The Far Traveler - Nancy Marie Brown (6 page)

Iceland’s odd earth does hold one advantage over Denmark’s: Whatever you do find can be easily dated. Dig a trench in an Icelandic hayfield, and the trench walls will be conveniently stratified, striped like a layer cake with volcanic tephra—a term for anything that spews out of a volcano and is light enough to travel through the air some distance. Tephra has a different color and texture from soil or sand. When you run the edge of your trowel over it, it rings out, as if you had tapped a glass bottle. It feels grittier, almost spiky on your fingertips or, if you’re not certain, on your tongue. (Archaeologists taste a lot of things they pick out of the dirt: Putting a sample on your tongue is also the best way to tell pottery from bone.) In northern Iceland, the tephra layer from an eruption of Mount Hekla in 1104 is particularly thick: It looks like a line of white icing between the dark cakes of soil. Equally obvious are two honey-brown lines from eruptions two thousand and four thousand years ago, well before the first people came to the island. An eruption from about the year 1000 left a greenish-gray layer that can be traced in most spots. A darker line, often swirled with charcoal or organic matter, marks the settlement of Iceland in the year 870. Archaeologists call this tephra the
Landnám
or Settlement Layer; they have found no signs of human culture beneath it. The historical dates—870, 1000, 1104, plus or minus a year or two—are quite secure. Not only is the 1104 catastrophe mentioned in church records (it wiped out several farms in the south of Iceland), signs of all three eruptions can be seen in the cores drilled from the Greenland ice sheets to study climate change. It’s a lucky accident that the eruptions in 870 and 1104 frame the Viking Age in Iceland.

To make use of this tephrochronology, John’s colleague Douglas Bolender, a doctoral student at Northwestern University, devised a soil-sampling protocol. He and his assistants walked back and forth across each modern farm-field on the Langholt ridge in Skagafjord, an area that encompasses both farms named in Gudrid’s saga: Reynines (now called Reynistadur, or “Rowan Stead”) and Glaumbaer. They logged their coordinates with GPS, reading off two satellites and a transmitter on a lighthouse offshore. They flagged each site, creating a grid of colored plastic flags 50 meters (a little more than 50 yards) apart. At each flag, they punched in their steel soil-coring tube and recorded the depth and quality (including tephra layers) of the soil. If the soil was shallow, they passed it by—only deep soil will preserve a turf house. Shallow soil means the wind has already eroded everything of interest. They examined pockets of deep soil (20 inches is as deep as the corer goes; poking it in twice, you can sometimes reach 40) for marks of a kitchen or garbage midden: flecks of charcoal, peat ash, or burned bone. If they found these below the white 1104 tephra layer, the spot was worth testing further. (Logical on paper, this protocol broke down somewhat in reality: A farmer making hay mowed down the plastic flags before the soil samples could be taken; a bull chased the soil corers through two fences and into a muddy drainage ditch, fouling all their equipment, though they preserved the precious data sheets.)

To places with interesting soil cores, John and Brian Damiata, a UCLA geophysicist, brought a variety of machines that can see beneath the earth. Adapting these remote-sensing devices to Iceland’s wet soils was tricky, as was adjusting any magnetic effects for the nearness of the North Pole. The gadget that finally did work was originally designed for plumbers to detect problems in buried pipes. Called the EM-31, it measures how well the soil conducts electricity. It is unwieldy (a 12-foot-long tube carried by a strap over the shoulder), heavy (30 pounds), and temperamental.

“You have to get all the neighbors within half a mile to turn off their electric fences,” John said. “Sometimes they gave me a pretty short window.” At one farm, a band of young stallions seemed to know exactly when the current to their electric fence was off and would break through the wire to rumpus with the mares.

The results were disappointing. “We’re making these maps,” John said, “and we couldn’t see anything on them.” But then Tim Earle, John’s mentor, now teaching at Northwestern in Chicago, suggested John “rough up the data.” John wrote a computer program to look at the
differences
among readings and discovered a series of anomalies. On the new map, each showed up as a colored cluster of squiggles that stuck out from its surroundings. Choosing one, they dug a test trench. The hole came up empty: no sign of a turf house.

“Later,” said John, “we found out that this machine’s coordinates were one meter off”—about a yard. (John, like all scientists, thinks in meters; I still see things in inches, feet, and yards.) Once they corrected for that yard-long error, they started finding things in their trenches. “We got a lot of landslides. Then one of these anomalies wasn’t a landslide, it was a wall. Once we identified that signal—that it was a turf house—we were flying high!”

The buried turf wall was in the hayfield behind the Skagafjord Folk Museum at Glaumbaer, a collection of historic houses on a busy road that runs the length of the valley to the town of Saudarkrokur, population 2,600. Beside the road is a stone statue of a stout-armed woman balancing a tiny boy on her shoulder—Gudrid and Snorri, the first residents—but the museum was not established to honor them alone. On a low mound, beside a trim white church that is still active, is a rambling turf farmhouse, its walls and roofs of sod forming a jumble of lumps much like a collection of hobbit holes connected by tunnels. Its wooden gables and doors are painted mustard yellow. A house has been on this site, the history books say, since saga times, a thousand years ago. The current structure, begun in 1750, was lived in continuously until it passed to the museum in 1948.

Two other historical buildings were moved onto the museum grounds in the 1990s. One houses a coffee shop and galleries. It had been built in 1886 to be a girls’ school (though “this never came to pass,” a museum brochure relates). The other building, a little white wood-framed house with a green grass roof, provides office space for the museum staff. Its claim to fame is its track record: The house had been dismantled and moved six times since it was built in 1862, logging over 120 miles by ice, sea, and road. Arriving at Glaumbaer in 1996, it was very nearly placed on top of the turf wall yet to be discovered beneath the hayfield.

That wall, John found, was topped by the shiny white tephra from the eruption of Mount Hekla in 1104. Curious, John angled a soil corer into the mound on which the 1750s turf house sits, right in front of the mustard-yellow kitchen door, where the family would have thrown their fireplace ash and scraps. “That ash sits exactly on the 1104 tephra layer, and under that is sterile soil,” he said. “For the house we found down below in the hayfield, everything is
under
the 1104. So the main house at Glaumbaer moved about 1104.”

John’s Icelandic colleagues, including museum curator Sigridur Sigurdardottir and archaeologists Gudny Zoega and her supervisor at the time, Ragnheidur Traustadottir, were skeptical. They asked Gudmundur Olafsson to come up from the Icelandic National Museum in Reykjavik to take a look. Gudmundur, who has excavated more Viking Age longhouses than anyone else, spent several days on the site, dug a long trench, and was also not convinced.

No one doubted that John had found something made of turf and older than 1104. But was it a longhouse? None of the histories, censuses, or tax records showed a house there. And even if a longhouse
did
exist in the Glaumbaer hayfield, another line of reasoning went, it was just an oddity. In general, both history and archaeology agreed, turf houses didn’t move. When fashions changed, the new house was simply built on top of the remains of the old. John argued, on the other hand, that if there was one invisible Viking house in the valley, there might be many.

The Icelanders suggested they take another farm—one not mentioned in the sagas—and survey it using both methods. John chose the farm of Stora Seyla (“Big Marsh”), about three miles south of Glaumbaer. Its turf house had stood until the 1920s atop a complicated mound cut through by a stream, and the nearby fields had not been bulldozed flat or plowed. With the historical records in hand, the Icelandic archaeologists marked every feature on the landscape that looked man-made. John’s team tested each by taking soil cores: According to the tephra lines, none of the features was older than 1104. “So then we cored the whole place on a 50-meter grid,” John told me, “past the modern boundaries of the farm Stora Seyla until we hit the fjord. One core came up with charcoal, and an adjacent one had some very deep soil.” They tested that area with the EM-31.

John paged through the 2005 grant proposal to the EM-31’s output: a map of Stora Seyla from the surface to six feet down, color-coded by how well the soil conducted electricity. “If you fuzz your eyes,” he told me, “you see a tract of light blue and dark blue. We imagine this is the limits of a structure.” It was 115 feet long. Based on this map, John’s team dug test trenches. They found a turf wall, a floor, hay, and a burned birch-bark roof, perfectly preserved—and older than 1104. The main house at Stora Seyla had also moved.

John was, he admitted, a little too pleased with their results. “I was pretty arrogant,” he said with a rueful smile. “I essentially attacked the Icelanders for being incompetent.”

But if this technique—soil coring on a grid combined with the EM-31—worked so well, I asked, why was he bringing a new remote-sensing device—the untested ground-penetrating radar or GPR—to Skagafjord in 2005?

“Because I’m
not
an Icelandic archaeologist,” he said. “I don’t want to spend years excavating these sites. I have a few basic questions. How big is the farm? How much hay did they store? I want a shortcut to digging. GPR is the best remote-sensing technique available to archaeologists. What we need it for is to
not have to excavate.

“Now we can
find
sites with the other technique, but we can’t tell what were looking at without digging into them. And we keep chewing into the wrong places. This pisses off the Icelanders no end. It’s hard to convince them I’m even sort of competent. I don’t know which wall is which. I can’t answer all this ambiguity in the remote-sensing data. We’ve learned that the biggest anomaly is usually a corner, but we’ve learned that only by chewing into it—and we just about destroyed that corner of the house.”

“You mean Gudrid’s house?”

“The house at Glaumbaer. But it had to be done. There was no other way. We had to calibrate our readings.”

“You destroyed the corner of
Gudrid’s house?”

He looked uncomfortable. “I think the case for saying the longhouse at Glaumbaer is the referent for the story in the sagas is true,” he said. “Whether the
story
is true is another question.”

On a warm winter day in March 2005, we were in Iceland, on the site of the storied house, and John was looking even more uncomfortable.

“Why do you think you can use a backhoe here?” Sigridur Sigurdardottir, known as “Sirri,” rolled a heavy glass paperweight between her broad hands. She had served us tea and coffee when we arrived at her office at the Skagafjord Folk Museum, and had even unwrapped a box of chocolates; but now she was installed behind her desk, taking on the full authority of the book-lined office with its unsettling touches of practicality: refrigerator, microwave, spinning wheel. Her gaze was firm and unapologetic.

“He has already hired one,” Gudny Zoega told Sirri in Icelandic.

John backtracked in his ninety-second introduction to his archaeological protocol, vainly trying to rephrase his argument. With the backhoe he intended to quickly strip the top layer of turf off the hayfield. Using trowels and shovels, his archaeologists (and unskilled volunteers, like me) would then expose the tops of the buried longhouse walls to check if the remote-sensing devices had drawn the floor plan accurately. That would be it for the 2005 field season. John was in northern Iceland now to work out where his crew of fifteen would eat and sleep, where they could have lab space, whether he could get a free car for five weeks in July and August. He had been in the country only one day and had spent much of that time searching out a certain kind of Danish backhoe that he thought was excellent for archaeology. The previous afternoon, he had found just what he wanted. Though the owner spoke no English, and John no Icelandic, they hit it off right away. “He keeps his backhoe inside, he likes it so much,” John had crowed to Gudny, when we met her in her laboratory in Saudarkrokur later that evening. “It will be
perfect
He had already gotten permission, he confided to her, to use a backhoe at Glaumbaer. The director of the national Archaeological Heritage Agency, from whom he got his official permits, had said it would be okay.

Those permits, I could see now, were useless.

Sirri’s eyes narrowed. “I see,” she said in English, and John fell silent.

I wrote in my notebook:
No backhoe.

“Isn’t it easily cut with a spade?” Sirri said, putting on her phone headset. She smiled at us and nodded toward the chocolate box, as if to say
enjoy!
Hospitality is a prominent cultural value in Iceland, as important now as it was in the saga days, so I leaned past Gudny and took a couple. Gudny took some, too.

After a rapid conversation in Icelandic, Sirri reported that two or three men could remove the turf in twenty-five hours. Her brother Helgi would arrange it.

“That’s as fast as a backhoe!” John said.

“Yes, I know.” Sirri’s smile was simultaneously smug and patient. “People who know how to do it are as fast as a backhoe. The people are a little bit expensiver, but not much. I like it better not to have a machine on the field. When do you want to start?”

That settled, Sirri went to the refrigerator and brought out a plate of cheese, crackers, and grapes. Gudny and I helped ourselves; John suggested a bit of show-and-tell. He had a movie about his latest remote-sensing device to show them, to explain the new procedure he’d be using that summer. He slipped his laptop out of its case and opened the lid. Nothing. He closed it, opened it, wiggled it. Still nothing.