Stanford University Home

Stanford News Archive

Stanford Report, November 12, 1997

Earthquakes toppled ancient cities: 11/12/97

Don't blame the Trojan Horse: Earthquakes toppled ancient cities, Stanford geophysicist says


Around 1200 B.C., the great Bronze Age civilizations of the Eastern Mediterranean toppled like a string of dominos. One by one, over a period of 50 years, dozens of bustling centers of scholarship and industry, including Troy, Mycenae and Knossos, collapsed into rubble. Today, crushed skeletons and scattered debris are all that remain of the powerful cities. What force could wreak such widespread destruction?

Historians and archaeologists have long battled over this question, citing civil war, invasion and pestilence as possible causes. When Stanford geophysicist Amos Nur examines the evidence, he sees another possibility: The earth moved. A string of massive earthquakes could have knocked down one city-state after another, bringing the Bronze Age civilizations to a premature end. Earthquake activity also may be at the root of the biblical prophecy of Armageddon, the site of the final conflict between good and evil. According to Nur, the repeated destruction of the city Megiddo probably inspired the author of Revelation to script his haunting prediction of the Apocalypse.

Related Information:

For over 20 years, Nur has been studying the role of earthquakes in the Eastern Mediterranean. By poking around the ruins of Megiddo and other ancient cities, he has gleaned fundamental information about the patterns of large quakes. Amidst the rubble of the ancients, Nur discovered clues that may help modern geophysicists understand when and why earthquakes occur. His findings suggest that earthquakes are episodic ­ periods of greater earthquake activity are bracketed by periods of relative quiet. Large quakes may trigger other large quakes, in a domino effect that zips down a fault line, knocking down cities along the way.

Nur is the Wayne Loel Professor of Earth Sciences, director of the Rock Physics and Borehole Project, and the current chair of Stanford's geophysics department. He is an expert on the physics of large-scale earth movements, including earthquakes. In the early 1970s, he began studying the temporal and spatial patterns of historical earthquakes in order to identify indicators of future tremors. He chose the Eastern Mediterranean, the Holy Land, since it has the oldest, most complete record of earthquake activity. "Although human history in this region provided evidence for bygone earthquakes, it was recent advances in our understanding of plate tectonics which imparted unexpected insights about the destruction of ancient cities," he said.

According to the plate tectonics model, the crust, or outer layer of the earth, consists of about a dozen large, irregularly shaped plates that slide over, under and past each other. As the plates move relative to each other, sudden dislocations in segments of the crust create fractures, or faults. Because faults reflect zones of weakness in the crust, earthquakes tend to occur at the plate boundaries in these regions. Those fault zones often form the mountain passes and river valleys most used by humans in their migrations.

For five millennia, the city of Megiddo stood at one of the most important junctions in the ancient Near East, the Nahal Iron Pass. This pass was the only means of traversing the Carmel-Gilboa mountain range on the road from Damascus to Egypt. By controlling this route, Megiddo commanded the course of trade and the march of armies in the Holy Land. Excavations suggest the city was repeatedly devastated by some large force. Archaeologists believe that warring factions were responsible for this destruction. Nur is certain that earthquakes were partly to blame.

Megiddo is located very near the Carmel-Gilboa fault system, which is a branch of the larger and more dangerous Dead Sea fault system. The latter system accommodates motion between two plates, the Arabian plate to the east and the Mediterranean plate to the west. Given Megiddo's proximity to a fault zone, "it is beyond doubt that Megiddo, along with its neighboring territories, must have experienced earthquakes strong enough to cause significant or total destruction," Nur said.

Archaeological and historical data support Nur's hypothesis. According to written records, the Holy Land has been shaken by 11 devastating earthquakes since 1400 B.C. At Megiddo, three layers of destruction cannot be explained by the invasion of foreign armies. In addition, the excavation of sites far to the north and to the south suggests that additional cities were damaged at the same time as Megiddo. This regional pattern of destruction is consistent with a massive earthquake along the Carmel fault.

The most compelling evidence for Nur's earthquake hypothesis is also the most gruesome: crushed skeletons found trapped under the collapsed rubble. The tortured positions of the bodies indicate that these people were struck by a sudden and massive load. The amount of debris found in adjacent areas suggests that the wall's collapse was not an isolated incident. It is unlikely that these people died in an invasion, given the presence of pottery shards and precious metals in their immediate vicinity. Why would conquerors destroy valuable objects instead of looting them?

Also, there exists at least one biblical reference to seismic activity at Megiddo. John of Patmos, the author of the book of Revelation, appeared to know of Megiddo's frequent destruction by earthquakes when he wrote, "And they assembled them at the place that in Hebrew is called Armageddon and there came a violent earthquake" (Revelation 16:16). The word Armageddon is a Greek transcription of the Hebrew Har Megiddo, which means the Mount of Megiddo. It seems likely that John used the recurring desolation of this one particular city to symbolize his vision of the Apocalypse to come.

Nur and his colleague, Hagai Ron of the Israel Institute of Geophysics, reported their data on Armageddon's earthquakes in the 1997 edition of the journal International Geology Review.

The fall of Troy ­ and more

Recently, Nur expanded his geophysical analysis of the Holy Land to include cities besides Megiddo. His current work suggests that earthquakes may have played a large role in the collapse of at least 50 great cultural centers, including Troy, Mycenae and Knossos, at the end of the Bronze Age. He presented his data in July at a conference on the destruction of Bronze Age civilizations held at Cambridge University.

Because it took 50 years, from 1225 B.C. to 1175 B.C., for the major cultural centers to collapse, it is unlikely that the end of the Bronze Age was caused by a single historical event. However, a string of earthquakes could have destabilized society enough to wipe out the economic, social and political structures. "The end of the Bronze Age may actually have been a period of recovery following a string of severe earthquakes," Nur said.

According to Nur, seismic records indicate that large earthquakes are temporally clustered. Short periods of very intense earthquake activity are preceded and followed by long interludes of relative quiet. Geologically, these episodes may be explained as follows: When a plate ruptures in one place, it strains another part of the plate boundary and may cause its collapse a short time later. This cascade of activity occurs until the entire plate boundary ruptures. This period of intense activity is followed by longer time periods when the whole plate is strained but doesn't quite give. Eventually the strain builds up and the cycle begins again.

Nur points to measurements of Turkey's North Anatolian Fault as evidence of episodic earthquake activity. In this century, between 1939 and 1967, a series of earthquakes ruptured the entire plate boundary along Turkey's North Anatolian Fault, causing a slip on the order of 2 to 4 meters. Historical records indicate that seismic crises also occurred in the 4th and 8th centuries.

According to Nur, the cities destroyed at the end of the Bronze Age were located in regions that historically have experienced high seismic activity. He has calculated the intensities of recent earthquakes and shown that the modern-day regions that experience heavy damage overlap with the ancient ruins destroyed at the end of the Bronze Age. As Nur states, "Earthquakes have been happening in this region for thousands of years. There is no way that these places could have escaped severe ground shaking. It is impossible."

It is not too hard to imagine how earthquakes might have caused the collapse of ancient societies. Given their limited technology, it would have been difficult for societies to rebuild their magnificent temples and houses. In the wake of such a catastrophe, skills like reading and writing could have disappeared if people were concerned with more important activities, like survival. "It probably took many years to recover from such an event," Nur said.

Straining geophysicists' ideas about quakes

Nur believes his studies of the Bronze Age civilizations may cause geologists to rethink the forces that generate earthquakes. Do earthquakes really occur in episodic bursts? If so, geologists will need to reevaluate how and when strain is released along plate boundaries. The traditional view is that strain is periodically released at each segment of the plate boundary. The evidence obtained from Nur's analysis of the Eastern Mediterranean suggests that strain is released in episodes via a sequence of earthquakes. A tremor at one segment of the plate boundary appears to trigger a chain reaction of displacements along the rest of the fault. According to Nur, "The entire plate boundary gets unzipped by this sequence of large earthquakes."

Nur's findings may have an impact on the way geophysicists predict the likelihood of future earthquakes. By charting past episodes of strong earthquake activity, it may someday be possible to design models that predict active and quiescent periods. According to Nur, such predictive algorithms are a long way off. He says, "Right now earthquake patterns aren't regular enough to tell us anything."

Still, based on seismic measurements, it's pretty clear that a strong earthquake will hit the Eastern Mediterranean some time in the future. Who knows? If John of Patmos' description is correct, Megiddo may be the site of the "next big one." As it was written in the Book of Revelation: "into a place called Armageddon . . . there was a great earthquake such as was not seen since men were upon the earth. . . . And the great city was divided into three parts, and the cities of the nation fell. . . . And every island fled away, and the mountains were not." SR

Ellen Licking is a science writing intern at Stanford News Service.