The most notorious of these wartime studies, at the University of Minnesota’s Laboratory of Physical Hygiene, involved thirty-six conscientious objectors—men who had refused on principle to serve in the armed forces but had volunteered instead for a grueling experiment. Led by Ancel Keys, the influential researcher who had developed the K-ration for soldiers and who went on to propose a link between dietary fat and heart disease, the Minnesota Starvation Study put the volunteers through six months of “semi-starvation,” eating on average 1,570 calories in two meals each day while working for 15 hours and walking 22 miles per week.
In previous VO2max studies, scientists had trusted that they could simply ask their subjects to run to exhaustion in order to produce maximal values. But with men who’ve been through the physical and psychological torment of months of starvation, “there is good reason for not trusting the subject’s willingness to push himself to the point at which a maximal oxygen intake is elicited,” Keys’s colleague Henry Longstreet Taylor drily noted. Taylor and two other scientists took on the task of developing a test protocol that “would eliminate both motivation and skill as limiting factors” in objectively assessing endurance. They settled on a treadmill test in which the grade got progressively steeper, with carefully controlled warm-up duration and room temperature. When subjects were tested and retested, even a year later, their results were remarkably stable: your VO2max was your VO2max, regardless of how you felt that day or whether you were giving your absolute best. Taylor’s description of this protocol, published in 1955, marked the real start of the VO2max era.
By the 1960s, growing faith in the scientific measurement of endurance led to a subtle reversal: instead of testing great athletes to learn about their physiology, scientists were using physiological testing to predict who could be a great athlete. South African researcher Cyril Wyndham argued that “men must have certain minimum physiological requirements if they are to reach, say, an Olympic final.” Rather than sending South African runners all the way across the world only to come up short, he suggested, they should first be tested in the lab so that “conclusions can be drawn on the question of whether the Republic’s top athletes have sufficient ‘horse-power’ to compete with the world’s best.”
In some ways, the man-as-machine view had now been pushed far beyond what Hill initially envisioned. “There is, of course, much more in athletics than sheer chemistry,” Hill had cheerfully acknowledged, noting the importance of “moral” factors—“those qualities of resolution and experience which enable one individual to ‘run himself out’ to a far greater degree of exhaustion than another.” But the urge to focus on the quantifiable at the expense of the seemingly abstract was understandably strong. Scientists gradually fine-tuned their models of endurance by incorporating other physiological traits like economy and “fractional utilization” along with VO2max—the equivalent of considering a car’s fuel economy and the size of its gas tank in addition to its raw horsepower.
It was in this context that Michael Joyner proposed his now-famous 1991 thought experiment on the fastest possible marathon. As a restless undergraduate in the late 1970s, Joyner had been on the verge of dropping out of the University of Arizona—at six-foot-five, and with physical endurance that eventually enabled him to run a 2:25 marathon, he figured he might make a pretty good firefighter—when he was outkicked at the end of a 10K race by a grad student from the school’s Exercise and Sport Science Laboratory. After the race, the student convinced Joyner to volunteer as a guinea pig in one of the lab’s ongoing experiments, a classic study that ended up demonstrating that lactate threshold, the fastest speed you can maintain without triggering a dramatic rise in blood lactate levels, is a remarkably accurate predictor of marathon time. The seed was planted and Joyner was soon volunteering at the lab and embarking on the first stages of an unexpected new career trajectory that eventually led to a position as physician-researcher at the Mayo Clinic, where he is now one of the world’s mostly widely cited experts on the limits of human performance.
That first study on lactate threshold offered Joyner a glimpse of physiology’s predictive power. The fact that such an arcane lab test could pick the winner, or at least the general gist of finishing order, among a group of endurance athletes was a tantalizing prospect. And when, a decade later, Joyner finally pushed this train of thought to its logical extreme, he arrived at a very specific number: 1:57:58. It was a ridiculous, laughable number—a provocation. Either the genetics needed to produce such a performance were exceedingly rare, he wrote in the paper’s conclusions, “or our level of knowledge about the determinants of human performance is inadequate.”
By Day 56, the relentless physical demands of Henry Worsley’s solo trans-Antarctic trek were taking a toll. He woke that morning feeling weaker than he’d felt at any point in the expedition, his strength sapped by a restless night repeatedly interrupted by a “bad stomach.” He set off as usual, but gave up after an hour and slept for the rest of the day. “You have to listen to your body sometimes,” he admitted in his audio diary.
Still, he was more than 200 miles from his destination and already behind his planned schedule. So he roused himself that night, packed up his tent, and set off again at ten minutes after midnight under the unblinking polar sun. He was approaching the high point of the journey, slogging up a massive ice ridge known as the Titan Dome, more than 10,000 feet above sea level. The thin air forced him to take frequent breaks to catch his breath, and a stretch of sandy, blowing snow bogged his sled down and slowed his progress for several hours. By 4 P.M., having covered 16 miles in 16 hours, he was once again utterly spent. He had hoped to cross from the 89th degree of southern latitude—the one closest to the South Pole—into the 88th, but he was forced to stop one mile short of his goal. “There was nothing left in the tank,” he reported. “I had completely run empty.”
The next day was January 9, the day that Shackleton had famously turned back from his South Pole quest in 1909. “A live donkey is better than a dead lion, isn’t it?” Shackleton had said to his wife when he returned to England. Worsley was camped just 34 miles from Shackleton’s turnaround latitude, and he marked the anniversary with a small cigar—which he chomped with a gap-toothed grin, having lost a front tooth to a frozen energy bar a few days earlier—and a dram of Dewar’s Royal Brackla Scotch whiskey, a bottle of which he had hauled across the continent.
Of the many advantages Worsley had over Shackleton, perhaps the most powerful was the Iridium satellite phone he carried in his pack, with which he could choose at any moment to call for an air evacuation. But this blessing was also a curse. In calculating his limits, Shackleton had been forced to leave a margin of error due to the impossibility of predicting how the return journey would go. Worsley’s access to near-instantaneous help, on the other hand, allowed him to push much closer to the margins—to empty his tank day after day, after struggling through the snow for 12, 14, or 16 hours; to ignore his increasing weakness and 50-pound weight loss; to fight on even as the odds tilted further against him.
Eventually, it became clear that he wouldn’t make it to his scheduled pickup. He’d been trying to log 16-hour days to get back on schedule, but soft snow and whiteouts combined with his continuing physical deterioration to derail him. He contemplated a shorter goal of reaching the Shackleton glacier, but even that proved out of reach. On January 21, his seventieth day of travel, he made the call. “When my hero Ernest Shackleton stood 97 [nautical] miles from the South Pole on the morning on January 9, 1909, he said he’d shot his bolt,” Worsley reported in his audio diary. “Well today, I have to inform you with some sadness that I too have shot my bolt. My journey is at an end. I have run out of time, physical endurance, and the simple sheer ability to slide one ski in front of the other.”
The next day, he was picked up for the six-hour flight back to Union Glacier, where logistical support for Antarctic expeditions