(Tk., “sparrow harbor”),

a small, protected bay (36°34′25″ N, 28°03′08″ E) with a deep, narrow entrance on the southwest coast of Turkey, opposite the island of Rhodes and just northeast of the better-known bay of Loryma (modern Bozukkale), with its excellently preserved Hellenistic fortification Serçe Limanı was almost certainly the Portus Cressa of Pliny (Natural History 5.29) and the Kresa of Ptolemy (5.2); it served the town of Kasara, probably a deme center attached to Kamiros on Rhodes.

Underwater surveys show that the seeming tranquility of Serçe Limani attracted ships to an anchorage at the north-eastern end of its entrance passage from the Early Bronze Age until modern times. Its busiest period, as might be expected, was during the Hellenistic heyday of Rhodes. Ships that did not quite reach the safety of the harbor have left their remains outside it. These include a steamboat, perhaps from the nineteenth century CE, and a sixth-century BCE wreck that has not been located, although it has yielded strap-handled amphoras to sponge draggers' nets.

Ships that did find shelter in this natural harbor still faced peril, for sudden winds channeled from different directions by valleys at the bay's ends make Serçe Limani more dangerous than it seems. Two ships sank along the eastern side of its entrance passage, and two others just beyond these where the bay widens. One of these, just inside the harbor mouth, about 30 m deep, is a Roman amphora carrier of the first century BCE or CE. Farther inside the harbor, at a depth of 35–37 m, are the remains of a Hellenistic wine carrier. Already looted of almost all the amphoras that had been visible when archaeologists from the Institute of Nautical Archaeology (INA) discovered it in 1973, it was partially excavated by the INA and Texas A&M University in 1978–1980. The excavation was interrupted when it was discovered that much of the site had been covered by tons of boulders from a rockslide that took place at an unknown date after the ship sank, making the excavation more difficult and dangerous than anticipated. Nevertheless, they uncovered about six hundred Knidian-type amphoras, in two sizes, some with Zenon Group stamps that date the wreck, with its glazed and plain wares, to the middle of the first half of the third century BCE. Other finds include the two components of a stone hopper mill, and the lower grinding platform of a quern. The ship itself was at least partly lead sheathed, in the custom of the day, and yielded a lead pipe that may be the earliest evidence of a bilge pump.

Next inside the harbor, approximately 150 m north of the third-century BCE wreck, is the eleventh-century CE wreck described in detail below. East of it, closer to shore, appear to be the remains of another Hellenistic wreck, badly broken and scattered in shallow water.

The wreck for which Serçe Limani is best known is that of an eleventh-century CE merchant ship with a cargo of Islamic glass excavated by the INA between 1977 and 1979. The ship was about 15 m long with a beam of about 5.2 m. Its pine hull and elm keel, now reassembled in the Bodrum Museum of Underwater Archaeology in Turkey, provides the earliest dated example of a seagoing vessel built in the modern, frame-first manner. Although only a few of its frames (ribs) were erected before the planks were installed, this framework was now the primary structure; the ship-wright, however, was still essentially shaping the hull with planks. The ship's flat-bottomed, boxlike shape would have allowed it to navigate in shallow waters, including rivers. It carried two metric tons of ballast in the form of just over one hundred boulders, with an additional half ton of cobbles. Its eight iron, Y-shaped anchors, with removable wooden stocks, were forged from short iron rods; three were bowers—two to port and one to starboard—with the remaining five stowed just forward of midships. [See Anchors.] A surviving halyard block and the hull design suggest that the ship sported two lateen-rigged masts. There appear to have been three living areas: one at the bow (probably covered by a deck), an open space on the midships deck, and a stern cabin.

The ship probably sailed from the Black Sea or somewhere around Constantinople in about 1025 with Bulgar merchants carrying in the hold about one hundred of their own individual amphoras that had been fired in kilns around the Sea of Marmara. The voyage took the ship to the neighborhood of Caesarea, in modern Israel, where it took on its cargos of Islamic glass and glazed bowls, and perhaps jewelry and copper and bronze vessels of Islamic origin; the glazed bowls and an earring of delicate filigreed gold find their closest archaeological parallels at Caesarea, where apparently similar glass is also found. [See Caesarea; Glass.]

Although much that the ship carried was of Islamic origin, it seems that many or most of the crew and merchants on board were Christian. Three of four Byzantine seals recovered have Christian scenes impressed in their lead: the Virgin Mary with the Christ child, the ecstatic meeting of Peter and Paul, and a warrior saint. One of the seals protected the writing of someone named Peter; the fourth had not yet been used. The Christianity of those on board is also indicated by the pig legs found among the food remains, less likely to have been eaten by Jews or Muslims, and the crosses (and in one instance the name Jesus) inscribed or molded inside some of the nine hundred-odd lead fishnet sinkers used during the voyage.

In the forward part of the ship's hold there must have been a cargo like soda because it left no archaeological clue as to its nature. The wine amphoras were stowed in the center and aftermost parts of the hold. In the after third of the hold three tons of glass cullet included two tons of raw glass and one ton of broken glass. The broken glass was largely waste from an Islamic glassworks that has not been identified. Cullet is needed not only for making glass vessels in places that do not produce their own glass, but also for making new batches of glass from its basic ingredients. Mending the glass cargo from between half a million and a million shards during a ten-year period revealed more than two hundred shapes, some unique, including varieties of cups, bowls, bottles, plates, lamps, beakers, jars, jugs, and ewers. It is estimated that these are the remains of between ten thousand and twenty thousand vessels, many of which had engraved or molded designs. Altogether they form by far the largest and best-dated collection of medieval Islamic glass known. The colors are mostly various shades of green and purple, but dense greens and blues also occur. Above the cullet, the ship carried cargos of raisins and sumac, the latter believed to have been exported through Caesarea at about the time the ship sank.

Serçe Limani

SERçE LIMANI. Figure 1. Collection of Islamic glass vessels. (Courtesy G. F. Bass)

view larger image

Eighty intact Islamic glass vessels were also on board, perhaps once in bundles of merchants' goods; these were, without exception, in the bow and stern living quarters (see figure 1). The living quarters were identified by cooking and eating wares as well as by the bones of sheep or goat and pig; other food remains included almonds, apricots, plums, and olives, although which of these were for shipboard use and which for trade has not been determined. In the stern living area were eighteen Islamic glass weights that date the wreck to about 1025 CE, a date supported by the Fatimid gold coins and Byzantine copper coins of Basil II recovered; this has allowed the closest dating of some of the types of Islamic glazed bowls on board. Weighing instruments included sets of both Byzantine and Islamic weights, three pan-balance scales, and a small Byzantine steelyard. Also at the stern were the ship's carpenter's tools, including bow drills, hammers, axes, adzes, saw, files, and the earliest known caulking tools. Those who shared this stern area played chess, some of whose wooden pieces survive, whereas what may well be a backgammon counter found amidships suggests a less intellectual game for the common sailors on board. Lastly, those in the stern were the only ones on board who had pork and fish for food at the time the ship sank.

It seems that three fishing nets, each about 40 m long, based on the numbers, size, and distribution of more than nine hundred lead net sinkers, were being mended on deck with net needles on the final voyage; eight bone spindle whorls of probable Syrian origin found near the net sinkers suggest that the sailors spun their own cords for this task. There was also a smaller casting net.

The ship was heavily armed, with fifty-two javelins, eleven spears, and at least one sword (with a feathered bird on its bronze hilt, inside a wooden scabbard); the weapons were issued in sets of one spear and five javelins wrapped in burlaplike cloth. Personal possessions on board included a grooming kit comprising a wooden delousing comb, a razor, and scissors; found nearby, near another comb, were piles of orpiment, a substance still used with quicklime as a depilatory. The ship's ultimate destination is not known, but it may have been returning to Byzantium with its Levantine cargo.

Bibliography

  • Bass, George F. “The Shipwreck at Serçe Liman, Turkey.” Archaeology 32.1 (1979): 36–43.
  • Bass, George F. “Archaeologists in Wet Suits.” In Science Year 1982: The World Book Science Annual, pp. 96–111. Chicago, 1981.
  • Bass, George F. “The Million Piece Glass Puzzle.” Archaeology 37.4 (1984): 42–47.
  • Bass, George F. “The Nature of the Serçe Limani Glass.” Journal of Glass Studies 26 (1984): 64–69.
  • Bass, George F., et al. “Excavation of an Eleventh-Century Shipwreck at Serçe Liman, Turkey.” National Geographic Society Research Reports 17 (1984): 161–182.
  • Jenkins, Marilyn. “Early Medieval Islamic Pottery: The Eleventh Century Reconsidered.” Muqarnas 9 (1992): 56–66.
    Describes how the excavation revolutionized accepted views of early medieval Islamic art
    .
  • Koehler, Carolyn G., and Malcolm B. Wallace. “The Transport Amphoras: Description and Capacities.” American Journal of Archaeology 91 (1987): 49–57.
    To accompany Pulak and Townsend (below)
    .
  • Pulak, Cemal, and Rhys F. Townsend. “The Hellenistic Shipwreck at Serçe Limani, Turkey: Preliminary Report.” American Journal of Archaeology 91 (1987): 31–49.
    To be accompanied by Koehler and Wallace (above)
    .
  • Steffy, J. Richard. “The Reconstruction of the Eleventh-Century Serçe Liman Vessel: A Preliminary Report.” International Journal of Nautical Archaeology 11 (1982): 13–34.

George F. Bass