FRANK OLSON had been behaving oddly. He was a civilian biochemist at the US Army鈥檚 biological warfare lab at Camp Detrick in Maryland. Suddenly, he began to talk too freely about his work. And then, just as suddenly, he was dead. Around 2.30 am on 28 November 1953, Olson plunged to his death from the 13th floor of the Statler Hotel on New York鈥檚 Seventh Avenue. After a quick postmortem, the body was embalmed and taken back to Maryland for burial.
But that was not the end of the Olson affair. More than twenty years later, in 1975, his family received another shock. Evidence emerged during a Congressional inquiry into the activities of the CIA which suggested that the family had not been told the whole truth about Olson鈥檚 death. The CIA, it emerged, had been dabbling with mind-bending drugs, and, unknown to Olson, he was one of its guinea pigs.
An extra kick
Nine days before he died, Olson and some colleagues had enjoyed an after-dinner glass of Cointreau. That evening, the drinks had an extra kick: they had been spiked with LSD, a drug the CIA was investigating for its potential during interrogation.
Advertisement
Olson started to behave so irrationally, the CIA told the inquiry, that the agency decided to take him to New York for medical treatment. After that single dose of LSD, the CIA鈥檚 only concern, it stressed, was to help Olson. In 1975, President Ford apologised to the family for the years of deceit, and that seemed to be the end of the story.
Until 1994, when Olson鈥檚 wife died. Still convinced that they hadn鈥檛 been told the whole truth, Olson鈥檚 two sons decided the time had come to find out exactly what had happened in Room 1018A at the Statler Hotel that night.
They called in a crack team of forensic investigators to search for new clues about how their father died. After minute examination of the evidence and a battery of new forensic tests, Jim Starrs, professor of law and forensic science at George Washington University in Washington DC, believes there is more than enough evidence to question the original verdict. At the annual meeting of the American Academy of Forensic Science in February, Starrs鈥� team laid out all the pieces of the puzzle.
Evidence from the original investigation was scant. Olson had been sharing his room with Robert Lashbrook, a CIA scientist. Lashbrook told the New York City police that he had seen nothing, but was woken by the sound of breaking glass. The autopsy was perfunctory, says Starrs. The two-page report noted obvious external wounds and fractures, and that tests for alcohol had proved negative. Death, it said, was the result of 鈥渏umping or falling鈥� from the window, and bore all the hallmarks of suicide.
In June 1994, Starrs and his team set to work. The most important question the scientists had to try to answer, says Starrs, was 鈥渄id he go out of the window of his own choice or was he given an unscientific shove鈥�? He was unlikely to have 鈥渇lown鈥� under the influence of LSD given to him nine days earlier. 鈥淧eople don鈥檛 go out through windows during LSD flashbacks,鈥� says Starrs.
A search through documents from the time proved frustrating, with much of the detail blanked out by the CIA and vital papers missing. The findings of the original postmortem, however, suggested that Olson had gone head first through the window. Lashbrook鈥檚 claim to hear glass breaking implied that it was closed at the time. If Olson did go head first through the closed window, the evidence should still be there on his embalmed body. With the family鈥檚 agreement, Starrs had the body exhumed. For a body that had been in the ground for 40 years, it was remarkably well preserved.
As the first of the new postmortem examinations began, Starrs and five colleagues travelled to New York to collect information about the scene of Olson鈥檚 death. Adding such detail to the pathologists鈥� findings would, they hoped, allow them to reconstruct the night鈥檚 events. They carefully measured the room, which had changed very little. The sash window had been replaced, but Starrs found an original window in another part of the hotel. The drop from the 13th floor was 52.7 metres. And, according to the hotel鈥檚 night manager at the time, the body landed on the pavement exactly 2.8 metres from the foot of the building.
At the time of Olson鈥檚 death, the exterior of the hotel was being steam-cleaned and the base was fenced around with a green, wooden barricade. The original postmortem noted traces of green-painted wood buried deep in a large chest wound, indicating that Olson struck the fence before he hit the ground.
Moment of impact
With this information, Steven Batterman, a bioengineer at the University of Pennsylvania, plotted the trajectory of Olson鈥檚 body. He calculated that Olson must have hit the barricade at a speed of around 32 metres per second, 3.2 seconds after crashing out of the window. He reached the pavement less than 100 milliseconds later. This would mean his horizontal velocity as he left the window was 0.6 metres a second-less than half normal walking speed. 鈥淭hat鈥檚 not very fast,鈥� says Batterman. 鈥淐rashing through the glass and the window shade would have slowed him down a bit, but clearly this is a speed at which he could have jumped.鈥� These figures, then, suggested nothing especially odd.
Back in the laboratory, after X-raying the body, the next step was to repeat the postmortem and compare notes with the original. Jack Frost, a forensic pathologist and deputy chief medical examiner for the State of West Virginia, found little sign of the multiple cuts about the face and neck described in 1953. Changes in the skin, which becomes dark and leathery after death, might have obscured fine cuts, but Starrs thinks it highly suspicious that so few signs of injury remain.
The large chest wound, acquired when Olson hit the barricade, was much as described, and Frost could feel that many ribs had been broken by this impact. Both legs were twisted, indicating breaks to both the upper and lower leg bones. The lower bones of the right leg had rammed right through the skin and out at the heel-showing that Olson landed feet first, with the right foot hitting the floor a fraction of a second before the left.
The skull was fractured in several places, but the wound that prompted most speculation was a bang on the forehead, just over the left eye. At this point, blood had formed a pool beneath the skin, leaving a dark stain on the skull. This was the second piece of evidence that didn鈥檛 tally with the original version of events.
Then there was the question of drugs. Had Olson been plied with LSD more than once? The task of answering this question fell to Bruce Goldberger, a toxicologist at the University of Florida College of Medicine in Gainesville. Goldberger screened for a range of drugs, both medicinal and hallucinogenic, in a variety of organs. He also sent hair samples to Japan for state-of-the-art analysis. All tests proved negative.
鈥淏ut negative results don鈥檛 rule out the ingestion of drugs,鈥� stresses Goldberger. LSD breaks down within hours if it isn鈥檛 kept under the right conditions. 鈥淪o we shouldn鈥檛 expect to have found it,鈥� he adds. Also, Olson could have consumed drugs no one thought to screen for. 鈥淭he CIA was also experimenting with other drugs we don鈥檛 know about,鈥� says Goldberger.
With drugs tests yielding no clues, it looked as if the bones might offer the best chance of finding out what really happened. After the postmortem, Olson鈥檚 body was sent to John Levisky at York College of Pennsylvania. Levisky stripped away the soft tissue to expose the shattered skeleton. He pinpointed every fracture and drew a map of the breaks. Three clear groupings emerged, each consistent with a particular impact.
A set of breaks in the upper arm, shoulder and ribs fitted with the notion that Olson had grazed the green barrier before hitting the ground. The focus of the second group of breaks was the pelvis and legs. Levisky confirmed what the autopsy had shown-that when Olson hit the pavement he shattered his leg bones. The impact was so huge it sheared the heads right off the femurs, and the pelvis cracked. The third group of breaks, fractures to the skull, suggested that after hitting the ground, Olson fell back and hit his head on the pavement.
If, as claimed, Olson went out of the window head first, he must have rotated through 270 degrees to land feet first. According to Batterman鈥檚 calculations, the drop from the 13th floor was probably too short to allow more than one complete rotation before landing feet first. Nor, he says, could Olson have been tipped over when he hit the fence: there was not enough of a drop from the fence to the ground, and such an impact with the barricade would have resulted in a different pattern of injuries. So Olson must have started to rotate as he left the window. This suggests that his feet pushed off the floor or clipped the top of the radiator beneath the window as he went through the glass.
So, according to the bioengineering studies, Olson could have jumped and he could have gone head first. But Starrs remains doubtful. Two pieces of evidence still don鈥檛 fit with this version of events. First, says Starrs, the relatively unscathed skin of the head and neck is surprising. 鈥淚f he went out head first through the glass his head must have been a bloodbath,鈥� he says. Second, and potentially the most important piece of new evidence, is the mysterious stain on the bone of the forehead. Blood tends to form a pool or 鈥渉aematoma鈥� beneath the skin only if the heart is still pumping. 鈥淪o it couldn鈥檛 have happened at street level鈥� says Starrs. 鈥淚t must have happened some place upstairs.鈥�
If Olson had hit the window with his head before his fall, the few seconds before he hit the ground might conceivably have been enough to explain the pooled blood. But it would mean that at the moment of impact with the glass, Olson was holding his head up to see where he was going. 鈥淚f he got it going through the window his head must have been in a very odd place,鈥� says Starrs. 鈥淗e would have raised his head and looked directly at the glass as he was about to go through it. Who鈥檚 going to do that?鈥�
In a departure from more traditional forensic techniques, Starrs enlisted experts from Engineering Animation, a company based in Ames, Iowa, to reconstruct what might have happened from the data now available. The animation team built a computer model of the Statler Hotel, and tried several variations in exit speed and trajectory that might leave Olson in the right spot on the pavement after hitting the fence-and still produce the same pattern of injuries. 鈥淎nimation can鈥檛 prove the theory,鈥� says Matt Loozis of Engineering Animation. 鈥淏ut it can show if it鈥檚 in the realm of the possible. We ended up showing that it could have been a suicide-or someone else could have been involved. We could not eliminate either theory.鈥�
Four strong men
With such equivocal evidence, there is some dissent about what happened in room 1018A. 鈥淚 think he dove,鈥� says Frost. 鈥淎t the speed he went out of the window it would take four men swinging him back and forth. Why would they throw him through a closed window?鈥� Batterman is inclined to agree. He sees nothing sinister emerge from his calculations. 鈥淭here are lots of smells in this case but we only know what happened after he left the room. We can never tell what happened in the room.鈥�
But in Starrs鈥檚 view, there is no forensic evidence to confirm that Olson dived head first through a pane of glass. So how did he acquire the bump on his forehead? 鈥淗e could have been thrown or shoved into the window and out,鈥� he says. 鈥淎nd we can鈥檛 exclude the possibility that he was banged into the wall.鈥�
Starrs and his team鈥檚 suspicions will never be laid to rest unless further experiments are carried out to test them. For example, what sort of force would it take to smash through both the glass and the window shade? 鈥淚f he hit the window, he would have had to hit it at significant velocity. Could he have run up fast enough in a room as small as 1018A?鈥� asks Starrs. 鈥淲e could test this. And if it鈥檚 not feasible then there must have been some third-party force . . .鈥� Did he even go through the glass? The pathological findings suggest perhaps not. It is possible, Starrs suggests, 鈥渢hat the window was broken by the CIA鈥� after Olson was thrown through the opened window.
鈥淭here is strong enough likelihood of foul play to warrant further investigation,鈥� says Starrs. Now it鈥檚 the turn of 鈥減eople with sufficient power鈥� to demand access to censored files and answers from uncooperative officials and agents. But why would the CIA have committed such a crime? 鈥淭here was a motive for murder,鈥� says Starrs. 鈥淭hey wanted to stop him talking about his secret work.鈥�