Visit the blog
announces a new imprint

Search Articles

Search Authors

Advanced Search

Join our Mailing List
Thursday, August 01, 2002


The Canal of Xerxes in Northern Greece: Fact or Fiction?

There are few examples of canals in the ancient Mediterranean world. The Romans tried to build one across the isthmus of Corinth in Greece in the first century CE but gave up, resorting instead to a slipway along which ships were dragged, parts of which are still visible today. In earlier times, it was the Phoenicians who were the most proficient marine engineers: their hallmark was the cothon, the short stretch of canal linking an inner harbor to the sea, best seen at Carthage and Motya in Sicily. Against this background, the detailed description by Herodotus of a canal built in northern Greece on orders of the Persian king Xerxes in 480 BCE to allow the Persian fleet safe passage into the Aegean in advance of its invasion of Greece seems extraordinary: We read that it was some two kilometers long and wide enough for two triremes to pass side by side. If this account is true, the construction must have been a remarkable engineering feat for its time, surpassing in audacity the pontoon bridges reputedly built across the Hellespont earlier in Xerxes’ fraught campaign of 480 BCE.

In view of Herodotus’ descriptions, the Canal of Xerxes, which is situated at the narrowest point on the easternmost peninsula of Chalkidiki close to the famous holy community of Athos in northern Greece, attracted the attention of nineteenth-century travelers. Although they found little evidence of it on the ground, they firmly believed Herodotus’ account. On the other hand, there was an alternative hypothesis, first voiced by Demetrius Skepsius (reported by Strabo) that the terrain at the southern end of the isthmus would not have permitted construction of a canal, with the result that there was a slipway for a short section from the sea connecting with the canal that was built over the remainder of the isthmus. Furthermore, there are reports that during the Greek war of independence, boats were dragged across the isthmus.

Today, the only physical evidence of a canal is a shallow linear depression in the central part of the isthmus. Visitors traveling southward to Ouranoupolis may notice a large sign proudly announcing the presence of the canal, but, as they travel along the modern road running parallel to the canal, they are unlikely to notice much. The surrounding land is agricultural; nothing is visible of the canal at either the northern end, in a marshy area close to where the present-day village of Nea Roda lies, or the southern end at Tripiti. Moreover, perhaps surprisingly, there is an apparently total lack of remains of any installations – such as quays, breakwaters, or buildings that would surely have been necessary during the time the canal was in operation – on or close to the coast. Pottery shards found on the surface across the isthmus belong to various periods. If the canal had existed, it would now in all likelihood be deeply buried. In sum, there have continually been sufficient questions and hypotheses regarding the possible existence of the canal to merit an archeological investigation.

Over the last decade, this goal has been achieved by employing a combination of modern techniques, mainly geophysical, that would sense any archeological presence beneath the ground surface without recourse to traditional excavation. The British-Greek project was a multidisciplinary effort, carried out under the auspices of the British School of Archaeology at Athens, with a permit from the Greek ministry of culture. Initiated by Dr. Ben Isserlin (University of Leeds), the project comprised myself, Dr. John Uren (University of Leeds), Dr. Vassilis Karastathis (National Observatory, Athens), Professor Stavros Papamarinopoulos (University of Patras), and Dr. George Syrides (University of Thessaloniki). Financial support came from various sources, chiefly the British School at Athens, the British Academy, and Professor Papamarinopoulos’s GEOTECH project funded by NATO’s Science for Stability program.

  The project’s first task was to map the probable course of the canal based on present-day topography and the presence of the shallow depression in the center of the isthmus. This showed that construction of a canal would not have confronted flat terrain: the ground surface in fact rises to over 15 meters above present sea level in the center, an observation that has implications for the scale of the building operation. In the geophysical survey that followed, several techniques were employed, some of them experimentally. In the event, the results of electrical soundings and ground-penetrating radar were not as informative as those from seismic surveys, which until now have found little application in archeology elsewhere. Seismic refraction and reflection measurements provided decisive evidence of the canal’s existence in the central sector of the isthmus, where its depth was estimated to be 14-15 meters below present ground surface, with top and bottom widths of 25-35 meters and at most 20 meters, respectively. The high-resolution seismic-reflection profiles agree remarkably well with Herodotus’ account of the canal. Seismic surveys elsewhere along the canal were no less successful. At the southern end, the results indicated that, while the canal’s outline could not be identified, there would have been no physical obstacle to its construction, and thus no need (pace Demetrius Skepsius) to invoke the slipway hypothesis from the coast at Tripiti for 200-300 meters inland. From the recent seismic survey at the northern end, important findings are emerging suggesting a revision of the canal’s course: it met the sea at a more easterly position than was previously thought, which is in agreement with what can be discerned from a recently obtained satellite image of the Athos peninsula. In this same area, a magnetic survey identified some building remains that it would be tempting to associate with the canal’s “infrastructure,” but, in the absence of excavation, this is speculative.

In tandem with the geophysical survey was the analysis of cores recovered from bore holes drilled at positions along the likely course of the canal. Within the central sector, two classes of sediment were found, compact red beds lying below silty coarse sands. Most significantly, the point of contact between the two classes was roughly 14.5 meters below present ground level, representing the probable bottom of the canal; thus the red beds were presumably the original material through which the canal was cut, and the silty coarse sands formed the subsequent infill of the canal when it went out of use. This may well have occurred rather rapidly because no marine organisms were detected in the sediments in the central sector. (If the canal had remained open for some time, such organisms would have found their way from the sea to the middle of the canal.) During the canal’s short lifetime, it was probably not maintained, but instead rapidly became blocked as a result of its sides collapsing. The sediments examined in the cores contained some organic material that was suited to radiocarbon dating (carried out at Demokritos research center in Athens); unfortunately but not unexpectedly, the dates turned out to be much earlier than the fifth century BCE because the sediments are effectively contaminated with organic components containing older carbon present in the soil before sedimentation starts. As a result, it has thus far been impossible to obtain an objective chronological “handle” on the canal.

At this point, the question posed in the title above can be returned to and answered with some confidence: Modern scientific techniques in archeology have revealed without recourse to excavation a canal-like structure built across the isthmus whose main features resemble those described by Herodotus. The canal, which is now deeply buried, had a width at the top of 26-30 meters and slopes ranging from 20 to 30 degrees. The depth of water in the canal is difficult to gauge owing to the rise in sea level since its construction, but was probably only a few meters.

More interesting perhaps is Xerxes’ motivation in ordering the canal’s construction. Herodotus gives a vivid account of the Persian king’s uncertain state of mind, and the trials and tribulations he confronted in advance of the invasion of Greece. There was the practical factor brought about by the calamity that befell the king’s navy led by Mardonios a few years previously, when it was destroyed rounding the head of the Athos peninsula in treacherous weather. Xerxes would have vowed that that could never happen again. He embarked instead on a remarkable engineering project led by Phoenician engineers and manned by a large (local) workforce working in sections along the canal, as Herodotus describes, shifting an estimated quarter of a million cubic meters of earth. Impressive though this estimate may sound, the operation could have been accomplished in a matter of months rather than years since, in contrast to the canal at Corinth, it was dug not through rock but sands compacted with clay and pebbles, as the sedimentological analysis has clearly shown.

Aside from the practical issues, however, there is the likelihood that the canal had an additional purpose: prestige and a show of strength. After all, history (even modern history) shows that many buildings and structures owe their existence to the same motivation. All the indications are that once the major engineering operation was completed and the Persian fleet had successfully passed through the canal, it was soon abandoned, never to be reused. This would go some way toward explaining the apparent lack of building structures and harbor installations.

R. E. Jones is senior lecturer in the department of archeology at the University of Glasgow. His work focuses on the production and exchange of prehistoric pottery in the eastern Mediterranean, as well as on archeological geophysics.
Page 1 of 1 pages