Johnson families ever did.

Cooper’s body, per the family’s wishes, was transported to Kansas. It arrived in a metal coffin, shipped inside a plain wooden crate.

The coffin was buried in the cold December ground in El Dorado. The empty crate sat for decades in the garage of Cooper’s parents, who couldn’t part with it.

The results of the full autopsy were sealed by the judge. But he released the cover page, which noted the cause of death. Would “investigative narrative” be an appropriate description for “Ghosts on the Glacier?”I suppose. I’m in a cozy position where I don’t have to categorize my stories. This was “investigative” — aren’t all stories, in their way? “Narrative” is a journalism word that I don’t quite understand.

It was not exposure, not pulmonary edema, not even the mysterious wound to the abdomen that plunged through five layers of clothes.

Causa de la defunción: Contusión cráneo encefálica.

Cause of death: Cranial contusions. Injuries to the skull and brain. The plot twists. Could you describe the process of writing and revision that enabled you to wait until now to make this dramatic reveal?It's less an intended plot twist than a chronological piece of information. This is how people experienced the news at the time, and it seemed right that readers should have some semblance of that.

The judge made only one declaration: We need Janet Johnson’s body.

What method do you use to organize all of the information you reported?My Google Drive is a mess of files, from photos to research, interviews, travel plans and various drafts. Typically, I have a Google Doc for research for each story (I call it "notes"), and I'll start throwing interview transcripts into it. When I write, I open a new Doc, named "draft," and sometimes there might be a "draft2," "draft3" and so on, until I get to "final," which I then put into the NYT system. One organizational challenge here, beyond the volume of material, was all the visual elements — hundreds of photos, memos, letters, newspaper clippings. At some point, my head spinning, I spent hours changing all the names of them to start with the date, so that they fell in chronological order and were easier to find.

FINDING JANET JOHNSON

ALBERTO COLOMBERO WAS 17 when he and two others found Johnson’s body. He keeps the photos from that day in a small box.

It was Feb. 9, 1975. Colombero was climbing Aconcagua with his father, Ernesto, and Guillermo Vieiro, both experienced Aconcagua climbers, now both deceased. A storm forced them to abort a summit attempt. The three decided to come down the Polish Glacier. They knew the story well. They knew Johnson’s body might be somewhere. How did you locate Colombero?The magic of Nicolás Garcia and teamwork. He and I scoured newspaper accounts, memos, even the margins of Cooper's diary, and recorded every name with a connection to this story. In this case, we knew that Guillermo Vieiro was widely credited for finding Janet's body; he was a well-known figure around Aconcagua and spoke to the media about it. We knew he was with a friend named Ernesto Colombero, who was less known. Both are dead. But somewhere we learned that there was a third person, rarely mentioned, on that trip — Colombero's son. Nicolás found him near Buenos Aires.

Colombero saw something reddish to his left. It was obscured by knee-high penitentes, the ice pillars characteristic of Aconcagua, and partially covered in fresh snow.

The men thought it was a tarp, a tent, maybe a backpack.

They found Johnson face up. Her face, blackened from two years of exposure, was battered in three places. White bone stuck out of her nose, her forehead and her chin, where skin hung down like a flap. There were blood stains on her face and jacket.

A crampon was missing from one foot. Ropes were tangled around her. Her hands were bare, her light jacket unzipped. They could not find her ice ax.

The slope was shallow. Didn’t Zeller say that he and Johnson had a long fall together? There was no way this was where they fell, they thought.

Colombero’s memory holds one other striking detail: a rock sitting on top of Johnson. Her body was in a field of ice. Your prose has a relentless quality that rivets the reader. How do you achieve that?I don't know that I'm conscious of my style, but I am conscious of how it sounds in my head and ear. I would love to think of a reader nodding at a certain line, or making a face of surprise, or even laughing, the way certain music and lyrics do that to me. A lot of newspaper writing bores me. I don't want to bore myself.

Colombero said that he was too young and inexperienced at the time to draw conclusions. But the older men, for the rest of their lives, were sure Johnson was murdered, Colombero said.

“They thought everything was planned,” he added. “That it wasn’t an accident, that someone had hit her and tried to make it look like she rolled down the hill in exhaustion.” This strikes me as a particularly damning conclusion on their part. What do you think? A: Nice try! In the minds of those who found Janet, it was a crime scene. That's all I'm trying to convey, without judgment.

Their discovery and version of events were soon highlighted in the Mendoza newspapers, along with gruesome photos they took. Johnson’s body was just 20 meters from where Cooper’s body had been found, the reports said.

The three men were unprepared to bring Johnson’s body down. So they dug it out and shifted it so that a future recovery expedition would see it.

They found a ring with a cloudy brown stone on Johnson’s finger. They removed it and passed it to an American climber named Allen Steck, who happened to be on the mountain at the same time. In April 1975, he sent it to Abrahamson, Johnson’s sister.

“I am enclosing the ring that Janet was wearing when we examined her,” he wrote. “We did not find anything of her equipment or her camera (assuming she had one).”

The ring is the only possession from the trip that Johnson’s family received for 50 years.

IN FEBRUARY 1976, William Montalbano, the Latin America correspondent for The Miami Herald, wrote two articles about the deadly mysteries of Aconcagua.

The second focused on plans for the recovery of Johnson’s body.

“How Did Janet Johnson Really Die?” the headline read.

“There is sufficient mystery and enough unanswered questions surrounding the death of Janet Johnson and NASA engineer John Cooper on the same 1973 expedition to have raised the suspicion of foul play,” Montalbano wrote.

The article focused on Ramon Arrieta Cortez, the lead investigator, who “must establish if Aconcagua killed Janet Johnson or if she was murdered,” Montalbano wrote.

Soon after, a team of men, mostly police officers in Mendoza with climbing experience, found Johnson’s body. Her face was darker, far more mummified than a year earlier, because of recent exposure to sun and wind. They found no other belongings.

The men struggled to extricate Johnson from the ice. They crudely sliced her left arm at the shoulder and left it, with a broken watch still on her wrist.

“We had to dig the ice to get her unfrozen from the glacier,” said Rudy Parra, one of the men, now a retired police officer. “It was like taking a piece of the glacier off the mountain.”

The room where the Cooper and Johnson autopsies were performed in Mendoza is still in use today. It sits in a worn, one-story stucco building that looks like a barrack. It is equipped with stainless steel tables, electric tools that hang from the ceiling and concrete floors that slope to drains. Why do you include this description?This is an example of something we discussed earlier. It would be easy and economical to lose this graf. The reader would never notice. But I was there. I saw this room. I feel a deep need to bring the reader along to places I've been, to help them see it. It can be a fine line between details and narcissism. So I made the line as thin (and the graf as quick) as possible.

Daniel Araujo was a medical student and an assistant to the medical examiner, Dr. Carlos DeCicco, on the Cooper and Johnson autopsies. Today he is a neurosurgeon in Mendoza.

He still remembers Cooper because of the skull fracture and, especially, the tubular hole to his abdomen. It was like a bullet hole, perfectly round. The wound was so deep that it reached all the way to Cooper’s spine. Araujo always suspected an ice screw.

Johnson’s autopsy stands out because of the damage to her face — bone exposed in three places. Araujo recalled deep cuts on her boot that made him think someone had taken some hard whacks at her.

Johnson’s autopsy report, with photos, was submitted to the judge. Like Cooper, she officially died of contusión cráneo encefálica. A brain injury.

Araujo has been haunted by the memory of those autopsies for most of his life.

“They were killed,” he said. “Both of them. These kinds of injuries were not self-inflicted.” He states this so-matter of factly in a video interview. Are you convinced?Nice try again! His views are his, and we've explained his expertise in the matter.

Was that the consensus of the examiners in the room?

“Yes,” he said. “No doubt about it.” Did Araujo’s conclusion that the deaths were homicides raise any libel concerns? While all the climbers were dead, survivors can sue for defamation. Was your story lawyered? A: I believe our lawyers read the piece, yes.

News media coverage did not go that far. In “forensic circles,” Los Andes reported, “it appeared to be a crime, though the police had not made any accusations.” It left the case open for public interpretation yet again.

“Were the wounds to the head from a fall or deliberate?” Los Andes asked. “Perhaps the truth will never be known.”

That is where any serious consideration ended. On March 24, 1976, Argentina’s government under Isabel Perón fell to a deadly military coup. Argentina was turned inside out, and tens of thousands are thought to have died in the seven-year upheaval. Did any of your sources conjecture that the case might have been solved were it not for the coup?I'm not sure if any think it would have been "solved" conclusively, though some still hold that slim hope. The presumption is that the focus would have remained on it, at least for a while longer, at the behest of an engaged public, the media, maybe the families and U.S. government.

Any formal investigation into the American expedition was handed over to the collective imagination. The mystery seemed to freeze in place.

Days before the coup, the body of Johnson was buried in a small mountaineer’s cemetery near the trailheads to Aconcagua. No one from her family came. But a flower bouquet rested on her coffin. “De Tu Madre,” it read — From Your Mother.

Among the two dozen witnesses were members of the police group that recovered her body, including Arrieta Cortez, the lead investigator. (According to his son, Juan, Arrieta Cortez died in 2017 and never reached a conclusion in the case.)

“Under the sky of America, we bury a daughter here on Argentine soil,” Arrieta Cortez said at the graveyard gathering. How do you know what was said? A: There were reporters there, and he was quoted.

Representatives from the American embassy had made the 650-mile trip from Buenos Aires. The ceremony took 15 minutes.

“I wish to inform you that your daughter, Janet Johnson, was buried on March 19, 1976, pursuant to your request, at the Mountain Climber’s Cemetery at Punta del Inca,” the embassy wrote to Johnson’s mother. “Funeral services at the gravesite were very dignified and impressive.”

A man arrived late, rushing to the service just as it ended. It was Miguel Alfonso, the guide, there to pay his last respects.

THE CAMERA

FOR NEARLY 50 years, a Nikomat camera, carried by an American woman, sat frozen in a high-altitude time capsule. But it was not frozen in place.

Where the camera was dropped may not be where it was found. The Polish Glacier has been shrinking and shifting, cracking and moving downhill by the pull of gravity and with the change of seasons.

And on a sunny day in February 2020, the heart of the Argentine summer, the camera sat on a stocky penitente, like a museum piece on a pedestal.

It was Marcos Calamaro, a young porter, who brought it down to camp. It was Ulises Corvalan, the experienced guide, who recognized the name stamped on the bottom.

At the camp that day was a photographer named Pablo Betancourt. He recognized that the film inside might be evidence to be preserved, as it had been for most of five decades. He put the camera into a case and stuffed it with snow.

He contacted The New York Times, wondering whether such a discovery might be of interest. How and why did he contact you?Pablo sent a tweet with my Twitter handle, asking for an email address. He later said he knew me from writing “Snow Fall,” and thought this story had elements that might interest me. And he wondered what else the melting glacier might be revealing.

Johnson’s arm was found, in a red jacket sleeve, near the glacier’s edge. Then her knapsack, filled with gear and two more aluminum canisters, with film inside.

In Oregon, Johnson’s only surviving immediate family member got a surprise call, sharing the news of the discovery.

Abrahamson’s response was clear. Yes, develop the film. Find out everything you can. Please. Why didn’t you quote  her directly?As a practical matter, I wasn't interviewing her or recording her. I was just in conversation. So I don't know the exact quote. Even if I did, I probably wouldn't have used it, because it's so transactional. I like the quote I used just below, taken from a subsequent conversation.

“She’s still my sister,” she said. “I still want to know what really happened to her.”

INDIAN HEAD, SASKATCHEWAN, is about an hour east of Regina. Its tallest structure is a grain elevator. There is not a mountain in sight.

On a downtown corner is a former bank, a two-story brick structure from the 1800s. Today it is home to Film Rescue International, run by a man named Greg Miller. What logistical challenges did the story pose for you and your visual journalism colleagues?The first big challenge was a blend of timing and travel. The camera was found in February 2020, at the onset of Covid. The specialty photo lab where we wanted to take this (shout out to colleague Josh Haner for steering us to Film Rescue International) was in Indian Head, Saskatchewan. We wanted to hand-deliver the camera and film rolls, to avoid airport X-rays and the chance of anything getting lost. Should Pablo bring it? Should I go get it? I almost got on a plane in late February, but I was afraid I might get stuck in South America as Covid was shutting the world down. We decided to pause a few weeks for Covid to blow over. (Ha.) In the meantime, the lab instructed Pablo to keep the camera and film in the same type of glacial setting where it had been for 47 years. He put it in his freezer in Mendoza, Argentina. It stayed there for two years.In the meantime, I began a deep research dive, mostly trying to find anyone associated with the expedition who was still alive, both in the United States and Argentina. The Times’ research staff helped get me started. A journalist in Mendoza named Nicolás Garcia, a friend of Pablo Betancourt’s, chased down leads and became our fixer and translator.With this story on ice (groan), I mostly did other things. I covered the Summer Olympics in Tokyo and the Winter Olympics in Beijing. It was in China in 2022 that I told my friend and colleague Emily Rhyne, one of our great video journalists, all about the Aconcagua story. She saw the visual, multi-media possibilities and enthusiastically joined the team. We started to move. Pablo Betancourt met us in Regina with the film and camera.Once we had the images, we made plans to do in-person investigations and interviews. The first stop was Oregon, to see Janet Johnson’s sister and show her what we had. Then I made a 10-day trip to Argentina, to work with Nicolás, knock on doors, dig through archives, interview people and make plans for a return trip with the video crew. We did that in early 2023. Among the other trips were one I made to visit the brother of John Cooper, whose family had loads of information about the expedition. And all of us went to visit John Shelton, the only one of the eight Americans on the expedition who was still alive. He died in November.

His small team of technicians receive and process old or damaged undeveloped film from all over the world — rolls abandoned in attics, reels discovered in shipwrecks, the forgotten Instamatic found with film inside. Who chose Miller’s company? A: Josh Haner, a colleague, friend and Pulitzer-winning photographer, is now our senior editor for photo technology. He quickly began searching for the best development option.

Now Miller was holding a camera that had been locked in a glacier at roughly 20,000 feet for almost five decades. The camera was intact; the only crack was inside the lens. The mechanisms worked. The leather holster screwed to the camera bottom had probably protected it from leaks.

It turns out that a glacier on Aconcagua is not a bad place to preserve film. Humidity is always a detriment, but the Andes are notably dry. High-altitude radiation can be a concern, but the camera had been entombed in ice. Cold temperatures are much better for film than hot ones.

Miller took the camera into a dark room, flicked on an infrared light that would not expose the film and clicked the back of the camera open.

“I think we’re going to see something,” he said. Is this reconstructed or were you present?I was there when they developed it. We also had a photographer (Max Whittaker) and two videographers (Emily Rhyne and Pablo Bettancourt), capturing things inside the dark room.

The processing responsibility fell to Erik LaBossiere, a 35-year-old part-time pro wrestler and metal-band guitarist with a bald head, a soft voice and arms covered in tattoos. What a vivid character description. How do you get details like that from your notebook onto the screen?As I'm interviewing people, I often write down descriptions — what they're wearing, facial hair or color, things that might be useful later.

He was nervous. There was only one chance to do this.

Under infrared light, LaBossiere moved the rolls of film into lightproof drums. The drums went into a machine that washed the film in a cycle of solutions, precisely timed — an automated version of the dunk-and-soak method from old photographic development. Why did you describe this part of the process as “dunk-and-soak?”I'm sure the method has a real name. But how would I explain this to, say, my wife? A friend? This is a sign of my experience on high-school yearbook staffs many years ago.

When LaBossiere emerged from the dark room, he seemed pleased.

Had he not known the origin of the film — trapped on a glacier in Argentina for decades — LaBossiere “would have assumed it was on a shelf somewhere,” he said.

After more machines and more solutions, LaBossiere unspooled the film and held a strip to the light.

“Yeah,” he said. “Mountains and people.”

Johnson was a good photographer. The photos are beautiful, haunting, marred only by streaks of moisture that color the frames, some more than others. They turn ordinary landscapes into something closer to art.

One of the rolls was unused. Johnson had carried it toward the summit with the apparent expectation that she would need it.

Another, found in a canister, had 36 exposures. The first frame was shot from a valley just short of base camp, an ethereal image of snow-covered mountains. Then came lots of penitentes and snow-covered peaks. They chronicle the expedition’s up-and-down method of moving from one camp to another, acclimatizing and hauling gear.

There is one photo of Johnson, having handed her camera to someone else. She is smiling, wearing a floppy hat and heavy-duty aluminum-framed glacier glasses. She has an ice ax in her right hand and an overstuffed pack on her back.

The roll found inside the camera had 24 photographs.

The seventh photo was taken near camp at the foot of the Polish Glacier. Only Johnson, Cooper, Zeller and McMillen went higher than that. Johnson snapped pictures from the glacier. Footprints dent the soft snow.

Around midday, the sun high and the shadows short, Johnson took a photo of one of the other climbers, who was downhill and sitting on the glacier.

Afternoon shadows got longer with each photograph. Soon the four climbers would dig a cave to sleep in. Cooper would head downhill the next morning while the other three continued up.

Johnson took more photos after Cooper was gone. The 21st photograph showed either Zeller or McMillen climbing ahead of her, into the afternoon sun, each step making deep holes in the snow. Why did you devote so much attention to Johnson’s photographs?Without them, we wouldn't be telling the story at all. And they are beautiful and revealing, helping us confirm facts and see the world that she saw.

Published in the Mazamas annual later that year is the opposite photograph, taken by Zeller — downhill, of Johnson hiking up on the summit ridge, at about 22,000 feet.

Johnson wore her floppy hat. Her coat was unzipped and her mittens dangled from strings at her sleeves. She held her ice ax in her right hand.

Before dark, Johnson snapped three more photographs of the surrounding Andes. If she was oxygen-deprived or delirious, she still knew how to focus the lens, compose the frame and hold the camera steady to take clear photographs.

That is where the film ends. That is where the legend begins.

The film does not solve the mystery. Does this dilute the tension that built throughout the story?In short, yes. I was aware of this tension — will the photos solve the case? — but careful not to overpromise. I would have loved for a photo to definitively answer the question of what happened. But, really, how reasonable is that expectation? So how upfront should we be? When I began writing, I saw this as a classic yarn. If you told this around a campfire, how would you tell it? When would you reveal which bits of information? I trusted that readers would see that the story isn't really about the photos, is it? It adds to it — it tells you what Johnson saw in her final hours — but not how she felt. Not how she died.

Not every discovery leads to revelation. Some just make you want to know more.

THE MYSTERY

IF JANET JOHNSON and John Cooper were still alive, they would be in their late 80s.

All of the Americans from the expedition to Aconcagua are gone. Dafoe, the leader, died in a car crash on a rural Montana highway in 1975. Zeller died in 2003, McMillen in 2011. Shelton died in November, leaving behind a collection of old photos, Mazamas memos and newspaper files.

“It remains the greatest mystery of Aconcagua,” said Moran, the Argentine journalist who covered the expedition and its aftermath. He is 80 now. “This story had nearly faded from popular memory, but there are enough reasons for doubts and arguments to make the mystery persist.”

Folklore happens when facts are short and time is long. After all these years, this story is not about long-gone Americans on the mountain, but the unknown that lives in those who remain. It is less about certainty than memory and imagination.

A question arises again and again among those familiar with the story: Just what are the possibilities? An “accident” is a tidy catchall, a useful way to move on. What if it was something else?

Corvalan, a dean of Aconcagua guides, with 59 successful summits, first heard the stories from old-timers when he began climbing the mountain 35 years ago.

There were theories and embellishments, dots connected with fuzzy lines.

A love triangle gone wrong. A stash of money that was never found. Cooper as a government agent. Assassins who crossed the nearby Chilean border. Is that why Loren McIntyre, an American, had shown up, as if from thin air, to find the bodies? Why was he taking so many photographs?

Corvalan studied Johnson’s photos from 1973. He noted the shallow slope and the uncharacteristically soft snow on the Polish Glacier that year. A long fall and a deadly slide down the ice were improbable, maybe impossible, he said.

But something else bothered Corvalan. He has seen bodies ravaged by even short falls. Bones are broken. Clothing and equipment are shredded.

Why, Corvalan wondered, did so little of that appear to have happened to Johnson and Cooper? Why was the damage confined mostly to their faces?

Corvalan thought about it. He is a mountaineer. He has been atop the Seven Summits. He knows what experience and common sense tell him: an accident. But more than before, Corvalan believes that — maybe — there was foul play. Why include “maybe” as aqualifier “maybe”? A: I sensed that his ambivalence grew over the months I interacted with him. Maybe I wanted to protect him. He really doesn't know what to think. Which is illustrative, I think.

Foul play. It is a persistent, vague euphemism in this story. Negligence? Manslaughter? Worse? How? Why? Is it even possible at such an altitude, with such fatigue?

Corvalan shrugged.

Roberto Bustos, the base camp manager, is now 76. He has a file of yellowed clippings and photos at home. He has a rope that belonged to Shelton that he holds as a treasured memento.

Johnson’s newly developed pictures stir memories but do not change his mind.

He sees what happened to Johnson and Cooper as “a mountain accident,” he said, but he does not dismiss the possibility of something violent. Norms shift at high altitudes, he said. Desperation toys with right and wrong.

One thing that has not changed in 50 years, on mountains from Aconcagua to Everest, is the notion of ethics and responsibility. They get squishy at high altitudes, amid the dangers and limits of the moment.

“It’s a different world at 6,000 meters, with different laws and rules,” Bustos said. “And the behavior — you would go down to 5,000 meters and think these people are crazy.”

If their climbing partners did all they reasonably could to help Cooper and Johnson, wasn’t that enough? If they abandoned their colleagues to save themselves, or somehow did harm to them, could they be blamed? Why did you inject your questions here?You may notice in this story that I never say what the possible "foul play" could be. It's your imagination, not mine. I want readers to draw conclusions on their own. This section is almost like an invitation for discussion. I wanted to ground people. What I was thinking was this: Even if you think someone could have done this — do you? — isn't there a deeper, squishy question about responsibility? We see this in mountaineering, as climbers step over dying strangers on their way to the top of something like Everest. Porters abandon clients to save themselves. Where does blame lie? Where do ethics and responsibility merge or divide? Don't the questions get more complex the more you know?

Zeller’s widow, in her 90s, said through her son that she did not want to talk about the expedition and requested no further contact.

“As a state policeman, he is precise, exacting and careful,” the local newspaper wrote of Zeller in 1973. “When he speaks he says only what needs to be said. There are mysteries of the mountain he cannot explain. He is not used to that.”

McMillen’s family said that he continued to climb mountains for the rest of his life, including Denali twice, even after he was diagnosed with multiple sclerosis. He had more than 100 dairy cows, and would give slide presentations of his climbs to friends and family in the barn.

His children recall McMillen talking about how he and others were held and questioned in Argentina because of the deaths. They know little about any speculation of foul play, of the stories spun in Argentina. It seems impossible to them.

Judge Victorio Miguel Calandria Agüero never made a ruling in the case.  Shortly before he died in 2022, he was asked about the American expedition by a local journalist, who said that readers had followed the coverage “like a novel” and raised the specter of murder.

“None of that was ever proven,” the judge said.

And then, from the ice, came Johnson’s camera. And whatever ghosts had been laid to rest were stirred back up again.

IN OREGON CITY, ORE., Judie Abrahamson had not gone through her sister’s belongings for years. They were stashed under the house, ignored if not forgotten.

None of it made a lot of sense — these slides of mountain landscapes and strangers in climbing gear, those yellowing newspaper clippings in Spanish where her mother crossed out every suggestion that her daughter ever wanted to die alone. What’s the purpose of this extended litany that reveals what happened to the climbers’ survivors — first Johnson’s sister, then Cooper’s widow and son?It's to bring the story back to real people, today. Without it, it's just mythology. With it, it's heartbreak. It's a way of saying: Let's not lose sight that this had a real impact on real people, and they're still feeling it.

To Abrahamson, Janet Johnson was not an accomplished climber in Colorado or the haunting name that echoes in the Andes. She was not a someone else’s legend or anybody else’s mystery.

She was Janet, a brainy 10-year-old who asked for a little sister and welcomed her to the family with a doll. She was an overachiever who grew into a woman that her mother could not understand.

She was just a big sister, Aunt Janet to Abrahamson’s kids, who set out to prove that she could do anything she wanted, even climb the highest mountains.

Abrahamson thinks about her sister and wonders how she might have grown old, might have climbed more mountains, might have come out, might have felt … accepted, even celebrated.

In Kansas, Joy Cooper is nearly 90, the older sister who remembers John Cooper as a little boy with so much wanderlust that their father had to build a fence to keep him in.

She remembers when people filled the church for his funeral, and they buried her little brother in the cemetery right after Christmas. Her parents were never quite the same after that.

In Texas, Randy Cooper, son of a NASA engineer, raised by a widowed mother who has since died, does not remember much of anything about his father. But he has been told that they share some of the same mannerisms, like the way they pop their knuckles. The section suggests Randy was interviewed. Why did you write about him in the third person without quoting him?I suppose I didn't love any of the quotes. And you'll notice — there are no quotes in this entire section. This was me, resuming my role as narrator — the one telling the campfire story.

As Randy got older, he decided to go by his first name: John. And when people asked about his father, he would tell them the only thing he really knew: My dad died mountain climbing.

The Johnson and Cooper families never learned much about what happened on Aconcagua. They just knew that things went wrong, and that Janet and John were gone.

The details — the newspaper stories, the letters, the official documents, all the questions and regrets — were swallowed up by sadness, then by time. Why did you choose to end the story this way? Were other approaches considered?This was near the end of the first draft, which I left a bit loose not knowing what more I might learn or what the multimedia might entail. I had a few more grafs after this. But Mike Wilson had a thought: What if you just ended it here? As usual, he was right.Does the ending fulfill the promise of the lead?I'm not sure what the lede promised other than information about a camera found and its relationship to a bit of high-altitude folklore. In that respect, yes. And I hope it gives readers something they weren't promised and weren't expecting. Those are my favorite kinds of stories.

* * *

Chip Scanlan is an award-winning writer who taught at the Poynter Institute and now coaches writers around the world. He is the author of several books on writing and the newsletter Chip's Writing Lessons.

The post A reporting team trekked back 50 years to explore an unsolved climbing mystery appeared first on Nieman Storyboard.

Every beat has its own allure. The dramatic stories of crime, criminals and their victims draw reporters to the cops beat. The best science journalists revel in data, discoveries, evidence and the challenges of determining their veracity.

For more than a decade as a writer and editor, Rachel Corbett has mined the world of art for compelling stories. “I’ve always been interested in art,” she told me about her abiding interest in the specialized beat. “My friends were always artists. Plus, the art world is full of great characters and power struggles: the tradeoffs between artists and collectors, everyone jockeying for money and status.”

Beat reporters combine their command of the worlds they cover with a keen eye for news pegs that allow them to strike when a story becomes relevant. During the many years Corbett has been writing about the art world, the Wildensteins, a French family “legendary for having built an extraordinary multigenerational art-dealing empire,” fascinated her. When she learned in 2021 that a French court ordered them to go back on trial for an alleged billion-dollar fraud scheme, Corbett realized that the time had come.

After eight months of international travel, bilingual research and reporting, and writing, editing and fact checking, Corbett’s story about the case appeared in August in The New York Times Magazine. “The Inheritance Case That Could Unravel an Art Dynasty” is an intriguing blend of a detective story crossed with an art history class, featuring offshore tax havens, shell companies and trusts, the masters from Vermeer to Picasso, Nazis and a super-wealthy cast of villains and victims.

To report the story, Corbett flew to Paris after the magazine’s editors bought her pitch and agreed to underwrite her travel expenses and the cost of interpreters. There she connected with the story’s protagonist, a tenacious lawyer named Claude Dumont Beghi. Twenty years earlier, Sylvia Roth Wildenstein, the widow of legendary art dealer Daniel Wildenstein, approached Dumont Beghi for help after the fortune she assumed she would collect disappeared into the hands of Wildensteins’ two sons from a previous marriage. Sylvia died in 2010 without proving her case, but Dumont Beghi never gave up trying to establish that the Wildensteins were guilty of defrauding her and the French tax authorities in a massive criminal scheme.

Precision and persistence marks Corbett’s reporting process. Steeped in art history but with a limited knowledge of French, she brought interpreters  along with her to meetings with the lawyer. To decipher books and a voluminous court record in France, she relied on Google Translate, backed up by native speakers who checked the parts she wanted to use. Her investigation culled documentation from museums in Europe and America. Piercing the veil of the stealth tactics employed by the Wildenstein family of fine art collectors took time and patience to convince people to talk, “given the family’s stature and influence in the art world.” In both cases, she said, “the solutions just came down to spending the extra time — translating and making call after call.”

Most of the narrative traces Dumont Beghi’s journeys around the world and the extraordinary lengths she took to uncover the Wildenstein dynasty’s well-kept secrets cultivated over three generations that evaded tax laws and ex-wives.

With clarity and grace, Corbett delivers easy-to-digest explanations of the complex system of wills, trusts and the other havens where wealthy art patrons stash fortunes, especially the free port of Geneva, Switzerland. She describes it as  “a prisonlike complex of high-security storage facilities that is said to contain more art than the Louvre.”

Corbett admits she’s normally not very organized, relying on  notes spread out through multiple files and notepads. To keep track of this case’s serpentine history through the years, she “drew up a giant timeline of all the court orders and financial transfers and family milestones.” The result is a mostly chronological structure that keeps the reader on track from the start of Dumont Beghi’s quest for justice to its stunning finish.

“I felt that it had everything you could hope for in a good narrative: a dynastic drama, a high-stakes court case, a billionaire family torn apart by greed,” Corbett told me. “The legal saga also provided a clear example for readers about how the art market works, and how it can be manipulated by the ultra-wealthy to hide fortunes and evade taxes.” 

Corbett, who has also written for The New Yorker, The Atlantic, and New York Magazine, among others, is the author of “You Must Change Your Life: The Story of Rainer Maria Rilke and Auguste Rodin.” The book won the 2016 Marfield Prize for nonfiction books about the arts written for a broad audience.

In an email annotation for Storyboard, Corbett unspooled the reporting strategies and writing process behind her story and the challenge of making sense of the byzantine world of French law and financial instruments that made the Wildenstein’s fraudulent schemes possible. Our conversation has been edited for length and clarity. 

ANNOTATION: Storyboard’s questions are in red; Corbett’’s answers in blue.

The Inheritance Case That Could Unravel an Art Dynasty

How a widow’s legal fight against the Wildenstein family of France has threatened their storied collection — and revealed the underbelly of the global art market.

By Rachel Corbett

The New York Times Magazine, August 23, 2023

Twenty years ago, a glamorous platinum-blond widow arrived at the Paris law office of Claude Dumont Beghi in tears. Someone was trying to take her horses — her “babies” — away, and she needed a lawyer to stop them.

She explained that her late husband was a breeder of champion thoroughbreds. The couple was a familiar sight at the racetracks in Chantilly and Paris: Daniel Wildenstein, gray-suited with a cane in the stands, and Sylvia Roth Wildenstein, a former model with a cigarette dangling from her lips. They first met in 1964, while she was walking couture shows in Paris and he was languishing in a marriage of convenience to a woman from another wealthy Jewish family of art collectors. Daniel, 16 years Sylvia’s senior, already had two grown sons when they met, and he didn’t want more children. So over the next 40 years they spent together, Sylvia cared for the horses as if they were the children she never had. When Daniel died of cancer in 2001, he left her a small stable.

Then, one morning about a year later, Sylvia’s phone rang. It was her horse trainer calling to say that he had spotted something odd in the local racing paper, Paris Turf: The results of Sylvia’s stable were no longer listed under her name. The French journalist Magali Serre’s 2013 book “Les Wildenstein” recounts the scene in great detail: Sylvia ran to fetch her copy and flipped to the page.  Sure enough, the stable of “Madame Wildenstein” had been replaced by “Dayton Limited,” an Irish company owned by her stepsons. That’s when she called Dumont Beghi. The story grabbed my attention from the very start. You used the hallmarks of a suspense novel: colorful characters, a world of immense wealth, and a cheated widow who launches a quest for answers to a mystery. What prompted that approach?For some reason, the image I had in my head going in was like a film noir scene — a tearful woman meeting a detective (or in this case, a lawyer) in the wake of a terrible betrayal. I chose to start here because I felt that the reader could watch the story unfold through the eyes of the lawyer, who was also coming into the situation cold.

To the lawyer’s surprise, Sylvia showed up to their meeting with no proof of ownership for the horses and no information on her late husband’s estate. “She didn’t have any — any — documents at all,” Dumont Beghi says. Sylvia mentioned that she signed some papers shortly after her husband’s death, but she didn’t know what they said, nor did she have copies. “I put that in the corner of my mind,” Dumont Beghi says.

Why would a widow draped in diamonds and furs have no records from her wealthy husband’s estate? Dumont Beghi got the feeling there was more going on than a dispute over horses. But she went ahead and gave Sylvia the good news: She could simply decline to transfer the horses to her stepsons. Dumont Beghi sent a letter, halting the transaction.

Dumont Beghi recalls an almost instant kinship with Sylvia, who discovered that they were both Scorpios and lived in the same building complex in the posh 16th Arrondissement. What were you signaling with the verb “recalls?”I’m just telling the story from her perspective 20 years ago, so she’s flashing back. After Dumont Beghi saved her horses, Sylvia trusted her completely, and she began to explain to Dumont Beghi the complexity of the situation. Daniel had fallen into a coma for 10 days before he died, and while he was under, his sons, Alec and Guy, showed up at the hospital along with lawyers from Switzerland, the United States and France. She recounted how, a few weeks after the funeral, her driver took her to the family’s 18th-century hôtel particulier, Why didn’t you define this French term?I think because it’s generally understood among English speakers and it conveys to me a more specific image than a word like “mansion” does. which housed an art research center, the Wildenstein Institute. Her stepsons told her she needed to hear something important. They had reviewed their father’s estate and discovered that he died in financial ruin. As his next of kin, Sylvia was about to inherit debts so large they would ruin her too.

Sylvia was stunned. She had never heard anything about money troubles from her husband. For 40 years, she had lived with chefs and chauffeurs, in at least five homes on three continents. But what did she know? She never signed the checks. Daniel, intellectual and rigid, ran the business, while Sylvia, who was light and cheerful, played the nurturer in the family. She was known to dote on Alec and Guy’s six children, whom she considered her grandkids. She trusted her stepsons completely, so when they told her that she must renounce her inheritance at once or face “catastrophe,” she didn’t blink. “I signed all the papers they presented to me. I signed, signed, signed” — even the ones written in Japanese, she later told Serre. Why did you use Serre’s book as a source? Were you able to independently verify Sylvia’s account?Because Sylvia died many years ago, secondary sources were the only way to include her direct voice in the piece. Serre, an accomplished journalist and documentary filmmaker in France, had access to Sylvia while she was alive. I independently verified everything I could. 

They promised to take care of her financially and even offered to pay her 30,000 euros a month out of their own pockets. Sylvia was grateful. But then, over the next few months, the reality of what she had done set in. Sylvia told Dumont Beghi how movers came to her apartment and took a beloved Pierre Bonnard painting off the wall. Then they came back for the furniture, because, she was told, it belonged to her husband’s business, which was now run by his sons. A letter came notifying her that Daniel’s 69 thoroughbreds were now owned by Guy and Alec’s stable. Her household staff stopped being paid. Soon, her stepsons told her she would have to move from her home on Avenue Montaigne to another apartment. (Alec died in 2008; Guy declined a request for an interview, though a representative answered some questions provided by The Times.)What did you have to do in your pitch to win this assignment?I sent an overly long initial pitch and then I revised it with the editor’s feedback on narrowing the scope of the narrative. I did some more reporting and went back with an angle that focused specifically on the widow Sylvia Wildenstein, who is the genesis of the current tax case, and her lawyer, Claude Dumont Beghi.How much reporting and writing did you do before you pitched the story?I did maybe two or three interviews before I officially started, to make sure I understood certain key details and also to gauge how much people would be willing to talk. I needed to make sure that at least Dumont Beghi would be on board.Is it unusual for an editor to work with you on a pitch?Usually there are conversations or back-and-forth on the idea before you begin. But his questions about the narrative provoked me to rethink it a bit so I rewrote that part of the pitch, both for him and I assume also for his colleagues.What role did your editors play throughout the reporting and writing?My editor played a big role in thinking through structure with me before I began and throughout the process. He challenged me to write more analytically at times, using my background knowledge about the art market. She stopped receiving invitations to celebrate holidays and birthdays at the family’s ranch in Kenya or their castle in France. Guy shipped back her clothes and belongings from their British Virgin Islands compound, where she had vacationed for years with Daniel and their chef and pastry chef. As Sylvia spoke, two things became increasingly apparent to Dumont Beghi: One, Sylvia had renounced her inheritance. “She had no freedom.” she says, and “no proof. Not a shred of evidence.” No bank account, no income, no independence. It was as if “she died at the same time as her husband,” Dumont Beghi says.

The other thing that struck her was that the Wildensteins were more than merely rich.

“When she first came to me, I didn’t know anything about the family,” Dumont Beghi told me when I visited her this past winter at her office in Paris. To my left, a bronze bust of a panther stared from a pedestal at eye level. Behind her glass desk hung a print of a leopard prowling in a tree. What kind of notes do you take during an interview?Everything I see in the room and any observations I make about a person — how they look, their mannerisms and what they’re saying, of course. I also keep a running list of topics I want to follow up on. Dumont Beghi is also the personal attorney for President Ali Bongo Ondimba of Gabon, who is widely considered a strongman, and often describes herself as a lone warrior woman in a jungle of male adversaries. She had never heard of the Wildenstein dynasty of art dealers. In fact, outside elite niches of the art world, few had, which was how Daniel wanted it. Dumont Beghi was about to find out why. What’s the narrative purpose of this sentence?To set up that we’re about to go on a journey with her, and that she’s going to have to go to extraordinary lengths to uncover these very well-kept secrets.

First, she drew up a list of known assets, which soon zigzagged into a chart of far-flung bank accounts, trusts and shell corporations. Over the course of several years, she would fly around the world to tax havens and free ports, prying open the armored vaults and anonymous accounts that mask many of the high-end transactions in the $68 billion global art market. What do you base that figure on?It’s from the Art Basel UBS annual report on the art market, which is seen as one of the more reliable sources in the industry. Multimillion-dollar paintings can anonymously trade hands without, for example, any of the requisite titles or deeds of real estate transactions or the public disclosures required on Wall Street. She would learn that the inscrutability of the trade has made it a leading conduit for sanction-evading oligarchs and other billionaires looking to launder excess capital. The Wildensteins were not just masters of this system — they helped pioneer it.

Over 150 years, the family has amassed an art collection estimated to be worth billions by quietly buying up troves of European masterpieces that would be at home in the Louvre or the Vatican, holding their stock for generations and never revealing what they own. When Sylvia realized the magnitude of her stepsons’ deception, she devoted the rest of her life to unraveling the family’s financial machinations, and even left a will asking that Dumont Beghi continue her fight from beyond the grave. Were you able to read the will?Yes.

Sylvia and her lawyer were never able to win the settlement they thought she deserved while she was alive. From the start, in 2004, a judge rejected Dumont Beghi’s attempt to cancel Sylvia’s renunciation of the inheritance; a few years later, a court rejected a subsequent claim that she was entitled to €450 million worth of art and assets, a figure the judge called “pharaonic.”Did you rely on French court records as part of your documentation? If so, how does the process to gain access work, especially compared with the availability of records in this country?Yes, and it’s actually not terribly difficult to access some of them, for instance those in the Court of Cassation, which is the court that ordered the Wildenstein retrial.They were mostly digital, available on the court website, or PDFs that various parties sent me. I was finding out about all kinds of decisions and hearings throughout my reporting so I was acquiring, reading and translating records the whole time. There were a lot of legal challenges over the years, involving a number of Wildenstein family members and their lawyers, trustees and business associates. I spoke to French legal experts about aspects of them that I couldn’t comprehend on my own.

The representative for Guy notes that, early on, Sylvia was awarded approximately €15 million, based on the value of Daniel’s French estate. “Dumont Beghi continued to litigate for several years, seeking to have certain trusts settled by Daniel Wildenstein included in the estate,” the representative says. “During this protracted litigation, Dumont Beghi made numerous, unsubstantiated allegations, but the court ultimately ruled against her client.”

Now, more than a decade after Sylvia’s death, their efforts have landed the Wildensteins before France’s highest court. The evidence she and Dumont Beghi brought forth has persuaded prosecutors that the Wildensteins are a criminal enterprise, responsible for operating, as a prosecutor for the state once put it, “the longest and the most sophisticated tax fraud” in modern French history.

A trial this September will determine if the family and their associates owe a gargantuan tax bill. The last time prosecutors went after the Wildensteins, several years ago, they sought €866 million — €616 million in back taxes and a €250 million fine, as well as jail time for Guy. The consequences could do more than topple the family’s art empire. The case has provided an unusual view of how the ultrawealthy use the art market to evade taxes, and sometimes worse. Agents raiding Wildenstein vaults have turned up artworks long reported as missing, which fueled speculation that the family may have owned Nazi-looted or otherwise stolen art, and spurred a number of other lawsuits against the family in recent years. Financial distortions have saved the family hundreds of millions of dollars, prosecutors allege, but their treatment of Sylvia could cost them far more — and perhaps lead to the unraveling of their dynasty.How would you describe the purpose of the previous paragraph? Do you consider it the nut graph?Yes. I don’t always like obvious nut graphs, but my editor and I felt one was needed here. It sets up what’s at stake for this family and why they’re important to a broader understanding of how the art market aids and abets all kinds of dubious financial dealings.When you’re working on a story do you consider theme?I’ve actually never once thought about that term until now. But, yes, I guess what I just said before reflects my consideration of a theme.

IN ORDER TO PROVE that Alec and Guy misled Sylvia about her husband’s estate, Dumont Beghi first needed to know what assets they did report. But because Sylvia had renounced her inheritance, she didn’t even have a right to that information. “Every deed, every bank statement, every inventory item in the estate and every document related to the succession of Daniel Wildenstein is in the hands of Guy and Alec,” Dumont Beghi says, and they did not intend to turn them over.

Dumont Beghi’s first step, then, was to ask a court to nullify the agreement Sylvia signed giving up her inheritance. Only then could she access details about Daniel’s estate. Fortunately, she had a compelling precedent to show the judge. Sylvia wasn’t the first wife the Wildensteins had tried to cut off by pleading poverty: Jocelyne Wildenstein, Alec’s first wife, was similarly cut out of the family’s fortune during her 1999 divorce, with Alec claiming he was an unpaid personal assistant to his father. Documents revealed at court in New York — where the couple primarily lived — valued the family’s art collection at about $10 billion. The judge in the case said that Alec’s income statement “insults the intelligence of the court”; he settled for a rumored $3.8 billion — which would be the largest divorce settlement in New York history. (Jocelyne denies that the settlement was $3.8 billion but did concede that it was “huge.”)How did you access these records?I found some in a book by one of the lawyers. That case was also such a media circus that some of it is in previous coverage.How did you reach Jocelyne and persuade her to talk with you for the story?I was only able to reach her by the end, during the fact checking process, but she was very helpful.How would you describe your interviewing style?I don’t really have any strategies. I just try to go in open-minded and listen super carefully for all the potential threads I may want to follow up on.

Dumont Beghi argued that if the family was worth billions then, there was reason to doubt that Daniel, who orchestrated the deal between Alec and Jocelyne, died in ruinous debt just two years later. The French court ordered Guy and Alec to hand over the declaration of Daniel’s estate. It included some properties in France, a few cars, paintings and bank accounts, altogether totaling €42 million. Dumont Beghi didn’t believe that figure was anywhere near the estate’s true value, but still, “It’s not nothing, for someone who died broke.” And it showed, Dumont Beghi concluded, that Sylvia had renounced her inheritance under false pretenses.

Dumont Beghi’s next move was to get her hands on Daniel’s medical records.

She learned that he spent his final days in an unresponsive, vegetative coma — and yet apparently signed a contract selling his 69 thoroughbreds (including Sylvia’s) to his sons for a bargain price. In 2005, a court granted Sylvia’s request to nullify her renunciation. It was only the beginning of what Dumont Beghi has called her international “treasure hunt” for every stashed masterpiece, undeclared property and offshore account left out of Daniel’s estate.

Her next order of business was to locate Sylvia’s beloved Bonnard nude, a gift from Daniel that his sons had removed from her wall. Dumont Beghi knew it was included in a trust that Daniel had set up for his wife in the Bahamas,How did she know this?It was part of a trust known as the Sylvia Trust, which contained only Bonnard paintings. Dumont Beghi was at some point able to obtain the inventory of that trust and I reviewed it, too. but when she asked the trustee for information about its contents, management and regulations, she received no response.

Dumont Beghi decided to do her own research on Daniel’s collection of Bonnards. She learned from his memoir, “Marchands d’Art,” published two years before his death, that he considered their acquisition “the biggest coup” of his life.
[caption id="attachment_178936" align="alignright" width="300"] "Nude Washington Feet in a Bathtub" by Pierre Bonnard 1924. The Wildenstein family owned more than 180 Bonnards.[/caption]

When Bonnard died in 1947, he left behind an enormous estate of some 700 paintings and thousands of drawings. Daniel learned that all of it was set to be inherited by three estranged nieces-in-law of the artist, and it gave him an idea. He approached another Bonnard relative who Daniel believed could also lay claim to the estate and told the man he would pay him $1 million to buy his inheritance rights. Then he armed the man with a “battalion” of lawyers to fight on his behalf.

 After more than a decade in court, Daniel walked away with nearly 500 paintings; the nieces were left with just 25. How did you learn the inside story of this “coup?”Daniel spoke about it a few years before he died.Without your making any value judgements and hewing rigorously to the facts, Daniel and his sons come across as, greedy, odious, rapacious, conniving. Were there any libel concerns and if so, how were they addressed?I addressed it by making sure everything in the story was as airtight and accurate as humanly possible. This story took nearly a year to report and write and the magazine had three fact checkers assigned to it. There are limits to what I can ever know without being able to speak with the family members themselves, of course, but a representative for them weighed in on some of the factual components.

(Daniel promised them some more to prevent further litigation.) In his memoir, Daniel revealed that he still owned 180 Bonnard paintings — and not just any Bonnards but “the most beautiful. The most magnificent.” He added that the great Bonnards were worth between $5 million and $7 million each. (Today they can sell for twice that.) How do you determine the current valuation?By looking at recent auction data, which provide the only publicly available prices in the art market.

Dumont Beghi flew to the Bahamas to find out what other paintings by the artist Daniel may have left for Sylvia. She received a court order to open up the trust and found that Daniel had bequeathed no fewer than 19 Bonnards to her client. Though the trust was nominally in the Bahamas, the Bonnards were being held at the Geneva free port, a prisonlike complex of high-security storage facilities that is said to contain more art than the Louvre.

Independent of any national jurisdiction, free ports allow traders to ship and store property without paying taxes or customs duties. If a dealer buys a painting in one country, he can ship it to a free port without paying import taxes; then, when he is offered the right price, he can sell it there too, without paying capital gains.Can you describe the reporting, writing and revision that enabled you to convey such a clear and concise description of free ports?I’ve never seen a free port myself, but I’ve seen pictures and read accounts by people who have, and I spoke to Dumont Beghi about her experience visiting the one in Geneva. The legal language around their tax advantages can be complicated and dry, so I just decided to write a concrete example of how some people use them in order to make it more accessible.

It has been estimated that $100 billion worth of art and collectibles are held in the Geneva free port alone, to say nothing of those in Zurich, Luxembourg, Singapore, Monaco, Delaware or Beijing.

Dumont Beghi flew to the Geneva free port, which is the size of 22 soccer fields, This is an effective analogy that helps readers visualize something they'll likely never see.How did you settle on and compute it? It’s actually not my analogy! Others have said it, too. along with an appraiser to examine the Bonnards in person. Bonnard is “light,” Daniel wrote of his favorite artist, who is known above all for his radiant use of color. But when Dumont Beghi descended two flights down into the gloomy bunker, she found the paintings locked behind an armored door, including Sylvia’s “Pink Nude in the Bath,” its warm glow extinguished in the dark.

An acquaintance in the art world explained to Dumont Beghi that hundreds, if not thousands, of Wildenstein works are held in museums, but that the labels often identify their owners simply as “private collection.” So she wrote to the major museums — the Louvre, the Hermitage, the Prado — to ask whether Daniel Wildenstein ever lent or donated works to them. Surprisingly, she says, a few wrote back. The National Gallery in London told her that Daniel lent it valuable paintings by Poussin and Boucher. The Prado had recently bought a Velázquez portrait from Wildenstein & Co Inc. for €23 million.

Then Dumont Beghi made perhaps the most important stop of her tour: the Metropolitan Museum of Art, where she stood before a painting she loved, Caravaggio’s late masterpiece “The Lute Player,” labeled on loan from a “private collection.” She searched the New York State Department’s records to see whether Wildenstein & Co. had ever borrowed money using works in its collection as collateral. Dozens of names were listed — Cézanne, David, Degas, Manet, Monet, Matisse, Rembrandt, Picasso and Rodin among them. And then there it was: “The Lute Player,” valued at upward of $100 million.

At that point she realized, “The company is titanic.”

"IN MY FAMILY, we have elevated discretion to the level of muteness,” Daniel wrote in his memoir. Here you break from Dumont Beghi’s hunt for answers to the Wildenstein saga. Why did you shift gears?I felt that by now we’d want to know more about who these people are that acquired such a monumental art collection and fortune. Dumont Beghi didn’t have much contact with them outside of the courtroom, so we needed to jump out of her perspective to learn on them. “We do not speak. We don’t tell. We don’t talk about one another.” This code of omertà has been the governing principle of the Wildenstein art dynasty since its founding five generations ago. A dealer “is not allowed to talk about his stock,” Daniel said. “Why? Because it’s the stuff of dreams. Every art dealer must maintain the illusion of the masterpieces he owns or does not own.” Many believed that his grandfather, the founding patriarch Nathan Wildenstein, for example, owned 10 Vermeers; he actually had just one. No one knows today whether the family still owns it, and that question is meaningful to art history. Experts believe Vermeer made about three dozen paintings in his life, and as many as nine could be missing.

[caption id="attachment_178933" align="alignright" width="400"] Daniel Wildenstein, left, and his son, Alec, in 1965 at the family's private sculpture garden in an old firehouse in New York.[/caption]
A tailor from Alsace, Nathan had no training in art when, in the 1870s, a client asked him to sell some artwork she owned. He “holed himself up in the Louvre” for 10 days, according to Daniel, and came out a believer. He sold the art and used the 1,000 francs he earned to buy two more pictures, by the Rococo artists François Boucher and Maurice-Quentin de La Tour, which he resold. At the time, Nathan could afford 17th- and 18th-century French art because no one else wanted it, so he amassed passé — but to his eye, beautiful — paintings. He began dressing in embroidered waistcoats and top hats to pitch collectors and critics.

Soon Nathan was selling his taste to Rothschilds and Rockefellers, in Europe and the United States. While Nathan was grooming his young grandson to enter the family business, he took him to see a silent film about a man who wore a hat that everyone initially mocked; by the end of the movie, the whole town was wearing one. Nathan explained to Daniel that this was their family’s calling: “Find the guy’s hat and wear it before the others.” For Nathan, that hat was French art of the 18th century: Fragonard, Watteau, David. These are now among the most famous names in art history, but at the time they were synonymous with the French Revolution and the aristocrats it overthrew — a period the public wanted to put behind them, especially as they began to embrace the avant-garde era of Impressionism.

In 1905, Nathan bought an hôtel particulier in the center of Paris to house Wildenstein & Co. He expanded into Renaissance art and Impressionism and, when his son, Georges — Daniel’s father — was old enough to join the business, a bit of Modernism. Nathan bought a space down the street for Georges and his friend Paul Rosenberg to set up a small operation. The pair gave two of its floors to Picasso, whom they agreed, in 1918, to pay a generous salary in exchange for first pick of the artist’s works. Georges installed a red telephone in his office that had two direct lines: one connected to Rosenberg, the other to Picasso's studio.

Upon Nathan’s death in 1934, Georges steered the family into an era of unprecedented prosperity by building an infrastructure around his artists’ markets. He organized exhibitions, edited an art journal and published definitive catalogs of works by artists in his inventory — Ingres, Fragonard, Chardin. (Daniel would later do the same with Monet, Manet and Gauguin.) The books were well respected and helped market their artists to museums. They also gave the family final say over authentication questions. Today anyone who thinks he or she owns a Monet that’s not in the Wildenstein book needs a nonprofit co-founded by Guy to sign off on it. (When the Wildenstein Institute handled the authentications directly, it developed a reputation for being unaccommodating.) Why did you use a BBC story to support this assertion?There are several examples, but the BBC’s was a particularly thorough examination of a man’s attempt to get Guy to authenticate a Monet that many other experts thought was real.

But Georges’s ruthless instincts also contributed to the “dark aura,” as one dealer put it, that would come to surround the Wildenstein name. Hitler’s personal curator told an Allied intelligence agent in an interrogation after the war that Georges did brisk business with the Nazis after fleeing to Provence, in the unoccupied zone. Once there, he helped the Germans locate important collections in occupied France in exchange for sparing his own. Profits from his newly “aryanized” gallery in Paris were said to be sent to New York, where he had opened a branch. (The representative for Guy denies this.)How did you get hold of the copy of the intelligence report and the summary of the interrogation where the Nazi art dealer claimed he worked with Georges?From the U.K. National Archives. I also found a lot of related wartime materials related to the Wildensteins at the Getty library in L.A.

Other art-dealing dynasties have since sprung up in the Wildenstein mold. They buy up huge quantities of blue-chip art and store it for years, until they effectively corner their own niches of the market and control the prices. The billionaire Nahmad brothers and their sons, based among London, New York and Monaco, reportedly bought more works by Picasso than any other family in the world (except the Picassos) and, for the most part, have locked them up in the Geneva free port for years while they accumulate value. The Mugrabi family of Pop Art dealer-collectors, led by its patriarch, Jose Mugrabi, and his two sons, have done the same with Andy Warhol, stockpiling some 1,000 works by the artist and keeping prices high by bidding his art up at auction, even if they don’t intend to buy. (The Mugrabi family did not respond to a request for comment.)

Those who complain that the art market today operates more like the stock market often blame these families, who shifted a value system once driven by connoisseurship to one based on the law of scarcity. (“Monet and Picasso are like Microsoft and Coca-Cola,” David Nahmad once said.) Their dominance derives from the fact that they’re family firms, bolstered by internal secrecy, pride and lifetimes of experience. As the Wildensteins proved, families can be structured like corporations, where the profit principle governs even relationships and succession plans. The few people who seem capable of undoing them are themselves. For the Wildensteins, the weight of the family legacy seems to have cracked the younger generations. Your story reflects such mastery of the subject matter. What makes it possible to write about about a rarefied world with such confidence?I think because I’ve been writing about it for like 15 years. The art market can be a complex, opaque and bizarre system, but it’s also very small, so you see the same machinations and players emerging again and again.

Even though Daniel described Georges as a “bad father,” he parented his own children in similarly severe ways. He enforced his father’s business tactics — extreme secrecy, consolidation of wealth in the bloodline — as laws of family life, too. Daniel tried to seclude his two children, Alec and Guy, at home and unmarried, to protect the family from publicity and divorce. They lived as if in another era — the French 18th century — with opulent floral décor, heavy drapes and footmen who stood behind their chairs during meals. As children, Guy and Alec commuted to the Lycée Français de New York by limousine, and they rarely had a play date. Alec was forbidden to play sports and attend university, Guy was prevented from pursuing acting and both were required to learn their father’s trade. Daniel was particularly strict with Alec, his elder son.  According to a 1998 Vanity Fair article, he started taking Alec to brothels at age 15 in the hope that he would find prostitutes a satisfying alternative to a wife. When Alec defied his father and married Jocelyne, he did so secretly in Las Vegas with no guests. Eventually, Daniel’s sons and wives and children all lived together in his New York townhouse.

Those who know Daniel have said that he infantilized and humiliated his sons and that they’ve gone on to treat the women in their lives similarly. Guy, Sylvia believed, was jealous of Daniel and took it out on her; Alec blamed Jocelyne for the humiliating headlines generated by their divorce. (The New York Post dubbed her the “Bride of Wildenstein” for her apparently extensive plastic surgeries.) Alec, who wore bold pinstripe suits, was the flashier brother; Guy kept a lower profile but played on the Diables Bleus polo team with aristocratic friends, like the future King Charles III, the godfather of his eldest child. You paint such a vivid picture of the family dynamics. What did you rely on to report and verify those details?It’s an amalgamation from a lot of sources: interviews with people who know the family personally, details from past media reports, interviews the family members themselves gave over the years. Colleagues remember that the brothers would sit quietly in meetings. Guy married a Swedish model named Kristina Hansson, who has never appeared in a tabloid. In fact, he once boasted that “hardly anyone knows what my wife looks like.” So when Daniel died in 2001, Guy was the clear successor to the family art empire,while Alec took over the horse business.

Guy, who is now 77, is the family’s patriarch and president of Wildenstein & Co. But mounting lawsuits and scandals have begun to drag him down. So far he has avoided any serious consequences — a fact some critics attribute to well-positioned friends like former President Nicolas Sarkozy or to the fortune at his disposal for defense counsel. But now that the family is on trial, Guy, it seems, may have taken the legacy of silence too far. The Wildenstein policy to preserve confidentiality at any cost may ultimately expose the family’s secrets. How do you view this device — some might call it a cliffhanger — that draws the reader to keep reading?I think it’s the moment where we see how Guy’s personal story led to his present legal situation, so it was a natural transition. But perhaps I did want to nod to the stakes at hand before asking readers to continue on through another section about the legal maneuverings.

IN 2009, after a long string of setbacks, Dumont Beghi had a breakthrough. Over the years she had sent Liouba Wildenstein, Alec’s second wife, multiple summonses for information about the family’s assets. Unsurprisingly, she ignored them. But after Alec died of prostate cancer at age 67 in 2008, Liouba, a former model from Russia, found herself in trouble. According to Serre’s book, Alec owed €12 million in back taxes, and his father’s estate was still tied up in litigation with Sylvia. Guy offered to lend Liouba the money to help pay his brother’s debt — all she had to do in return was give him access to a trust Alec had set up for her, supposedly so Guy could reimburse himself later. But after the deal was done, Serre’s book recounts, Guy didn’t pay Liouba the millions he promised. He sent only small, sporadic sums — not enough to pay her tax bill or to live on. Liouba found herself in a situation much like Sylvia’s: cut out from the family, with no money and no recourse. (The representative for Guy says that he did issue the loan.)

That’s when Dumont Beghi’s phone rang. Liouba had finally decided to answer her third summons. She told me recently that she felt she had no choice but to take action: “Many women in the family had to fight for their rights,” she said. How were you able to reach her?I emailed her and corresponded for a while with her lawyer until she agreed to meet with me in Paris. “The women want to be respected.” Twenty-four hours later, a lawyer would deliver Dumont Beghi dozens of documents that Liouba had found on Alec’s personal computer — contracts and letters about the family’s expansive network of offshore trusts — which would reveal what Dumont Beghi and Sylvia had long believed without being able to definitively prove.

The documents mapped how the Wildensteins had structured their patrimony, and hidden their wealth, for generations. Daniel’s estate, Dumont Beghi learned, included several hundred artworks — including the 180 Bonnards, hundreds of 16th- and 17th-century French paintings and dozens of works by old masters including Caravaggio, Velázquez and Fra Angelico. Then there was the real estate: multiple homes and buildings across France and the United States, the 58,000-acre ranch in Kenya and the 18-acre Virgin Islands compound. There was a Gulfstream IV jet, a yacht and the thoroughbred stable, which was registered to multiple intermediaries in England and Ireland. The art was held in shell companies and trusts in tax havens, including two previously unknown entities in the Cayman Islands and Guernsey. These were “operational structures specializing in tax evasion,” Dumont Beghi wrote, which also helped the family shield assets from divorce. (The representative for Guy disputes the accuracy of this recounting of the estate.) Given the response from Guy, how did you verify what his estate included?A lot of it is in court records, or is verifiable through property records. Some of it is from Dumont Beghi’s investigation, which I double checked as thoroughly as I could. Guy’s rep did not specify which part was inaccurate, nor did he provide any counter information.

According to Dumont Beghi, two trusts named Sylvia as a beneficiary, something Sylvia said she was unaware of. Also revealed was a letter from Guy and Alec’s Swiss lawyer seeking to remove Sylvia as beneficiary from one of the trusts. Investigators also discovered $250 million in art that Daniel had apparently ordered airlifted out of the United States while he was in his coma. (The representative for Guy denies that this is true, calling it “illogical.”)

Dumont Beghi rapidly began to issue new summonses and build an appeal for a review. But time was running out. Sylvia had been diagnosed with ovarian cancer, which was spreading. She was running out of resources, too. “I have no more money,” Serre recounted her saying. “This procedure has brought me to my knees.” She had paid more than €10 million in legal fees over the past eight years and had resorted to pawning her jewelry and relying on help from wealthy friends. In her final interview, she said of her stepsons, “They robbed me, and now they are waiting for me to die.”

Dumont Beghi continued on, believing that Sylvia was entitled to a settlement of $300 million. She filed a new criminal complaint against Guy and the heirs of Alec — his two children and Liouba — as well as their business associates, using the new information she had received. To her surprise, this time the government responded. The police raided the Wildenstein Institute and the family’s apartments on a court order to identify any assets that might have been concealed from Sylvia. In the basement, officers discovered vaults filled with hundreds of drawings, paintings and sculptures. Some of the frames were inscribed with swastikas.

Officers seized about 30 lost works by the likes of Degas and Berthe Morisot. Some had been reported stolen by a Jewish family during the war, and others were reported lost by families who had involved Daniel in the management of their estates. (Some of the looted works had been reunited with the family and then entrusted to the Wildensteins; others, according to one heir, were never reunited with the family and first rediscovered in the vault.) Guy pleaded ignorance: He never inspected that vault. And who could prove otherwise? The family took such pains to protect their inventory that no one knows what they really have, perhaps not even them. (A representative for Guy said that all the works stolen from the Jewish family during the war had been first reunited with them and then entrusted to the Wildensteins. Litigation with the family is ongoing.) How did you decide where and how many times to include comments from Guy’s representative?We didn’t include the rep’s dispute of facts that were verifiably true, but I think we otherwise included every point of contention they raised. This was important since they declined interviews and we did not have much of their side in the story.  

Dumont Beghi’s involvement in the Wildenstein affair officially ended on Nov. 8, 2010, when she called Sylvia for the last time to wish her a happy 77th birthday. Five days later, Sylvia died at home in Paris. She was buried in the Wildenstein tomb next to her husband, but Guy had her maiden name, Roth, etched into the marble tombstone. Without a client, Dumont Beghi’s case was closed for good.

But the lawsuit was far from over for Guy, as the state picked up where Dumont Beghi left off. She had mapped for the government the global system through which the family moved money among nine companies registered in Ireland, four trusts on three islands, a handful of galleries and real estate companies and bank accounts in at least four countries, possibly depriving the French public of hundreds of millions of euros. How was this figure arrived at?The prosecutors allege they owe that much in unpaid taxes.  In addition to the Swiss free port and the Paris vault, they had art in a nuclear bunker in the Catskills, a former fire station in New York and many other far-flung places. “I mean, there are pictures I have never seen that my great-grandfather bought,” Alec told Vanity Fair in 1998. They were, he said, “in vaults and crazy places, in back of other things.”

OVER THE NEXT DECADE, the Wildenstein tax case wound its way through the French courts. At the same time, public outrage over tax loopholes for the wealthy was growing, and the government passed what is popularly known as the Wildenstein law to crack down on tax evasion via foreign trusts. Still, the family won two controversial acquittals, first in 2017 and then again in 2018. How much writing and revision did it take to compress the events of a decade?I probably started with like 1,000 words on those earlier lawsuits and it was just all cut. There were just so many legal cases in the end that we needed to breeze through some of it to keep things moving.

But then, two years ago, France’s attorney general and tax authorities brought concerns about the decision to acquit the Wildensteins of tax fraud and money laundering to the Court of Cassation, France’s highest civil and criminal court. The lead judge in the 2017 case had said that the family displayed a “clear intention” to hide their wealth, but the tribunal let them off because, at the time, foreign trusts fell into a legal gray area. In reopening the case, the Court of Cassation disagreed, saying the lower court “disregarded” the facts.

“It’s really uncommon,” Dumont Beghi says of the upcoming retrial. She believes the path to victory will be much tougher for Guy and his co-defendants this time. Prosecutors will argue that the Wildensteins were, in fact, required to report their foreign trusts at the time of Daniel’s death, and later Alec’s. They also contend that the trustees improperly took orders from the family in violation of the rules of irrevocable trusts, which must be independently managed.

The extreme lengths to which the family went to obscure their wealth led French media to dub them “the Impressionists of finance.” But in reality many of their practices are commonplace in high levels of the art trade, which a 2020 U.S. Senate subcommittee called the “largest legal, unregulated market.” Unlike financial institutions, art businesses are not expressly subject to the Bank Secrecy Act, which requires firms to verify customers’ identities, report large cash transactions and flag suspicious activity. A study from the U.S. Department of the Treasury last year cited a figure estimating that money laundering and other financial crimes in the art market may amount to about $3 billion a year. (Britain and the European Union, however, have implemented anti-money-laundering regulations that require stricter due diligence in art transactions there.)

According to a report by Art Basel and UBS, auction houses did about $31 billion in sales last year. They say that they know who their clients are, but those may just be the names of art advisers or other intermediaries. And collectors’ insistence on anonymity, long framed as genteel discretion, hasn’t budged. The buyer of the most expensive artwork ever sold at auction, Leonardo da Vinci’s $450.3 million “Salvator Mundi,” registered at Christie’s a day before bidding with a $100 million down payment, identifying himself as one of 5,000 princes in Saudi Arabia. A few weeks later, it was revealed that the true buyer was Crown Prince Mohammed bin Salman — who was reportedly displaying the painting on his superyacht — and that a little-known cousin of his bought it as a proxy. It was billed by Christie’s as the “last Leonardo da Vinci painting in private hands,” but it’s only the “last” Leonardo until someone reveals another one, like the Madonna and child the Wildensteins sold in 1999 to an anonymous collector, who is still believed to own it.Throughout your story, you write about a complicated subject with great clarity. How much revision was involved? What’s your response when you need to revise?I think I had maybe five drafts, starting at 5,000 words and then ending with a little over 7,000. Throwing away work you’ve already done and then redoing it is never fun, but it’s necessary. Each revision definitely made the story better.What is your writing process?I report and gather as much information as I can before writing anything. Then I write a really sloppy first draft. Then I just revise and revise.

For a business that routinely transacts in secrecy jurisdictions, literally in the dark and underground, scarcity can be manufactured, and value is dictated by whatever someone is willing to pay. “A client’s privacy should be an art dealer’s primary concern,” Daniel wrote, calling it a matter of “respect.” But secrecy is also a core competitive advantage in a profession predicated on insider knowledge — a model the Wildensteins themselves relied on. The gallery kept a legendarily detailed directory of where every coveted painting in the world was located using intelligence sometimes gathered by spying on rival dealers — even, one competitor alleged, Why is this source anonymous?I honestly don’t know! It’s a matter of public record, so it must have been a narrative choice not to name them.  tapping phones. That system of ultra-insular knowledge and extreme scarcity is why, today, the dealers who bought “Salvator Mundi” for $1,175 at a New Orleans auction house were able to resell it for a reputed $80 million, and then, in the span of five years, see it flipped for $127.5 million to the collector who ultimately sold it to the Saudis for the record-breaking $450 million.

Younger dynasties like the Mugrabis and Nahmads have similarly been accused of strategically obscuring ownership of their assets to shield them from divorce or other legal claims. When a Frenchman accused the Nahmads of possessing a Modigliani painting, once estimated to be worth up to $25 million, that Nazis looted from his grandfather, they said it was owned by a company called International Art Center. A couple years later, the Panama Papers revealed that David Nahmad owns International Art Center, a holding company whose assets are stored in Geneva. (A representative for the Nahmad Collection says the case has “no merit.”)

“Many of these very wealthy families do sort of act like cartels,” says Christopher A. Marinello, a lawyer who recovers lost art. “We’re still dealing with these Nazi-looted-art cases because the art market hoped they would outlast the heirs.” The Wildensteins, too, he says, have handled “problematic” pictures, though none that he is currently pursuing are in their possession. Whenever he asks the family for information that might aid in his search for stolen pictures, they take a very long time to respond, he says, and are reluctant to provide information. “They’re just looking the other way,” he says. “It’s just this unwillingness to lift a finger and do anything.”

I MET DUMONT BEGHI once more in New York, where she had come to visit galleries with her son, an artist and designer. How much time did you devote to  on-the-ground reporting versus research and reviewing documents? What was the balance?Probably 60% interviews and 40% research.

At a windy table outside Harry Cipriani’s food hall on the Upper West Side, she told me that she plans to attend every day of the Wildenstein trial this fall. It will finally mark the end of the defining case of her career. “It’s my professional life, it’s my personal life,” she said. “I start something, I finish it. I will go every day. I want to see it through.”

Her long entanglement in the case created legal troubles for her too. Guy Wildenstein sued her for defamation in 2016. A few years later, she was convicted of tax fraud and money laundering for depositing $5.1 million she received from Sylvia in an undisclosed HSBC account in New York. She is currently pursuing a partial appeal and has suggested that the $5.1 million was a “customary gift.” (Guy dropped the defamation suit two years ago.)

In 2012, Dumont Beghi published a book about her seven years on the case, “L’Affaire Wildenstein.” In the opening lines, she describes it as “a story of two women alone facing the establishment,” run by privileged and powerful men like the Wildensteins — “a universe where women are omitted.” Some have questioned whether Dumont Beghi was really representing her client’s best interests in pursuing the costly, yearslong battle. But regardless of her motives, it’s obvious that the saga has become personal for her. Her eyes welled up when she spoke of Sylvia’s death. “She wanted the world to know that as a woman she wanted to be respected.” She described tax fraud as a crime that disproportionately deprives women. This is what she and Sylvia were fighting for. “It may be hard to understand the depth of our relationship,” she told me.

With a potential billion-dollar guillotine hanging over its neck, the house of Wildenstein is in unprecedented peril. Even before this latest legal trouble, its influence waned for years as the market for the historical art it sells declined, and museums are by now fully stocked. As Daniel reached his eighth decade, he started waking up in the mornings asking himself, “How long are we going to last?” The profession his family dominated for most of the 20th century had been overtaken by a new guard of contemporary-art dealers selling status baubles to Wall Street millionaires. These collectors weren’t interested in Rococo or Neoclassical art; they were spending millions on living stars like Damien Hirst, whose market the advertising tycoon Charles Saatchi has dominated since he bought up vast quantities of the artist’s early work. Daniel tried to get in on the frenzy by forming a joint venture with Pace Gallery in 1993. But its contemporary clients generally didn’t convert to Impressionism or old-master collectors, and vice versa. “It was a mistake,” Pace’s founder, Arne Glimcher, told me. “I think we did it because we were so flattered.” Pace bought back its shares plus inventory from Guy in 2011.

Now the family appears to be liquidating some assets. In 2020, Guy and his wife put their Tudor estate in Millbrook, N.Y., which they spent a reported $50 million renovating, on the market for $20 million. Around the same time, their son, David, and his wife, the jewelry heiress Lucrezia Buccellati Wildenstein, listed their Connecticut equestrian compound for $6.9 million. The Virgin Islands property is up for sale, too, for $48 million. In 2016, while facing his initial tax trial in Paris, Guy listed his Sutton Square townhouse in Manhattan — Corcoran blurred the paintings on the walls, naturally — for nearly $40 million, only to finally offer it at a loss in March for $29.5 million. “I see the end of this empire,” the old-masters expert Eric Turquin says. “The organization is too heavy for a market that has shrunk. The market is one-tenth of what it used to be for 18th-century French art.”

Some market insiders have noticed that the family seems to be selling off more art lately, too. Though paintings are often sold at auction anonymously, provenance histories can reveal ownership information. In the past two years or so, “they sold a lot of paintings at auction, at Christie’s, not under their own name,” says Robert Simon, one of the old-masters dealers who rediscovered “Salvator Mundi.” “But when they’re cataloged, you can see that they’re shown by Wildenstein in previous shows or were acquired here and there.” He adds, “And then they’ve kind of shed their staff as well.” The mass liquidation of assets suggests that the family could be anticipating a large expenditure, like an overdue tax bill.

In 1932, Georges Wildenstein hired the society architect Horace Trumbauer to design the family’s majestic limestone gallery on East 64th Street, with marble floors, gilded wood paneling and lead vaults. “It was the grandest gallery in New York,” Simon says, recalling the heavy drapes the Wildensteins would pull back to reveal paintings to clients. It’s where they sold one of Raphael’s most treasured Madonnas, Caillebotte’s iconic cityscape “Paris Street; Rainy Day” and Cezanne’s largest and most lyrical “Bathers.”

Guy’s son, David, who is vice president of Wildenstein & Co., has described the building as “the soul of this company and the soul of this family.” Yet he helped sell it in 2017 for $79.8 million, then the highest price ever paid for a townhouse in New York. The contemporary-art gallery Lévy Gorvy Dayan has since taken occupancy of the space while Wildenstein & Co. has moved into a 15-story commercial building in Midtown, open by appointment only. “It’s like an office,” one dealer told me. “A small office.”Why did you decide this was where you would end the story?The Wildensteins felt to me like a family stuck in a different time. They’re so insular they never really adapted to changes in their industry, or in society. In the end, I felt they’d been left in the past, in the shrinking world they made for themselves. Since the story was told largely chronologically I wanted to end with some idea of what that world looks like now as they’re forced to contend with this lawsuit.Some writers say they need to know the last line of a story before they begin writing. Are you one of them?No, the last line of the story is usually the last line I write. It just emerges — usually on deadline!

* * *

A correction was made on Sept. 1, 2023: An earlier version of this article used an outdated name for the gallery located in a townhouse formerly owned by the Wildenstein family. It is now known as Lévy Gorvy Dayan, not LGDR.

A correction was made onSept. 22, 2023: An earlier version of the article described incompletely the dispute between the Wildensteins and the family of a Jewish collector whose art was stolen by the Nazis and then discovered in a Wildenstein vault. Some of the artworks were first reunited with the family and then entrusted to the Wildensteins as estate managers; others were never recovered, according to an heir who has ongoing legal disputes with the family. It was not the case that the bulk of the lost artworks had been missing since they were looted in the war. How did this correction come about?A representative for the family felt that this passage unfairly implied that ALL of the Nazi-looted works uncovered in the Wildenstein vault were improperly gained, when some of them had first been returned to their rightful owners before they ended up with the Wildensteins. The other ones are still in dispute. While the facts in the story were accurate, the editor felt it was important to explain the story more fully, especially given the nature of the allegations.

* * *

Chip Scanlan is an award-winning newspaper reporter who taught writing at The Poynter Institute for 15  years. He publishes Chip’s Writing Lessons and has two books, ”Writers on Writing” and “33 Ways Not To Screw Up Your Journalism."

The post How an arts reporter unraveled a controversial and opaque family art dynasty appeared first on Nieman Storyboard.

Samantha Michaels was reading The New York Times one day in 2019 when she read a story about a case where the punishment seemed vastly disproportionate to the crime.

The situation involved Tondalo Hall, an Oklahoma woman who was leaving prison after her sentence was commuted. “She got 30 years behind bars because her boyfriend broke bones in her children’s legs and she failed to stop him,” Michaels recalled. “He’d gotten a much shorter sentence for committing that violence — two years in jail.”

Michaels, a senior criminal justice reporter at Mother Jones, soon learned that Hall’s case was just one striking example of an undercovered story: Harsh sentences that women, many of them Black, receive in so-called failure-to-protect cases. Theoretically, she told me, failure-to-protect laws “require mothers and fathers to stop someone else’s violence against their children. But in reality, it’s almost always mothers who are blamed.”

Last summer, after a year-long investigation, Mother Jones brought the hidden story to light.  On Aug. 9, 2022, the magazine published Michaels’ 8,200-word story, “She Never Hurt Her Kids. So Why Is a Mother Serving More Time Than the Man Who Abused Her Daughter?” The project, partnered with a short documentary, won a 2023 National Magazine Award for best video and was a finalist for best reporting. It also won the 2023 John Bartlow Martin Award for Public Interest Magazine Journalism from her alma mater, Northwestern University Medill School of of Journalism.

As a vehicle to drive the story, Michaels used the current case of  another Oklahoma woman. Kerry King is serving 30 years for failing to protect her 4-year-old daughter from her boyfriend, who used violence against her and her child; his sentence was 18 years — little more than half of hers.

The narrative recounts King’s journey from abused woman to prison inmate. It is powered by dramatic, at times horrific scenes and buttressed by dialogue and vivid details. Michaels weaves the personal story with an investigation of the injustices at the hands of what she says is a sexist, racist and classist criminal justice system.

Michaels amassed a mountain of information: court documents, including audio, video, photos, transcripts and the police video of King’s interrogation. The state corrections system sat on her application to visit King in prison for months until the story was almost finished. Much of the reconstruction relies on more than 25 hours of phone conversations with King. Michaels also visited King’s three oldest children, who were living with her ex-husband, who had gained custody of them despite his own history of violence in their marriage.

The reporting was anchored by an estimated 160 hours of fact checking by research editor Ruth Murai and supported by a grant from Columbia University’s Lipman Center for Journalism and Civil and Human Rights.

Before joining Mother Jones in 2015, Michaels worked as a copy editor and reporter in Indonesia and then Myanmar. While at Irrawaddy news magazine in Myanmar, she traveled with rebel soldiers and other insurgents. Her colleagues — many of them political dissidents and some who spent time in prison — modeled formative lessons for a budding investigative journalist.

“Watching them do their jobs in rough conditions — sometimes with no electricity or internet, sometimes sharing a single phone in the office, sometimes trekking into war zones, and almost always covering violent atrocities and an unjust military dictatorship — reminds me what we’re capable of as journalists and why this work is so important,” she said.

In email conversations with Storyboard, Michaels described her reporting, interviewing and writing strategies, the granular approach she took to document her findings and how she structured a complex narrative. Our conversation has been edited for length and clarity and is followed by an annotation of the story.

Why did you become a journalist?
I’ve always loved writing. When I was a little kid, I wanted to be a children’s book author and illustrator. By middle school I was thinking about becoming a novelist. But I wasn’t sure how financially stable life as a novelist would be, so in sixth or seventh grade I started imagining a career in journalism. (How naïve I was to think journalism would be financially stable!) For a school assignment, we had to apply to our dream college and I chose Northwestern’s Medill School of Journalism. I ended up there and I soon realized that I loved reporting as much as I loved writing. I considered myself shy and enjoyed having an excuse to reach out to people I might not otherwise approach, and to ask them questions about their lives without the pressure to talk a lot about myself. When I got into a good interview, it was like I stopped noticing time pass, and I felt energized and engaged in a way I had never felt before. Journalism appealed to my adventurous side, my curiosity and my desire to help people, even if that just meant listening and helping them feel heard.

You’re now an award-winning writer focusing on critical social issues. What formative influences and/or people led you to this moment in your career path?
Maddie Oatman, the editor I’ve worked with for nearly nine years, is my biggest professional mentor. She helps me feel grounded during stressful moments and has probably taught me more about writing than any other teacher. I also owe a lot of gratitude to my former Burmese colleagues, easily the most badass reporters and editors I’ve ever met, and some of the most generous.

I also would not be where I am today without my family. For years after returning to the United States, I struggled with fears that I might get stuck copy editing and never become a full-time reporter. My family, especially my parents, stood by me through that anxiety and encouraged me to keep going. Another close family member is one reason I pursued criminal justice reporting in particular. As I was starting out, stories about violence, trauma and punishment struck a deep chord with me partly because of his prior experiences: He was stabbed in the chest by a stranger when I was in college, and I found myself thinking about that night—and the flood of adrenaline and fear it unleashed in me—when I started to interview crime survivors for my early stories at Mother Jones. I also thought about him when I began reporting on juvenile justice and the people serving life in prison for crimes they committed as teens. He was prosecuted in adult court when he was a kid, for a minor offense. He likely avoided a conviction and any lifelong consequences because of our privilege; my family is white and we could scrape together enough money for an attorney. When I began interviewing incarcerated people, I thought often how unfair it is that other people don’t receive second chances like he did.

To that end, some of my biggest influences have been my sources—the survivors of crime, police brutality and prison brutality who spent months opening up to me about their traumas. Each of them taught me about the ways that our law enforcement, courts and correctional facilities fail to keep communities safe, and about the capacity we all have for growth and redemption.

Are there writers whose work inspires as you produce narrative nonfiction?
So many! I read the Marshall Project’s criminal justice newsletter almost every morning. It helps me generate story ideas or learn more about topics I’m covering. Some of my favorite writers on this beat are Keri Blakinger at the Los Angeles Times, Jennifer Gonnerman and Jelani Cobb at the New Yorker, and Ken Armstrong at ProPublica. At Mother Jones, my editor Maddie encourages me and her other reporters to meet once or twice a month to discuss one long-form piece and dissect what techniques worked well or didn’t. Before writing my story about failure-to-protect laws, Maddie and I read Pamela Colloff’s award-winning feature “The Innocent Man,” about a father who was wrongfully convicted for his wife’s murder. The piece was beautifully written and helped inspire the lede to my story; I liked how Colloff started with a scene inside the prison, and she used the man’s letter-writing to quickly convey some of his central motivations, especially to see his son again. Kerry’s driving desire was similarly to see her children again.

What was your goal for the failure-to-protect story?
I wanted to show the sexism and racism inherent in these prosecutions, and how the punishments were often having unintended effects: Lawmakers wanted to protect kids from abuse, but the tragic consequence was that moms were being blamed and punished for violence they didn’t even commit. Sometimes the kids, ripped from their mothers, were sent to live in foster care or with other abusive relatives. I wanted my story to elicit an emotional response from readers, whether angry or sad, and potentially spark them to speak out and try to change this system.

This is a multi-media project involving complex social issues, multiple layers of sourcing, sensitive ethical considerations and graphic descriptions of violence. Credits at the end of the story show there were many colleagues and editors involved. How did all this come together wtihout losing the focus of the story? In addition to helping me prioritize and focus, Maddie Oatman was a sounding board during the reporting process as I navigated interviews with Kerry, her young children and their father, a man who also had abused Kerry. She helped me coordinate with Ryan Little, a data reporter who was instrumental in figuring out how common these prosecutions are in Oklahoma. Filmmaker Mark Helenowski wove together our reporting with heartbreaking footage of Kerry’s kids for the film; this was my first time working on a documentary, and Mark patiently coached me through things like script-writing and voiceover recording. Deputy editor James West shepherded the film, too, and former editorial director Amanda Silverman gave additional top-edits to the article, as did editor-in-chief Clara Jeffery. Finally, fact-checker Ruth Murai put in weeks of time to make sure the story was accurate.

Annotation: Storyboard’s questions are in red; Michael’s answers in blue.

THE BIG FEATURE

She Never Hurt Her Kids. So Why Is a Mother Serving More Time Than the Man Who Abused Her Daughter?

Failure-to-protect laws are incarcerating women all over the country—for other people’s violence.

Story by Samantha Michaels, Video by Mark Helenowski

August 9, 2022

A WEEK BEFORE Christmas last year, Kerry King helped three of her children build gingerbread houses in a prison visitation room in Oklahoma. King wanted to make the holiday special for the kids, even under the circumstances. But as the Black 35-year-old spread frosting on a graham cracker while dressed in her orange jumpsuit, her hair braided for the occasion, the mood still felt bittersweet. Why do you identify her by race?We decided this was relevant because we later reveal that Black women are disproportionately prosecuted for failure to protect. We also later identify that the officers who arrested her were white. Since she was incarcerated six years earlier, her kids could only visit once a month, and soon it would be time to say goodbye.

“Do you love me?” Lilah, 10, the most outgoing of King’s children, asked her mother as the visit was ending.

“Of course I do,” King answered.

Lilah thought a moment, her brown eyes serious, and then said something that caught King off guard. “Do you still love him?” she asked. “Because if you still love him, I’ll never forgive you.” Were you in the visitation room? If not, how did you reconstruct this scene?I wasn’t. For this scene I relied on interviews; I spent more than 25 hours on the phone with Kerry during the project. She mentioned the visit with her kids a couple of days after it happened, and I asked a bunch of specific questions like: “How did you decorate the gingerbread houses? What time of day was it? Where in the prison were you, and what did the room look like? How did you feel, getting to spend time with your kids before Christmas? What were you wearing and how did you have your hair? You mentioned Lilah made a comment about never forgiving you; do you remember how she started talking about that? What facial expression did she make as she said it, and what was her tone? Were you both sitting, or standing? How did you feel when she said that? Had she ever said anything like that before?” A few weeks later, I asked Lilah about the visit during an in-person interview. I fact-checked details about the scene with Lilah’s older siblings, who were there, too. What recording tools do you use to capture scenes, dialogue and details?If I’m recreating scenes that I didn’t witness, I’ll record interviews with my sources and type notes as we talk; I’ll also look for court documents, videos, and photos that can give me extra details. If I’m witnessing a scene, I’ll use my phone’s audio recorder app to capture dialogue and its Notes app to type quick notes, jotting down things my recorder can’t capture, like where someone is standing when they say something and what their facial expression is. And I’ll take photos and videos of what’s happening around me. The first break I get after the scene, I’ll try to flesh out my notes with extra information from memory. At the hotel that night I may flesh things out more, writing down why I thought that scene was important and where I could see it fitting into the story.

King’s heart dropped.

 Her ex-boyfriend had abused them both, years ago. He was the reason King was in prison now. But her daughter had never said anything like that to her before. And there wasn’t enough time to have the long conversation they both craved. Why did you open your story with this scene and would you describe the thinking, writing and revision process it took?It captured the heart of the story quickly. Kerry is in prison, mostly separated from her kids, and Lilah is still trying to process what happened to them. The quote about never forgiving Kerry is also poignant because it introduces this idea of Kerry, the mother, being an object of blame.It took several revisions to write. Initially, I started the with the night of the abuse. At first, that scene felt like a natural launching point because it led to Kerry’s incarceration and I had so many gripping details. But my editor, Maddie Oatman, pointed out that starting there felt formulaic and could lead the reader to associate Kerry above all else with victimhood. We wanted to open with a present-day scene that helped the reader understand who she is and what her driving motivation is: to be a good mother, to get out of prison and to reunite with her kids.

Back in her cell, King agonized over whether a letter to Lilah would suffice. She had been mothering her children over letters and phone calls for too long. No matter what, King wanted to tell her daughter, I love you more than I could have loved anyone else, any man. And you should never, ever have to even consider whether I do. How do you know what was going on in her heart and mind in her cell?I asked her what she did when the kids left and how she felt in that moment. I also asked, “What are the things you wish Lilah would have known? What do you want to tell her?”

“I am not guilty,” King had said to me over the phone, months before the Christmas visit. How did you discover Kerry King?I learned about Kerry in 2019 when I asked an ACLU attorney in Oklahoma to share a list of failure-to-protect cases that were on her radar. I looked up Kerry’s court docket, read one of her appellate brief, and decided I wanted to know more. I called her trial lawyer, who was no longer in touch with her, and her appellate lawyer, who had a phone number for her mom, Lela. I called Lela and explained why I was hoping to speak with Kerry; Lela put in a good word for me. Then I wrote Kerry a letter. She called me soon afterward.  “I just wanna go home. I wanna my see my kids so bad. It kind of eats you up.”

KING HAS SPENT countless nights tossing and turning in her cell, replaying the night that pulled her away from her family. (The following descriptions are confirmed in court records and testimony.) Why did you include summary attribution here?This is a legally sensitive section, so we wanted to show the types of sourcing we used. Referencing court records can provide us some legal protection as a magazine.

In January 2015, in the small yellow house she shared in Tulsa with her boyfriend and a roommate, King had been bathing Lilah, then 4, when she saw bruises on the girl’s legs and arms. Purdy “was mean,” Lilah said, sitting in the soapy water. John Purdy was King’s boyfriend at the time; he’d recently lost his telemarketing job and sometimes watched Lilah while King worked at a gas station. He’d been physically abusing King for more than a year, often high on heroin, which he’d recently forced her to try. But she had never seen him harm her children. When she asked about the bruises, he claimed Lilah had slipped on the ice.

Two days later, King woke up in the middle of the night and noticed Purdy was not in bed. She was groggy—he had ordered her to shoot up again—but saw a light on in Lilah’s bedroom. When she entered the room, Purdy was holding Lilah’s shoulders, with his fingers wrapped around her little neck. He said they were having a pillow fight, but Lilah whimpered beneath his grasp. “I thought he was hurting her,” King later told detectives. So she tried to free her daughter the only way she knew how, by clenching her fist and punching Purdy in the face. What makes you think you can hit me? she remembers him telling her, turning toward her. Why did you put Purdy’s challenges, here and below, in italics?I think it adds drama, but the primary reason was because King was recalling the quote; she hadn’t recorded it and it wasn’t something I heard myself. He slammed her head against a wall, insisting Lilah needed to be spanked.

King agreed to hold Lilah down like Purdy demanded, hoping that if she complied it would be over faster. But when she saw how hard his blows were, she threw her body over her daughter’s, receiving Purdy’s belt lashes on her own back. Purdy pulled King off Lilah, and King tried to run out of the house to get help from the neighbors or the police—but he blocked her at the door. You ain’t going nowhere, she recalls him saying. “I was scared. I didn’t know what to do,” she later told me, adding that it felt like she “had been in chains.” Purdy dragged King to the master bedroom by her hair and threatened to kill her if she didn’t stay there. He reentered Lilah’s room and locked the door, leaving King outside, listening to her daughter’s cries.

Purdy held King’s phone in his pocket, making it hard for her to call for help. She begged him to come out of Lilah’s room, and then asked him to take a shower with her, anything to distract him. Around 6 a.m., he directed her to lie down in bed. Still high, King reluctantly fell asleep. When she woke up, he wouldn’t let her or her daughter leave the house.

It wasn’t until later that day, when a contractor came to work in the yard, that King’s housemate snuck out and asked the man to call 911. The police found Lilah in a locked room, sobbing. Bruises covered her forehead, cheeks, ears, and neck, and belt lashes cut into her back. Chunks of her curly brown hair were missing. This passage puts the reader in the apartment that night. How you were able to write it with such granular detail?I interviewed Kerry extensively, but almost every sentence in this passage is based on court records: I had thousands of pages to pull from, including a 900-plus-page trial transcript, felony filings, warrants, affidavits, bond documents, sentencing documents, Purdy’s guilty plea, pre-sentence investigation reports, letters Kerry sent to court, and emergency protective orders. I also had trial audio, recordings of King’s jail calls, videos from her police interrogation, photos of the house and her injuries, drawings of Lilah’s injuries and audio of the 911 call.Writing this section was fairly straightforward, though emotionally difficult, because we had so much material. I tried to draw readers into the scene with sensory details and dialogue. I also tried to convey urgency and drama by using strong active verbs (clenched, punched, slammed, dragged, begged, etc.).

The police arrested Purdy. King, who had bruises on her own body, helped investigators take her daughter to a hospital. Then she returned home, exhausted. But her nightmare was far from over.

Less than a week later, King was sitting inside in her pajamas when a police car pulled up to her house around 11 a.m. Two white officers asked her to come with them for an interview. “I didn’t want him to hurt my baby,” she told them. “I was trying to prevent this.”

But the officers didn’t buy it. They wanted to know why King waited so long without seeking help for her daughter. Why had she held Lilah down for the beating? Why didn’t she call 911?

They arrested King and put her in jail, where she would stay for more than a year awaiting trial. They accused her of child neglect and permitting child abuse. Were you able to talk with the police about the case?No, but I obtained lengthy videos of their interrogation of King and watched how they questioned her. Two officers testified at Kerry’s trial and I had those transcripts. I also had police affidavits and their findings of probable cause. By the time it was all over, though she had never laid a hand on Lilah, she would be sentenced to 30 years in prison—12 more years than Purdy, the man who had assaulted both her and her daughter. The story is propelled by a singular focus through the prism of one Black woman’s experience. How did you settle on that approach?First, we had to choose a location. These prosecutions are common nationwide, and I initially pitched the idea of traveling to multiple states. We decided to focus on Oklahoma because it has one of the harshest laws in the country: Parents can be sent to prison for life for their supposed failure to protect, with no exception for moms who were victims of domestic violence themselves. The ACLU had also done a lot of prior work in the state to identify problems with the law.Next, we had to find a central character. Kerry King was an obvious choice; I couldn’t stop thinking about her story after I read some of her legal documents. In particular, I found it compelling that she physically threw her body between her boyfriend and her daughter, trying to protect the girl, and that she received a much harsher punishment than he did. Her case had a lot of court documents and evidence, including videos, photos, and audio, which would be helpful for fact-checking such a legally sensitive topic. Plus, while some other women were nervous about talking publicly about their trauma, Kerry wanted to share her story.Even after narrowing down the location, character and central premise, focus was difficult. My first draft was more than 13,000 words and the second more than 14,000, not counting annotations and notes. My wonderful editor Maddie Oatman encouraged me to tighten the piece by thinking through not just the surface-level plot (what happened to Kerry) and policy-related thesis (the law is unfairly incarcerating mothers), but also the deeper emotional question or, as reporter Brooke Jarvis put it in another Nieman Storyboard interview, the “philosophical grand narrative.” Mine was simple: “What makes a good mother, and how do these laws get in the way of that?” Keeping this question in mind during revisions helped me choose scenes and quotes and cut away words that didn’t contribute toward answering it. Were there others you auditioned before setting on King as your central character?The ACLU sent me 14 names, including Kerry’s. I googled them all and searched for them on Oklahoma’s state court website, reading whatever documents I could easily find. All these cases were heart-wrenching, but Kerry’s jumped out because she got such a long sentence and because the court records contained so much detail. Still, I didn’t reach out to her right away and considered other options. I spoke with a Tulsa public defender who was helping five other women with similar cases apply for commutations. He tried to connect me with two of them, but it never panned out. I think some survivors are wary of sharing their trauma with reporters, especially if they’ve been blamed for it previously. In the end, Kerry’s story was the one I couldn’t get out of my head, it had never been told before, and she was open to working with me.

IN A PARALLEL UNIVERSE, you could imagine the police leaving King in the care of a women’s shelter. But the detectives did not view her as a victim. That’s because of Oklahoma’s “failure to protect” law, which requires parents to shield their kids from physical harm if they’re aware or “reasonably” should have known that another adult was abusing or might abuse the child. Because of this law and how it’s interpreted, King was blamed for what happened to Lilah.

The law is “inherently problematic,” says Megan Lambert, the legal director of the ACLU of Oklahoma, who studies these cases. “A lot of times, motherhood is used as the grounds that they ‘should have known,’ simply because they are the child’s mother.” And mothers in violent relationships are especially vulnerable to prosecution: If they were abused by their partners, juries often believe they should have realized their children might be in harm’s way too. “Folks who are charged often haven’t actually engaged in any harmful behavior,” says Lambert. “They were put in impossible situations and were not able to act fast enough.”

Most states have similar laws, opening the door to anywhere from a few years to decades behind bars as a punishment. But Oklahoma, which incarcerates more women for all crimes than almost any other state, has one of the harshest penalties: Moms can be sent to prison for life for their supposed failure to protect, with no exception for women who were abused themselves. The ACLU estimates that Oklahomans convicted of the offense receive an average sentence of about a decade behind bars.

These types of laws aren’t talked about very much, but they are used to punish parents nearly every week. Last year, I found local news reports of 53 people across 29 states who were, within the span of just three months, arrested, prosecuted, or convicted for similar crimes. Why did you introduce yourself in the narrative here?It’s something we regularly do at Mother Jones. Our style is fairly conversational, so we’ll often write that a source “told me” something, rather than saying the source “told Mother Jones.” This sentence also refers to data work that I did by myself, as opposed to the data work referenced a few sentences later, which I did alongside reporter Ryan Little. Many more cases go under the radar. There are no national data sets to show how many parents have been convicted of failure to protect—in part because their convictions are often labeled as “child abuse” or “child neglect,” making them difficult to track down. But if Oklahoma is any indication, an enormous number of families have been ripped apart. When my colleague Ryan Little and I conducted a groundbreaking review of the state’s court records, we identified hundreds of people who were charged under the law since 2009, when a new version of the statute went into effect.

While the language of these laws refers to parents, prosecutors overwhelmingly target mothers, not fathers. Since 2009, at least 90 percent of the people incarcerated for the offense in Oklahoma were women. Attorneys in multiple states who specialize in this area of law tell me they have never seen a man prosecuted for failing to stop someone else’s violence against a child; Mother Jones found relatively few examples. “It’s sexism,” says Lambert. “It’s the assumption that women are responsible for all the goings-on in the home.” In Oklahoma, the vast majority of women convicted for failure to protect had no prior felony record.

Women of color are disproportionately prosecuted. Black people make up 8 percent of Oklahoma’s total population but 19 percent of those found guilty under the statute since 2009. These types of laws are “really enforced in a racist and classist way,” says Stacey Wright, a women’s rights activist who has also studied these cases. Why do you bring in multiple voices to buttress your assertions?Outside experts can connect dots and help readers make sense of all the numbers. Sometimes people’s eyes glaze over when they see a lot of statistics, so it’s nice when an expert can analyze the numbers and just hit the nail on the head: This is sexism; this is racism.

Not only are the laws used to prosecute women who, like King, are themselves victims of abuse, but there’s no proof these laws are successful in protecting kids. Separated from their mothers, children affected by failure-to-protect convictions sometimes end up in foster care or with abusive guardians, according to several attorneys who are familiar with the statutes.

These laws also create an impossible dynamic that makes survivors less likely to report what’s happening to police. When someone calls 911 after being abused by a partner, some cops open a child welfare investigation if there are kids in the family. So if a mother calls 911, she risks losing her kids; if she doesn’t, she risks being prosecuted for failure to protect. As one legal expert suggests, there’s no way to win. “It creates another barrier for domestic violence victims to seek help, because now they are also threatened with criminalization and incarceration, which also means losing their children,” says Lambert.

In essence, the criminal justice system makes these mothers ultra-culpable, blaming them for things that are largely outside their control. Mothers are punished not only for their partners’ violence, but for the violence that has been inflicted upon them—for the sexism that leads to domestic abuse, for the poverty that makes it hard to escape, for the racist policing systems that don’t protect them, for the circumstances that leave them with few options. As an untold number of women sit in prisons for these supposed crimes, their kids in someone else’s care, maybe the real question we should be asking ourselves is: Who is failing to protect whom? In this section, you stepped back from Kerry King’s narrative. Why?It was important to offer context for how Kerry’s story fits into a bigger picture and why we should care about it. The section functions like a long nut graf: It’s my attempt to answer focusing questions like: “Why does the story matter? What’s the point? What does the story say about life, the world, and the times we live in?” One sees this hybrid model, weaving dramatic scenes and expository context, in The New Yorker and from ProPublica writers such as Rachel Aviv and Lizzie Presser. Why do you think it’s so popular?It’s a simple structure that packs a punch: The scenes draw you in and give you a character to get attached to — a reason to ask, “What happens next?” The exposition offers the history and information to put that character in context. It justifies why I’m asking you to read this singular story and get emotional about it, and it helps convince you that maybe this story isn’t so singular after all.Your prose is so compelling. Could you describe the process you used to write and revise your story?Thank you. This story went through a lot of drafts. The writing process was atypical, in that I needed to file a first draft before I’d finished reporting or traveled to Oklahoma. I started by making a timeline of Kerry’s life and relevant events. I also made a short outline focused on structure—sketching out how I envisioned the beginning, middle and end. Then I made a MUCH more detailed outline (like...28,000 words): I read all my source material, copying the relevant bits and pasting it underneath the section headings where I thought it would fit. Then I refined that outline into a series of progressively shorter, better-organized outlines, eliminating material that seemed redundant or less interesting. Eventually I started drafting, going through each section of the outline and writing them out like an actual story. As I did this, I cut more material and continued playing with structure. I waited and wrote a kicker after my reporting trip to Oklahoma. The revisions were intense: My editor, Maddie, did a first pass focused primarily on big-picture top edits, like rethinking the lede, encouraging me to find a “philosophical grand narrative,” offering me some focusing questions. We went back and forth a couple of times, trimming the piece down, then sent it to editor Amanda Silervman and editor-in-chief Clara Jeffery, who weighed in. The story probably went through five substantial rounds of edits, plus several final rounds that were more focused on small line edits, fact-checking, and legal tweaks.

AT THE POLICE STATION, the detectives reprimanded King. During the videotaped interrogation, a female officer brought up the moment when Purdy locked the door to Lilah’s room. “You should have went and got help,” another female officer told her. How did you obtain the video?It was submitted at trial as evidence. I requested it from the Tulsa district court.

“You guys don’t understand,” King said, her voice quivering. “I was so scared myself.” She ran her fingers through her hair, near the scar on the back of her head where another man—her ex-husband and the father of her three older kids—once hit her with the butt of a gun. She had another scar from him over her eyebrow. And another near her wrist, where Purdy had cut her. “I didn’t know what to do,” she told the detectives. “I wanted to get in there and grab her away from him and hold my baby.”

“The problem is, you already held her once…while he whipped the shit out of her,” the first officer said calmly. Why did you use this adverb to characterize the officer’s demeanor?There was something jarring about the juxtaposition—how calm the officer’s tone was, but how sharp and accusatory her words were. It was like the officers wanted to convince Kerry they were gentle, that she could trust them, but they also seemed to have already made up their minds against her. “You should have ran for help.”

“That’s why you’re going to jail today,” the second officer said. “Because of the things you didn’t do. Your job as the mom is to protect your child.”

“And you failed,” both detectives said in unison.

They handcuffed her and prepared to lead her away. “I love my kids,” King said, crying. “I’m not a bad mom. I’m not.” How did you decide what to select from this extended interrogation?I tried to highlight main themes: Kerry’s quotes and actions show how she’s a victim. The officers’ quotes show how they blame her—and use language that sum up what “failure to protect” laws are all about. And Kerry’s final line comes back to her driving motivation to be a good mom.

THE OFFICERS DID NOT seem to care about the abuse King had endured, abuse that began when she was just a child. She grew up in Stillwater, Oklahoma, the daughter of a Black mother who worked as a social worker and at the financial aid office of a junior college, and a white father who was a math instructor at Northeastern State University. King didn’t meet her father until she was 4.

When King was about 6, her mother, Lela Owens, grew suspicious after King accidentally peed in the driveway one afternoon while playing basketball. How do you know this?Primarily through interviews with Lela and Kerry. Owens took the girl to a therapist, who surmised she had been sexually abused multiple times. King remembers being molested by a preteen in a park when she was about 4.

As a kid, King liked animals and dreamed of becoming a veterinarian or a doctor; she wanted to take care of others. But by high school, King had lost much of that self-esteem and began hanging out with boys who used her for sex. “I wanted somebody to love me so bad, I didn’t care how they treated me,” she recalls. When she was 16, she met Ali Jordan Lalehparvaran, whom she would later marry. He was a couple of years older and seemed so sophisticated and kind. He would take her out for dinner or to the candy store for fudge. He bought her a diamond necklace. She’d never been treated that way by a boy before.

Months after they got together, Lalehparvaran learned that King had slept with other people before meeting him, and something seemed to snap. According to court records, one weekend they went boating and he smacked her on the head with an oar so hard she needed stitches. Another time, he broke her arm. How did you obtain these records?This came up during Kerry’s trial; I got the trial transcript by calling the courthouse and asking for the former court reporter’s contact information, then reaching out to her directly. I also obtained some of Kerry’s protective orders from the court. But “I felt like I deserved it,” she says. “Like there must be something wrong with me…I had been very promiscuous and felt like I couldn’t do any better.”

At 20, she got pregnant with their first child, Persia. She was thrilled—she and Lalehparvaran had been trying to conceive for a while. “I always wanted a family,” she tells me. She loved the feeling of the baby moving inside her. “It’s just the most beautiful thing ever,” she recalls. She imagined putting her child in colorful dresses and fixing her hair. “I was really excited about it. I could see myself being able to take care of a girl.” She hoped that having a daughter would calm Lalehparvaran down.

They married and had two more kids, William and Lilah, and the violence escalated. While she was pregnant with Lilah, Lalehparvaran pushed King up against a wall and broke her clavicle. She left him and went to stay with her friends. But because Lalehparvaran had paid her bills, a relative told her to go back to her husband, and she did. Another night, Lalehparvaran, who was drunk, smacked a gun into King’s head and shot up their home with an AK-47. As the bullets flew, King told the kids to lie on the ground, covering them with her body. “I was in survival mode. I wasn’t thinking of anything except, I have to protect my kids,” she recalls. Lalehparvaran was sentenced to seven years in prison; they divorced in 2013.

Afterward, King struggled. She had a job at a pharmacy that paid $10 an hour, but she had just $800 in her bank account, not enough to cover the mortgage and keep the power and heat on. She tried to help her young children, who were missing their father. She showed them how to bake banana bread and solve math equations. She regularly drove them an hour away to visit their dad’s mom. On birthdays she organized parties and baked special cakes. “She was a good mother,” says Kathleen Araujo, Lalehparvaran’s aunt, who still spends time with the children. “If she loved you, she would do anything for you,” adds Melissa Williams, Lalehparvaran’s mother. How many people in King’s orbit did you interview? How did you persuade them to talk with you?About 9-10 people, including her attorneys, her mom, three of her kids, her ex-husband, and Kathleen and Melissa. Initially, everyone seemed eager to talk because they believed Kerry was punished unfairly, and they hoped sharing her story would give her a better shot at a commutation or encourage lawmakers to amend the law.Her ex-husband, Lalehparvaran, was the toughest to convince He disliked that I was investigating his prior abuse of Kerr, and he feared I had a negative impression of him. With Kerry’s permission, I told him that his point of view was important if I wanted to understand the full picture. I said my reporting so far suggested that he had changed since he got out of prison, that he wanted to be a good dad and was doing the best he could for his kids. I also said it would be important for me to talk with him and the kids so I could capture how Kerry’s incarceration affected them. He relented and agreed to let me visit them, because he said his kids missed and needed their mom.

“I think a good mother is someone who can nurture them, loves them, and points them in the right direction,” King tells me. “Somebody that they can talk to, that they can rely on, that will always be there.”

She was 26 when she met John Purdy. Then 19, Purdy was handsome and athletic, with big brown eyes and chiseled, tattooed arms. And he made her laugh, a relief after such a volatile marriage. She sympathized with the challenges he’d overcome: As a boy, he’d been abused too. The bulk of your story is a profile of Kerry King. What advice would you give writers trying to create such a multi-faceted portrait?It’s important to build trust with the person you’re profiling and their loved ones. You can do that by spending a lot of time together, doing what you say you’ll do, being available when you say you’ll be available, showing compassion for how the interview process might affect them, and being kind and honest with them. Sometimes it helps to schedule regular check-ins: Even if you don’t have anything to discuss for the story, you can learn what’s new in their life and get a better sense of them as a person. Sometimes I’ll even ask how they describe their own personality or demeanor. It can be helpful to see old letters they wrot, or diaries they kept about a relevant event, and to talk with their family and friends. Before you start writing, ask yourself focusing questions, like: What is the main quality, ambition, skill or desire that readers must understand about this person? What is it about their story that is relevant to larger policy debates? In choosing which parts of their background to include, which anecdotes or details relate to themes you’re trying to convey?

In 2013, King got pregnant with their child, Trinity. “I was excited but kind of scared, because I wasn’t sure how he felt about it,” she says. Soon, court records show, the relationship took a turn. Purdy falsely accused her of getting pregnant by someone else. He started controlling her—dictating everything from her hairstyle to when and how she could use her phone.

Violence followed, she would later testify: One time he sliced through her calf with a kitchen knife and left a 2-inch cut. Sometimes he backhanded her or choked her. He did heroin in the house and demanded she join, even holding her arm down while his friend injected her. How do you know these things are true? What steps did you take to verify King’s story?These allegations came from court records: trial transcripts, restraining orders, etc. King’s story was also consistent over time—she told me the same version of events that she shared with law enforcement. Her mom offered corroboration. I didn’t speak with Purdy, but he admitted in court to abusing heroin; he did not admit to cutting King, but transcripts of his jail calls showed he threatened to carve his name into her skin. I tried to interview their roommate but never heard back. We ultimately attributed this graf to King’s testimony, rather than stating it as a matter of a settled fact.

She imagined kicking him out but worried he would retaliate. “I was really scared, more than anything. I felt kind of trapped,” she says. In fact, experts say that leaving an abuser can be dangerous for women, and that many mothers with kids struggle to get away.

Purdy sometimes apologized, vowing to be better. “I believed him when he said he was never gonna do it again,” King later told investigators.

But he did.

AFTER PURDY was arrested, Lilah went from the hospital to foster care and then to her paternal grandmother’s house, where she would stay as the investigation continued. King was devastated: “I was just in shock, just completely distraught, like I just didn’t know what to do,” she recalls of the separation from her kids. (Lilah’s older siblings, who had been living with King’s mom near Chicago, were allowed to stay there.) “It’s like my greatest fear came true.”

In jail, King was put on suicide watch, and she shivered and sweat as she withdrew from heroin. Her housemate bailed her out a few weeks later, but a court soon terminated King’s parental rights to Trinity, who was just 1 year old. Purdy also lost parental rights, but Trinity was adopted by his friend, a man who has prevented King from talking with the girl. Were you able to interview the friend to get his side of the story?Unfortunately I wasn’t, but I did get some corroboration from King’s mom. And I should note that after the story came out, Purdy’s friend allowed Trinity to get back in touch with King. It was “like my world was crushed,” King says. Unable to pay for a lawyer, King lost track of her court schedule and missed a hearing. As a result, she landed back in jail.

If she had any hope of getting her children back, King had to think carefully about her legal options. Her mom urged her not to plead guilty and instead to go to trial. It’s something she’s regretted ever since. “I gave her the most awful advice ever,” Owens tells me now. “I knew she wasn’t really guilty of anything, except being used and abused…[But] I should have told her to take the plea bargain.” At trial, “she went through hell, and nobody cared.”

FAILURE-TO-PROTECT laws sprang from changes to child abuse protocols in the 1960s, as doctors became obligated to report signs of mistreatment. The idea was to compel parents who witnessed violence to take action. But the provisions didn’t become commonplace until after a high-profile child abuse case in the late ’80s led to a media frenzy and one of New York’s first televised trials. How did you learn the history of the laws?I interviewed a lot of subject-matter experts, especially attorneys and advocacy groups, and I read law review articles and other scholarly work. I also read previous journalism articles; BuzzFeed had a fantastic investigation from 2014. How long did the story take, from concept through final revisions? After getting the idea in November 2019, I made a few calls but couldn’t prioritize the story because I was already working on another investigative piece. When the pandemic struck in March 2020, there was so much breaking news to cover that I had to put the idea on hold indefinitely. In the summer of 2021, when I finally had more time, I applied for and got the grant. I then spent a year reporting, writing and revising the article and documentary. We published the package online in August 2022.

In 1987, according to court records, 6-year-old Lisa Steinberg died in New York City after Joel Steinberg, an attorney who had illegally adopted her, beat her unconscious. His girlfriend, Hedda Nussbaum, a former Random House editor who helped care for Lisa, remained with the dying girl for about 12 hours without calling the police; Steinberg meanwhile freebased crack cocaine in their Greenwich Village apartment. Nussbaum, also high, said she believed Steinberg had supernatural healing powers. Only when Lisa stopped breathing did Nussbaum finally urge Steinberg to dial 911.

Prosecutors initially charged them both but dropped the charges against Nussbaum when the abuse she’d endured became evident, both by her deeply misshapen face and by X-rays and exams that revealed her to be anemic and malnourished, with broken bones and chronic infections. Doctors and other witnesses testified that years of beatings from Steinberg had left Nussbaum traumatized and physically incapable of wounding Lisa, or of intervening to protect the girl.

After watching Nussbaum testify, the evidence of her abuse clearly on display, the public became deeply divided. Some saw her as a victim, but others viewed her as a co-conspirator. A People magazine cover showed an image of young Lisa and the question “How could any mother, no matter how battered, fail to help her dying child?” Nussbaum’s critics wondered why she had covered up Steinberg’s violence, and why she hadn’t cried when the police arrived at their brownstone for Lisa. In court, she even testified that she “loved Joel more than ever” while Lisa lay dying.

“Why was Hedda Nussbaum given a walk?” Washington Post columnist Richard Cohen wrote. She was “a mother who did nothing as her daughter was brutalized—who put up with the most incredible indignities herself and who, even as Lisa was taken to the hospital in a terminal coma, attempted to provide Steinberg with an alibi.” Why did you rely on court records for such a  highly-publicized case? How did you obtain them?We cited court records because of the legal sensitivity; plus, it’s nice to fact-check with primary sources rather than relying on news clips. I found these court records on CaseText.com from a Google search.How did you learn about the People cover and Cohen’s column?I read as much news coverage as I could find, including this Slate article that mentioned both.

The public backlash against Nussbaum likely helped spur lawmakers and prosecutors to ramp up passing and enforcing failure-to-protect laws, says Karla Fischer, an attorney and expert witness in Illinois who has assisted the defense of women in dozens of these cases around the country. Why did you insert “likely” as a qualifier?Fischer and other experts couldn’t prove what led to the uptick in prosecutions, but this was their hypothesis. Fischer believes prosecutors are especially hard on these mothers because of political pressure, perceived or real.

Oklahoma approved its first failure­to-protect law in 2000. It’s one of several states, including Texas, West Virginia, and South Carolina, that allow maximum sentences of life in prison for the offense. But these prosecutions are not just a red-state phenomenon. They’re “a problem nationally,” says Colby Lenz of the nonprofit Survived & Punished, who points to California and Illinois as two Democratic-leaning states where parents are often imprisoned for similar crimes. What role did such advocacy groups play as you worked on this story?As I was starting to report, I spoke with experts from five or six advocacy groups, plus scholars and attorneys, to get a sense of the landscape. I wanted to know what they perceived as the main problems, where they thought these prosecutions were particularly common, which cases jumped out at them and what reporting challenges they thought I might encounter. I also asked for advice about how we might collect data. The ACLU of Oklahoma was particularly helpful, since its researchers had already done a lot of data work in their state, but the other groups offered good insights, too. I found recent cases across the country, from Massachusetts to Michigan to North Dakota to New York.

Today, at least 29 states have laws that explicitly criminalize parents for failure to protect against abuse. How did you nail down this figure?In 2014, BuzzFeed compiled a list of each state’s failure-to-protect laws. My fact-checker and I went through that list, looked up all the laws, and ensured they were still on the books today.Could you describe the fact-checking process?Mother Jones has an intense fact-checking process for magazine stories like this one. I started by annotating a draft of my article — basically inserting footnotes that showed my sourcing for just about every sentence or explained how I reported it, so Ruth could retrace my steps. That first annotation was about 40 pages. I also organized all my source material and shared it with her. She went through the piece and verified every word, making a great effort to ensure that legally sensitive sentences could be traced back to privileged documents like court records. She and I also did some fact-checking interviews with my sources. We reviewed the article with a Mother Jones attorney as well. All told, Ruth spent an estimated 160 hours on fact-checking.In many places that don’t, prosecutors take similar actions under more general laws—like charging a woman with murder even if her boyfriend killed the child, and then using legal theories about failure to protect to convict her. Some prosecutors are now extending similar logic to fetuses, charging women who self-abort. “There seems to be a strange obsession with our lawmakers when it comes to asserting control over women’s lives,” says Wright, the women’s rights activist.

And it is almost always women who are held responsible. Alexandra Chambers, an adjunct professor at Vanderbilt who is tracking these cases in Tennessee, sees a religious underpinning of these prosecutions—springing from the Christian myth of Eve, who was blamed for the fall of Eden, and sexist traditions that dictate women be the “moral center” who rein in men’s worst impulses. “Women are judged by what their partner does in a way that men aren’t,” she says. “And it can be seen as a moral failing that she didn’t have the moderating influence” to stop the abuse.

“It becomes insurmountable, the number of things [women] have to do in order to be in compliance with what we think is a good mother,” Colleen McCarty, an attorney who worked on commutations in these cases, told Tulsa People in 2019. Why did you use a secondary source instead of interviewing McCarty yourself?I actually did interview her — she makes a big appearance in the documentary. But I loved this quote and she didn’t say anything quite like it to me.  Clorinda Archuleta, an Oklahoma mother who’s serving a life sentence for neglect and permitting abuse while her boyfriend serves 25 years for the same charges, believes she was punished so harshly because she didn’t appear sorry enough; according to a local news organization, the prosecutor described her emotional demeanor as “flat.” Did you speak with Archuleta?No. I spoke with one of her attorneys, looked at her court records and pulled from news clips.

Juries are also much more likely to deem Black women as bad mothers, reflecting a structurally racist legal system. Black women face higher incarceration rates than white, Hispanic, and Asian women, and they’re more likely to experience domestic violence and poverty. Black parents also face stricter scrutiny from the child welfare system, which investigates more than half of all Black kids nationally, according to a 2017 study in the American Journal of Public Health. “Black families get scrutiny that white families don’t get,” says Cindene Pezzell, the legal director of the National Clearinghouse for the Defense of Battered Women.

On top of all this, failure-to-protect laws ignore how often abuse of a child overlaps with abuse of a parent. One 2006 study sponsored by the Justice Department found that kids are more likely to face mistreatment by either parent if the mother is being beaten by her partner. Your reporting is so thorough. How did you keep track of the information you gathered?I collected thousands of pages of court records, plus all the other court evidence I’ve mentioned (photos, audio, and videos), interview notes and audio from dozens of sources, plenty of studies and academic articles, state statutes and a lot more that I’m probably forgetting. I organized these things into folders (within folders within folders) on my laptop, trying to label things carefully so I could find them later.We also did a lot of data reporting. My former colleague Ryan Little, who specializes in data, scraped more than 1.5 million court records from Oklahoma’s state court websites so we could figure out how common these prosecutions are and whether they are disproportionately leveled against women. I’m not sure how he organized the raw data—he stored it on his own computers and did a lot of the technical analysis himself, while I helped guide him by asking outside experts for suggestions about our methodology, pointing him to particular sources of data, and giving him focusing questions for what we were looking for.

I also read hundreds of news clips about child abuse to find dozens of recent failure-to-protect cases outside of Oklahoma. I organized those links into a spreadsheet, which I color-coded, with information about the location, defendant, type of charge and time peg. For the documentary, filmmaker Mark Helenowski kept a lot of the raw footage and audio on his computer. I organized other footage (mostly from court evidence) into folders on my own. This project was a big team effort with so many contributors. Memos, Zoom meetings, and shared Google Drive folders helped us all work together and stay on the same page. A survey of 6,000 American families found that half of men who frequently assaulted their wives also frequently harmed their children. So it’s not surprising that so many mothers locked in prison for failure to protect are also victims themselves: In Oklahoma, roughly half of the women convicted under the law between 2009 and 2018 were experiencing intimate partner violence, according to an ACLU analysis of 13 of the state’s counties. “The law should treat someone who’s a co-victim as a co-victim, as someone who is in need of support and resources, and not as a co-defendant, as someone who is in need of prosecution and incarceration,” says the ACLU’s Lambert.

Only a handful of states make exceptions for domestic violence survivors. “We weren’t thinking about domestic violence,” former Oklahoma state Rep. Jari Askins, who wrote the state’s failure-to-protect law, told BuzzFeed News in 2014. Why didn’t you interview her?I reached out to her but didn’t hear back. Askins argued that women in abusive relationships could tell the court about that history to receive leniency. In reality, such history is often used against them, as prosecutors convince juries that suffering through years of abuse without leaving is a sign of bad parenting. What do you base this conclusion on?This is something I heard from a lot of attorneys, scholars and people directly impacted by these prosecutions. It’s also something I saw firsthand in the transcript for King’s trial.

“It’s hard for people who are on the outside looking in to understand how someone could harm their kids or submit to an abuser’s demand to do that, but survivors know what the consequence of each choice is,” says Pezzell. A mother, for instance, might follow orders to hold her daughter down or even hit her child because she believes that obeying his commands will keep him from going harder against the kid. And for these women, dialing 911 can be dangerous. “If the police don’t believe her and send her home, then she feels he’s gonna kill [her]. Why did you put this word in brackets?I tweaked that word for clarity. Originally, Fischer said “me” rather than “her,” which made sense during the conversation but seemed confusing on the page. And if she’s dead, he has unfettered access to the child,” says Fischer, the attorney in Illinois. “Battered women are forced into a position where they think differently about what’s safe and what’s not.”

King says society has unrealistic expectations. “They think we’re supposed to be like he-man women, like super strong and able to beat men down whenever they come at us,” she says. “It doesn’t make any sense to me, how you could expect us to be able to take down a man.” (When survivors do kill their abusers, they frequently end up in prison.)

Courts often don’t consider the catch-22 that such moms find themselves in. In 2015, a man in Frederick, Oklahoma, fractured a toddler’s skull, and the boy later died. The jurors convicted the man and recommended a 17-year prison sentence. Another jury recommended that his girlfriend, who was the boy’s mother and had been asleep during the injury, go to prison for life. “A lot of the disproportionate treatment is rooted to our psyche in this state: We tend to get madder at the woman who we felt like didn’t protect the kid than we do the man who abused the kid,” says Tim Laughlin, executive director of the Oklahoma Indigent Defense System, which defended the mother in Frederick. “It seems to be part of our collective consciousness,” he adds. In Oklahoma, at least 15 women accused of failure to protect received longer sentences than their male partners who were accused of abuse. On average, these women got 20 years of prison or probation, while the men got less than eight.

KERRY KING AWOKE before dawn on the first day of her trial, on Halloween in 2016. “Inside I was like an earthquake” of nerves, she recalls. A guard handed her a plum-colored dress that her attorney had sent for her to wear to court. Nobody had asked for her size, and it was far too small, the fabric clinging tightly to her hips and thighs. “I felt immodest, like I was stuffed into it,” she says. “Like, how would that look to the jury? I was so upset about that dress.”

When she arrived at the Tulsa County courthouse in the plum dress, King faced prosecutor Sarah McAmis, who was seeking a decades-­long sentence. McAmis had told Tulsa World that it disturbed her when mothers entered romantic relationships with men they didn’t know well and then chose not to intervene against abuse after warning signs. “If you bring a child into this world,” she said, “you must do everything necessary to protect the child.” She said she believed child murder victims “are watching from heaven and…know we are able to get justice for them.” And previously she had employed an extreme tactic to win in another trial, punching and kicking a doll in front of the jury to demonstrate the injuries to a child, even though there was no evidence the child had been kicked in real life. Why use a secondary source here? McAmis declined to speak with me. The Tulsa World article from 2016 was not specifically about King’s case, but it conveyed McAmis’ general approach to child abuse prosecutions.

Her boss, Tulsa District Attorney Steve Kunzweiler, told The New Yorker in 2018 that a prosecutor’s job was to “teach people the morals they either never learned or they somehow forgot.” His approach to criminal defendants, he added, was similar to the way he disciplined his daughters: “There are times when your kids need a lecture, times when they need a grounding, and times when they need a spanking.”

McAmis began her opening statement to the three men and nine women of the jury, all but one of whom appeared white, by painting King as complicit: “[W]hen [Purdy] came back with the belt, this defendant held her daughter down.” Were you present for the trial? If not, did you reconstruct the events?I wasn’t there, but I read the entire trial transcript and listened to a lot of the trial audio. I also interviewed people who were there, including one of the jurors. The prosecutor declined to talk with me, but I interviewed a spokesperson at her office. McAmis soon called Kristi Simpson, a child welfare investigator, to the stand, and asked whether King should have done more after discovering the bruises on Lilah in the bathtub. King had stayed home from work afterward to watch her, but “it was not enough,” Simpson replied. At the hospital after the final attack, Lilah could not stay calm while she received a CT scan or an MRI. “We were trying to explain it to her,” Simpson recalled of the procedure. “And she was crying and saying, ‘Please don’t. I’ll be good. I’ll be good. I’m sorry.’”

“Myself and the medical professionals, we teared up,” Simpson added. “We had not seen anything like this.”

At a table beside her attorney, King cringed. The prosecutor’s questions, she recalled, felt “like a battering ram.” McAmis emphasized to the jury that King did not call 911 to save Lilah. And it was true, she never did. But King had tried to get help.

As King would later share in court, she had asked Purdy to shower with her when he emerged from Lilah’s room around 4 a.m. After the shower, Purdy briefly left King’s phone within her reach. When he walked away, she texted, “Help me,” to her mom, then deleted the message from the phone so he wouldn’t see it. Since you have already recounted the horrors of that night, how did you choose what to include in this section that returns to it?I liked the idea of transitioning from this moment where Kerry and Lilah couldn’t speak—Purdy’s hand was literally over Lilah’s mouth—to the next section, where Kerry is taking the stand to finally share her side of the story. I also thought it was important to elaborate on the question of why Kerry didn’t call 911.

Moments later, the phone rang. When Purdy answered, Owens, King’s mother, who’d been roused from a deep sleep by the text, was on the other side, asking for answers. Purdy hung up and then pushed King up against the linen cabinet. What did you say to your mom? he asked before hitting her in the face. What is your source for this?Most of this graf comes from the video of King’s police interrogation. (I also got some corroboration from Owens.) King told me the detail about Purdy’s hand on Lilah’s mouth. Purdy demanded that she call Owens back to say they were all just fine. So she picked up the phone and dialed.

Everything is just fine, she told her mother, trying to hide her terror as he sat on the couch opposite her, holding Lilah’s mouth closed so she wouldn’t cry out.

ON THE FOURTH DAY of the trial, King took the stand, hoping to finally share her side of the story. But during the cross-­examination, McAmis “got under Kerry’s skin,” her attorney Brian Boeheim says. “And that nice, sweet girl that was testifying got angry and got snippy with the prosecutor. That was it for the jury. They saw her as someone who could be aggressive and could allow somebody to do this to a child.”

Testifying can be an agonizing experience for survivors of domestic violence, forcing them to recall painful moments in vivid detail. Prosecutors sometimes use the same intimidation tactics that abusers employ, retraumatizing them. How do you know this?Partly by interviewing attorneys, partly by reading Kerry’s trial transcript and speaking with her, partly by speaking with other people at the trial, and partly by reading law review articles and scholarly work, including a 2020 paper in the Arizona Law Review and a 2021 article connected to the Boston University School of Public Health. McAmis suggested that King was dishonest because a few parts of her testimony did not perfectly match statements she’d made immediately after the assault; decades of research show that some survivors can recall certain moments of abuse in great detail but, because of the fear response in their brains, may have only fragmented memories of other moments.

Throughout her questioning, McAmis often framed the abuse King had endured as evidence that she’d been a bad mother to her children. “Were they having fun after they saw you get pistol-whipped and the blood gushing from your head?” the prosecutor asked her, referring to the night when Lalehparvaran shot up their home while the kids were inside. And when King brought up the ways she had also been a victim, McAmis suggested she was unsympathetic to her daughter’s injuries. “Are you telling this jury you have had a more traumatic experience than what Lilah experienced that night?” McAmis asked, referring to Purdy’s beating. “Surely you are not minimizing what your daughter went through?” she added later. (McAmis did not respond to requests for comment.)

McAmis’ approach, says the ACLU’s Lambert, is “heartbreakingly indicative of the attitude behind failure to protect. Prosecutors are using this tool for the noble purpose of protecting children from child abuse, but with the horrific method of criminalizing and vilifying the nonabusive parent who is often the victim of that same abuse.”

Before King left the stand, she kept looking over at the judge, wondering, Are you for real gonna let her say these things to me?

“This is exactly what survivors fear will happen when they tell someone they are abused,” Lambert adds. “To not be believed, and even worse to be blamed.”

During closing statements, McAmis focused on how Purdy repeatedly called King from jail and King kept talking to him.

“Those jail phone calls, they’re disgusting,” McAmis told the jurors. Then she mockingly offered her interpretation of King during the calls: “‘Bummer for you, Lilah. Mommy is still in love. You’re on your own, kid.’” On the phone with Purdy, King said she loved him and even wanted to have another child with him.

That “sunk the ship,” Boeheim, her attorney, recalls. “You could just see the jury fold up.”

Boeheim had tried to tell the jury that during the calls, Purdy incessantly begged King to say she loved him and threatened to hurt her if she didn’t. At one point he said he’d take the baby away if King didn’t stay in a relationship with him. According to a court document, he said he’d shoot bullets at her kids, kidnap her, tie her up, and rape her:

Tell me you love me. When I get out of here I’m going to drop kick you. I’m going to kick your teeth in. I’m just kidding, baby. I’m just kidding. Tell me you love me. You want to marry me. Come on, just go run down to the courthouse and you can marry me. You can marry me. Come on, we’re soulmates. Tell me you love me. I’ve got to hear you love me. I’ll carve your face. I’ll carve my initials in your face. Come on, you know I’m just joking, baby. Quoting Purdy seems complete and word-for-word from the court document. Why did you set it off in italics?I like the look of it, but there are some other reasons: In the trial transcript, Boeheim reads this graf to the jury, quoting Purdy. When I went back to listen to the jail call recordings myself, I knew Purdy didn’t say all these things one right after the other; he was having conversations with Kerry.

King talked with Purdy initially because she “wanted to understand why he did what he did,” she tells me. She doesn’t know why she kept talking with him after that. But experts say it’s not uncommon for survivors to continue engaging with their abusers, and that it can be difficult to cut ties. “I really thought I loved him,” she says.

During his closing statement, Boeheim tried one last time to convince the jury that his client did not deserve punishment because she had not been the true perpetrator. “He’s the monster. He’s the predator,” he said of Purdy.

But he may have been fighting an uphill battle. In Oklahoma, “juries are extremely punitive toward women who have children being hurt in their home,” McCarty, the attorney who has worked on commutations, tells me. “It’s a cultural more, especially in the South or the Midwest, that our job as women is to take care of our children.”

The jury deliberated for less than an hour. King’s muscles tensed as the foreman read the verdict—guilty on both counts—and recommended 30 years in prison for each. King dropped her head and started to cry.

“It was a travesty,” Boeheim says.

“I was flabbergasted,” says Araujo, King’s ex-husband’s aunt, who attended the trial.

“It was like my life was over. Like I’m not gonna see my kids grow up,” King recalls.

Only later did the jury learn that Purdy took a plea for 18 years in prison—12 years less than one of King’s sentences. “I was sick to my stomach,” juror Misty Reed, now 37 and with children of her own, recounts. “If I would have known that the boyfriend had gotten way less years, it would have definitely changed my mind” about the proposed sentence for King, she says. “And I guarantee it would have changed many others on the jury as well.” (Other jurors did not respond to my calls or declined to comment.) How challenging was it to get Reed on the record?Challenging! I cold-called her (and texted and emailed) after finding her name in the trial transcripts. She texted back but then ghosted. I left another voicemail when I got to Oklahoma, adding that I wondered if I had some information about the case that she might not know. Then she called back, asking what I knew. She seemed nervous, especially about going on camera for the film, but she also wanted to help; she was upset about the disproportionate punishments and had spent years thinking about the trial. How would you describe your interviewing style?It really depends on the source and situation. I often start by briefly explaining what the story is about and why I want to talk with a source, and asking if they have any questions for me; I want to put them at ease. If I’m asking them about trauma, I’ll let them know they can take breaks or decline to answer questions or end the interview whenever they want. Again, the idea is to give them power and help them feel comfortable, and I don’t want to retraumatize them. I’ll have a detailed list of questions that I use as a guide, though I often depart from it or skip questions that seem less relevant. I may reiterate what I’m hearing, as a way to help them feel heard and ensure I’m actually understanding. I’ll embrace silences, giving them a chance to add more if I think that might lead to good material. I’ll normally end by asking if there’s anything else they want to add that I didn’t ask about, or if there’s anyone else they think I should talk with. Of course, my style is different if I’m interviewing an academic or an antagonistic government source; I might be quicker or focus less on giving them power and helping them feel heard. All interviews have slightly different goals, and so my interviewing style changes to reflect those goals.

With the trial over, King braced herself for the sentencing hearing. “All I ask is that you don’t make this the end of my life with my children,” she wrote to the court. The judge took some mercy and allowed her sentences to run concurrently. It was positive news, but little consolation: By the time she gets out of prison, she will be about 60 years old, and her children will be grown.

But for now they were still young and needed a permanent home. King had lost parental rights to her youngest daughter, Trinity, who was adopted by Purdy’s friend. What about the other three? The state of Oklahoma sent her daughter Lilah to live with Lilah’s father and King’s ex-husband, Lalehparvaran, despite knowing about his violent criminal record and history of substance abuse. Lilah’s older siblings, Persia and William, then 9 and 7, soon went to live with him too.

That state officials would condemn King as a bad mother for the way she had dealt with her ex-husband’s violence, and then send her young children to live with him while locking her away in prison, felt like the epitome of a double standard. “My heart sank,” King says. “I lost all faith in justice. I lost everything. What was I ever thinking in trusting a system that was not for me? I didn’t realize the depth of how much it wasn’t for me.”

KING ENTERED Mabel Bassett Correctional Center in McLoud, Oklahoma’s biggest prison for women, in May 2017. They handed her an orange jumpsuit that was too small, and a prisoner ID number and told her to memorize it. She would have one locker for her belongings: a bottle of shampoo, a tiny toothbrush with toothpaste, a bar of soap, two pairs of socks, two bras. Why was it important to report these details?Sometimes the more specific or intimate you get with the details, the more universal it feels, as Nieman Storyboard editor Jacqui Banaszynski put it in a writing class recently. If you just say she didn’t have a lot of belongings, that might not strike a reader, but everyone can imagine how they would feel if they only owned two pairs of socks.  Within a week, she watched someone get beaten with a shower rod.

In Oklahoma and around the country, the vast majority of incarcerated women are mothers: National data is scant, but studies have found that upward of 90 percent of them experienced physical or sexual violence before landing behind bars.

In 2018, King learned that she wasn’t the only person at Mabel Bassett serving time for failure to protect. Attorneys and advocates from the nonprofit Oklahomans for Criminal Justice Reform, frustrated by lawmakers’ lack of action to reduce mass incarceration, had begun a commutation campaign to help incarcerated people petition the state for shorter sentences. Lambert, the ACLU lawyer, joined the effort and introduced King to several other women in the prison who had been convicted under the same law. All were women of color. “It made me see how messed up this state was,” King tells me. “There are so many of us in here that really should not be in here.”

In November 2019, one of them received good news. Tondalo Hall, now 38, had already served half of her 30-year sentence for enabling child abuse. Her ex-boyfriend Robert Braxton broke several bones in two of her children’s legs; he served only two years in jail before he was released on probation. Oklahoma Gov. Kevin Stitt, a Republican, agreed to shorten Hall’s sentence as part of a mass commutation of more than 500 incarcerated people around the state—the biggest commutation in US history.

When Hall learned that she would be freed, she could hardly believe it. It had been more than a decade since she’d spent much time with her kids; two of her children were younger than 2 when she was first incarcerated. When she walked out of the prison gates later that week, she started crying as she hugged her family, rocking back and forth with her son Robert, now a teenager. Did you witness the reunions?I watched a video of the reunion (filmed by a local news crew) and interviewed Tondalo about it. From inside the prison, inmates cheered Hall’s reunion loud enough for her to hear them. “I love all them ladies,” Hall said to TV cameras as she looked back toward them. “I’m coming back to help a lot of them.”

But since Hall’s release, Oklahoma’s pardon and parole board has rejected all the other mothers with similar cases who applied for a commutation. Other potential avenues for relief have also closed. In 2019, advocates pressured Oklahoma lawmakers to amend the failure-to-protect law to make it more lenient for survivors of domestic violence. Yet lawmakers balked after prosecutors claimed they needed the law’s life sentence provision to properly prosecute child abuse and to use as leverage to convince mothers to testify against their abusers. Around the same time, a committee charged by the legislature to reexamine the maximum punishments for a wide array of crimes recommended that people convicted of permitting child abuse be limited to 40 years. But that change has not yet been approved, and even if it were, it would not be retroactive, doing little for women like King.

In states like California, Illinois, and Tennessee, activists are trying to help incarcerated women in similar situations, but their cause hasn’t gotten much traction. “These are hard fights,” Lenz of Survived & Punished says of the moms pushing for mercy around the country. “There’s a fear for politicians of looking like they are excusing child abuse.”

Oklahoma continues to pursue failure­to-protect cases at a swift pace. In November, a Cleveland County jury convicted 30-year-old Rebecca Hogue of first-degree murder after her boyfriend beat her 2-year-old son to death while she was away at work. Her boyfriend later died by suicide in the Wichita Mountains Wildlife Refuge, where investigators found the words “Rebecca is Innocent” carved into a tree next to his body. The judge did not allow a photograph of the carving to be shown to the jury, deeming it “hearsay” on behalf of the boyfriend. The jury recommended that Hogue go to prison for life with the possibility of parole. How is this possible?She was convicted under the failure-to-protect law. I wrote another story about it, which you can read here.

DURING KING’S first year in prison, her children were not allowed to visit her. She was furious that Lalehparvaran had received custody of them. She didn’t think he would physically abuse them—he never had—but she worried he wouldn’t properly care for them. When they lived together, it had always been on her to hold the family together, to make sure the kids had warm coats and gloves for the winter. How and why did you decide on what appears to be a near-chronological structure for Kerry King’s narrative?Sometimes writing chronologically is the simplest; it’s easy to follow and how a lot of us naturally tell stories.

When mothers are convicted of child abuse crimes, some children are sent to foster care and others to live with relatives, but it’s not unheard of for them to go to a partner or ex who was previously abusive, or to the ex’s family, according to attorneys who monitor these cases. After a mother named Ashley Garrison went to prison in Oklahoma about a decade ago for child neglect, her abusive ex-husband got full custody of the kids and prohibited Garrison from contacting them, according to Tulsa People. Garrison, in handcuffs, gave birth to another child the day she was sentenced, but the baby soon went up for adoption and her parental rights were terminated. She attempted suicide in one of the prison showers. How many of these kinds of cases did you turn up in your investigation? How did you decide which ones to use?This graf is mostly based on comments from experts; I didn’t track where the kids went for the hundreds of failure-to-protect cases we found. Beyond King’s case and Garrison’s, I’d heard of one other woman in Oklahoma whose kids went to live with her co-defendant’s mother. I thought Garrison’s case was compelling and tragically demonstrated how much pain it could cause to separate a mother from her children.

Fortunately, Lalehparvaran seemed to have turned over a new leaf. “After prison, he’s changed; he’s a hard worker,” said Williams, his mother. Lacking a driver’s license, he rode a bicycle to the classes he needed to get certified for his HVAC job. He made sure the kids did their homework and the chores. “I think they like living with their dad. They feel safe,” Owens, King’s mother, told me. The kids said the same thing.

But even when there’s a decent guardian to step forward, losing a parent to prison can have ripple effects. According to a study by the federal government, kids with incarcerated mothers or fathers are more likely to struggle with mental health conditions like anxiety and depression, and they face a higher likelihood of medical problems as adults, from asthma and migraines to substance abuse. Teenagers with incarcerated mothers are significantly more likely than other teenagers to drop out of school and be arrested themselves.

King tried to stay in touch with her kids however she could. She called them several times a week and wrote them letters. When Persia eventually started her period, King drew diagrams to show her how to use pads. She knitted them socks and blankets with yarn she’d bought at the commissary.

But Trinity’s adoptive guardian, Purdy’s friend, doesn’t answer when King calls. King heard that Trinity likes unicorns, so she sent her a unicorn stuffed animal. No reply. She hasn’t been able to learn much else about her daughter.

About a year into her stay at Mabel Bassett, King’s older kids were finally approved for visitation rights, and they’ve been going to meet her about once a month ever since. After Christmas, King looked forward to seeing them in January, when she hoped she and Lilah might continue their conversation. But then King and Lalehparvaran had a disagreement over the phone, and he refused to bring them. She was devastated, she tells me. After all these years, she still had so little control in their relationship and much of the rest of her life.

The day after their missed visit, I met the kids at their house in Stillwater, the same city where King grew up. That evening, they planned to celebrate Lilah’s 11th birthday. In the afternoon, they sat side by side on the couch as they flipped through an album with old family photos, giggling about how they looked as babies.

But Lilah’s cheerful demeanor faded a few hours later, when, in the privacy of her bedroom, she began to talk with me about her mother’s incarceration. It still confused her. “Like, I kind of know why she’s in jail, but I know she’s not supposed to be there,” Lilah said, beginning to cry.

“I just really miss her. I just want to talk to her,” she said, imagining what it might be like to see her mom every day, not just once a month. “You can’t really make memories on a phone.” How do you gain a child’s trust so that she will confide in you? Did such an interview raise any ethical questions in your mind?I wanted to know how the kids viewed King’s incarceration and how it changed their lives, but I worried about interviewing them because they were so young. I think I gained their trust by gaining the trust of their parents. King and her ex-husband approved of me meeting them and asked the kids if they would talk with me. When I arrived at their house, I spent some time just hanging out and getting to know them, asking for a tour of their rooms, asking them about their hobbies, before we started formal interviews. As I would do when interviewing an adult about something traumatic, I asked if they were okay talking about the issue and told them they could skip questions or take breaks or stop the interview at any time. I tried to physically position myself at their level; for one interview I sat on the floor, because King’s son was sitting on a very low bed. I did not ask them to rehash any of the violence they witnessed, because I didn’t want to retraumatize them, and I focused mostly on how they felt about their mom being in prison. I asked if they were okay with me writing about their family’s story and naming them, and they said yes because they hoped it might help their mom get a commutation or encourage lawmakers to change the law. There were a lot of ethical questions, including whether to name them in the story. We decided to use their first names but not their last, with their parents’ permission. Using fake names didn’t make much sense because they were appearing in the film, so they would be recognizable.

She composed herself and went back downstairs, where she ate dinner with her family and joined a game of hide-and-seek with her siblings. Then they all gathered around the table for cake. Lilah blew out the candles and smiled as they sang to her. Her brother, William, snuck a taste of the white icing speckled with rainbow sprinkles.

Dozens of miles away, King sat in prison, missing it all. In May, her ex-husband, Lalehparvaran, was arrested and charged with burglary and reckless handling of a firearm. The kids went to stay with another relative.

Lately King wonders whether someone might take mercy and grant her a commutation, let her “get out of here, be there for them, and be the mom that I always knew that I was.” Reed, the juror who spoke with me, recently got in touch with King and apologized for recommending such a long sentence.

During one of our many conversations, I ask King what she wishes people would understand about her case. “I was just as much of a victim as my daughter was, and they should not just look at that one day,” she says. “They should look at every day and see all that I went through, and how I had gotten to the point of that day.” Why did you choose this comment to end your story?We considered many alternatives. In early drafts, I ended with Kerry in prison, missing her children, but used different kicker quotes that emphasized her desire to be a good mom and her sadness at being separated. I also toyed with ending on a quote from Lilah, explaining how much she missed her mom. Ultimately, we chose to give Kerry the power of the last word because this is really her story, and we used her quote above because it emphasized two key points: that she is a survivor, too, and that those who condemn her should remember everything that led to her incarceration, and all the ways that cops and courts and everyone else failed to protect her when she needed it.

* * *

Chip Scanlan is an award-winning newspaper reporter who taught writing at The Poynter Institute for 15  years. He publishes Chip’s Writing Lessons and has two books, ”Writers on Writing” and “33 Ways Not To Screw Up Your Journalism."

The post How a reporter documented the failure of “failure-to-protect” laws appeared first on Nieman Storyboard.

It was the stuff of great narrative, a dramatic saga with conflicting storylines and no clear resolution:

In October 2020, British authorities and the media reported that seven stowaways from Nigeria were aboard a mammoth oil tanker as it approached the British coast. The seven Nigerians had braved the elements for days, huddled on an uncovered platform above the tanker’s rudder before they were discovered and brought on board. Days later, the ship’s captain broadcast a distress signal: The stowaways had hijacked the ship.

British commandos stormed the Nave Andromeda and took the stowaways in custody. Prosecutors, however, declined to press charges, arguing there was no evidence of a violent altercation. The men were kept in immigration custody and then relocated across England as they started the lengthy process of seeking asylum.

Little about it made sense, and British freelance journalist Samira Shackle wanted to know more. But to get past the official and obscure accounts, she and David Wolf, editor at the Guardian Long Reads, knew they would need to get to the people closest to the action.

“So we decided that I’d have a go at finding people who were on board the ship, both crew and stowaways,” Shackle told me.

But how? Three months had passed since the primary events, and the story had gone cold. Only two of the stowaways had been identified by name in the media. The Nave Andromeda was back at sea and the identities of its crew were not included in public reports. British authorities were tight-lipped.

Those challenges just added to Shackle’s determination to get to the bottom of the story: “I have a long-running interest in revisiting stories that have been hyped up in Britain’s particular media and political culture and then promptly forgotten about.”

She also has learned from previous stories that the passage of time can be an advantage, giving sources “more time and space to reflect on events.” That proved true in this case. The stowaways and three crew members she tracked down provided vivid details about what actually took place on the tanker — a narrative at odds with the histrionic coverage when the story first broke.

Her 5,700-word story, “Seven stowaways and a hijacked tanker: The strange case of the Nave Andromeda,” ran in the Guardian in June 2022. It brims with authority, based on numerous interviews with eyewitnesses, academics and authors familiar with hijackings on the high seas and the byzantine world of maritime finance and regulation.

Shackle’s reporting was interrupted for several months by pregnancy. Rather than let the story drop, she created a solution useful for writers at any time:

“When I realized I wouldn’t be able to finish it before having the baby, I wrote a detailed bullet-point plan highlighting where there were reporting gaps, so that I would have something coherent to return to rather than a mass of disorganized notes and transcripts.”

Back at work after maternity leave, she turned to organization, a crucial part of the writing process. She plotted out the story in five sections, following the structure of classic narrative: exposition, rising action, climax, falling action and resolution. Narrative specialists often rely on these elements, especially when intensive reporting captures a wealth of dramatic scenes, characters and dialogue that leads, as Shackle’s story does, to revelatations in the public interest.

“I find this very useful when thinking about how to turn reporting into a story,” she said.

Shackle is a regular contributor to the Guardian. She is the author of “Karachi Vice: Life and Death in a Divided City,” an account of the lives of five ordinary people in Pakistan’s largest city.

Shackle answered questions by email about the reporting challenges she confronted, the value of “draft zero” and how she folds exposition into places where it illuminates the story. The interview has been edited for length and clarity, and is followed by an annotation of the main narrative.

How does the story fit into your reporting portfolio?
Over the years I’ve taken this approach on a wide range of subjects, returning to a story that hit the headlines in an inflammatory or hysterical way, then speaking to as many people on the ground as possible to build up a sense of what really happened. These include a 2017 report on the Trojan Horse scandal in Birmingham schools (which has since been the subject of a “Serial” podcast), which took 18 months to report; a 2018 look at the UK government’s attempt to investigate British soldiers over alleged war crimes in Iraq; and a 2020 investigation into the Gatwick drone. The way that this alleged hijacking had been portrayed as a terrifying threat to British shores fit in with the increasingly hysterical way that migration and particularly asylum seekers are portrayed in the media and by politicians, so I was interested in going back and reporting it in a more considered way.

Could you describe your writing process?
When I write for the Guardian Long Read, I typically gather a lot of material, and then when I think I’m mostly done with the reporting, I put together a detailed bullet-point plan. This is usually set out in five sections, and I bullet point in detail what will be included in each. I always start this process thinking I don’t have enough material and then end up writing a plan of 3000 words for a story that’s only meant to be 5000 words. Then I send this to my editor to look over and give feedback on, and I use this as the basis for writing the piece. In some ways, this is the stage of writing that I find most difficult, when you have a vast mass of material and you’re faced with a series of decisions about how best to order it and make it hang together.

A story like this has a clear timeline, so that made it slightly easier to structure. I don’t always end up sticking exactly to my plan when I sit down to write, but in this case I mostly did, except for the fact that when I wrote the plan I had yet to speak to Michael, and he ended up featuring very heavily.

Once I actually start writing, I work more instinctively. It’s been quite interesting for me to reflect on these questions about why I made particular structural or word choices, because I often don’t think too much while I’m doing it. I tend to write, fairly quickly, a very rough draft — which I call “draft zero” — with the aim of simply getting all the words down on a page in roughly the right order, without worrying about making it good. This draft zero is usually way too long and very messy. Then I spend some time going through that, tightening it up, moving things around and cutting words out, until it’s formed into an acceptable first draft.

How long did it take you to report, write and fact-check this piece story?
Inn all it was about 16 months, but with a gap of several months in the middle while I was on maternity leave. In terms of the reporting itself, the first stage was to see if I’d be able to find anyone at all who had been on board the ship. My editor said we needed to speak to an absolute minimum of three people to make the story viable, so I had something concrete to aim for. This was not an easy task as, other than the captain, the names of the crew were not in the public domain. At that stage, I didn’t even know their nationalities. And only two of the seven stowaways had been named in the press. So the first stage of identifying who I was looking for and then trying to track them down took quite a bit of time. I’d almost, but not quite, completed the reporting as my due date approached. The crucial thing that was missing was an interview with a second stowaway; at that stage I had only spoken to John. I took some time off work, but early in the new year I returned to the story, and managed to find Michael on Facebook. I met him in the spring. I also traveled to Manchester to interview John in person, as previously we’d spoken only on the phone. I completed a few other missing interviews, and wrote the piece up soon afterwards.

Are there writers you look to for inspiration as you work on your narrative nonfiction?
I love Patrick Radden Keefe’s work in The New Yorker; he’s so good at telling a really compelling story that completely immerses you. I’m lucky to have a number of brilliant friends and colleagues who also write for the Guardian’s Long Reads section, and I always enjoy and learn from their work, especially Sophie Elmhirst, Samanth Subramanian and Daniel Trilling. 

Annotation: Storyboard’s questions are in red; Shackle’s answers in blue.

Seven stowaways and a hijacked oil tanker: the strange case of the Nave Andromeda

In October 2020 an emergency call was received from a ship in British waters. After a full-scale commando raid, seven Nigerians were taken off in handcuffs – but no one was ever charged. What really happened on board?

By Samira Shackle

Shortly after 9 a.m. on 25 October 2020, the captain of the Nave Andromeda sent out a distress call. The crude oil tanker was situated six miles off the coast of the Isle of Wight, close enough to be visible from the pebble beaches that edge the island. In Greek-accented English, the captain, Antonis Perros, said that seven stowaways who had boarded the ship in Nigeria had escaped from the cabin where they were locked: “I try to keep them calm but I need immediately, immediately agency assistance.” For their safety, he said, most of the 22 members of the crew were now locked into a secure area of the ship known as the citadel. You begin the story in the middle of the action, after the stowaways have sneaked onto to tanker’s rudder. Why did you open your narrative here?I decided to open with the story that readers, at least in the UK, might be familiar with. It was only in the news for a few days, but it totally dominated headlines and evening news broadcasts for those days. This was almost two years before my story came out, so I wanted to remind readers of that with a dramatic tick-tock account of how it played out in the media. I also wanted to set out the overblown political response so it was in the readers’ mind before going back t0 offer a different version of events.

The local police force on the mainland, Hampshire constabulary, began coordinating a response. Policing the seas is complex, and they were in communication with the Maritime and Coastguard Agency and the UK Border Force. A three-mile exclusion zone was established around the ship. How did you verify that the reconstruction is accurate? A recording of the captain’s distress call was in the public domain. Other than that, I based this section on a combination of news reports from the time and what I’d found in my reporting. (Eg. I knew the size of the crew because I had a copy of the crew list). I also spoke with Hampshire constabulary’s press office who explained some background about policing the seas. I was keen to get an interview with one of the police officers involved, which they declined, but they did offer some useful information about how responses are coordinated.

At about lunchtime the story broke in the media. Isle of Wight Radio reported “an attempted hijacking”, and soon afterwards Hampshire police confirmed that there was an “ongoing incident”. By 3.45 p.m,, coastguard helicopters were circling the Nave Andromeda. The vessel was moving aimlessly, raising fears on shore that the captain had lost control. What were the sources for this paragraph that enabled you to use direct quotes and describe the state of the tanker?The direct quotes are from contemporaneous media reporting and from the police statement which is still online. The movements of any commercial vessel are tracked on specialist websites. This story had played out in real-time in the media in October 2020, and journalists on live broadcast and national newspaper blogs had been tracking the ship’s movements using these online tools. All stated that the ship appeared to be drifting and moving in zig-zags.

Hampshire police told journalists that the stowaways had made “verbal threats” to the crew. Apart from that, not much was known. Within government, there was anxiety. “There’s all sorts of directions this could have gone in: the ship’s crew assassinated, the ship damaged in some way and hitting the coastline, or itself being used as some form of weapon to drive in and hit a port,” Tobias Ellwood, Conservative MP and chair of the Commons defence select committee, told me 18 months after the incident. This seems like s a critical interview. How did you land it?I’d originally tried to get interviews with the defence secretary Ben Wallace, or anyone at the Home Office, but they said a firm no. So I was relieved that I was able to get the official perspective through this interview with Ellwood, who is not a minister but still had some firsthand knowledge of events. It took several months of back and forth.  “We’re talking about minute-by-minute decision-making.”

The police requested military assistance, and later that afternoon, home secretary Priti Patel and defence minister Ben Wallace gave the go-ahead for an operation by the navy’s Special Boat Service (SBS). At around 7.30 pm, the operation began: 16 elite troops from the SBS stormed the tanker by sea and air, backed by airborne snipers. Commandos fast-roped on to the deck from Wildcat helicopters and scaled the ship’s hull from high-powered inflatable boats. “Fast-roped” is a great action verb, but not a typical one. How did you come to use it?I’d read it used in descriptions of military operations and thought it conveyed it well. The operation – which took more than 10 hours to coordinate – was over within nine minutes. Before 8pm, the ship was secured, the stowaways handcuffed and awaiting arrest. Soon afterwards, the Nave Andromeda was brought into dock at Southampton. The seven stowaways were arrested on suspicion of “seizing or exercising control of a ship by use of threats or force”. They were led off the Nave Andromeda in handcuffs, past a sign announcing: “Welcome to the port of Southampton, gateway to the world.” What were the sources you used to reconstruct this military operation?I relied on the extensive news reports from those days, and where possible I corroborated details with interviewees. I visited Southampton Port and was shown the spot where the Nave Andromeda docked, which is how I knew about the sign.

The government ministers involved in the operation were keen to highlight its speed and success. “In dark skies, and worsening weather, we should all be grateful for our brave personnel. People are safe tonight thanks to their efforts,” said Wallace in a statement released 40 minutes after the ship was secured. Patel thanked police and armed forces for their “quick and decisive action”.

Although it was still not clear exactly what had happened on the ship – that evening, lawyers for the company that owned the Nave Andromeda said it was “not a hijacking” – Wallace told journalists the next day that there was “a clear threat to life on the ship”. The Daily Mail reported that “stowaways smashed glass on board and made threats to kill”. Former Royal Navy Rear Admiral Chris Parry told Good Morning Britain: “Next time they could be terrorists.” This paragraph appears to bring to a close your tick-tock reconstruction of the events of Oct. 25, 2020. Why did you wrap it up here?I wanted to demonstrate the tenor of some of the coverage of the story before stepping back to explain the context of the media and political environment that it took place in. And this was also more or less where the story ended in the eyes of most people; the January 2021 report about all charges being dropped got much less coverage than the original ‘hijacking’.

The story of a heroic mission to defeat a hijacking in the Channel seemed to arrive at just the right time. In the summer of 2020, the number of migrants crossing the Channel had rapidly increased; in July and August more than 2,000 migrants had attempted the crossing, compared with 500 in the same period in 2019. Politicians and the media had seized on these crossings as evidence of Britain’s failure to protect its borders. On one occasion, the BBC deployed a journalist to stand at the cliffs and literally count the approaching boats. Just weeks before the Nave Andromeda incident, Patel had pledged to crack down on small boat crossings. According to documents later leaked to the press, proposed solutions had included building floating walls across the Channel – one of the world’s busiest shipping routes – and using water cannon to create waves that would push boats away from British shores. You step back here to put the incident on board the Nave Andromeda into a larger context of migrants attempting to cross the channel to England. Why?To me the two things seemed inextricably linked: a political culture which has a serious interest in emphasising and sometimes exaggerating the threats to Britain’s shores, and the way in which this incident was blown out of proportion. One of the advantages of returning to a story some time after it took place is that you can see these connections more clearly. And an advantage of writing at this length is being able to take a step back and look at the context.

By January, all charges against the stowaways on the Nave Andromeda had been dropped, after the Crown Prosecution Service (CPS) said that new evidence “cast doubt on whether the ship or the crew were in fact put in danger”. The story faded from public view, and the seven stowaways were left to make their way through Britain’s asylum system. But what really happened on board? While the tale of marauding Nigerian pirates took hold, accounts from the ship’s crew and the stowaways themselves suggest a very different chain of events. Why did you use a qualifying verb?Legally we needed to qualify throughout since some parties maintain a different version of events.It appears the story will go on to debunk official and media accounts of the incident. How did you arrive at the conclusion stated in the final sentence of this paragraph?Through months of reporting! The key point for me was that the stories told to me by the two stowaways and three Filipino crew members I spoke with all corroborated almost exactly, even though these people were not all in touch with each other. And those corroborated accounts undermined the official narrative and that maintained by the Greek officers on the ship.

* * *

MICHAEL DID NOT INTEND to stow away. Before 6 October 2020, the day he found himself climbing off a fishing boat 10 miles from the Nigerian coast and into the rudder of the Nave Andromeda, he had never even been on water. This is a brief but dramatic character sketch. Did you ever consider making it your opening?Briefly, but I was fairly certain that the best approach would be to start with the most dramatic version of the military raid as it had played out at the time.

A softly spoken man, now aged 26, Michael has a serious face with high cheekbones. On one cheek he has a prominent scar, inflicted during a violent initiation into the gang he joined as a teenager in his home city of Lagos. You do another brief but vivid description of Michael’s appearance, one of only two characters for which you do do. Why?I didn’t manage to interview everyone in person - they were scattered in Greece, the Philippines, and at sea - so that’s why they don’t all have physical descriptions. Michael and John are in the UK so I was able to meet with them in person. Soon after, in 2015, Michael was ordered to commit a robbery. When he refused, senior gang members threatened to kill him. Realising he couldn’t leave the gang without deadly reprisal, he ran away from home. He spent the next five years living on the streets, picking up odd jobs where he could. In the summer of 2020, Michael heard that gang members had shot and killed his mother. At that point, he felt he had no option but to flee the country. Who’s the source for this paragraph and how did you verify it?Michael is the source, and unfortunately I wasn’t able to verify it as much as I’d have liked to, in terms of ground reporting; I simply didn’t have the resources or time to fact-check his background in Nigeria, which would be very challenging. But he connected me with his case worker and lawyer in the UK so that I could confirm that this was what he’d told them.

An elderly man who had recently offered Michael work hauling fish said he could help him sneak on to a tanker at Lagos’s main port, a bustling complex in the south of the city. Michael did not know which ship he would get on or where it might be heading, only that it would take him away.

A few days later, Michael sat with the elderly man on the small fishing boat that would take him to the Nave Andromeda, which was anchored near the port, ready to set off to Russia. “The sea was so rough,” Michael recalled. He leaned over the side and vomited. As they approached the oil tanker, Michael was puzzled by the sight of it. The Nave Andromeda is 228 metres long, and he had never seen such a big ship. “I was thinking, what kind of boat is this? Is it a house?”

When they reached the ship, Michael’s friend helped him climb on to the rudder. The rudder is a massive steel plate attached to the outside of the vessel. A large pole links it to a steering room inside the ship, and around the pole is a recess called the rudder stock, a space less than two metres wide. Such a concise description! Did it take much revision?I’m gratified that you asked about this. I think this was the most stringently and repeatedly revised section of the entire piece! It was so difficult to get the description right, to make it clear without being bogged down in detail. I was tying myself in knots with it, so my editors should really take the credit for how concise it is. Michael clambered into this space. Right above the massive propeller fan and exposed to the wind and waves, the rudder stock is not designed for human occupation, and it is sealed off from the rest of the vessel. This was where Michael was stuck. And he was not alone. What was the intention of this last line?This section was closely told from Michael’s perspective and I wanted to signify that here we’d broaden it out to the others.

That day, six other men arrived separately on small boats. They all crammed into the rudder stock, clinging on to the central pole. Some of them secured themselves with ropes as best they could. One of them, a tall man with a narrow face and his hair in neat twists, was called John. Like Michael, John was escaping from a gang. An acquaintance had offered to help John on to a ship, saying that once he made it overseas, John could send back money to pay for his transit. John agreed, and three days later found himself climbing on to the rudder of the Nave Andromeda. “It was either I survive, or I die,” he told me when we spoke on the phone 10 months later. How did you locate John?I found him through Detention Action, a charity that had supported him while he was in immigration detention. John had spoken to reporters before, using a pseudonym, and the charity had been mentioned in that report, so I reached out to them. They were no longer supporting him as he was out of immigration detention, but they were extremely helpful in tracking him down and putting us in touch. I spoke to him first on the phone, and later traveled to the north of England to meet him in person.

The discovery of a stowaway is highly stressful for a ship’s crew. International waters are poorly governed, and it is usually unclear which country’s jurisdiction the stowaways are under. The ship’s voyage is often seriously disrupted as the crew and the shipping company must negotiate with port authorities, governments and police forces to try to find a port that will accept them.

Stowaways are also a financial problem. Most port authorities will detain stowaways and deport them, at the expense of the ship’s owner. Shipping companies usually have insurance to cover this cost, but, as Edward Carlson, a US-based maritime lawyer who has worked on numerous stowaway cases, told me: “It’s a big expense, and it’s a headache.” How did you find Carlson?He was quoted in Ian Urbina’s excellent book “Outlaw Ocean.” Once on shore, if they claim asylum, stowaways become the legal and financial responsibility of the state; but if their claim is later refused, the shipping company may be asked to pay for repatriation at that stage.

Any commercial voyage involves multiple parties – typically, private companies charter vessels from ship owners. Delays or diversions to the voyage cost money, as do security arrangements, flights and fines that governments impose for breaches of immigration law. “The vessel is constantly accumulating costs, and who is responsible for those costs is a source of endless debate and hours of legal work,” said Carlson. In recent years, stowaway searches have become routine. The ports in Lagos are chaotic, a fact exploited by established networks of people smugglers who push stowaways into ever-more dangerous hiding places: below the engine room, in the cargo hold, or the perilous area that hides the anchor chains. You break away from the dramatic action here for 266 words of exposition. How do you decide when to make this kind of shift?I usually try to fold the exposition into places where it illuminates the story in some way.What kind of reporting did it take for you to deliver this explanation?I had to really go deep into maritime conventions, immigration laws and shipping insurance. I interviewed scores of people who were not quoted in the final story, including people working at major marine organisations, port chaplains, shipping insurance executives, and academics researching stowaways. As always, I was incredibly grateful for their time and knowledge.

It was noisy in the rudder stock, between the roaring of the waves and the metallic thrum of the propeller fan. Salt water lashed at the stowaways’ skin, and as the Nave Andromeda pulled out of the port, the men lurched from side to side. They agreed to share their scant supplies of food – biscuits, powdered cassava – and bottled water. Who provided this information?Michael and John. They speculated that they might be there for one or two days. Night fell. Morning came. Michael tried not to think about falling in. How do you know this?Michael gave me a really vivid and moving account of the journey and the way he had felt throughout it, which is part of the reason I decided to focus on him. But with nothing visible apart from the deep blue of open water, it was difficult to think of anything else. Proper sleep was impossible. “If you fell asleep you would fall into the water and that would be the end of your life,” John told me. “It’s just open sea, so no matter how you swim you can’t survive it.” That’s such a stellar quote. How do you decide which ones to use from an interview?Some quotes, like this one, just really jump out at you. But there are always more than you’re able to use. Process-wise, I record all my interviews but I might also note down by hand anything that really hits me as someone is talking. Once I have a transcript of an interview, I read through it and highlight anything that seems like important information or that is a particularly powerful quote. That way, when I come to writing, the best quotes are already marked.

Night fell again. The ropes caused cuts and welts on the men’s skin, and their muscles ached from clinging on. Their surging adrenaline meant they didn’t feel much hunger, but thirst began to set in as sea water drenched them from all sides.

Nine days passed. On 15 October, Michael saw land. Gradually, buildings came into view. The Nave Andromeda had reached Las Palmas port, on the Spanish island of Gran Canaria, where it was scheduled to stop to restock on food and fuel. The stowaways saw a small tugboat sailing out from the port. This was their chance. They clambered out of the small recess where they were hiding and on to the top edge of the rudder. They shouted and waved, summoning the last of their strength to bang on the sides of the ship, desperate to be noticed and taken to land. Someone on the tugboat spotted them, and radioed the captain of the Nave Andromeda, who spoke to the port authority. The vessel came to a stop.

Soon afterwards, two rescue boats sped from the port to the rudder of the Nave Andromeda. The stowaways climbed off the rudder and on to the two small boats. “I did not know what was going to happen,” said Michael. “I was so scared.” But the rescue boats did not sail into port, as the men had hoped. Instead, they pulled up beside a ladder that led up to the deck of the Nave Andromeda. The stowaways were ordered to climb up the ladder and the rescue boats went back to port. The seven men were now the responsibility of the ship’s captain.

* * *

IN THEORY, THE MEN should have been allowed to get off the ship in Gran Canaria: maritime convention states that, if possible, stowaways should be disembarked at the vessel’s next port of call. But this was not what happened. Instead, as soon as they heard that there were stowaways on board, the port authorities denied permission for the Nave Andromeda to dock, and it remained anchored offshore. If a vessel doesn’t dock, it does not formally enter the country. Had the Nave Andromeda docked at Las Palmas, the authorities would have been obligated to allow the stowaways off, leaving them free to apply for asylum in Spain. This and other expository paragraphs ring with authority. How do you achieve that?As I mentioned earlier, I did a lot of reporting to be able to give these explanations - that involved numerous interviews with experts (and sometimes it took several interviews to even get to the right person to give the information), as well as wide reading around about the particular points mentioned, from academic papers to minutes of official meetings of the International Maritime Organisation. Shipping is not a world I really knew anything about going into this, so I wanted to make absolutely sure I got it right.

The decision by the port authorities in Gran Canaria should have come as no surprise: less than a fortnight earlier, on 6 October, an almost identical incident had taken place at the same port. The Champion Pula, another oil tanker, had left Lagos 10 days earlier. On reaching Las Palmas, four stowaways were discovered in the compartment above the rudder. The Spanish authorities would not keep them. That incident ended relatively quickly: the Champion Pula’s next scheduled port was Herøya in Norway where the men were allowed to disembark and claim asylum. But if multiple ports refuse to allow a ship carrying stowaways to dock, the vessel can effectively become trapped at sea.

“There are very few jurisdictions that accept stowaways with open arms,” one shipping insurance executive told me, adding that this is a particular problem in Europe and the US, owing to increased hostility towards migration. Why didn’t you identify the executive?The executive didn’t have authorisation from the company publicity department to speak with the press, and I didn't think that their name was important to the story so I didn’t push it. This is despite the fact that the numbers involved are relatively small: in 2018, the most recent year for which statistics are available, there were 90 recorded stowaway events on ships worldwide involving around 230 people. (The real number is likely to be higher, since this only includes cases in which an insurance claim was made. Nevertheless, stowaways on ships still represent a tiny proportion of migrants.) Why did you decide on this exposition here?I try to fold in the exposition in places where it illuminates the events being described. In this instance, I thought that the responses of European ports to stowaways, and the relatively small number of them, was relevant to understanding what was happening to Michael, John and the others.

Once it was clear that Spain would not allow the stowaways to disembark, Michael, John and the other five sat on the deck of the Nave Andromeda as the crew tried to work out where the men could sleep. Albert, a Filipino man in his late 40s, was a low-ranking member of the crew. He is a man of strong opinions who has learned to keep his views to himself during two decades of working in the strictly hierarchical world of shipping. The first thing he noticed was the smell – the stowaways reeked of sweat and seawater after their ordeal in the rudder stock – and the cuts on their skin. “It broke my heart, to be honest,” he told me. The seven stowaways were all devout Christians, and some murmured thanks to God and prayed as they sat on the deck. This struck a chord with Albert, who was also Christian. He saw other common ground, too: “If you’re poor, if you don’t have food, you might try to go to another place. That’s why I go to the ship to work, because I’d be poor if I stayed in the Philippines. It’s the same with them.” How did you find Albert?I tracked him down with the help of a journalist based in the Philippines, who found him on Facebook by searching for posts about the Nave Andromeda in their native tongue. We spoke on Zoom and unfortunately it was a terrible line, but we got what we needed. We stayed in touch via email afterwards but unfortunately I wasn’t able to speak with him on Zoom or the phone again, as he was away at sea for around 8 months after our call with limited phone data that — understandably — he didn’t want to use to speak with a journalist.

The crew brought the stowaways food, water and clean clothes, and dressed their wounds. Afterwards, the men were taken to a cabin containing six bunks, with a seventh foam mattress on the floor, and an adjoining bathroom. The door closed behind them and the lock clicked shut.

Later in the day, Perros, the captain, came to the cabin. The stowaways told him that they wanted to get on to land, and didn’t care which country they went to. Perros (who declined to be interviewed for this story but referred us to his employers, Navios) explained this wouldn’t be possible in Gran Canaria, but that he would try to make an unscheduled stop in France and let them out there. The next day, 16 October, the Nave Andromeda set sail.

Over the next five days, the stowaways settled into a strange sort of routine, mostly confined to their cabin. Someone from the crew brought food and water three times a day, and once a day they were escorted to the deck for fresh air. Most days Perros, or the superintendent Giannis Zafeirakis, came to their cabin to update them and check in. (Zafeirakis also declined to comment.) “They were so nice,” Michael remembered. “The captain was telling us stories. He gave us a TV so we could watch movies. He gave us everything we wanted.” Michael cried a lot, thinking of his mother’s death, and remembers the captain taking special care to ask if he was OK, even bringing him cigarettes. You alternate regularly between exposition, and dramatic action using cliffhangers to keep the reader captivated. Why?In a story of this length, I always try to fold in background information and context around the main narrative. This way you keep a sense of narrative momentum, and the information is used in service of illuminating the story, so the reader stays with you.

The Nave Andromeda is owned by the Greek company Navios Tankers Management. (Navios did not respond to a request for comment.) The ship’s captain and six other senior officers were Greek, while the rest of the 22-strong crew was almost entirely Filipino. There wasn’t much communication between the two groups, although they spoke English to each other. The crew was not allowed to spend time with the stowaways, and for the most part, only Perros and Zafeirakis talked to them. The handful of other crew members, including Albert, who briefly interacted with the stowaways, found them to be pleasant and polite. “They were very nice guys, with very big hearts,” said Albert.

On their long days inside the cabin, the Nigerian men talked in a way they hadn’t been able to as they hung on for their lives inside the deafening rudder chamber. They shared details of how they had ended up leaving Nigeria. But more often, they talked anxiously about getting off the ship. “Most of us felt sick and we were getting depressed being locked up,” John said. Michael was grieving – and his fear of the water had been exacerbated by the traumatic journey. “When I was in the big sea, I couldn’t see the land,” he said. “Let me see the land, I was thinking, where I can stretch my legs and know I am free.”

On 20 October, the Nave Andromeda reached Saint Nazaire in France. You also organize your story using the calendar. Do you rely on timelines?I always use timelines for long-form stories. I usually build this on a simple Excel spreadsheet while I’m still reporting, with fields for the date, what the event is, my source for the information (eg. from an interview,  or from documents, or news reports) and any extra notes. I usually have a colour coding system too, to indicate whether an event is part of the main dramatic action (in this case, the hijacking itself and the following legal processes) or if it relates to a personal story of someone I’m writing about (so that would be Michael and John’s own movement through the asylum system, for instance) or if it’s wider context (such as other comparable cases of stowaways, or the broader policy statements about immigration). I find this an invaluable tool when it comes to writing, since I’m usually reporting these stories over a long period of time and alongside other projects. When the captain informed the port that there were stowaways on board, permission to dock was denied.(Although Albert wasn’t party to the negotiations, he said he was told that the port authorities cited Covid restrictions and the fact that the stowaways didn’t have identifying documents.) When Perros and Zafeirakis came to inform the men, Michael said they looked visibly stressed. “The captain and the superintendent were telling us to stay calm and that they were looking for somewhere safe to drop us,” said John. “They also said that they needed us to be off the ship because they were running out of food.”

Crew on vessels that have been refused permission to dock have occasionally taken extreme action. In his book The Outlaw Ocean, the journalist Ian Urbina describes the practice of rafting, “whereby a crew, discovering such uninvited guests, sets them adrift in the middle of the ocean and leaves them to die.” How important was this book to your reporting?The book has a chapter on stowaways which provided incredibly useful context. I spoke to Ian Urbina early in the reporting process and he was very generous with his time and advice. This was before I’d tracked down Albert or anyone in the crew, and Ian suggested a number of ways in which I might try to do this. He also connected me with a Greek journalist with expertise in shipping who was able to help with the reporting in Greece. Through his Outlaw Oceans Project he even provided some contacts to different marine organisations. I also got some incredibly helpful advice from the British writer Rose George, who wrote a fantastic book on shipping. She was the one who suggested I contact the Southampton port chaplains, as she said they always know what’s going on at any port. I wouldn’t have thought of this myself and they provided a really useful gateway into the story early in the reporting process. I’m always so grateful to other journalists who are generous with their time, expertise and contacts. There have also been instances of stowaways, desperate to disembark, turning violent. In 2018, a group of Nigerian stowaways threatened the crew as the ship approached Tilbury docks in the Thames Estuary. They were later convicted of affray and making threats to kill.

On the Nave Andromeda, everyone was tense. The senior officers were in an awkward position. Their commercial voyage had been disrupted and the ship’s supplies were diminishing. Meanwhile, the stowaways were growing increasingly frustrated.

Even so, relations stayed cordial. “There was no aggression,” said Albert. I spoke to another crew member who corroborated Albert’s version of events, but did not want to be quoted. Perros took John and another man up to the bridge, the navigational centre of the vessel, and showed them a map. He pointed to where they were anchored, and said that he would try to get them off in Holland or the UK. The stowaways returned to the cabin and waited.

* * *

AFTER THE REFUSAL in France, the Nave Andromeda drifted for a few days while the senior officers and Navios, the shipping company, tried to work out what to do. Concerns about the financial implications filtered down to Albert and the rest of the crew. The stowaways’ meals were reduced from three a day to two.

On 24 October, Albert saw the six Greek officers gather on the ship’s bridge for a meeting, with Navios’ head office in Athens dialled in. (Another crew member confirmed that this meeting took place.) How many crew members did you interview? How did you locate them?On board the ship were seven Greek ‘officers’ as well as the rest of the crew, who were predominantly Filipino. The only name in the public domain was the captain, so one of the first things I did was to find him on social media and message. He replied straight away, initially agreeing to an interview when he got back to land, but he later withdrew this as he had accepted another job with Navios. I didn’t have any other names, but with the help of the Filipino journalist I’ve already mentioned I spoke to Albert, who shared a crew list. I looked up all the Filipino crew on Facebook and messaged all of them, but most didn’t reply or declined to speak - either because they were away at sea, or because they were concerned about their future employment prospects if they criticised a former employer. In the end, I directly interviewed Albert and one other crew member who didn’t want to be quoted directly, and my colleague spoke to one more. I wasn’t able to connect with this third person myself so I decided not to include this account, though he also corroborated Albert’s story. An excellent Greek journalist called Eleni Stamatoukou helped me to track down all the Greek officers, and we contacted them all, though they all declined to speak on the record, mostly citing the terms of their employment. Later that evening, according to Michael and John, Perros came to the cabin and said that at 10am the next day they would be dropped in Southampton, England. The ship’s chef was told to prepare breakfast early because the crew would be doing a drill.

The next morning, the crew ate breakfast together at around 7am. Straight afterwards, they were told to go to the citadel – the secure area where, in the event of an attack, the crew can shelter and wait for backup. Albert and another crew member told me that they were in the citadel by 8am. Shortly after this, at around 9am, the distress call went out to the British coastguard: “I need immediately, immediately agency assistance.”

Inside their cabin, the stowaways were watching the clock on the wall. They had no idea that anyone had called for help, but they knew something was amiss. Usually someone brought breakfast before 7am, but today no one had come with food or water. Even inside their locked cabin, they could usually hear clanking machinery and the footfall and shouting of crew going about their daily work. That day, everything was silent.

The men could tell that the ship was drifting, and they started to get scared. John wondered if they’d been abandoned at sea, and panicked about how they would ever steer the vessel. Michael worried that the ship was turning back to Nigeria. At lunchtime, they waited for a knock but no one came. Hungry, thirsty and confused, the men banged on the door. There was no response. Soon after, at around 2pm, they decided they needed to find out what was going on. They broke the door of the cabin. The corridor outside was abandoned. “We didn’t see nobody. We didn’t hear anything, even a rat,” said Michael.

Increasingly panicked, the seven men moved through the ship, shouting out to attract attention from the crew. There was no one to be seen. “Why would these people decide to lock themselves in their rooms for no reason? That’s how I knew something was wrong,” said John. They came to the deck. From here, they could see the bridge, where Perros was standing. Some of the men approached the bridge, hoping to talk to Perros. He held up a handwritten note against the glass, telling them to stay calm and that the authorities were coming. Since the tenor of some of the events appears to be in question, how did you verify what the men told you?This was all based on interview material; the accounts given to me by the stowaways and members of the crew corroborated each other in terms of the timing of the day (crucially, that the crew was in the citadel long before the stowaways broke out of the cabin). I didn’t include anything that wasn’t corroborated by others.

Not sure what to think, the men sat on the deck and waited. “We can see the captain in the glass. He is radioing, I don’t know who he is talking to,” said Michael. “That’s when we see helicopters coming.”

At 3.45pm, the coastguard’s helicopters circled above the Nave Andromeda. Some of the men had the idea of waving white handkerchiefs to show that they were peaceful. The helicopters continued to circle, but nothing else happened. As afternoon turned to evening, it got cold out on deck. Some of the stowaways went back into their cabin to wait there.

At around 7.30pm, the wait ended. The Special Boat Service operation began: armed commandos scaled the sides of the ship. Michael was on deck when they arrived. His heart beat fast as he saw a gun trained on him. “It was like a movie,” he remembered. “I was so scared. I have not seen that kind of gun in my life before.” He wasn’t even sure what country he was in.

John was in the cabin when the commandos arrived. “I knew that we were in big trouble because there were so many forces for just seven of us,” he said. As he saw their weapons, he was convinced he was going to die. “I had nothing to defend myself, and even if I did, what could I do against so many guys who looked like they should be going to a war front?”

The stowaways did not resist: in the space of the nine-minute military operation, they were all handcuffed and made to lie flat on the deck. How do you know how long it lasted?This came from news reports and official statements at the time. Michael had never left Africa before, and the British sea air chilled him to his bones. After several hours, the ship docked in Southampton. Almost three weeks after they had climbed on to the rudder in Lagos, the seven men were led off the Nave Andromeda in handcuffs and taken to Southampton Central police station, a red-brick building across the road from the port. On board the ship, they had felt like they were in prison. But their confinement was just beginning. Another cliffhanger. What’s the value of the device?Just what you’d expect — to keep the reader interested and eager to read to the next section.

* * *

DAVIOS LATER SAID in a statement that Perros sent out the distress call because he “was concerned for the safety of the crew due to the increasingly hostile behaviour of the stowaways”. This is certainly possible. But according to the accounts of some crew members and stowaways, it seems that the distress call was made several hours before the stowaways broke out of their cabin – the event cited as the trigger for the call – and that the crew were locked in the citadel long before this happened. Why did you qualify this with “seems?”To reflect the fact that the events are in question.Given that, how did you draw this conclusion?Through corroborating testimony from the crew and stowaways.

Whatever the reasoning behind the distress call, as soon as it was made, the incident took on a life of its own. The British press, and British politicians, quickly described it as a hijacking. But hijacking is a specific crime: the term refers to the seizure of a commercial vehicle by force or the threat of force. It is not a term that Perros used in his distress call. Nor is there any evidence that he or anyone else from Navios suggested that there was a risk of the crew losing control of the ship, or indeed any attempt by the stowaways to seize it. In fact, that very evening, Navios’s lawyers reportedly told the BBC that the incident was “100% not a hijacking”.

British ministers could have used more neutral language – they might, for instance, have referred to events as a “security incident” – but they chose not to. The next day, Wallace told journalists that the stowaways were “threatening to do something with the ship” that would have caused a “threat to the environment and, more importantly, to the lives of people on the ship … something the state can’t tolerate”. It is not clear what these statements were based on, but it added to the impression that something extremely violent had taken place. “We had government ministers practically saying that the allegations are true before any investigation or trial,” said James Wilson, deputy director of Detention Action, a charity that supported John and one other stowaway while they were in immigration detention. How did you find the charity that supported the men?It had been mentioned in an earlier news report in which John spoke to journalists. They were critical in locating John, but they had only supported him and one other stowaway, and they were not able to get in touch with the second. “It just looks like a massive overreaction.”

John, Michael and the other men spent three nights at Southampton police station, before being released on bail and transferred to Colnbrook immigration removal centre, a drab building close to Heathrow airport. Did you visit the center?No, but I looked at pictures online. They were confused about what had happened on board, and all denied the allegations of violence. John panicked that he would be sent back to Lagos and murdered by the man who had put him on the boat; he still owed money for his passage. Michael relied on his religious faith, telling himself that God would not allow him to be punished for a crime he did not commit. “I know I am not a hijacker,” he said to himself. How did you know what Michael told himself?He told me. This is almost verbatim from an interview transcript.

The vessel sat at Southampton port for a few days. No one was allowed on or off the ship because police were treating it as a crime scene. Police questioned the captain and crew. Were you able to obtain any police records?Unfortunately not, despite my best efforts. I am always very jealous of U.S. journalists who can relatively easily access court and police records. We don’t have the same kind of transparency in the justice system in the UK. I was particularly keen to get hold of the Crown Prosecution Service records because in their decision to drop the case they alluded to new evidence such as mobile phone footage and witness testimony contradicting the notion that there was a violent struggle. But there is no mechanism for a journalist to request these kinds of records from the CPS or police, so you can usually only get them from a lawyer. I tried contacting the legal aid solicitors who represented John and Michael in the criminal case (I’m called once a week for several months, sent a letter and went to their office to ask in person) and eventually had to admit defeat. The only other way to get police records is for a person involved in the case to make what is called a “subject access request” — but given the chaotic nature of Michael’s and John’s lives, being shunted around temporary accommodation, to the point it was difficult for me to keep in regular contact with them, this just wasn’t feasible. Albert said their questions focused narrowly on whether he had been injured or had his property stolen. “They didn’t ask what really happened,” he told me. He wanted to tell someone what he told me eight months later: “It’s not a hijack, it was just a drama. It’s a story only. A big fucking lie.” Some publications don’t print profanity. What’s the Guardian’s style?The Guardian’s style is to never use asterisks, which the style guide describes as a “cop out”, but only to use swear words where they’re essential to the facts of the piece or demonstrating someone’s character. I think it adds something here. (He no longer works for Navios.)

After a few days in Southampton, the Nave Andromeda sailed on to Antwerp and then continued on its planned voyage to Russia. The senior officers never discussed the events with the Filipino crew. In Albert’s view, the decision to withhold food and water from the stowaways – which led directly to them breaking down the door of their cabin – was a deliberate provocation. Of course, there is a reasonable explanation: the rations may have been reduced due to food shortages, and it is not unusual to lock stowaways into a cabin when approaching a port so that they don’t try to escape and swim to shore. But this does not explain the distress call. For all the puzzling inconsistencies, one of the Greek officers, speaking on condition of anonymity, described the incident as “piracy” and told me that “everything written about it was true”.

* * *

JUST BEFORE CHRISTMAS, two of the stowaways were charged with conduct endangering ships under the Merchant Shipping Act. They were taken to prison. The other five men remained in detention, waiting to hear if they would be charged, too. A few weeks later, the CPS dropped charges against all the men on the grounds that there was no realistic chance of conviction. Explaining this decision, the CPS said that “while initial reports had indicated there was a risk of destruction or serious damage to the ship”, video footage and further analysis had “cast doubt” on whether ship and crew had been in danger.

The Home Office publicly criticised the CPS decision, saying: “It is frustrating that there will be no prosecution in relation to this very serious incident and the British people will struggle to understand how this can be the case.” This was a highly unusual step. “It’s genuinely extraordinary for the Home Office as an institution to criticise the independent Crown Prosecution Service,” said Colin Yeo, an immigration lawyer. Why did you need a source to make this point?In our current political context, politicians make wild statements and overreach their roles all the time, to the point that we get desensitised to it. I wanted to speak to experts to check whether this statement was actually as extraordinary as I thought it was, and then I decided it was important to have someone in the field making the point that this was extremely unusual, rather than me making this claim. “I think it’s unprecedented.”

I spoke with John and Michael separately after the charges against them had been dropped. John was still in Colnbrook detention centre when he heard the news. “I was so happy,” he remembered. “They did good research and justice prevailed.”

But the stowaways’ ordeal was not over. Seeking asylum in the UK is a complicated and slow process: the Refugee Council estimates that it takes most people between one and three years. Over the next few months, the stowaways were released from immigration detention one by one, and given temporary accommodation around England.

John was sent first to a small town in the north-west, then to Manchester, where he is currently staying. He read news reports about the incident on the Nave Andromeda online. “I didn’t know why the Home Office was pressuring [the CPS] to send us to prison,” he said. Asylum seekers are not allowed to work or go to college, and so he spends a lot of time alone in his room. He misses his six fellow stowaways; they formed a close friendship on the ship, which developed while they were locked up in the same detention centre. They keep in touch, chatting on Facebook or on the phone, but they’re scattered around the country now, and it isn’t the same.

I met John in a park in Manchester on a cold, bright day in March. He wore a black bomber jacket and a camouflage headband that pulled the hair back from his face, highlighting a small scar on his forehead. “When someone lays a [false] allegation on you, of course you’re going to get angry, but you have to control yourself,” he said. “The most important thing is for people to find out that it’s true you’re innocent.” He spoke in a flat voice about his experience on the Nave Andromeda, lightening up only when conversation turned to a recent Champions League match; football provides a welcome escape from the daily monotony of sitting in a hostel room, with barely enough money to buy food.

Michael was released from detention a couple of months before John, and spent some time in hotels in London before being moved to Coventry. The conditions in one hotel were so poor – a filthy room with a strong smell of faeces – that he went to the local police station and asked to be arrested so that he could sleep in a cell. He finds it difficult to manage his mental health and often sleeps all day.

While John anxiously read the media coverage of the Nave Andromeda incident, Michael did not, and had not fully understood the status of the criminal case. We met over a year after the CPS had dropped all charges, but he handed me a crumpled bail notice from the police and asked me to explain it. When I said that the case had been dropped, pulling up a BBC report from the previous January on my phone, he began to cry. “I didn’t know that,” he said.

Michael goes to his local church every Sunday, and occasionally plays football with members of the congregation, but otherwise he keeps to himself: he is afraid that the gang who killed his mother will track him down in the UK, which leaves him reluctant to make new friends.

It has been weeks since either of them had any update on their asylum claims; they have no timeline for when a decision might be reached. Although such delays are not uncommon, it is difficult not to worry that the allegations are playing a part. “They say: ‘You are sea pirates, hijackers’,” said Michael. “But we are not sea pirates. We came only for survival.”

After the high drama of a triumphant military operation to keep Britain’s shores safe, this is what remains: seven men, alone in bedsits, waiting. Your story ends on a note that sounds ironic but also is poignant. It highlights the unfairness of what happened to the stowaways. Why did you close this way? Did you consider alternate endings?I wanted a sense of pathos, to contrast the bombast of the way this story was originally portrayed with the reality I found. For me, this was always the core of the piece — the totally disproportionate reaction to a very small, very vulnerable group of men. Although there were many unusual things about this case in particular, that disproportionality is emblematic of many of our asylum policies. So I wanted to end on this note, although I think in my original draft, this paragraph and the one before it were the other way around so it ended with “we came only for survival”. My editor switched them around, which actually works much better.

* * *

Chip Scanlan is an award-winning newspaper reporter who taught writing at The Poynter Institute for 15  years. He publishes Chip’s Writing Lessons and has two books, ”Writers on Writing” and “33 Ways Not To Screw Up Your Journalism."

The post When a migrant story trail goes cold, fresh reporting finds a deeper truth appeared first on Nieman Storyboard.

Journalism, by its very nature, focuses on the now — the events and people making the news today. But powerful stories can be found by mining the past to add fresh material and context to what had once been a sketchy breaking story, abandoned after a few news cycles. Revisiting such a story can pay off with new and unexpected results.

For years, Lauren Smiley, who writes about humans in the tech age, had kept track of Rafaela Vasquez. In 2018, Vasquez was at the wheel of an Uber self-driving Volvo when the car struck and killed Elaine Herzberg, who was crossing a darkened road in Tempe, Arizona. She became the world’s first pedestrian fatality involving a self-driving car.

“I often find my best projects by circling back to news events and chasing the deeper story,” Smiley told me. Her trip back in time produced “‘I am the operator’: The aftermath of a self-driving tragedy,” a gripping, 10,000-word narrative for Wired magazine published in March 2022.

Smiley displayed remarkable patience as she kept track of the post-crash investigation. Vasquez was charged negligent homicide in connection the crash (the case is scheduled to go to trial later this month, but could be further delayed) but had never spoken about it with a journalist. Two years after the accident, Smiley reached out to Vasquez’s defense lawyers, who gave the okay to sit down with their client.

To prepare, Smiley sent out a flurry of records requests, riding herd on them until agencies delivered. She studied a critical report of a voluminous investigation by federal transportation safety officials. After Uber declined to talk, she used LinkedIn to cold call dozens of past and present employees, many of who spoke on or off the record about the company’s flawed self-driving program. She reached out to Herzberg’s family to learn the impact of the crash on the victim’s mother and daughter.

The story focuses relentlessly, however, on Vasquez, a transgender woman who grew up with a religious, unaccepting father and suffered repeated sexual violence when a judge assigned her to serve a sentence with male inmates for a botched armed robbery attempt with her boyfriend. “I’d never had to have anal stitching before,” Vasquez told Smiley during a seven-hour interview, “but I had it in jail.”

Smiley funneled all of this data into multiple timelines, all annotated for the Wired fact-checker, a backstop that enabled her to reconstruct you-are-there portions of the narrative with limited attribution. Guided by her outlines, she put her inner critic on hold as she began to write. “First drafts are about infrastructure, not beauty!” she said.

Smiley weaves Vasquez’s story before and after the crash with a granular account, at times down to the hundredth of a second about how the confused computer on board the self-driving car misidentified Herzberg, a shadowy figure jaywalking on the road, and an authoritative look at the now-humbled state of the self-driving movement in the United States. It’s a compelling read, its relentless pace driven by carefully chosen descriptive verbs, a braided structure and elegant writing — elements that won it a SABEW Best in Business Award for features.

Smiley, a self-described “boots on the ground tech reporter,” is a contributor at Wired. She has written human-centered stories about technology for The Atlantic and New York Magazine. She was an investigative reporter for the “Broken Harts” true crime podcast and has worked on staff for Matter and SF Weekly.

In an email interview and annotation with Nieman Storyboard, Smiley discussed her methods for juggling multiple storylines, her education and writing influences and her editors’ essential role. In an annotation of the story, she goes into detail about how how she tracked the story for almost five years, how she gained access to a range of sources, ethical considerations about identifying sources and the story’s structure. Our conversation has been edited for length and clarity

Your bio says you write about humans in the tech age. Why that focus?
I live in the country’s tech capital, but I’m not the reporter who obsesses over gadgets or corporate brinkmanship. I am drawn to the philosophical questions embedded in tech and how tech has put humans in all sorts of weird situations, like driving strangers in their own car for a job; or believing a wild rumor on social-media and taking startling action; or surveilling the person stealing their Amazon boxes off their porch and reporting it on Nextdoor. I really enjoy the luxury of deep interviewing, when you can spend gobs of time understanding someone’s world in hopes of transmitting that to the page. It’s the human messiness that brings me to the game. My great ambition is that the life of the person I write about feels as real and complex as my life does to me.

How did you learn to write narrative nonfiction?
From my days at my high school newspaper and reporting for the University of Iowa college paper, I was always drawn to features, but it took me a while to realize you could apply feature-writing methods to serious material. In college, I started devouring the “Best American Magazine Writing” books compiled by ASME. I re-read Anne Hull’s “Divided Feast” 20 times, amazed that she’d unearthed weighty class observations by embedding in a supermarket. Michael Paterniti’s “The Most Dangerous Beauty” made me tearful in admiration as I realized that writing could actually elevate the reporting to a boiling point that it shape-shifts into art.

Throughout college, I honed my skills at feature internships, capped by a summer on the Dallas Morning News features desk, a wonderfully writerly culture circa 2005. (Chip, you should know that my editor, Michael Merschel, had me complete your questionnaire to figure out what a story was really about — an exercise that I still do).  The News sent me to the very first Mayborn conference. Listening to nonfiction writers talk about their process all weekend crystallized that I’d found my people and my place.

Before I showed up for my first reporting job at a San Francisco alt-weekly, the editor told me to read Jon Franklin’s “Writing for Story.” I’ve carried its lessons with me: how to pick a good tale; how to nod at the piece’s themes via word choices; how to implicitly ask a question in the lede that the piece seeks to answer. I then sprinted through a brutal gantlet of five sink-or-swim years at the alt-weekly, producing a 4,500-word cover feature every five weeks or so. That destroyed any preciousness about the journalistic process. The job acted as my grad school, teaching me how to get records, work with sources, pace a longform feature. The way I can still blaze through a first draft is a callback to that era, though thankfully I usually no longer have to churn them out in a terrifying weekend while chugging Red Bull.

Nina Martin, who edited several of my magazine pieces once I went freelance, also imparted the wisdom that you can write about complex people in equally complex situations with both empathy and fairness; those two things aren’t mutually exclusive. I find that framing useful for the stories I’m doing these days.

The learning continues, of course. And the crazy thing about journalism is all your heroes put their work for you to study right there, on the page!

Could you describe your writing process?
One of the most amusing writing scenes I’ve ever watched was in a documentary on the Los Angeles food writer extraordinaire, Jonathan Gold. He opens his laptop, types an entire column with dazzling prose, speaking the words in a voice-over as they hit the page, jaunty music playing, then closes his laptop with a satisfied look and walks out. A perfectly linear and publish-ready first draft!

I wish!

Here’s my way-less-entertaining montage: From all those outlines, I sketch out a short outline, keeping it to a page or two, to map and distill the sections. Then, blast-off! I type out a rough draft as quickly as possible, even when it means the roughest sentences you can imagine. That big messy document frees me of the fear of not having words on the page. There’s no reason to get overly wedded to sentences or a particular transition until you see how the whole story lays out.

Then I copy-paste that beast into a new doc and start refining.  Drafting is the loneliest time; no one can write that first draft but you. Only after many rounds of refining – each one copy-pasted into a new doc so I know the old one is always there like an old friend should I need it – do I start getting to that point of lift, when the sentences start to bounce off the page a bit. I try to get it there before I send it to my editor.

 At one stage, I try to read my draft aloud — that age-old advice. Clunky phrasings and redundant words leap out; I often shamefully realize I’ve written the same word three times in the same sentence. Also, I do a search-replace for common words in the piece, subbing in synonyms or tweaking the sentence to not even go there. Tighter, more interesting writing always emerges. (I just did that on this very document and deleted 25 uses of “story,” case in point!)

At the bitter end, I get really nit-picky about flow; each sentence should be easy for the eye and mind to sail through, right into the next one. My editors are really good at this verbal Zamboni-driving, too. No way any reader is ever going to give you 45 minutes of their time if reading feels like a chore.

Are there any writers, stories or books that inspire you as you worked on this or other stories? Sometimes I will read a great piece of journalism once I’m into the draft refining process, and it helps jolt me into trying to match that level of mastery. But I also try to apply the adage, “Don’t compare your insides to other people’s outsides.” Published stories have already been through the whole editorial ringer. Be inspired but realistic that your early drafts will not match the sparkle of someone else’s final product.

I was interested in The New Yorker’s feature on accidental killers. I was interested in how my mentor Elizabeth Weil wrote about a defendant at the center of another fatal tragedy, and in Matthew Power’s profile of a drone operator, which had haunted me for years.

What role do your editors play in your work?
I like talking through reporting decisions with my editor, and Wired’s executive editor Maria Streshinsky was a generous coach and sounding board over the long reporting road. She also ensured that we were keeping the pedestrian victim top of mind. Later in the revision process, when we were feeling a bit stumped by the structure in the muddled middle, Wired’s feature editor John Gravois had the insight of flipping the order of a few of the sections. Everything started flowing better from that point. Once I file a first draft, the editors copilot every editorial decision until the magazine closes.

Annotation: Storyboard’s questions are in red; Smiley’s answers in blue.

‘I’m the Operator’: The Aftermath of a Self-Driving Tragedy

In 2018, an Uber autonomous vehicle fatally struck a pedestrian. In a WIRED exclusive, the human behind the wheel finally speaks

By Lauren Smiley

March 8, 2022

RAFAELA VASQUEZ LIKED to work nights, alone, buffered from a world she had her reasons to distrust. This a powerful example of foreshadowing. Why did you include it in the first sentence of the story?The headline and lead photo indicate that this piece is about the person you’ve never heard from in this highly publicized crash, but I also wanted to telegraph right away that we’re going to hear Vasquez’s perspective. Of course, this story is concretely about Vasquez getting charged with a crime. But in the feature-writing mental exercise of “What is this really, really about?” for Vasquez’s personal story, I landed on a theme of seeking acceptance but the reality of being cast out by society throughout her life. I’m surfacing that theme right away in the first sentence. I also want to foreshadow right now that this isn’t going to be just another night on the job.

One Sunday night in March 2018, Uber assigned her the Scottsdale loop. She drove a gray Volvo SUV, rigged up with cameras and lidar sensors through the company’s garage, past the rows of identical cars, past a poster depicting a driver staring down at a cell phone that warned, “It Can Wait.” I’m not familiar with lidar sensors. Did you consider stopping to describe them?The lede’s goal is scene-setting and drawing the reader into this tense progression of events. It’s about the image in the reader’s mind of the futuristic car, not technical details. As long as the reader hears “fancy tech car,” I’m content. The clock ticked past 9:15, and Vasquez reached the route’s entry point. She flipped the Volvo into autonomous mode, and the car navigated itself through a blur of suburban Arizona, past auto dealers and Zorba’s Adult Shop and the check-cashing place and McDonald’s. Then it jagged a short stint through Tempe to start the circuit again. It was a route Vasquez had cruised in autonomy some 70 times before.

As she was finishing her second loop, the Volvo blazed across a bridge strung with bistro lights above Tempe Town Lake. "Blazed" is the first of many vivid verbs that stud the story. How do you arrive at them — during drafting or revision or during the reporting?Strong verbs make lively writing, but we have to pick an accurate and fair one. She’s not driving the car — it’s driving itself — and we know from records that the car was traveling at about 40 mph on a pretty open stretch of road. So I selected “blazed” in my first draft. Neon signs on glass office buildings were reflected in the water, displaying the area’s tech hub ambitions—Zenefits, Norton LifeLock, Silicon Valley Bank. How were you able to name the office buildings, and why do you do so?A sense of place is an important part of strong narratives so, while in Tempe reporting, I walked around the crash area, taking photos and videos to remind myself of what I saw when I got back to my desk. I later checked that those companies were there at the time. By selecting these details about tech companies, I’m foreshadowing that Arizona’s emergence as a tech hub is relevant to what is happening. Vasquez is surrounded by evidence of that ambition and the Uber test program she’s part of is also part of that ambition. Beyond the bridge, the car navigated a soft bend into the shadows under a freeway overpass. At 9:58 pm, it glided to a forlorn stretch of road between a landscaped median and a patch of desert scruff. Four signs in the median warned people not to jaywalk there, directing them to a crosswalk 380 feet away.

The Uber driving system—which had been in full control of the car for 19 minutes at that point—registered a vehicle ahead that was 5.6 seconds away, but it delivered no alert to Vasquez. Then the computer nixed its initial assessment; it didn’t know what the object was. Then it switched the classification back to a vehicle, then waffled between vehicle and “other.” Again you use a descriptive and not typical verb: "waffle."My first draft said “vacillated.” I tweaked to “waffle” in a revision. Don’t use the ten-cent word if the two-cent one is more effective. The machine is changing its assessment, so “waffle” was the right choice. At 2.6 seconds from the object, the system identified it as “bicycle.” At 1.5 seconds, it switched back to considering it “other.” Then back to “bicycle” again. The system generated a plan to try to steer around whatever it was, but decided it couldn’t. Then, at 0.2 seconds to impact, the car let out a sound to alert Vasquez that the vehicle was going to slow down. At two-hundredths of a second before impact, traveling at 39 mph, Vasquez grabbed the steering wheel, which wrested the car out of autonomy and into manual mode. It was too late. The smashed bike scraped a 25-foot wake on the pavement. A person lay crumpled in the road. How were you able to drill down to reconstruct such a scene down to the hundredths of a second?The NTSB public report with all its supporting documents on this crash was an incredible resource. Uber had turned over detailed information to the authorities on what identity the car was assigning the pedestrian in the road, broken down to fractions of a second. While corporate Uber was pretty tightlipped with me, the NTSB report allowed us to be precise throughout.

Vasquez did what Uber had taught its employees in the test program to do in case of emergencies: She pulled the vehicle over and called 911. “A bicyclist, um, I, um, hit a bicyclist that was in the road,” she told the dispatcher, her voice tense. “They shot out in the street … They are injured, they need help, paramedics.”

“I know it’s pretty scary,” the dispatcher said in soothing tones.

She told Vasquez to breathe.

Within six minutes of the crash, cops started to arrive. Paramedics too. One cop scanned a flashlight over the person on the ground. A paramedic kneeled down and pumped the victim’s chest. How do you know what the cop and paramedic did when they arrived at the scene?I received hours of police body cam footage in a public records request, which showed these early moments at the crash site. Much of this was backed up in the police report as well, confirming the identities of who at the scene was doing what.

Dashcam footage from Rafaela Vasquez's self-driving Uber, pulling out of the garage on the night of the crash.

A couple of minutes later, an officer walked up to the Volvo, where Vasquez sat behind the wheel. He asked if she was OK. “Yeah, I’m just shaken up,” Vasquez said. “Is the person OK? Are they badly hurt?” Back by the figure who lay on the ground, a woman began wailing. Vasquez asked the officer, “Is that the person screaming?” He answered: “No no, that’s probably some people that they know.”

For the next two hours, Vasquez waited, doing what the police asked. Uber reps arrived. In the early minutes after the crash, one jogged up to Vasquez’s car, and an officer asked him to let the cops talk to her first. Eventually Vasquez moved to sit in a supervisor’s car. She asked for updates about the victim. And she learned that the person with the bicycle had died. How did she learn this? Why did you choose to summarize rather than present it as a scene?The body cam footage shows Vasquez telling a police officer at one point at the crash scene that she’d been told that the victim had passed away. I didn’t have footage of that moment of Vasquez finding out, or any information about who exactly it was who told her, so I stuck to what I knew.

After midnight, Officer Kyle Loehr approached Vasquez and asked if she was the driver in the crash. “I’m the operator,” she said. He asked her to get out of the car, and, body camera running, explained that he was going to run her through some sobriety tests: “This protects you, protects the city, protects the company,” he said. “It’s just literally a box we need to check.” Vasquez tracked Loehr’s green flashlight with her eyes, then his finger, then looked up to the sky and told him when she thought 30 seconds had passed. Sober. Why use a one-word sentence fragment here?Let’s change up the pace. It’s sharper than saying “The officer decided she was sober.” Fewer words are usually stronger. About 10 minutes later, Loehr came back with more questions. His voice was congenial and chipper. “I’m trying to just lighten the mood a little bit,” he said at one point, “because I know it’s stressful, and it’s crappy.” He told her he had to read her Miranda rights to her. That’s what happens, he added, when someone is no longer allowed to leave a scene. Gently, he went on: “Let me walk you through what happens with any of these cases when there’s a fatality.”

“Oh God,” Vasquez whispered. “That word.” Were these and other exchanges between Vasquez and Officer Loehr drawn from your transcription of the body camera footage?Yes, this was from the body cam footage, which I believe was part of the full police report that I requested through the Tempe police department.

Multiple blunt-force injuries. That’s what the medical examiner would put down as Elaine Herzberg’s cause of death. Manner of death: accident. Herzberg had lived in Arizona her whole life and had resorted to camping in the streets near Tempe. The 49-year-old often carried a radio playing the local rock station; she collected frog mementos and colored to relax. She had struggled with addiction. That March night, she became the first pedestrian killed by a self-driving car. How did you verify this fact?This feature was published four years after the fatality, which was widely reported as the first of its kind in the young self-driving industry, so we (me, the editors, the fact checkers) were comfortable with stating this plainly without a qualifier word of the “first-known,” or some such.Why did you choose this scene to open the story?My first draft also started at the crash scene but a different entry point, a moment after the crash, and then talked a bit more about Vasquez in the second graf. My editor Maria Streshinsky urged me to stay in chronology— to start with the lead up to the crash while sticking close to Vasquez’s perspective. It’s more elegant for the scene to spool out chronologically, less jostling for the reader.

Herzberg’s death is the kind of tragedy the autonomous driving industry claims it can prevent. In the US, car accidents kill more than 38,000 people a year, more than 90 percent of them at least in part due to human error. By taking sleepiness, inattention, drunkenness, and rage out of the equation and replacing them with vigilant, precise technology, self-driving cars promise to make the roads dramatically safer. But to reach that purported future, we must first weather the era we’re in now: when tech is a student driver. Terrific analogy. When did you come up with it?My editor John Gravois came up with the student-driver concept in our many iterations of the nut graf, which is one of the very last parts of the story that locked into place. I was deep in the frenzied march to publication, and sometimes it takes an editor who still has that big-picture view to surgically cut through the haze with the perfect phrase. That means gangly fleets of sensor-bedecked cars sucking in data on millions of miles of public roads, learning to react to our flawed and improvisational ways. And inevitably, as experts have always warned, that means crashes.

Questions of fault when things go wrong have been settled over a century for human driving. But they are still largely the stuff of thought experiments for the cyborgs now roving our streets: vehicles controlled by a machine brain, programmed by human engineers, and usually overseen on the road by some other person behind the wheel. This is an ideal example of the power of three.I didn’t consciously think about the power of three! I wanted to get the unsettled question of fault in a self-driving crash high up, which was the philosophical dilemma that drew me to this subject. Also to explain that this isn’t really a matter of man vs. machine; the machine is doing what humans programmed it to do, and those corporate and engineering decisions are also going to be a big issue.  For years, researchers and self-driving advocates had anxiously prognosticated about how the public and the legal system would react to the first pedestrian death caused by a self-driving car. How do you decide when to insert a hot link in your narrative?The edit team decides what links to add, and they space them out throughout the piece with the hope of adding context and backup for certain facts and at certain moments.

The crash in Tempe ripped those musings into reality—forcing police, prosecutors, Uber, and Vasquez into roles both unwanted and unprecedented in a matter of seconds. At the scene that night, Vasquez stood at the center of a tragedy and a conundrum. She couldn’t yet fathom the part she was about to play in sorting out where the duties of companies and states and engineers end, and the mandate of the person inside the car begins. Do you consider the previous three grafs what you consider the nut graf of this story?Yes!Why did you decide to step back from the narrative of the crash to put it in context here?I first learned in Franklin’s “Writing for Story” that the lede asks a question, either literally or implied, that then pushes the reader to keep going in order to find the answer. So this nut graf is widening out the stakes, showing those bigger questions for us as a society, as well as the specific question of what will happen to Vasquez.Were those passages a challenge to write?A strange phenomenon is how when I go back and read a published piece, the sentences and structure have the air of inevitability, when, in truth, these decisions were only arrived at through an arduous creative process. We really labored on the nut graf for this complicated, multifaceted story, a story that still has no conclusion, since this criminal case is still pending. The exact language only locked into focus in late revisions.

“I’m sick over what happened,” Vasquez confided to the police as her mind spun in the hours after the crash. She said she felt awful for the victim’s family. She also grieved the event in a different way—as a loyal foot soldier of the self-driving revolution. “Oh God, this is going to be a setback for the whole industry,” Vasquez told Loehr. “Which is not what I want.”

At the time, Vasquez was an Uber defender. She had come a long way to this job. Over the previous few years, she’d acquired a dizzying track record of doing hidden work for highly visible companies—moderating grisly posts on Facebook, she says; tweeting about Dancing With the Stars from ABC’s Twitter; policing social media for Wingstop and Walmart. But her position with Uber’s Advanced Technologies Group had offered new stability, and after years of turmoil as a transgender woman navigating a hostile society, she was careful not to jeopardize it. Did you intend for this sentence fulfill the promise of the lede’s foreshadowing?We’re giving a bit more detail about the “distrust” we were talking about in the first sentence, parsing out enough information to start to feel we’re getting to know Vasquez. But even this is still summarizing and foreshadowing: we’re going to get more into the detail later. Vasquez had even removed the box braids of colorful yarn that had defined her look since she was young. At a new job, she had come to think, “the less attention I bring to myself, the better.” During her nine months of work as an operator, the viselike grip of everything she’d endured as a child and teen and adult had slackened just a bit. As she trudged into her forties, Vasquez had felt her life, finally, relaxing into a kind of equilibrium. How do you know this?Vasquez said this in our interview.

Now, as she and Loehr sat in a victim services van near the Tempe bridge after midnight, grappling with Herzberg’s death, the vise was tightening again. She found herself asking, “Do I need a lawyer?” How did you get on to this story in the first place?There was this widely viewed video in the seconds leading to the crash, so many people had seen Rafaela Vasquez’s anguished face in that moment. But she hadn’t given any interviews, so little was really known about her. With some Googling, I was able to find out that Vasquez had worked tech-related jobs for years, in tech support and customer service. At one point she had even been featured on a blog as an avid beta tester, talking about how she loved tech. It’s always a good sign when an idea gets more interesting as you learn more. I was also intrigued, philosophically and legally, in the question of fault when both humans and self-driving tech (itself designed by people, don’t forget) play a role in a car crash. The fact that only one person was being held criminally responsible, and yet the terrain here was so complex, also seemed ripe for further digging.Did you have a goal when you set out and worked on the piece?I wanted people to understand Vasquez, and what she was at the center of — that this crash isn’t just about her. I also wanted to tie together the many complicated threads swirling around this tragedy in a readable way and unearth public records that might add to the public knowledge of this case.

ARIZONA WELCOMED UBER’S self-driving program to Tempe with feisty, high-profile panache, after a long courtship. Business-­boosting governor Doug Ducey, the former CEO of Cold Stone Creamery, took office in 2015, promising to yank the state out of its post-­recession doldrums. He wanted to lure Silicon Valley companies over the Arizona border, pitching his state as the anti-California with trollish flamboyance. He axed restrictions on Theranos blood testing, welcomed an Apple data center, ended local bans on Airbnb, and pressured officials to let Ubers and Lyfts roll up to Phoenix’s largest airport. Once again you back off the narrative to present context. How did you arrive at the story’s architecture?It would have been valid for this section to continue with the early hours and days of the investigation to keep the plot rolling. I sometimes think some longform stories cut to background a bit too early before the reader has gotten totally invested. But in this story, I wanted to start showing the greater landscape this crash happened upon early, so we cut back in time here and then start working our way back to the crash chronologically.

Along the way, Ducey’s office and Uber entered a mutual embrace. At one point, a Ducey staffer emailed Uber and referred to a 2015 Arizona law that regulated ride-sharing as “your bill.” At times, Uber suggested tweets for the governor’s office account and talking points for press events. How were you able to describe Ducey’s courtship of Uber in such detail?I did a public records request for the emails. I try to always brainstorm what public records exist about my story topic, even on pure features. Records have brought me so many incredible details over the years, things I wouldn’t have learned by other means. In June 2015, Uber opened a customer service center and pledged to hire 300 Arizonans. And in August, Ducey signed an exuberant executive order allowing companies to test self-driving vehicles on public roads.

All of that was good for Uber. At the time, its CEO, Travis Kalanick, saw the development of robotaxis as an existential battle, particularly with Google. The company had to at least tie for first place in the autonomy race, or else, he said in an interview, “Uber is no longer a thing.” Why did you rely on Business Insider for this quote?They did the seminal interview with Kalanick at that critical juncture, in which he stated this existential fear. Self-driving car companies were making a lot of claims in the press in those days, adding to the fervor. Human drivers could never compete on cost. But by the beginning of 2015, Kalanick’s company was way behind. So Uber poached 40 experts from Carnegie Mellon’s robotics department to create something called the Uber Advanced Technologies Group and tasked it with turning the company into a self-driving force. Uber shoveled hundreds of millions of dollars into the self-driving unit, which would, over the next three years, grow to more than 1,000 employees across five cities. Your story is written with compelling authority. Some of what you present, such as most of the opening scene, is not attributed or the attribution is delayed. Why did you choose this narrative approach, rather than including information about your reporting or sources, either in the story or a footnote or sidebar, as some narratives packages do? And how did you decide what to attribute directly?Wired has an extensive fact-checking process. I annotate nearly every factual assertion with its source, which the fact-checker is then going to vet. For a piece this size, the process takes weeks. That rigor, the promise that the words here have been fully scrutinized, is part of the magazine’s covenant with the reader. We tend to sprinkle attributions in gingerly. Still, I’m a fan of the sidebar approach that many publications are doing these days on “how we got this story,” instead of gumming up the reading experience with tons of attribution.

In 2016, with Google’s Waymo and GM’s Cruise already piloting prototype self-driving cars around the Phoenix area, Ducey spotted yet another way to remake his state as the nation’s self-driving test capital. That December, California revoked the registrations on Uber’s test cars after the company refused to get a testing permit. Within hours Ducey tweeted, “California may not want you; but AZ does!” The next day, Uber’s Volvos were loaded onto semitrailers bound for Arizona. At the time, federal regulators were standing back, suggesting that companies voluntarily report their safety practices, and recommending states do the same.

In a secretive industry, miles driven in autonomous mode were a key signal of a program’s vitality. So throughout 2017, as Arizona became the largest site for Uber’s testing, employees recall company leaders demanding that the operators “crush miles”—hundreds of thousands, then millions, of them. “These were pretty purposely outrageous goals,” says Jonathan Barentine, a former employee who trained the human backup operators. “We were trying to ramp up really quickly, which at the time was what Uber was good at—or able to do.” How did you find Barentine, who proves to be such an important source, and get him to talk to you on the record?I’m not a beat reporter with a long list of existing sources, so I cold-contacted dozens of former Uber ATG employees, finding most of them on LinkedIn. By late 2017, Uber boasted that it was racking up 84,000 miles a week.

Soon Uber was running 40 cars across thousands of Arizona miles on up to eight shifts a day, with human pilots rescuing the fledgling robots when they went awry, and regulators barely watching. When Arizona welcomed Uber’s audacious program, Bryant Walker Smith, a leading scholar of self-driving policy, told the San Jose Mercury News that Ducey would symbolically “own” the company’s self-driving future—whether that be success or a high-profile crash. In California, Smith had recommended to the state’s officials that they revoke Uber’s registration; as for  Arizona’s quick embrace of the same program, he warned, “There are risks to that level of permissiveness.” Would you describe the research that undergirds your story?Extensive consumption of past news coverage, gathering court records and police reports, poring over the NTSB report on this crash, filing public records requests (both to learn new information and corroborate facts I got elsewhere), interviews with ex-Uber employees and self-driving experts and reaching out to many others close to this case who declined comment. In a long reporting process like this one, where your knowledge builds so much over time, I find it important to re-read documents at different junctures; there’s always some detail that will take on new meaning as you learn more.How long did it take to complete the research, reporting, drafting and revision?Once the attorneys agreed, it was time to start on the vast amount of research it was going to take to put an interview into context.  I sent out a bevy of records requests and pestered many of those agencies throughout 2021 to fulfill them.  I also read everything I could find about the crash and Uber’s testing program. I’m copy-pasting everything that seems important into a timeline — the primary tool I use on nearly every story, especially ones about criminal cases, where chronology is especially important. Even if all those events don’t make the cut, I need that backbone to know the contours; I need it when outlining and writing and later annotating the piece for the fact-checker. (I always include the original source for each entry so I can find it again.) I spoke with Vasquez in August 2021, and I started writing that fall while continuing to report, including cold-contacting dozens of former Uber employees. The editors spent time with the drafts, giving me the ability to step away from the piece and return with a little bit of distance from my words. Then I plunged in again that fall and winter. The piece published in March 2022 — so about a year and a half after I first learned I’d be able to speak with Vasquez.

OVER THE COURSE of 2017, the Advanced Technologies Group brought on hundreds of test operators in Arizona. You weave multiple timelines: the crash, Vasquez’s life before and after her Uber career, the rise and fall of the self-driving industry, among others. How did you keep track of them?My master timeline was crucial — and was color-coded by what storyline each nugget pertained to.  Jonathan Barentine, a friendly and precocious program manager who was just a few years out of studying liberal arts at Cornell, was posted in Tempe to oversee training for the new recruits. He remembers that Vasquez, hired that summer, took the training so seriously she appeared stressed. “It seemed like a bit of a big break for her,” he says. “She really cared about making sure that she could do her job.”

For many of the new operators, coming off work on cleaning crews or as delivery drivers or regular Uber drivers, walking into the Advanced Technologies Group’s Tempe headquarters was like entering a Silicon Valley Shangri-la. The facility came with an alluring nickname—Ghost Town—from its days when few employees reported to the sprawling office-park building. The name had stuck even as exploding ranks of workers dropped in for car assignments, free catered meals, and a break room packed with Red Bull and snacks. The operators earned full benefits and about $20 to $24 an hour—solidly middle-class wages in the area. Vasquez’s coworkers were buying houses and booking vacations. Workers marveled at the latitude and trust of the company’s culture: Everyone gave feedback at regular debriefs, and managers let workers take breaks as needed to stay sharp on the road. Some stayed after hours to play video games. What was the reporting you did that made it possible to describe Ghost Town’s culture with such specificity?Other staffers I reached out to filled me in with precise details about their jobs and workplace. Remember that not every source has to tell you every detail: I didn’t need to learn what Ghost Town looked like from Vasquez when there were hundreds of other people who worked there. And they worked at the vanguard of tech. Flavio Beltran, an operator in the Tempe program, says, “I felt like, wow, I’m a part of history. I felt a very huge sense of pride.” How did you find Beltran?Beltran was one of the few who had appeared in some news reports about this crash. ContactOut is a web extension that will surface emails from LinkedIn.