Middle Pleistocene sea-crossings in the Aegean

The possibility that humans reached the Mediterranean islands in the Palaeolithic has been the subject of discussion for decades. Until recently, the consensus has been that seafaring—narrowly defined—did not emerge until the Terminal Pleistocene, ca. 12,000 BP, a consensus challenged, at least for the Greek islands, by the discovery of early Palaeolithic stone tools on Alonnisos in the Northern Sporades, Gavdos and Crete in the southern Aegean, the western Ionian islands of Kephalonia and Zakynthos, and Melos and Naxos in the Cyclades.

If one assumes that some of these islands were separated from the mainland during much if not all of the Pleistocene, Palaeolithic hominins would have made open sea-crossings to reach them. These recent Palaeolithic discoveries have suggested to some scholars that maritime activity in the Mediterranean began in the Middle Pleistocene.

Nevertheless, the distances to be crossed are difficult to calculate, ranging from as little as 5 to as much 30 or 40 km, and near-shore islands may have been extensions of mainland home ranges, visited perhaps as part of a subsistence strategy that included aquatic resources in coastal environments. For some scholars, this ‘‘triggered a slight ‘stretching’ of behaviour”, but for others the suspicion is that deepwater ‘oceanic’ islands required the use of watercraft to reach them.

Crete and Naxos have recently produced significant numbers of Palaeolithic stone tools. They would have required watercraft to reach them in the Pleistocene. On Crete, early Palaeolithic assemblages have been reported from three locations: Loutro, Plakias, and Mochlos. At Plakias, more than 200 artifacts in vein quartz and quartzite were collected from nine find spots. They are dated to greater than 130,000 BP by their direct association with raised marine terraces dated by 14C and local uplift rates, and a highly developed paleosol dated by Optically Stimulated Luminescence (OSL). The geological contexts for these artifacts are secure, as they were extracted from within the marine terraces and the paleosol outcrop. Nor is the artifactual nature of the artifacts in question. A detailed study of morphotypological characteristics of the Plakias sample identified forms such as bifaces, picks, cleavers, and scrapers that are typical of early Palaeolithic traditions in the region.

The other site, Stelida on Naxos, is associated with a large chert outcrop that served as a source of raw material and atelier for the production of early Palaeolithic artifacts. The surface distribution of artifacts has been surveyed, and, crucially, excavations carried out in 2015 produced artifacts of early Palaeolithic type in stratified deposits.

We conclude that there is prima facie evidence that Palaeolithic hominins reached some of the Greek islands.

What were the environmental conditions in the Aegean in the Middle and Upper Pleistocene? The reconstruction of the palaeoenvironment requires more data than are presently available, but it is widely accepted that the Pleistocene climate was predominantly glacial with fluctuating sea levels. Indeed, sea-level lowstands reconfigured shorelines around the Aegean basin, creating new landmasses like the central Cycladic Island, and reducing the distances between offshore islands and the Greek mainland.

Acknowledging the limitations of the models, our selection of the Greek islands considered to be ‘offshore’ during the bulk of the Pleistocene is based on older reconstructions of palaeoshorelines.

Even in the absence of fine-grained palaeoshoreline reconstructions, it is evident that some Greek islands remained separated from the mainland for long periods of time. This ‘oceanic’ isolation is inferred from the presence on these islands of endemic fauna, such as hippopotami and elephantids, that were nanized as the result of ‘insular dwarfism’ and an absence of mainland predators.

In the Mediterranean there is no physical evidence for prehistoric watercraft, but experimental research has provided much food for thought. For example, an experimental sailing raft of canes made the crossing from Kythera to Crete in 48 hours.

Despite criticism of the significance of this project, especially the use of anachronistic methods and technological strategies during the construction phase of the raft, it demonstrates one simple possibility: Mediterranean canes bound into bundles and lashed together with cords can be paddled across open water from the Greek mainland to Crete in two days carrying ten persons. In another experiment, a replica papyrus reed boat—or ‘papyrella’—was paddled successfully from the mainland to Melos and other Cycladic islands in 1988.

Other types of watercraft include skin-covered boats, which were well-known in antiquity, e.g., the river craft in Mesopotamia, and in the Po Valley in northern Italy. Wooden dugouts are represented by a well-preserved specimen from the lakeside Neolithic settlement at La Marmotta (ca. 7500 BP) in central Italy. This ten metre-long dugout was hewn from the trunk of a single oak and was capable of making sea-crossings of considerable length. An experimental replica with a crew of eleven achieved rates of travel of 32 km a day in the open Mediterranean.

These experiments suggest that three technologies are required to build simple watercraft: sharp-edged cutting tools, fire, and cordage. Were these technologies available to archaic hominins (a question separate from whether they used them for this purpose)? The answer is yes. Stone tools capable of cutting reeds and wood, striping cane, and scraping skin are the earliest of all our known technologies, and Lower Palaeolithic cutting tools such as handaxes, cleavers, choppers, picks, and scrapers can be used to cut reeds or work timber, and transverse-edged cleavers could be used to hollow out a firecharred log for a dugout canoe. Fire also has an early pedigree, and evidence for the controlled use of fire is documented at least one million years ago.

Finally, cordage from plant fibres, animal hair or skin required few tools other than sharp stone flakes for cutting and scraping. Indeed, skin working and cordage are probably as old as stonetoolmaking, and knowledge of fibres has significant time depth in the Palaeolithic. We note that research is pushing the origins of fire and the manufacture of fibre cordage back in time, and technologies that were once thought to be the exclusive products of AMH in the last 50,000 years appear to have much earlier origins. There is good reason to conclude that early hominins had the technical elements necessary for constructing watercraft in the Middle Pleistocene.

Aegean waters are unpredictable, and in the Pleistocene, when straits were narrowed by glacial sea-level lowstands, there were correspondingly faster and more unpredictable currents. To reach an offshore island, even one in plain sight across a narrow channel, it would have been necessary to control the direction of travel—a point of importance if one planned to return to the point of departure. The growing evidence for Palaeolithic presence on Crete points to repeated visitation of the island, as opposed to a one-off ‘sweepstakes’ colonization event. A few individuals stranded on the island more than a hundred thousand years ago would be invisible archaeologically; repeated crossings would have been necessary, as they were in the Early Neolithic.

Prevailing winds played an important role in establishing maritime networks in historical periods, including the formation of trade routes, but for early mariners sea currents were equally important. Small, lightly-built watercraft that were rowed or paddled would have been slower and more susceptible to the movement of surface currents and swells than to winds, giving them more independence of movement.

Voyaging to the large oceanic islands of Cyprus and Crete required sailing at night and a degree of astronomical knowledge. The rising and setting of the sun and moon and the discernible patterns and rotations of the stars were no doubt observed, along with the constellations revolving around a celestial null point in the night sky where the stars rotate around the celestial pole, specifically, the closer the stars are to the pole the less they change position. These circumpolar stars served as reliable indicators of relative north for early navigators, a so-called ‘star path’ to help maintain a heading across open sea. Prior to the use of the magnetic compass, navigators set their headings by reference to these celestial bodies, by day fixing their position by the dawnstar, and holding the rising sun on their starboard bow for a northerly heading, and by night, fixing their position and heading by the evening star and the polestar. Early mariners in the Mediterranean would have studied the rising and setting of the moon, the motion of the constellations, and followed the star drift of the Milky Way.

Navigation out of sight of land was less important in the Aegean. Many islands are inter-visible for the duration of a voyage and mariners approaching land line up prominent landmarks on the nearby shore to set a course, such as the cloud-capped Cretan sierra. A landscape outline visible from the sea is called a ‘recognition view’ and such views enabled mariners to maintain a heading when returning to the same stretch of coast.

The physical environment of the Mediterranean provided a favourable coastal configuration for the development of early maritime activity in that region, particularly in the Aegean, where the unique regional conditions of intervisibility of land and narrow marine straits would have enticed Palaeolithic foragers and hunter-gatherers to explore the islands. It is time to consider the signs of Palaeolithic activity on the oceanic islands of the Mediterranean and the questions raised by their presence.

(Source: “Middle Pleistocene sea-crossings in the eastern Mediterranean?”, by  Duncan Howitt-Marshall & Curtis Runnels, 2016)

Our reader may be interested to read some additional information here: 1, 2, 3

Research-Selection for NovoScriptorium: Philaretus Homerides