One Archaeologist's Quest to Prove that Ancient People Practiced Trepanation

filed under: History
Image credit: 
Owen Freeman

Ephraim George Squier was a year into his journey across Peru when he reached the city of Cuzco. Evidence of his trek clung to his body: skin tattooed with mud, armpits ripe from the humidity, clothes so soiled their original color was obscured. It was 1864, and the 42-year-old American was hell-bent on finding something that would make him a legend among archaeologists.

What lured Squier to Cuzco is still a mystery—an invitation, a tip, instinct? But after months in the wilderness, the city must have beckoned to him. Squier compared the sight of the city’s palace to the decadent houses on the Grand Canal in Venice. Inside, amid the usual trappings of wealth, the walls were lined with statues, weapons, pottery. Squier had collected similar items in the jungle. But now he had stumbled into an abundance.

The head of the house, Señora Zentino, dressed the part: formal gowns, often augmented with a Peruvian scarf or, given her love of antiques, a necklace of Incan tokens. Next to such finery, the skull she cradled should have startled Squier.

The whole thing sounds like a fable: a wandering adventurer, a mysterious, aging beauty in a jungle palace, a stolen head. In a pulp novel, a skull such as this would have been cursed. But as Squier turned it over in his hands like a diamond in the sunlight, he marveled at its defining feature—a square-shaped hole on its top left side.

Archaeologists across the world had unearthed skulls with holes before, in quarries or mass graves. Some argued that ancient tribes had mutilated them postmortem, perhaps to make drinking vessels or amulets. Most academics simply dismissed the marks as the results of infections, birth defects, or animal bites.

But as Squier looked closer, he became convinced this hole was not natural: Nature doesn’t work in right angles. Peering at the squarish 15-by-17-millimeter hole, he could see healing scars and signs of new bone growth. This person had not only been alive during the cutting—he or she had survived. A startling idea occurred to Squier: Could this be evidence of ancient neurosurgery?

In proposing that Incans practiced brain surgery—something even the best European and American doctors struggled with—and positing that ancient American civilizations were as advanced as ancient Egypt and Rome, Squier was gambling with his reputation. At the time, there was a pervasive prejudice against Amerindian tribes, who had long been dismissed as loin-clothed savages wielding crude tools. This was tangible proof otherwise. And Squier was willing to do anything—cross colleagues, forgo his dignity, sacrifice his marriage—to make the rest of the world understand.

The Guano Problem

The funny thing is, Squier hadn’t traveled to South America to learn more about ancient neurosurgery or to combat stereotypes. No, Squier had come to South America to settle something far more serious: an argument about bird poop.

Throughout the 1800s, farmers used natural fertilizers to grow crops. The best fertilizers came from islands off South America, where mountains of guano had piled up over eons. Guano proved so important to global health and economics—millions of dollars were at stake, not to mention the health of millions of hungry people—that Chile, Bolivia, and Peru actually went to war over bird excrement in 1879.

For most of the 1800s, the U.S. imported thousands of tons of guano per year, and with the outbreak of the Civil War in the 1860s, securing fertilizer to help guarantee a steady supply of food became a necessity. But a number of international incidents (one involving Confederate pirates seizing Peruvian ships and destroying guano cargo) had angered Peru, and the government was threatening to cut off the pipeline. The situation forced Abraham Lincoln to address the United States’ guano deficit head on. He dispatched a delegation to Peru in July 1863, and Squier, who had served as a diplomat in Central America, was a natural choice.

Squier spent five months untangling the legal claims in Peru. Having succeeded, he then sent his wife, Miriam, home to New York and set out to explore his real interest, the country’s buried artifacts and vine-choked ruins. Over the next 18 months, he traveled everywhere from the coast of Peru to the peaks of the Andes deep in the interior. He saw mountain fortresses, llamas, and statues and artifacts of every kind. The trip culminated with a visit to Señora Zentino, which was when the idea of ancient neurosurgery grabbed him.

Squier saw the holes as evidence of trepanation—a procedure in which surgeons cut out a pocket of the skull to relieve internal pressure and remove sharp fragments of bone. Western doctors had been trepanning skulls as far back as ancient Greece, and it had a brutal reputation. Medieval doctors plugged patients’ ears with lint so they couldn’t hear their own heads being sawed open. Few people who endured trepanation survived. By 1700, most clinics had abandoned the procedure. As one British surgeon declared in 1839, any doctor who proposed trepanning someone “ought to be trepanned in turn.”

Squier’s claim, then, seemed iffy—if the best European surgeons couldn’t pull off trepanations, how could so-called primitive jungle tribes? But Squier was convinced. After returning to New York in 1865, he showed the skull to colleagues and outlined his theory. In the debate that followed, some sided with Squier, while others derided the idea.

Undeterred, Squier appealed to the highest scientific authority around, French neurologist Paul Broca. Broca had recently achieved worldwide fame by discovering the first-known language center in the brain, now called Broca’s area. The Frenchman shared Squier’s passion for archaeology, especially for skulls, and in 1867, he snapped up Squier’s offer to examine the Incan skull.

Broca’s conclusions were unequivocal: Squier was correct. The shape of the hole could not have been natural or accidental, and he confirmed that new bone had grown around the rim. Broca’s medical eye also found signs of inflammation, further evidence that the patient had survived.

Broca forced scientists in Europe to confront the possibility that ancient people had developed their own sophisticated medical practices. Soon, other archaeologists started to notice trepanned skulls in their collections, some potentially dating back 10,000 years. Most of the holes were small and circular, but some gaped as wide as five inches across. Many had rims completely smoothed over by new bone growth, indicating that the patients had lived for years after. Skulls with multiple trepanations even turned up. One unlucky Incan fellow had seven separate holes in his head, all perfectly healed. But as archaeologists embraced Squier’s theory, a bigger mystery emerged: Why were civilizations performing trepanations in the first place?

Trepanation: A How-To Guide

Before they could tackle why, archaeologists needed to understand how. Over the years, Incan pots with images of trepanations had turned up. Additionally, evidence from rural Kenya, New Guinea, and similarly remote areas showed that other tribes were also proficient in the practice.

The procedure looked something like this: Imagine a young warrior hit in the head with a slingshot stone, which left a crater of mangled bone. A surgeon would clamp the young man’s head between his knees, crack open a coconut, and pour the juice on the scalp. The doctor, meanwhile, would dab fresh-cut leaves on the wound to dull the pain.

Then he’d get to work, using a shark tooth or something sharp to cut into the skull, grooving it round and round the depressed fracture, carefully working the incision deeper. Throughout the process, the warrior would gulp alcohol or consume tobacco to quell the discomfort. He would feel almost nothing after the initial pain: just the friction of the shark tooth against his skull. At last, the warrior would experience a slight sucking sensation as a plug of skull bone came free. With bamboo fashioned into forceps, the surgeon would pick out the bone splinters and wash the wound with coconut milk. He’d sew up the scalp with a needle and thread made of bat bones and banana fibers. A dressing of leaves and a plaster of pepper, lime, and betel nut might seal the wound. Finally, the patient would be instructed to eat soft foods for a week and minimize the movement of his head.

As with today’s procedures, managing pain and infection were the biggest concerns, but surgeons had measures to combat these. The coca leaves helped anesthetize the skull. Similarly, wild plants like balsam killed bacteria, as did washing wounds with coconut milk. In fact, ancient surgeons did a remarkable job with sterilization: In one study of 66 trepanned skulls, just three showed any signs of infection. These surgeons had a better track record than their counterparts in industrialized countries. In one survey from London in the 1870s, 75 percent of neurosurgical patients died, mostly due to infections. Compare that to the New Guinea tribes, where surgeons lost just 30 percent of their patients.

THE SPIRIT THEORY

Why ancient cultures performed neurosurgery remains controversial. After years of studying skull holes, Broca concluded that doctors had trepanned skulls primarily to release spirits trapped inside the brain. Moreover, he hypothesized that they operated mostly on children, a claim he based on a macabre experiment. Using sharp glass, Broca managed to open the skull of a recently deceased 2-year-old in four minutes. Cutting a similar hole in an adult skull required 50 minutes, and his hand ached. Broca concluded that ancient surgeons lacked the patience and tools to cut through adult skulls and therefore must have limited the procedure to children, who grew up with holes in their heads.

But most scientists doubt Broca’s conclusion, partially because few trepanned child skulls have ever turned up. Broca’s theory that trepanation released evil spirits, however, proved enormously influential. This idea played into stereotypes of ancient people. And, in truth, many tribes—despite wildly different supernatural beliefs—probably did trepan people to treat epilepsy and hallucinations, maladies often associated with spirits.

Squier and other archaeologists always doubted the spirit theory, however. They promoted an alternative: that ancient neurosurgeons were removing bone fragments from injuries sustained during combat. Modern research has provided strong evidence for this, especially among the Inca. For one thing, far more males than females had trepanation holes, likely because most warriors were males. For another, the holes were usually located on the left side of the skull—where a right-handed assailant would aim a slingshot or smash his club.

From a modern medical perspective, the idea makes sense: Doctors today still trepan people to reduce pressure on the brain after injury. The practice is meant to reduce swelling and the buildup of blood and other fluids, which can kill brain cells.

In the end, Squier bested Broca in the debate over why ancient neurosurgeons cut open skulls. But while Broca continued to have a glorious career, Squier’s unraveled not long after his discovery—as if the skull really were cursed.

Sad State of Squier's Affairs

It all started when Squier sent his wife, Miriam, home from Peru after the guano affair. Alone and resentful, Miriam accepted a job editing magazines for publisher Frank Leslie, and the two became inseparable. After his divorce in 1866, Leslie moved in with her and Squier. This seemed suspicious enough, but things really turned nasty in 1867 when the trio took a trip to Liverpool. Squier had some outstanding debts in England, and, humiliatingly, the police arrested him the moment he stepped ashore. An “anonymous” tipster—likely Leslie—had wired ahead to alert his creditors. With Squier out of the way, Leslie and Miriam’s affair began in earnest.

In May 1873, Miriam finally divorced Squier after publicly accusing him of sleeping with two prostitutes. Free of her husband, Miriam married Leslie in July 1874—a betrayal that broke Squier’s spirit. Just one month later, he had deteriorated to the point that a judge temporarily committed him to an insane asylum. Squier died at his brother’s home in Brooklyn in 1888. He was 67.

It was a sad, sordid end for one of America’s greatest archaeologists. Still, Squier did accomplish his life’s goal. He hadn’t thought much of Lincoln’s assignment in 1863, grumbling that guano “has contributed more towards the corruption of [Peru] than any one other thing.” But his trip to South America—and his willingness to take seriously a funny-looking hole in an old skull—revolutionized our understanding of ancient medicine, showing the world that, sometimes, a hole in the head is a sign of sophistication.

This story originally appeared in an issue of mental_floss magazine. Subscribe here.

More from mental_floss...

June 27, 2014 - 1:00pm
submit to reddit