When Ivan Ghezzi first visited the ancient Peruvian settlement of Chankillo 6 years ago, the world was reeling from 9/11. It seemed appropriate to view the site, already famous for its thick walls, parapets, and restricted gates, solely as a fortress. Ghezzi, then a Yale University graduate student, had in fact received a grant to investigate just that possibility at the 2,300-year-old coastal-desert site.
However, he soon became fascinated with a 300-meter-long array of 13 towers arranged like a row of giant prehistoric teeth along an adjoining ridge. Archaeologists had proposed that these structures might have defended the area as the other Chankillo features had. Some researchers, however, had noted a hint of astronomical symbolism, as 13 is the number of lunar cycles in a year. But no one had followed up.
During his 2001 visit, Ghezzi revisited the idea that the towers might be an astronomical tool. Surveying the region around the ridge, he identified two structures—one just to the east and the other to the west of the string of towers—that seemed to be sites for observing the towers. Viewed from those sites, the placement and span of the 13 towers mark sunrises and sunsets from solstice to solstice. Ghezzi suggests that the towers served as a calendar accurate to within 2 to 3 days per year.
The 13 towers of Chankillo are the earliest known solar observatory in the Americas, according to a report in the March 2 Science by Ghezzi, now at the Pontificia Universidad Católica del Perú in Lima, and Clive Ruggles of the University of Leicester in England.
The Incas used solar calendars in another part of Peru during the 16th century, but the Chankillo towers were built 1,800 years before them, the authors say.
“The site seems quite unusual and is clearly earlier than other sophisticated [observing] sites in the Americas,” comments astronomy historian Owen Gingerich of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Mass.
The intricacy of the tower structure, as well as the dual observing sites, suggests that the design relies on as-yet-undiscovered solar observatories that date to an even earlier time, says Ghezzi.
Subscribe to Science News
Get great science journalism, from the most trusted source, delivered to your doorstep.
In a north-south line along the ridge, the flat-topped stone towers are evenly spaced, about 5 m apart. Though called towers, the structures are squat, varying in height from 2 to 6 m. Their rectangular tops are 11 to 13 m long and 6 to 9 m wide. Each tower has two sets of narrow stairs that lead to its top. From there, the eye takes in a landscape of foothills and mesas.
But it’s the view from two sites elsewhere in Chankillo that astonished Ghezzi. Soon after making basic measurements of the towers in 2001, he began to suspect that they served an astronomical function that would be apparent only from other vantage points. During the next 2 years, he conducted excavations nearby, and by 2004, he had uncovered two intriguing sites—one to the west and one to the east of the towers.
Each of the sites has a commanding view of the towers. Ghezzi suspected that by viewing the towers and noting the locations of sunrise and sunset, ancient observers kept track of the seasons.
One of the observation sites is located at the open end of a 40-m-long walled corridor that runs outside a building about 200 m west of the towers. Both the corridor and the building had been carefully constructed. A wall restricts the entrance to one end of the corridor. At the other, open end of the corridor, Ghezzi noted an abundance of pottery, shells, and tools that hadn’t been found at any other Chankillo entranceway.
The observation point 200 m east of the towers is an open-sided room in a building that stands apart from several others. A low wall restricts access to the room. Not far from the building, Ghezzi found scattered remains of offerings including ceramic panpipes and oyster shells. Nearby middens contained remains of serving vessels, panpipes, and maize.
Using modern technology—a handheld Global Positioning System device—Ghezzi determined the coordinates of the towers and each observing point. The results revealed that the sun’s full range of motion 2,300 years ago, from the summer solstice to the winter solstice, matches the breadth of the line of towers.
From the western site, an observer could mark the progression of sunrises as they move north to south for half the year and then back again during the other half. The gaps between towers near the center of the array correspond to 10-sunrise intervals, while the gaps between the outermost towers correspond to longer periods.
Similarly, from the vantage point that Ghezzi discovered to the east, observers could chart sunsets throughout the year. From this site, spaces between most of the towers correspond to 11-to-12-sunset intervals. Because the placement of the towers doesn’t follow an exact north-south alignment, the gaps between towers differ as observed from the eastern and western sites.
By 2005, Ghezzi had accumulated a large set of radiocarbon dates, which he used to determine the ages of the Chankillo site. He then contacted Ruggles, an astrophysicist who focuses on ancient astronomy.
Ruggles was initially skeptical when Ghezzi suggested that the 13 towers at Chankillo served as an astronomical observatory. After visiting many other sites proposed as observatories, “I am used to being disappointed,” Ruggles says. Chance alignments of stars with human-built structures are common, he notes.
The view from the twin sites, however, convinced Ruggles of Chankillo’s place in history as an ancient solar observatory. Ruggles own studies then confirmed Ghezzi’s results.
From either observing site, Ruggles and Ghezzi conclude in their report, “the various towers and gaps would have provided a means to track the progress of the sun up and down the horizon to within an accuracy of 2 to 3 days.”
Ghezzi and Ruggles propose that the relics found at the two viewing sites indicate that ceremonial feasts and rituals took place there in conjunction with viewing the rising and setting sun. The blocked access indicates that the observations may have been controlled by an elite group, perhaps a governing body at Chankillo.
The archaeological evidence of ritual use, most apparent at the better-preserved, western site, distinguishes these places from other sites at Chankillo, says Stephen McCluskey, a historian of science at West Virginia University in Morgantown. “That was, for me, the clincher” for a solar observatory, he adds.
Views of the past
Astronomer Ed Krupp, director of the Griffith Observatory in Los Angeles, says that he’s impressed by the sophistication that Ghezzi and Ruggles describe at Chankillo. With twin observing sites, residents had two chances each day—sunrise and sunset—to mark the location of the sun.
“Most horizon calendars only permit a sunrise or a sunset observation,” he notes. “The Chankillo horizon calendar provides [much] more observational flexibility.”
With the variability of coastal weather, a second opportunity each day to mark the calendar would be critical. “An overcast morning might otherwise mean waiting a whole day,” he says.
“There are many ways to devise calendars, but the purpose is coordination of goods and services,” Krupp says. Societies keep calendars to enhance survival. Planting crops only after the last frost has passed and knowing when to harvest, especially in a region of highly changeable weather, is essential, he notes. Other activities, including navigation, may also be seasonally modulated.
Krupp says that groups retained power by maintaining calendars that scheduled ceremonies and other seasonal events.
Some calendar functions could be accomplished by simply tracking the phases of the moon. “But because the sun is also obviously behaving seasonally, people may figure you need to integrate its behavior with the moon to ensure doing the right thing in the right way at the right time,” Krupp says.
For example, Southern California’s Chumash tribe and most other California Native American groups employed a moon-based calendar when anthropologists interviewed them in the 19th century. Nevertheless, these people “observed the sun on the horizon at the solstices to establish their occurrence for major rituals,” says Krupp. That activity may have dated to a few thousand years earlier.
“Solar observations are required to establish the solstices,” Krupp notes. “People use the solstices as references for ceremonies and rituals that fortify social cohesion through cooperative interaction with the cosmos.”
Some temples in ancient Egypt, built about 3,500 years ago, aligned with sunrise at winter solstice, Krupp adds. One of the symbols in Egyptian iconography shows the sun lodged in a mountain notch, “and this argues for a very old tradition of horizon-based sun watching.”
Among the Incas, evidence of sun worship dates to at least the 1500s, notes Ghezzi. An Inca king considered himself an offspring of the sun, and chroniclers described “sun pillars” in a pattern radiating from a central sun temple in Cusco, Peru.
These pillars reportedly marked planting times and marked seasonal observances. But the pillars vanished without a trace.
More recently, in the late-19th and early-20th centuries, the Hopi of Arizona and other Pueblo tribes of the southwestern United States used hillside markers to chart the passage of the sun, McCluskey says. “These records are sufficiently detailed that we can document the nature and precision of their observations,” he notes.
However, Krupp says that he doesn’t know of any other example of the dual-observation system seen at Chankillo.
Ghezzi proposes that charting the yearly migration of the sun in the Americas is probably much older than even the Chankillo towers. The knowledge of the heavens embodied by the towers and twin observing sites didn’t accumulate overnight, he notes.
People today can still use the towers as a sun calendar. Earth’s tilt varies by a few degrees in a 41,000-year cycle, and this shifts the exact positions of solstice sunrises and sunsets. The difference is small, however, so the vista today is nearly the same as that viewed by the residents of Chankillo in 300 B.C.
“You can bring someone here, and they can have the same experience as someone 2,300 years ago,” says Ghezzi.
“I don’t know of anyplace else where you can do that,” he adds. “This is truly remarkable.”