Deep into the Landscape TEXT AND PHOTOGRAPHY BY Richard Mertens MORE FROM RICHARD MERTENS Distance Archaelogy PRINT-FRIENDLY VERSION

In Turkey's Amuq Valley researchers from the Oriental Institute focus on big-picture patterns of the ancient world.

It was September in the Amuq Valley, and the cotton was bursting its husks. On both sides of the road stretched a sea of green flecked with white. Riding north in a big Ford van, Jesse Casana, AM'00, stared out the window, ignoring his colleagues' chatter. Whenever he traveled in the Amuq, he seemed to lose himself in the landscape. Sometimes he leaned his head out the window, like a dog muzzling the wind, to get a closer look.

He was not admiring the scenery. He was studying it, sizing it up, looking for clues to its past. He was doing archaeology on the fly. Casana is a Ph.D. student in Chicago's Department of Near Eastern Languages & Civilizations and a researcher at the Oriental Institute (OI). Usually his fieldwork was more systematic. He and his colleagues tramped along a predetermined course, scrambling up hillsides, cutting through olive groves, scanning the ground for the residue of ancient civilizations. But even from a passing car, a few scattered stones or broken roof tiles glimpsed in a field might reveal the site of an ancient village or hamlet. "There is so much out there," Casana explained, "a lot of our finds end up being opportunistic."

Chicago researcher Jesse Casana (right) and his Turkish colleague Merih Erek stand above the Amuq.

There are two kinds of archaeology. One is the work of pick and shovel, wheelbarrow and trowel. It is expensive and time-consuming. Excavating an ancient town or city requires dozens of workers to haul and sift. It commits the young archaeologist to weeks of toil in a hole ten meters by ten. It is what most people think of as archaeology: Heinrich Schliemann exhuming the burnt layers of Troy, Indiana Jones saving the Lost Ark. But some archaeologists find it, well, limiting. Instead of pouring their energies into one spot, they study the whole landscape, ranging widely, by foot and motor; increasingly they gaze down from above with the help of aerial and satellite photography. They are searching for deeper landscapes, and the land is a palimpsest that they read for the long-ago stories that humans have written and rewritten into it. They look for broad settlement patterns—not just the cities, but smaller communities and farmsteads, as well as the roads, canals, river channels, and fields that lay between. They investigate the forces that shaped the landscape, including erosion, deforestation, and the expansion and contraction of agriculture.

Even to an experienced eye, the endless rows of cotton revealed little. But Casana had a trick in his field bag. As Hanifi Topal, his cheerful Turkish driver, headed north, Casana switched on his Global Positioning System (GPS)—an instrument about twice the size of a cell phone that could fix his position to within about ten meters-and thrust it out the window. Back in Chicago he had pored over black-and-white satellite photos from the U.S. Geological Survey. The photos, taken by American spy satellites in the 1960s and 1970s, reveal blemishes—archaeological sites—on the valley's surface. From space even very small sites show up as variations of light and shadow or a slight discoloring of the soil. Site No. 290, on the photos a dark circle about 150 meters across, lay invisible under the cotton, but Casana was homing in on its coordinates. "We're about a kilometer away," he announced. The van crossed an irrigation canal, stopped, and everyone piled out.

Casana's team that day included three American graduate students, a Turkish professor of archaeology, and a Turkish undergraduate. Casana had gathered them for their diverse interests and expertise: one knew the Paleolithic, another ancient Rome, still another the later Islamic period. Following the GPS like a divining rod, Casana led them through a ditch and across a bare field toward the cotton's edge. He wore a floppy brimmed white hat, a long-sleeved white shirt, and loose khaki trousers. The heavy Amuq clay clung to his hiking boots. The GPS told the archaeologists they were 160 meters away. When at last they waded into the cotton, they found the ground littered with broken pottery. Casana stooped and quickly gathered a few pieces, piling them in a cupped hand. "This might be an early site," he exclaimed. Some sherds were dusky, others pale with daubs of orange or black paint. "Look at this," he said, holding out a gray, gently curved piece. "This is great. It's Chalcolithic pottery of the oldest period of occupation-6500 to 4500 B.C." It was rare to find a site this old on the Amuq, he said. Most lie buried beneath too many feet of sediment or at the bottom of mounds built up by later civilizations. He crouched in the cotton and picked through the scraps, cast up from the dawn of human history on the Amuq. He was deep into the landscape and loving every minute.

Bryn Mawr grad student Andrea DeGiorgi records sherds from Amuq sites.

The Amuq Valley lies in southern Turkey at the upper right-hand corner of the Mediterranean Sea. It is a small patch of the Fertile Crescent, the great arc through the Middle East where agriculture and cities first appeared. To the east, just over a range of hills, lies the upper Euphrates. The Amuq is actually more plain than valley, a 30-mile-wide expanse of cotton and wheat fields, irrigation ditches, and scattered villages, bounded on three sides by hills and mountains. The Orontes River loops through it, flowing from its Syrian headwaters to the Mediterranean, but the Orontes did not carve the valley. The Amuq is a rift valley, formed by the same seismic shifts that opened up the Jordan Valley and the Great Rift Valley of East Africa. Amuq, an old name, comes from an Arabic word meaning deep. (In Turkish the valley is usually called Amik Ovasi—ovasi is Turkish for valley.)

Breaking the monotony of the plain are many small, isolated hills. These are not natural features but the sites of ancient towns and military fortresses, built up slowly over hundreds and thousands of years. Some are low enough to be cultivated by local farmers; others jut 50 or 100 feet high. Not all are abandoned. In a few places Turkish peasants live in concrete-block houses on the buried ruins of Bronze Age towns. From almost anywhere on the Amuq a dozen or more mounds are visible.

Blessed with rich soil, plenty of water, and a mild climate, the Amuq has been prime real estate since the beginning of agriculture. Humans have lived here for perhaps 100,000 to 200,000 years, and it has been densely inhabited for at least the past 8,000 years. In and around the valley archaeologists have found the relics of Paleolithic hunter-gatherers, Neolithic villages, Bronze Age kingdoms, and outposts of the Roman Empire. At the plain's southeastern corner, where the Orontes flows beneath a steep mountain ridge, stood the ancient city of Antioch. Today it is called Antakya and is the bustling capital of Turkey's Hatay province. But until an earthquake destroyed the city in 526 A.D., Antioch was the greatest city of the Roman East. Its archaeological museum has one of the world's finest collections of mosaics, dug up from Roman palaces that still lie buried beneath the modern city.

Among archaeologists the Amuq lacks the allure of Mesopotamia, which lies hundreds of miles to the southeast and is known as the heartland of cities. But it has nonetheless attracted its share of attention. Between 1937 and 1949 the great British archaeologist Sir Leonard Woolley excavated Tell Atchana, the capital of a small Bronze Age kingdom. Woolley was drawn to the Amuq because it lay at an ancient crossroads. To the north were the Hittites; to the east, Babylon; to the south, Palestine and Egypt; to the west, over the Beylan Pass, the Mediterranean and the whole Greek world. "From the point of view of commerce," Woolley wrote in A Forgotten Kingdom (1953), the Amuq was "the meeting-place of the Great Powers."

Chicago also sent teams to the Amuq. Between 1932 and 1938 Robert Braidwood, PhD'43, the late director emeritus of the Oriental Institute, explored the valley, finding 178 archaeological sites and digging at eight. But the gathering storm of World War II forced Braidwood to abandon the Amuq. Woolley returned briefly after the war but left for good in 1949. For decades no archaeologist worked in the Amuq, in part because the political climate had changed. In the 1930s the region was a Syrian province administered by the French. After the war it became part of Turkey, and Turkish officials steered archaeologists farther east, where new dams on the upper Tigris and Euphrates Rivers—part of an ambitious irrigation project—threatened archaeological sites. "The Amuq Valley and Antioch were put on the back burner, and nothing was done," said K. Aslihan Yener, a Turkish archaeologist and associate professor at the Oriental Institute. "That entire region fell asleep."

Some sites, like the village of Terzihöyük, or Tell Terzi, are still inhabited.

The region returned to the front burner in 1995, when the OI started its Amuq Valley Regional Project. Directed by Yener, the project revived, and enlarged, the OI's interest in the Amuq. Working elsewhere in Turkey Yener had developed an expertise in ancient mining and metallurgy. The Amuq offered an opportunity to expand her studies to fresh territory. She wanted to see, among other things, what happened to the metals she had seen mined farther north when they reached their markets. But the project was envisioned as an ongoing effort that would encompass a range of archaeological investigation. In addition to excavating some of the valley's most important mounds—and revisiting Tell Atchana, the Bronze-Age mound that Woolley excavated—archaeologists would study the landscape to a degree that went far beyond Braidwood's early survey. This part of the project fell to Tony Wilkinson, another associate professor in the OI and one of a rising generation of landscape archaeologists.

Born in England and trained in Canada as a geomorphologist, Wilkinson has studied ancient landscapes in Iraq, Syria, Turkey, and Yemen, where aerial photographs helped him discover a Bronze Age civilization that no one suspected could have existed so far south on the Arabian Peninsula. In the Amuq Wilkinson is interested in long-term changes in population and settlement. For him one important story is the coalescing of settlement in the Bronze Age, when centralized kingdoms emerged and people began living exclusively in large fortified towns. He is also interested in seeing how the Amuq changed as the valley came under the sway of the great ancient empires, including the Assyrian and the Roman. "We're looking at settlement changes, changes in the ecological environment, changes in transport, and how they all relate," he said.

At 53 Wilkinson is white-haired, energetic, and given to expressions like "Right! Brilliant! Let's go!" In fall 2002 he and two Chicago colleagues began a project in Iran, which had barred Western archaeologists since the Shah's fall in 1979, and in many ways he seems to fit the Indiana Jones image of the intrepid archaeologist-adventurer. In fact he is a pioneer of a kind of archaeology that has become suddenly fashionable. "It's basically ancient geography," he explained. "Traditionally archaeologists have looked at specific archaeological sites and what's in them. That's a narrow perspective. Our perspective gives a much wider range of information on changing economies, and especially how different areas interact with each other, in terms of transport, agricultural production, and mineral production."

Landscape archaeology is not new. Even Schliemann's 19th-century study of Troy included the Trojan plain. But landscape archaeology did not emerge as a separate discipline until the late 1940s, when Harvard's George Willey mapped ancient settlements, canals, and field systems in Peru's Biru Valley. Then, beginning in the late 1950s, Robert McCormick Adams, PhB'47, AM'52, PhD'56, later director of the Oriental Institute, undertook a series of landscape studies in Iraq. Adams, who went on to become secretary of the Smithsonian Institution and is today an adjunct professor in anthropology at the University of California-San Diego, used aerial photographs to study remnants of the ancient Mesopotamian landscape. He was able to determine what the land looked like before and after cities began to form—how scattered settlements gave way to urban centers. "That was so popular it led other people all over the world to begin similar kinds of studies," said Curtis Runnels, a landscape archaeologist at Boston University and editor of the Journal of Field Archaeology.

Casana pulls a Paleolithic stone tool from a layer of gravel.

Casana, one of a half-dozen graduate students who work with Wilkinson, is 27 years old and grew up in Springfield, Virginia. "As a kid I was obsessed with trying to find old stuff in parks or woods or wherever," he said. "I was always carting home boxes full of rocks that I suspected might be stone tools, although I don't think I ever found any. Sometimes I tried to make maps of the places I went exploring. I always wanted to be an archaeologist. Sometimes it was a paleontologist." During his first season on the Amuq he took part in the excavation of an older mound, Tell Kurdu. After that he began to work under Wilkinson, gradually taking over much of the Amuq investigation. The work has taken him over a lot of ground, sometimes by car, more often on foot, in terrain that can be rugged and steep but always full of surprises. "It's fast and dirty and cheap," he told me over lunch one day in Chicago. He was in the midst of planning his September fieldwork, and he was brimming with enthusiasm. "We can go out there and in a few weeks find stuff that challenges conventional wisdom. We find a ton of things that people never knew existed."

I arrived in the Amuq in late August, crossing over the Beylan Pass in a bus from Adana, a city on the Mediterranean coast. The route was ancient but the road was modern—a divided highway that wound down from the pass to the flat green valley below. In the bus, which was as sleek and modern as the highway, a young man wearing a tie served Coke in plastic cups and squirted lemon-scented lotion into the hands of the passengers. In Antakya I found Casana and his colleagues living near the city center in an aging four-story French colonial house, a few doors down from the large buff-colored building that once housed the French provincial administration and is today a pornographic movie theater. Casana had rented the ground floor of the house. It was cool and airy and cluttered with gear: cots and sleeping bags, maps and satellite photos, reference books, and plastic bags stuffed with broken pottery collected the previous season.

While the grad student reaps clues to the past, an Arab family has come to the Amuq to pick cotton.

In the Amuq old civilizations reveal themselves principally through their ceramics. Where people have lived for long periods of time, the ground is usually littered with broken pottery—fragments of water pitchers, plates, bowls, cups, goblets, cooking pots, and large storage vessels called pithoi. As fashion and technology changed over time, so did the pottery, and by studying this ancient refuse, archaeologists can determine with remarkable accuracy when a site was inhabited. And while newer layers of settlement cover older ones, the old does not usually stay covered. After thousands of years, bits work their way to the surface. Worms, burrowing animals, wind and rain, tree roots, plows—these are some of the instruments that turn the soil and bring up what is buried. Without lifting a shovel or a trowel, an archaeologist can simply examine this surface litter and know a site's full range of occupation.

The day I reached town, Casana and his crew were getting impatient. He had arrived in mid-August with ambitious plans for two months of fieldwork. In the Middle East, fieldwork generally takes place in the autumn, after the summer heat has passed, and usually lasts only a few weeks. Casana had hoped to spend much of the field season exploring the hills around the Amuq, looking for the kind of small, dispersed settlements he and others had already found in abundance on the plain. But archaeology is politically charged and fraught with obstacles. Much archaeological effort, it turns out, is expended in a perpetual quest for permits and endless wrangling with officials.

Casana's permits had hung up somewhere in the Turkish Interior Ministry—he didn't know where or why. So instead of spending their days covering new ground, Casana's team stayed in Antakya and recorded the previous year's potsherds. It was dull but necessary work. They emptied plastic bags, measured the contents with calipers and diameter charts, and made careful sketches on pale blue graph paper. To a trained eye, the sherds were more than fragments; a small slice of a rim or crook of a handle might blossom in the imagination into a whole bowl or drinking cup. Laid out on a table, the fragments spanned 8,000 years of human history and many ages of the Amuq: Chalcolithic, Early Bronze, Middle and Late Bronze, Iron Age, Hellenistic and Roman, Late Roman and Early Byzantine, Middle Islamic.

Casana trods a freshly plowed field above the Amuq.

One morning Casana and his colleagues left on an excursion that was part reconnaissance, part tourism. They stopped at site No. 290, filling in that blank spot on the map of the Amuq. "I like to find a new site every day," Casana said. "Otherwise I feel I'm not really working." They also visited Tell Atchana, where Woolley's old dig house still stood, an abandoned two-story timber-frame building. Some of the ruins Woolley unearthed had been left uncovered: mud brick walls built in the Bronze Age and Woolley's "Stratification Pit," dug down more than 50 feet to the layers at the bottom of Tell Atchana. This coming autumn Tell Atchana will be bustling again as a team led by Yener and OI associate professor David Schloen resume excavations. A single guard stood watch. Then an old man hobbled up to see who the visitors were. He wore sandals, a red-and-white head wrap, and baggy charcoal-gray trousers with the crotch at the knees, a style favored by older Turkish men. Ali Yalçin introduced himself as one of Woolley's workers from the 1940s. "I was just a zambil," he said—a boy who hauled dirt in a basket. He leaned on his walking stick and grinned a broad, toothless smile. "It was good work."

The permit arrived at last, thanks to the intervention of the U.S. embassy in Ankara. The next day Casana and the others drove out of Antakya by 6:30, just as the city began to stir. On one street corner men with shovels gathered in the hope of work as day laborers. A man pushed a cart of round, pretzel-like breads along the street. The archaeologists drove east, following a two-lane road along the base of the low, brown hills bordering the plain. They stopped at the mouth of a small valley where a low hill, Casana explained, was the mound of an Iron Age city. "I want to get some early material, which might be hard to find and might not be here at all."

The students set to work, prowling the hill for bits of pottery. Casana led Merih Erek, a professor at Mustafa Kemal University in Antakya, down to a gully where a stream had cut through deep layers of alluvial gravel, exposing the stones of an old Roman dam or mill foundation and a later wall. Such discoveries were crucial; they would allow Casana to date the alluvium and the erosion that had produced it. The two archaeologists scrambled to the gully's bottom, stood before a cut bank about 20 feet high, and scanned the layers of gravel. Toward the bottom of the ban, Casana noticed a piece of flint, hard and shiny against the dull gray limestone gravel. He wiggled it free. It was a "lithic," an old stone tool. Sometime about 100,000 years ago, in the Middle Paleolithic era, a scalloped edge had been chipped into it by some ancient Amuqian. Casana looked more closely at the layer. He spotted more flints and pried them loose. The gravel was thick with lithics, washed down from somewhere above. "It's likely there's a Paleolithic site up the valley," he said. "It might be hard to find, but it's probably there." He turned to his Turkish colleague, who dreamed of finding a Paleolithic cave. None had yet been found in the Amuq. "What do you think?" Casana asked. Erek grinned. "I am a very happy man."

An hour later they were grinding up a narrow road past the squat, concrete-block houses of local farmers and up into the side valley. The hills were parched and brown, in sharp contrast to the irrigated fields of the Amuq below. It would be more than a month before the autumn rains arrived. Even the steepest slopes were a patchwork of olive groves and small fields with wheat stubble, scraggly cotton, melons, and tobacco. The previous year, Casana and Wilkinson had begun exploring these hills, looking for archaeological sites. In such rough, uneven terrain satellite photos were of little use; they work best on the plain, where only slight differences in uniformity stood out clearly. In hill county, archaeologists were forced to use an old-fashioned method—walking. Casana and the others spread out at 100-meter intervals and set off in a ragged line up the valley, scanning the ground as they trudged along.

Ali Yaçin worked as a dig boy for British archaeologist Sir Leonard Woolley in the 1940s.

"It's a lot more systematic than it looks," Casana said. "What we're trying to do is just to sample this area of the landscape to see if there are any sites on it. The problem is we have to cover all parts of a valley, so we know where sites are, but also so we know where sites aren't." They were looking for any scrap of a human artifact—a broken roof tile, a potsherd, maybe even a flint scraper. No matter how small, an archaeological site would show up as a patch of broken pottery, surrounded by a more diffuse "halo" of potsherds. "When you're near a site," Casana said, "you know it."

The ground was uneven, the slopes steep. The sun blazed down, and a dry, constant wind blew from the west. Communication became difficult. They were spread out too far, and the wind was too loud, for shouting. Casana tried hand signals, but mostly he used his cell phone. Everyone carried a phone; it was an essential piece of field gear. He dialed Alexandra Witsell, a Chicago grad student about 200 meters away. "If you see one sherd, look for more," he told her. "If you see more, we may be on to something."

The wind carried sounds up from below. Somewhere a rooster crowed. A sheep bleated, a child shouted. Suddenly Casana stopped and bent down. "Here's a little bit," he said. He walked a few feet. "Here's another little bit." He tossed aside two small orange chunks of badly worn pottery and gazed up the slope. "I feel there's something up there, and we're just skirting the edge of it." He climbed through rows of spindly cotton to the edge of an olive grove. A piece of a broken roof tile lay in the dirt; he kicked it with the toe of his boot. "These kinds of things are a lot more immobile," he said. He spotted a thick, slightly curved chunk of pale orange ceramic, probably the broken rim of a storage jar. The archaeological museum in Antakya had many of these giant pithoi—fat, deep, and bigger than refrigerators. Casana stood for a moment, puzzled.

"These are the kinds of places I hate," he said. There were too few potsherds to make a site, but too many to ignore. "Wait," he said at last. "Look here." Thirty feet away, on a patch of bare ground, lay a few small building stones, broken roof tiles, and little stone cubes of Roman flooring, called tesserae. "Yes, here it is," Casana said, relieved. The scattered potsherds extended into the olive grove, and he walked quickly around the perimeter of what seemed the densest patch. AS 291, as he named it in his notebook, was not large. "It's probably a little thing, a farmstead, 50 meters or so, but it's real," he said. "It's about the bottom end of what you find. It was probably just a few houses." He pulled a plastic bag from his shoulder sack and began to collect samples. "This is very typical of the sites we've been recording the past five years." He picked up a thin, delicately curved piece of orange pottery. "It's early Roman, probably the first or second century A.D.," he said, slipping it into the bag. Özlem Dogan, the Turkish student, hiked up to help. After ten minutes, they poured their bags out and selected samples to keep. These were the ones that seemed the most diagnostic—pieces that could be pinned to a specific period. "Good," Casana said as he fingered the sherds. "Very good." He held up a heavy chunk of a pot rim. "What's this?" he asked. "Pretty ugly." It went into the keep pile. "We have to be pretty aggressive about culling this stuff in the field," he said. "Otherwise we end up with massive amounts of material." Then he sat down in the shade and jotted some notes. He took coordinates from his GPS. His phone rang. Asa Eger, another Chicago student on the team, had found a second site in an olive grove a little way up the valley. They had already found two new sites, and it was barely lunchtime. "It's been a good morning," Casana said.

They ate a meal of feta cheese and bread in the shade of some olive trees, then continued their tramp up the valley, finding several more sites, collecting more potsherds, taking careful notes. The next day they returned and surveyed some of the higher slopes, struggling up a steep, windswept hillside that yielded nothing but a view. The Amuq plain stretched below, the cotton shining bright green in the morning light. "Our objective is to understand the whole landscape," Casana said, standing with his back to the wind. "We have to sample the whole area, the steep rocky hillsides, the streambeds. We have to demonstrate where there aren't sites. A lot of my colleagues don't like this part." The wind boomed across the hills, shaking the wheat stubble. "How do they even get a plow up here?"

A Cold War satellite map shows ancient mounds in the Amuq.

After surveying the first valley, they moved on to others. It was exhausting work. Tramping over the hills left them sore and weary. But they were finding new sites every day. Some were the size of hamlets and villages; others were, in Casana's words, "crappy little farmsteads." Slowly, however, they were filling in the landscape's story—a story of dramatic change. Until late in the first millennium B.C., most people in the Amuq lived in a few large towns and villages. Then they spread out, moving into smaller and more dispersed settlements, into villages, hamlets, and crappy little farmsteads. Casana and others had found dispersed settlements on the plain itself. Now, exploring the small valleys in the fringe of hills, they discovered that the expansion had been upwards as well. "Virtually all the sites date to the first century A.D.," Casana said. "By then, there's a huge population in these villages." The settlements seem to have flourished until the eighth or early ninth century. By the tenth or 11th, most were abandoned.

Another part of the story involved environmental change. As people dispersed on the Amuq and moved into the surrounding hills, there was an increase in erosion that lasted for hundreds of years. But Casana couldn't yet say why. Did farmers cause the erosion by cutting forests and clearing new fields? Or did erosion increase when farmers abandoned terraces on steep slopes? Many basic questions eluded explanation. Why did settlement patterns change in the first place? What compelled people who had lived for millennia in large walled towns to spread out into villages, hamlets, and farmsteads, and then to abandon them centuries later?

"There are a lot of unknowns," Casana said one afternoon, pausing on a high slope. "We have all these small sites. There are hundreds of years of occupation. It's hard to know what's going on." Later, over dinner in an Antakya restaurant, he explained, "The more you look into these questions, the more you realize you can't really answer them. Our knowledge and our methods are just too crude."

It was a problem that had long troubled him. "All archaeology is like that," he said another time. "What we want to know are things about action and belief in the past, and what we have are potsherds on the ground. Even if you dig them up it's still just more stuff, unless you have texts, and even then it is fraught with problems. How to reconstruct action from object is one of the old central problems of archaeology. I try in my own work to ask questions that the data can answer, but it's hard. I can address something like when did settlement become dispersed into the hills, but I can't really do much more than to speculate and argue about why."

I stayed about a week on the Amuq. By the time I left the cotton pickers had arrived in the valley. They were mostly poor Kurds and Arabs from eastern Turkey. They pitched canvas tents next to the fields and collected wheat chaff and old cotton stalks to burn for cooking fires. Far off in the fields whole families were already at work, specks of color bobbing in the green. Preoccupied with reading the Amuq's past, the archaeologists barely seemed to notice the present. They would stay on for another fortnight. Wilkinson joined them, and they worked harder than ever, going out in the morning, returning for lunch and rest, then going out a second time in the afternoon. There was, Casana said later, "So much to do, so little time."

Going to a site Casana and Erek (in cap) are, as always, accompanied by a Turkish official.

And so much ground to cover. They returned to site No. 290 and found, beneath the cotton, a rare agate stamp used to imprint the clay seal of a jar or basket. (The archaeological museum in Antakya later claimed it for its collections.) They crossed the Amuq and explored the slopes and valleys of the Amanus Mountains on the plain's north side, discovering settlements no higher than 600 meters, which happens to be the ecological limit of olive cultivation. Perhaps, they speculated, the dispersal of settlement around the Amuq had been caused by farmers seeking to expand olive production. They discovered that a site near the Beylan Pass was older than had been thought, confirming that the pass had been part of an ancient trading route to the Mediterranean. And they found a site in a hard-to-reach forested area that may have been a summer resort for wealthy Romans fleeing the hot summers of the plain.

Wilkinson left in the third week of September, heading to Iran, and Casana and the others ended their fieldwork. They had hoped to do more, but they no longer had permission to keep working. In any case they were exhausted. "The season was a great success," Casana said back on campus, where he is spending the year writing up his dissertation. "Everybody got along, and we got a lot done." But his thoughts were already racing ahead to his return and further explorations in the mountains on the plain's north side. "We found fascinating things up there, and there is a lot we don't know about them," he said. How dense were the settlements in the high valleys? What were people doing there, anyway? Were there mines, quarries, and roads? He also wanted to look harder for a Paleolithic site. He and his colleagues had found plenty of stone tools on the Amuq; maybe they could also find a cave where the makers had lived. Someday, too, he dreamed of doing a full archaeological survey within the city of Antakya, looking for ruins of ancient Antioch. The buried Roman city is thought to be inaccessible, but just driving around he had already found bits of it on Antakya's outskirts, where new construction is exposing-and destroying-archaeological sites. The search for ancient landscapes led in all sorts of directions, and to many more seasons on the Amuq.