This article originally appeared in The New York Times Magazine and is reprinted on Longform by permission of the author.
WE HAD BEEN on the island for about an hour when we found the first skeleton. It was a pile of yellow bones tucked inside a cardboard box. Mark Noah squatted down for a look. He is a stocky man of 48, with a light buzz of blond hair and the wind-beaten eyes of a lifelong outdoorsman. Since 2008, he has been traveling to the tiny Pacific atoll of Tarawa to search for the remains of more than 500 Marines who died there in World War II. Sometimes locals dig up their bones and leave them in his storage locker.
Noah reached into the box and pushed aside a fragment of cranium to remove a curved metal plate. “Wow,” he muttered. “Clearly a World War II burial with the helmet.” He passed it to the man crouching next to him, Bill Belcher, and added, “It looks American.”
Belcher nodded. “That’s what I thought when I saw it.” He laid the piece back in the box and picked up two sections of jawbone with the teeth still attached. They fit together into a complete lower mandible, which Belcher held close to his glasses, squinting. Noah pulled another hunk of metal from the box. “And this is a hand grenade,” he said. He shook his head and smiled. It was all pretty normal on Tarawa.
THIS WEEK MARKS the 70th anniversary of the Battle of Tarawa. It was a fight that lasted only three days, but they were among the bloodiest in 20th-century American history. By the time the battle ended, more than 1,100 U.S. Marines lay dead on the sandy earth and churning water.
By themselves, the islands held little value to the Japanese or the American government. They were situated about halfway between Pearl Harbor and the Philippines and were barely large enough to hold an airfield. But they served as an essential steppingstone across the Pacific: If American bombers wanted to reach Japan, they would need an air base in the Mariana Islands; to capture the Marianas, they would first need the Marshall Islands; and for the Marshalls, they needed Tarawa. To fortify the atoll, the Japanese sent in 3,800 imperial troops, along with 1,200 enslaved Korean laborers to be thrust onto the front lines. They spent a year building concrete bunkers and planting massive cannons along the beaches. The leader of the Japanese garrison, Rear Adm. Keiji Shibazaki, predicted that it would take “one million men, 100 years” to seize the islands.
American commanders selected the Second Marine Division for the job. In the fall of 1943, they boarded a convoy of battleships and destroyers, which the legendary war correspondent Robert Sherrod described as “the largest force the Pacific has seen.” Sherrod climbed aboard. As they drew close to Tarawa, he watched the ships unload 2,000 tons of explosive shells while American planes laid another 900 tons of bombs across the islands. It was, Maj. Gen. Julian Smith wrote, “the greatest concentration of aerial bombardment and naval gunfire in the history of warfare.” Into this inferno, the Second Marines disembarked. They descended into amphibious landing vehicles and raced toward the beach, but the tide was out, and the water was too shallow for their boats. The Marines found themselves stranded on reefs, hundreds of yards offshore, wading through waist-high water as Japanese gunners mowed them down. Those lucky enough to reach the shore crawled through a maze of corpses. “No one who has not been there,” Sherrod wrote, “can imagine the overwhelming, inhuman smell of 5,000 dead who are piled and scattered in an area of less than one square mile.”
After their victory, the Marines set about burying the dead. They wrapped the bodies in ponchos and folded them into shallow graves. Then they moved on, and military construction crews came in to raze the island flat. They expanded the airfield and built a network of roads and offices. By the time an excavation team arrived in 1946 to exhume and identify the dead — part of a global campaign to recognize the fallen — no one could remember where they were. Investigators spent three months searching, but they found only half the Marines. Today, 471 of the Tarawa Marines are buried by name in American cemeteries. Another 104 have been laid to rest in “unknown” graves at the National Memorial Cemetery of the Pacific in Hawaii.
And the rest — perhaps as many as 520 Marines — are still on, or near, Tarawa.
MARK NOAH HAD come to find them, though he couldn’t fully explain why. He is a private citizen with no formal connection to Tarawa or the Marines, but over the last several years, he has been consumed by a personal quest to locate the lost graves.
Noah was born in 1965 in Taipei, the son of a Foreign Service worker. By the time he turned 18, he had spent years in China, Korea, Finland, Thailand, the Philippines and Russia, where he became deeply conscious of the lingering effects of war. “Thousands of maimed Russian World War II veterans lived out on the streets,” he said. “A lot of the guys would still have their tunics on and all their medals, missing an arm or both legs, and living in the street in the middle of the winter in Moscow. And so I became very interested in the history.”
Today Noah works as a commercial pilot, but his passion for history defines him. In 2001, he bought and restored a 1945 Navy airplane called an SNJ-6. A couple of years later, he bought another and began offering rides to veterans who once flew them. Then he incorporated the operation under the name History Flight and began to expand his fleet. While researching vintage planes online, he was stunned to discover how many planes, and men, were still missing from the war. Nearly one in five American losses had never been found. “I had no idea,” he said, “because if you look at the standard mainstream history, it’s not in there.” Noah began to visit the National Archives in search of information about the missing troops, and in 2007 came upon a postwar report from Tarawa. It said that nearly half the Marines who died there had not been found. “I almost fell out of my chair,” Noah said, “because I knew that over 1,100 people had died.”
Since then, Noah has traveled to Tarawa more than a dozen times. He has canvassed the neighborhoods and alleys with ground-penetrating radar and magnetometers, completed the first RTK GPS survey in the region and launched a camera-mounted drone to produce a 62-foot-long aerial photograph. He has also hired a small team of archaeologists and historians who join him on the islands, along with a handful of local employees.
In the process, he has enraged parts of the U.S. military, including the man who stood beside him now.
On the surface, Mark Noah and Bill Belcher seem to have much in common. Each is physically imposing, wry, stubborn and occasionally prickly, with a penchant for colorful language. But Belcher wasn’t part of Noah’s team, or even really a friend. For 15 years, he has worked as a field archaeologist in the Joint P.O.W./M.I.A. Accounting Command, or JPAC, the military unit that recovers the remains of missing troops. And Noah’s work on Tarawa raised complicated issues for JPAC and Belcher. For months leading up to our October trip, the two men were locked in a feud with lasting implications for M.I.A. recovery around the world.
Noah first contacted JPAC in 2008. Over the course of a year he had assembled more than a hundred archival photographs that showed the original burials on Tarawa and had pieced together a theory about where the graves might be. But JPAC archaeologists are famously resistant to outside help. Enthusiastic civilians contact the unit routinely to offer information about sunken ships and planes and graves. The trouble is that many of those volunteers turn out to be little more than treasure hunters who have already pillaged the sites for artifacts, like guns, watches and coins.
“I always have a very, very deep, deep level of skepticism about people involved in this,” Belcher told me. “Particularly with World War II, because a lot of people want it for notoriety. They want to feel important. And the artifacts are actually worth quite a bit of money to collectors.”
When Noah delivered his initial Tarawa findings to JPAC, the unit largely ignored him. “They blew us off, and that was that,” he said. Noah tried to push back, issuing a news release about his project and reaching out to public officials. A Chicago alderman who served in the Marines put him in touch with Representative Dan Lipinski, and in 2010, under pressure from Lipinski, JPAC sent a team Tarawa to investigate Noah’s theories. But when the JPAC team dug four sites and found nothing, each side blamed the other: JPAC insisted that Noah’s research was worthless; Noah blamed JPAC for digging in the wrong places.
Noah kept going back. He began to experiment with new technologies. He bought a magnetometer to search for traces of metal underground, which might be a gun or a dog tag. He brought a cadaver-sniffing dog to search for the odor of decay. He took soil samples to study for subtle chemical clues. At one point in 2010, he says, the JPAC leadership tried to stop his research. An immigration official for Kiribati, the republic that includes Tarawa, emailed Noah that he would need embassy clearance to return to the island. Noah contacted Lipinski, who reached out to the State Department and, after nearly a year, was able to restore Noah’s access.
Noah had been in touch with the head of the veterans association for the Second Marine Division, a retired colonel named Dave Brown. In the summer of 2011, they flew to Hawaii to meet with the JPAC staff. It was a disaster. A lead archaeologist in the unit, John Byrd, attacked Noah’s methodology. “I got about two sentences into the first description, and Dr. Byrd was just crapping all over everything,” Noah said.
A month later, though, Noah received an unexpected call from Johnie Webb, one of JPAC’s senior leaders. Webb had been with the recovery unit since its beginning. Over the years, he led missions, commanded the unit and, after his military retirement, stayed on as a liaison with the families of missing men. Webb told Noah he was intrigued by his work and willing to bend the rules. “So he says, ‘Could you coincidentally be on Tarawa on the 5th of October?’ ” Noah recalled. ‘ “Because I coincidentally might have some of our people out there.’ ”
Noah spent four days on the islands with a JPAC analyst named Jay Silverstein, who immediately sensed that Noah’s work was serious. “It was really exceptional,” Silverstein told me. “Mark had put together a crack team. They were innovative. They were open to new ideas. Tarawa is one of the most complex battlefields we have, and Mark was using the right composite of scientific members to piece together that puzzle.” Back at JPAC, Silverstein began pushing for a new recovery team to return to Tarawa and excavate Noah’s target sites. He encountered fierce resistance from some of the unit’s archaeologists, but last year, JPAC committed to sending another excavation to Tarawa. This time, the lead archaeologist was Bill Belcher.
Belcher arrived on the island last fall with modest expectations. By the time he left three weeks later, he had successfully excavated three sets of skeletal remains that he believed to be Marines, from a neighborhood Noah calls Cemetery 25.
Then the relationship between Noah and Belcher fell apart. Belcher had promised Noah that if JPAC returned to Tarawa for another dig, Noah could join them. But early this year, Noah’s contacts on Tarawa informed him that JPAC was already there. Noah accused Belcher of lying, which Belcher denied vehemently by email, leading to a vituperative exchange that culminated when Belcher told Noah never to speak to him again.
For the next six months, Noah did not. Instead, he began to dig. Between February and July, Noah and his team were on Tarawa almost continuously. Finally, in August, Noah flew to Hawaii to reveal his results to Johnie Webb. “I didn’t even want him to come to JPAC, because I didn’t want him to get attacked again,” Webb told me. Instead, Webb brought the commanding general of the unit, Kelly McKeague, to meet Noah at a Chinese restaurant. Over lunch, Noah revealed that he had begun to dig on his own. He had filled seven large cases of material for JPAC — including dog tags, boots, ponchos, helmets, ammunition clips and the skeletal remains of at least 50 U.S. Marines.
Webb laughed when he recalled the conversation, but he said the prospect of telling Belcher filled him with dread.
Belcher’s response surprised him. “I said: ‘You know what? We can’t stop him,’ ” Belcher recalled. “ ‘Let’s see what he’s doing.’ ”
NOW BELCHER AND Noah were back on the islands, standing over a box of bones. We piled them into the car and set off down the road. From the size of the cranium and the shape of the grenade, both men were beginning to suspect that the skeleton was not American but Japanese. But they wanted to find out where the bones were found. Nearly everyone who lives on Tarawa has dug up bones at some point. The water table is only a few feet deep, so nothing is buried deeply. With so many skeletons being found and reburied, year after year, the bones can get mixed up. A Japanese cranium could be intermingled with American remains.
We drove through a series of shacks built from sticks and logs. Most were only a few feet square, with open sides and a slag of thatch on top to slow down the rain. Seventy years ago, none of this was there. The construction crews that razed the landscape and covered over the graves erected a number of stark geometric cemeteries lined with white crosses — but these were largely symbolic. The unit that came to exhume the graves in 1946 quickly discovered that most of the crosses weren’t placed over actual graves. “These ‘cemeteries’ were placed without any relation to the actual burials,” the leader of the ’46 recovery team, First Lt. Ira Eisensmith, wrote in his report on Tarawa. Many of the skeletons that Eisensmith and his team found were missing their hands and feet. Noah, who has studied the postwar report, questions its accuracy. “It’s a real C.Y.A. report,” he said. “It says that many of the Marines were buried without their dog tags, and that’s not true, because we’ve been recovering their dog tags. And it says that many of the dog tags were illegible, yet that’s not true — we’ve recovered many legible dog tags 70 years later.” Noah suspects that most of the Marines were intact when they were buried, and that Eisensmith’s team did incomplete excavations that left some of the remains in the ground. “I’m not one to make a big deal about it,” Noah said, “but the people that did this recovery work did a terrible job.”
The conditions on Tarawa today only complicate the process further. Tarawa is one of the most impoverished and overpopulated atolls in the Pacific. The main island, Betio, has more than 20,000 residents, crammed into about half a square mile, which is roughly the population density of Hong Kong. With this overpopulation comes environmental crisis. Garbage is strewn across the beaches and neighborhoods, along with human waste. Many islanders use the beach as a toilet, which pollutes the aquifer and the sea. Outbreaks of typhoid, hepatitis and tuberculosis, along with increasing incidents of leprosy, have struck in recent years, making life for islanders grim and recovery work all the more difficult.
Half a mile down the road, we turned into a dusty lot. This was the neighborhood of a man named Kautebiri Kobuti, who leads Noah’s crew on the islands. Kobuti was in the yard taking a shower with a jug of water. He explained that the bones in the cardboard box had been excavated by his friends. We drove to their house, where four men stood around the edge of a sandy pit while a fifth worked the bottom with a wood-handled shovel. Noah jumped into the pit and examined the ground. After a moment, he stood up with a small bone balanced on a leaf. “Just a fish vertebra, right?” he asked.
Belcher studied the bone. “No,” he said, “that’s a dog.”
Noah turned to Kobuti. “There’s more in here?”
Kobuti conferred with the men, who nodded vigorously.
“Can we scrape some of it away?” Noah asked, hunching down to study something lodged in the side of the pit. As he did, another man hurried from the house with a pink grocery bag filled with bones. A large femur protruded from the top.
“O.K.,” Belcher said, “those are definitely human.”
IT CAN TAKE thousands of hours and millions of dollars to bring home bones from a place like Tarawa, and even then the bones must be tested for DNA and matched with a surviving relative — all so that they can be buried again, closer to home. The closure that this gesture provides to families is difficult for outsiders to grasp, and only a handful of researchers have ever focused on the M.I.A. experience.
One of the first to do so, in the early 1970s, was a doctoral student at the University of Wisconsin named Pauline Boss. She was interested in the way that families cope with uncertainty. Whether it’s the sudden disappearance of a child or the slow erasure of a parent by dementia, the grief process is complicated by a disrupted narrative, because so much of grieving depends on the understanding and acceptance of what has happened.
When Boss presented her ideas, which she would later call “ambiguous loss,” at a conference in 1973, she was approached by a social scientist, Edna Hunter-King, who had been working with the relatives of missing men from the Vietnam War. “She said, ‘We have data on this, but no theory,’ ” Boss recalled, “ ‘and you have a theory with no data.’ ”
In the decades to come, they would develop a small literature on M.I.A. families. What distinguished the missing soldier from other combat losses, they found, was that the family was deprived of a clear explanation for their loss. “When you have someone missing, it does something to the human psyche,” Boss said. “There are no rituals for it. The rest of the community doesn’t know what to do. And grief therapy doesn’t work.”
Boss was also interested in the way M.I.A. grief passes down through generations. In many cases, a daughter, a son or a grandchild would become fixated on the loss of a man she or he had never known.
Over the past five years, while working on a book about a missing B-24 bomber from the Pacific war, I’ve met with dozens of M.I.A. families who have been profoundly shaped by the disappearance of a relative they never met. Lisa Phillips was born in 1963, but she grew up in the shadow of her uncle Joe, who was captured by the Japanese in World War II and killed in a prison camp. A military recovery team went to exhume his remains in 1947, but as they flew across the Himalayas on their way home, their plane went down. The remains of Phillips’s uncle, and of the recovery team that found him, are still there. Phillips said the loss has been part of her family ever since. “Every holiday, my grandmother would burst into tears,” she said. “She kept a room, his bedroom, the way it was when he left.” One of Phillips’s aunts could not accept that Joe was gone. Until the end of her life, she refused to move from her Connecticut house, saying, “I know he’ll be home someday, and I want him to be able to find me.” Today Phillips is the head of a group known as World War II Families for the Return of the Missing, which has encouraged the expansion of JPAC’s purview.
The modern recovery program, which includes JPAC and seven other offices spread throughout the Department of Defense, has operated in roughly its current form since 1973. For most of those years, the unit’s focus has been on Vietnam. That’s because advocates for the Vietnam missing are organized and politically savvy. But in recent years, M.I.A. families like Phillips’s have pressured Congress to widen the recovery mission. In 1994, legislators added the missing from Korea; in 1999, they folded in the aviators and airmen who flew over New Guinea in World War II; and in 2009, they added the rest of the World War II missing. In the space of only a few years, JPAC’s mandate ballooned from 1,647 personnel in the Vietnam area to some 83,000 M.I.A.’s around the globe.
World War II families see this as a matter of parity. They say there is no reason the missing from one war should be recovered but not those from another, because the legacy of loss persists in their families as well. “It’s been carried down,” Phillips said. “Even my kids are like, ‘You know, if something happens to you, I still want to get Uncle Joe back.’ ”
By 1998, Boss and Hunter-King had found so many M.I.A. families in which the grief haunted a second or third generation that Hunter-King was invited to draft a new chapter for the clinical handbook of multigenerational trauma, alongside entries on the effects of slavery, nuclear annihilation and the Holocaust. “Unlike the Holocaust,” she wrote, “mothers of M.I.A. children were not suddenly uprooted from their homes and deprived of their possessions, countries and cultures. They did not lose parents, siblings and husbands to programmed incineration. . . . On the other hand, most children of Holocaust survivors have not waited for over a quarter-century in a state of ambiguous grieving, wondering whether their parent is dead or alive, as children of M.I.A.’s have done.”
Many M.I.A. relatives are also burdened by a family story of how the missing man might have survived — perhaps he is lost, or has amnesia, or is being held captive. The son of a missing Army Air Force flier once told me of the great relief he felt at a meeting of M.I.A. families. “You go to these meetings,” he said, “and everybody has a story like mine. You know, somehow the guy survived. He’s still going to come home. You hear that from people who lost somebody in the Pacific, in Europe, in Africa, all parts of the world. As soon as they start talking, there’s something along that line. It’s a mistake. It’s not right. There’s something funny about it. Every time.”
This form of hope can seem corrosive and difficult to comprehend. Even within individual families, Boss explained, the persistence of survival stories can lead to profound conflict. “A family rift is almost predictable,” she said. “When you have no facts, no proof, then everybody makes up their own story — and the stories that family members make up often do not agree.”
This phenomenon makes the JPAC recovery effort not so much a military mission as a humanitarian cause. Deno Zazzetti was 13 when his brother died on Tarawa. The news arrived on Christmas Eve, 1943. “A Western Union kid drove up with his bicycle, gave the telegram to my sister and took off,” Zazzetti told me. “My sister yelled, and we found her on her knees.” That moment, he said, marked a breaking point in his life. “My ma used to sing all the time. She never sang again. She had pitch-black hair — a year later, she was solid white.” Today Zazzetti keeps an empty plot beside his mother’s grave. “For me,” he said, “it would be the best thing that I could do in my lifetime, if I could get my brother back to my mom.”
ON TARAWA, NOAH and Belcher wanted to sort through the bones, so we drove to a concrete government building where Noah rents an office. The room is tiny, about 150 square feet and lined with the portraits of missing Marines. A skeleton stands in the corner, by a bookshelf stacked with the white crosses that once marked Marine graves.
Noah set the box of bones on a workbench; then he and Belcher stretched on rubber gloves and began to sort through the pieces. Noah began with the cranium fragment. “Somebody already did the lifting-the-head ritual,” he said, referring to an island custom that involves pulverizing and then eating a skull.
Belcher rummaged through a jumble of vertebrae and pulled out a nasal bone. “They’re Japanese,” he said. “We have a peaked nose, and Asians have a relatively flat area. This nasal area is flat instead of peaked.”
The other evidence in the box — helmet, bullets and the hand grenade — also looked to be Japanese, so Belcher and Noah began to pack everything into an evidence bag. They would take this to the local police station, where Japanese officials could pick it up. The custom in Japan is to incinerate the bones in keeping with the Shinto tradition.
Next, Noah pulled out a bag of American artifacts. His team recovered so many Marine skeletons this summer, more than 7,000 bones, that they still had not found time to sort through all the other artifacts. For 30 minutes, Noah and Belcher pored over each item: a metal ring, the chin strap for a helmet, dozens of coins and buttons and a pair of dice.
Noah approaches his work on Tarawa with little sign of emotion, and he spends little time speaking to the descendants of the missing. One exception came this summer, when he traveled to Indiana for the funeral of a Marine named Manley Forrest Winkley, whose skeleton was one of three that Noah led JPAC to recover last fall. Winkley stormed the north beach of the battlefield on the first day of combat and was killed by a shot to the neck. When the JPAC team uncovered him, three feet below a pigpen, he was still wrapped in his service poncho.
On Aug. 24, Winkley’s family gathered around the coffin holding his remains in Nashville, Ind., then drove 60 miles in procession to a veterans’ cemetery. All the way, the streets were lined with veterans, children, firefighters and church groups who had come out to welcome home a Tarawa Marine. “There were probably about 4,000 people standing on the side of the road,” Noah said. “All these little towns were shut down. It made me feel like we had helped to put some of the America back in America.”
WHILE NOAH WAS in Indiana for the funeral, JPAC was under siege. All summer the unit was inundated with criticism about its management and methods. A report in July from the Government Accountability Office revealed the longstanding friction between JPAC and another agency of the recovery program, the Defense Prisoner of War/Missing Personnel Office, or DPMO. That same month, another report surfaced in The Associated Press, disparaging the unit’s historians. In early August, two congressional committees began inquiries into the reports. And in October, an NBC News story accused JPAC of holding “phony ceremonies” in which previously recovered remains were unloaded by an honor guard in full military dress.
Most people familiar with the recovery program knew these reports were thin. The tension between JPAC and DPMO goes back years, and it is typical of any interagency program. Likewise the report attacking JPAC historians was difficult to take seriously — it was commissioned by JPAC’s laboratory staff, which competes with the historians for influence and funding. Senator Claire McCaskill, who is leading one of the congressional inquiries, told me the tone of that report, filled with ad hominem attacks, made her instantly suspicious. “It was bizarre,” McCaskill said. “It was a giant, childish mess.”
The story of the “phony ceremonies” was equally peculiar. It described a purely symbolic ritual that takes place after each successful mission. Anyone who had actually been on a mission knew that there was another ceremony, which took place in the field, where the recovery team stands at attention as the human remains are loaded onto a transport jet, then flies 10 or 15 hours around the world, often arriving at JPAC headquarters after midnight. The idea that someone at the symbolic ceremony in Hawaii might believe the honor guard in dress uniform, carrying a flag-draped coffin, was actually a recovery team returning from six weeks in the jungle was absurd.
The real problems facing JPAC were more serious and complex — and they were embodied on the Tarawa atoll. With more than 73,000 new cases from World War II, many of them deep in the Pacific, on islands filled with homes and people and racked with environmental damage, and with a testy public-private partnership struggling to bring home what may prove to be the largest M.I.A. recovery in American history, Tarawa was no longer just a distant string of islands. It had become a symbol of the recovery program itself — its history, its mission, its problems and, maybe, some of the answers.
One day last month, I flew to Hawaii to meet with JPAC’s senior leader, Johnie Webb. I’ve known Webb for many years, and we met at a restaurant overlooking the beach to speak frankly about the challenges facing the unit. Webb has come to believe that outside groups like Noah’s can do some recovery operations as well as JPAC. The question was how to change the culture of JPAC so that those groups would be welcome, without abandoning the unit’s standards and oversight. “Everybody’s not going to be Mark,” Webb said. “Do we set up some type of program, training, whatever you want to call it, and then we certify these groups?”
Belcher, too, had come around on this point. He was planning to spend much of November and December coordinating with a private operation in the Himalayas, while a younger JPAC archaeologist would spend six weeks with Noah on Tarawa.
The bottom line, Webb told me, was that “there’s more work to do than we can get done.”
On our last day on Tarawa, Belcher and Noah decided to dig together. We drove to a tattered neighborhood with a small crew. The two men knelt on the ground to frame a rectangular space with twine; then they grabbed shovels and began to cut into the earth. They found diapers, chicken bones, a piece of rubber hose. As the sun went down, Noah dug while Belcher leaned against a tree. “Mark,” he said, “I’ve got your next project.”
“What’s that?” Noah said.
“Hürtgen Forest.” It was a battlefield in Germany where hundreds of Americans went missing. Belcher did an excavation for JPAC a decade earlier and has been pushing to go back ever since.
Noah’s face lit up. “I’ve already started on it,” he said.