According to Eric Walters, who was a trainer at Sealand from 1987 to 1989 while working toward a bachelor's degree in marine biology at the University of Victoria, the module was so tight that the orcas had difficulty avoiding conflict, and their skin would get scratches and cuts from rubbing against the sides. About once a week, Walters says, one or more of the orcas would simply refuse to swim into the module and would have to be left in the performance pool overnight.

  The orca show was performed every hour on the hour, eight times a day, seven days a week. Both Nootka and Tilikum had stomach ulcers, which had to be treated with medication. Sometimes Nootka's ulcers were so bad she had blood in her stool.

  Walters was interested in the science of training and was encouraged when Sealand brought in Bruce Stephens, a former SeaWorld head trainer, to make recommendations to improve Sealand's practices. Stephens gave each trainer a handbook, which warned, "If you fail to provide your animals with the excitement they need, you may be certain they will create the excitement themselves." He emphasized that killer whales needed constant change to keep them engaged and responsive, and made a series of recommendations for new learning sessions and playtime for Sealand's orcas. But within a month, Walters told me, Sealand was back to its usual routines. "They basically ran it like you would run McDonald's," he says. "It just can't be good for an animal that is so intelligent to do the same thing every day." (Wright still runs a marina at Oak Bay but declined to speak to Outside.)

  As Stephens had warned, bored killer whales look to make their own fun. If any unusual object ended up in the water, Haida, Nootka, and Tilikum would race for it and play keep-away with the trainers. Once the orcas took something, they were determined to hang on to it. Walters worried about what might happen if one of the trainers—who worked in rubber boots on a painted fiberglass deck—fell into the pool. Many marine parks try to defuse the danger with desensitization training that teaches the killer whales to stay calm and ignore anyone who falls in. The training might start with just a foot in the water (the orca is conditioned to ignore it) but ultimately requires gradually easing an entire person into the pool. According to Steve Huxter, who was the head of animal training and care at the time, desensitization was a Catch-22. After thinking about it carefully, "Bob [Wright] was not willing to take that risk."

  Each whale had a distinctive personality. Tilikum was youthful, energetic, and eager to learn. "Tilikum was our favorite," says Eric Walters. "He was the one we all really liked to work with."

  Nootka, with her health issues, was the most unpredictable. According to Walters, Nootka pulled a trainer into the water. (Walters quickly yanked her out.) Twice she tried to bite down on Walters's hands. Not even the audience was safe. A blind woman was once brought onto the stage to pat Nootka's tongue. Nootka bit her, too.

  Frustrated, Walters quit in May 1989. A year later he wrote a letter to the Canadian Federation of Humane Societies, to share with participants at a conference on whales in captivity. In it he detailed Sealand's treatment of its marine mammals and the safety concerns he had. In closing, he wrote, "I feel that sooner or later someone is going to get seriously hurt."

  On February 20, 1991, Sealand had just wrapped up an afternoon killer whale show. Keltie Byrne, a twenty-year-old marine-biology student and part-time trainer, was starting to tidy up when she misstepped and fell halfway into the pool. As she struggled to get out, one of the killer whales grabbed her and pulled her into the water. A competitive swimmer, Byrne was no match for three orcas used to treating any unusual object as a toy. "They never had a plaything in the pool that was so interactive," says Huxter. "They just got incredibly excited and stimulated." Huxter and the other trainers issued recall commands and threw food in the water. They tried maneuvering a life ring close enough for Byrne to grab, but the orcas kept her away from it. In the chaos and dark water, it was hard to see which killer whale had her at any one time. Twice she surfaced and screamed. After about ten minutes, she popped up a third time for an instant but made no noise. She had drowned.

  Bryne was the first trainer ever killed by orcas at a marine park. It took Sealand employees two hours to recover her body from Nootka, Haida, and Tilikum. They had stripped off all of her clothes save one boot, and she had bruises from bites across her skin. "It was just a tragic accident," Al Bolz, Sealand's manager, told reporters at the time. "I just can't explain it."

  Paul Spong, seventy-one, the director of OrcaLab, in British Columbia—which studies orcas in the wild—did part-time research at Sealand before Tilikum arrived. He is not so befuddled. "If you pen killer whales in a small steel tank, you are imposing an extreme level of sensory deprivation on them," he says. "Humans who are subjected to those same conditions become mentally disturbed."

  Byrne's death led to a coroner's inquest, which recommended a series of safety improvements at Sealand. The park responded, but according to Huxter, "the wind came out of [Wright's] sails for the business." In the fall of 1991, Sealand contacted SeaWorld to ask if it would like to buy Nootka, Haida, and Tilikum. Sealand closed in 1992.

  If you want to try to get an inkling of what captivity means for a killer whale, you first have to understand what their lives are like in the wild. For that, there's no one better than the marine biologist Ken Balcomb, sixty-nine, who has spent thirty-four years tracking and observing killer whales off the coast of Washington State.

  In early May I meet Balcomb in his cluttered yard on San Juan Island. He's trying to find the source of a leak on his Boston Whaler. His wood-framed house, which also serves as headquarters for his Center for Whale Research, sits perched atop the rocky shores of the Haro Strait, a popular orca hangout; Balcomb says he sees them about eighty days a year from his deck. Inside there's gear all over the place—spotting scopes, cameras, tool kits—from a recent expedition to California. In the middle of it all, on a table, sits an enormous killer whale skull that he picked up in Japan in 1975, when he was a flier and oceanographic specialist for the U.S. Navy.

  Balcomb, of medium build, with a ruddy, sun-baked face and a salt-and-pepper beard, has been carefully photographing, cataloging, and observing the Puget Sound orcas—also known as the Southern Residents—since he was contracted by the National Marine Fisheries Service in 1976 to assess the impact of the marine-park captures. Many people assumed there were hundreds of orcas around Puget Sound. After identifying each individual killer whale by its markings, Balcomb found that there were just seventy left.

  Since then he's become the Southern Residents' scientific godfather, noting every birth and death and plotting family connections. The population, he says, is now at eighty-five orcas, but he won't know for sure until they show up this summer. Talking on his sun porch, Balcomb stresses that one of the most important things to know about killer whales like Tilikum is that in the wild they live in complex and highly social family pods of twenty to fifty animals. The pods are organized around the females. The matriarch is usually the oldest female (some live to eighty or more), who has a wealth of experience and knowledge about where food can be found. Within the pod, mothers are at the center of smaller family groups. Males, who can live to fifty or sixty years, stay with their mothers their entire lives and often die not long after she does. According to Balcomb, separation is not a minor issue.

  The Southern Resident population is made up of three distinct pods. Each pod might travel some seventy-five miles a day, following the salmon and vocalizing almost constantly to keep the entire group updated on who's where and whether there are fish around. Killer whales are highly intelligent. They coordinate in the hunt, share food freely, and will help an injured or ill member of the pod stay on the surface to breathe. Most striking is the sophistication of their dialect. Each family group within a pod uses the same vocalizations, or vocabulary, and there are also shared vocalizations between pods. Balcomb says he can usually tell which pod is about to turn up simply by the sounds he hears through a hydrophone.

  The social and genetic con
nections that bind orcas in the wild are intense. There is breeding between the Puget Sound pods. Sometimes they'll all come together at once and go through a distinctive greeting ceremony before mixing. But they will have absolutely nothing to do with the genetically distinct, transient killer whales that sometimes pass through their waters. (Transients travel in much smaller groups over vast distances and mostly feed on marine mammals instead of fish.) "When you get born into the family, you are always in the family. You don't have a house or a home that is your location," says Balcomb. "The group is your home, and your whole identity is with your group." Aggression between members of a pod almost never occurs in the wild, he adds.

  Puget Sound is small enough that Balcomb used to run into Goldsberry from time to time. Despite their differences, the two men would talk killer whales, drink Crown Royal, and trade stories. Today Goldsberry, seventy-six, lives about one hundred miles away in a small, ground-level condo near Sea-Tac Airport. His only water view is of a man-made lake, and when I go to see him he's busy drilling a walrus tusk that's been made into a cribbage board. Goldsberry has a square head, with close-cropped white hair. His health is fragile, and he has an oxygen tube clipped to his nose. But he still has the beefy arms of a waterman, and he appears unmoved by the controversy of his hunting days. "We showed the world that killer whales were good animals and all of a sudden people said, 'Hey, leave these animals alone,'" he says, sipping a mug of vodka and ice. "I had to make a living."

  Goldsberry has mostly kept his mouth shut about his work for SeaWorld and doesn't much like talking to reporters. "I'm only speaking with you because those idiots out there, mainly the politicians, want to release all the killer whales," he growls. "You might as well put a gun to the whales' heads." He spends the next couple of hours telling me about his cowboy days in the orca business: how he helped build the global trade, how he kept one step ahead of Greenpeace and activists, and how he battled the media, dropping one TV newsman's camera into the water, asking, "I wonder if this floats?"

  Goldsberry says he always got the resources he needed to keep the killer whales coming, and he developed relationships with other marine parks around the world, which would often hold killer whales for him, many of which would eventually end up at SeaWorld. (Balcomb calls it Goldsberry's "whale laundry.") "I would go into SeaWorld and say, 'I need a quarter of a million' or 'a half-million dollars,' and they put it in my suitcase," he says with a grin. "It was good, catching animals. It was exciting. I was the best in the world. There is no question about it."

  Asked about Goldsberry's work for SeaWorld, Fred Jacobs, vice president of communications, denies that killer whales were laundered. "Any killer whale that entered our collection from another facility did so in full accordance with their export and our import laws," he says. "We have imported whales that were collected by other institutions, but they were not collected on our behalf and held for us."

  Goldsberry's last great haul of wild orcas came in October 1978, when he caught six off Iceland. (Five ended up in SeaWorld parks.) He continued to collect all sorts of other animals for SeaWorld for the next decade. When Goldsberry and SeaWorld finally parted ways in the late 1980s, Goldsberry says he was offered $100,000 to keep quiet about his work for two years. He happily took it. SeaWorld's Jacobs explains that Goldsberry's relationship with SeaWorld occurred under prior ownership. "I have no way of knowing if this is true or not," he says.

  Whatever his methods, Goldsberry had helped SeaWorld turn killer whales into killer profits. The company currently has parks in Orlando, San Diego, and San Antonio, which are visited by more than 12 million people annually. Most of those visitors, paying up to $78 each for an entrance ticket, come to see killer whales. Last year Anheuser-Busch InBev sold SeaWorld's marine parks—and seven amusement parks housed with SeaWorld under the Busch Entertainment umbrella—to a private-equity giant, the Blackstone Group. The purchase price was reported to be $2.7 billion.

  One of the keys to SeaWorld's success was its ability to move away from controversial wild orca captures to captive births in its marine parks. The first captive birth that produced a surviving calf took place at SeaWorld Orlando in 1985. Since then SeaWorld has relied mostly on captive breeding to stock its parks with killer whales, even mastering the art of artificial insemination. "Early in the morning, the animal-care crew would take hot-water-filled artificial cow vaginas and masturbate the males in the back tanks," says John Hall, a former scientist at SeaWorld. "It was pretty interesting to walk by."

  Tilikum's sudden availability in 1991 was a boon to the captive breeding program. While preparing to transfer Haida, Nootka, and Tilikum to Orlando, SeaWorld, one of only a few facilities with the expertise to care for them, discovered that Tilikum had already impregnated Haida and Nootka. A sexually mature male, even one involved in a dangerous incident, was a welcome addition. "It was not the only reason [SeaWorld] had interest but definitely a part of the decision," says Mark Simmons, who worked as a trainer at SeaWorld from 1987 to 1996 and was part of a team sent to Sealand to manage Tilikum's transfer. Media reports at the time pegged Tilikum's price at $1 million.

  If Sealand was like a McDonald's, SeaWorld Orlando was like a five-star restaurant, with 220 acres of custom marine habitats, thrill rides, eateries, and a 400-foot Sky Tower. There were seven different killer whale pools, including the enormous Shamu show pool, and 7 million gallons of continuously filtered salt water kept at an orca-friendly 52 to 55 degrees. There was regular world-class veterinary care. Even the food was a custom blend, made up of restaurant-quality herring, capelin, and salmon.

  The big question for SeaWorld was whether to teach Tilikum to perform with trainers in the pool. Called "water work," it has long been the most thrilling element of the Shamu shows. In contrast to Sealand's repetitive food-for-work equation, SeaWorld's training strategy was finely honed and based on intense variation. Daily activities were constantly altered, and the orcas were given a variety of rewards—sometimes food, sometimes stimulation (back rubs, hose-downs, toys, or ice), and sometimes nothing. "Variability makes the animals more flexible about what the outcome is and keeps them interested," says Thad Lacinak, who was SeaWorld's vice president and corporate curator for animal training when Tilikum arrived and who left in 2008 to found Precision Behavior, a consulting firm for zoos and other animal facilities.

  Lacinak believed that Keltie Byrne died because Sealand's killer whales had never been trained to accept humans in the water. So when she fell in, they treated her like any other surprise object. Lacinak had confidence that Tilikum could be trained for Shamushow water work. But he and SeaWorld's top management also knew that when it comes to killer whales (or any wild animal), there are no guarantees. Normally SeaWorld begins training inwater interaction when its killer whales are 1,000 pounds or less, but Tilikum was by now a very large bull. Plus Tilikum had been involved in a death. "If something did happen, you would look like a fool," Lacinak says. "It was too risky, and from a liability standpoint it was decided not to do [water work]."

  Some of the trainers at least wanted to desensitize Tilikum in case someone fell in. "There were several of us that pushed for water de-sense training. You don't run from the storm; you harness the wind," says Mark Simmons, who left SeaWorld in 1996 to earn a business degree and later cofounded Ocean Embassy, which consults on conservation and marine parks. "We wanted to make humans in the water so commonplace that it didn't elicit any response. And if that had been done, it would be very unlikely that we'd be having this conversation today."

  But SeaWorld faced the same vexing Catch-22 that had given Sealand pause. SeaWorld's head trainer, Flaherty Clark, says that it's impossible to prove or disprove what might have happened if Tilikum had been desensitized. "It's easy for former trainers to frame that as a hypothetical," she says, "but we viewed water work with him and all the conditioning that might have permitted it to be effected safely as simply too great a risk."

  Instead, SeaWorld focused on creating rol
es for Tilikum that showcased his size and power when no trainers were in the water. The sight of him rocketing into the air awed the crowds. One of his specialties was inundating the front rows—the "splash zone"—with a tidal wave pushed up by his enormous flukes. "He's a crowd-pleasing, showstopping, wonderful, wonderful wild animal," says Flaherty Clark.

  Keeping Tilikum from water work made sense for another reason: as long as SeaWorld had been putting trainers in the water with killer whales, trainers had been getting worked over by them. Since the 1960s there have been more than forty documented incidents at marine parks around the world. In 1971 the first Shamu went wild on a bikini-wearing secretary from SeaWorld, who was pulled screaming from the pool. For every incident the public was aware of (the ones that occurred in front of audiences or that put trainers in the hospital), there were many more behind the scenes. John Jett was a trainer at SeaWorld in the 1990s. He left to pursue a PhD in natural resource management in 1995, having grown disillusioned with the reality of keeping large, intelligent animals in captivity. He says that getting nicked, and sometimes hammered, was just part of the price of living the killer whale dream: "There were so many incidents. If you show fear or go home hurt, you might be put on the bench." Flaherty Clark says SeaWorld gives trainers wide latitude: "The safety of our trainers and animals is paramount. Our trainers are empowered to alter any show or session plan if they have even the slightest concern."