Version classiqueVersion mobile
OpenEdition Books

Géoarchéologie des îles de la Méditerranée

Matthieu Ghilardi

Partie 1. Anthropisation et mutations paysagères à la transition Paléolithique/Néolithique / Anthropization and landscape changes during the Late Paleolithic/Neolithic transition

Late Pleistocene to Early Holocene Sea-Crossings in the Aegean: Direct, Indirect and Controversial Evidence

Papoulia Christina


The oldest direct evidence of a boat from the Aegean is dated to the Neolithic and comes from a lakeside settlement in NW Greece. However, indirect evidence in the form of structures, artefacts and aquatic resources are present on islands and date from at least the Mesolithic. Obsidian procurement networks testify to seagoing journeys since the Upper Palaeolithic, while a number of studies have recently suggested that the initial sea-crossings took place during earlier parts of the Pleistocene. At present there is no consensus regarding the exact date of the earliest successful attempts to cross the Mediterranean, thus rendering the extant pre-LGM (Last Glacial Maximum) evidence highly controversial. This paper reviews the geoarchaeological evidence for the early prehistoric sea-crossings in the Aegean (Eastern Mediterranean). It stresses the differences in character and scale between the initial serendipitous crossings and the fully-organised maritime networks of the later parts of prehistory. In view of the limited direct evidence, the examination of indirect forms, together with the information gained from ethnography and experimental archaeology, enables us to propose specific hypotheses regarding the nature of the crossings, the routes and the types of vessels used.

Texte intégral


1Artefact dispersal accompanies human diasporas, and human diasporas are framed by natural or arbitrary borders. During the Pleistocene and early Holocene these borders were natural boundaries such as deserts, mountains, forests and the open sea (Bar-Yosef and Belfer-Cohen, 2013; Derricourt, 2006; Roberts and Petraglia, 2015). At times, and under specific circumstances, otherwise insurmountable boundaries may be crossed resulting in the exploration of unfamiliar territories (Dennell et al., 2014; Flemming et al., 2003; Rockman and Steel, 2003). But while some crossings may be successful, others may not. The different degrees of success depend on various factors, both social and environmental, which are related to skill, know-how, innovation and exchange of ideas, as well as an appreciation of the climatic conditions and an effective plan for overcoming hazardous conditions (Farr, 2006; Papoulia, 2014). Marine crossings include both organised voyages to known or unknown lands and serendipitous sea-crossings, and even unintended landings at unfamiliar shores due to natural hazards (Leppard, 2015; Ruxton and Wilkinson, 2012). Depending on the degree of intentionality of the crossing, as well as on cultural traditions, technical expertise, and available resources, different types of vessels are evident in the archaeological and ethnoarchaeological record (Anderson et al., 2010; Marangou, 2001). As one goes further back in time, the evidence decreases both in quantity and quality (Table 1). Thus, although for the Neolithic we have direct evidence of actual vessels and abundant indirect evidence of seagoing, the Greek Mesolithic record is restricted to a much smaller amount of indirect evidence from a limited number of sites (Galanidou, 2011; Mylona, 2014; Sampson, 2006 and 2010). Furthermore, the Late Pleistocene record comprises very few sites with robust indirect evidence dating from after the Last Glacial Maximum (LGM) (Laskaris et al., 2011), while the arguments for pre-LGM (> 20 000 BP) sea-crossings are based on a small amount of indirect evidence which remains controversial (Ammerman, 2013; Broodbank et al., 2014; Papoulia, 2016).

Table 1: Types of evidence (indirect and direct) for sea-crossings in the Aegean for each cultural period (pre-LGM = before the Last Glacial Maximum, ca. 20,000BP, LUP = Late Upper Palaeolithic, Meso = Mesolithic, Neo = Neolithic).

2This paper presents a review of the geoarchaeological evidence for the earliest episodes of maritime dispersals in the Aegean, spanning the Late Pleistocene to the earliest parts of the Holocene, up until the Neolithic. Aided by the information gained from the ethnographic record as well as from experimental archaeological projects (Tzalas, 1995; Tichý, 1999 and 2002; First Mariners, 2014), the relationship between the types of crossing and the types of vessel is discussed. The aim is not to provide an exhaustive inventory of the maritime archaeological record, nor to discuss in detail the arguments for or against the Palaeolithic sea-crossings (for this, see Papoulia, 2016). Rather, the intention is (a) to highlight the different degrees of familiarisation with the sea: i.e. to stress the differences in character and scale of the initial serendipitous crossings with those of the final Late Pleistocene and the early Holocene up until the more organised seafaring activities of the Neolithic which resulted in full colonization of insular territories; and (b) to emphasize the expectations that archaeologists should have when seeking the earliest indications of our Pleistocene ancestors’ attempts to confront the sea. As will be seen, although it is now widely accepted that the oldest marine crossings in the Aegean occurred sometime during the Pleistocene, their exact timing remains controversial.

The archaeological record

3In terms of prehistoric navigation, two kinds of evidence are available: direct and indirect. Direct evidence comprises all archaeological findings that are clearly associated with marine navigation, such as wrecks, boat remains, paddles, etc. The second category comprises all of the other evidence that only indirectly suggests the construction and use of sea-going vessels. This may include models of boats as well as iconographic evidence such as engravings with depictions of boats (Marangou, 2001; Televantou, 2008). Other types of indirect evidence include structures, artefacts, materials or even animal species (domesticates and small non-avian terrestrial mammals that are unable to swim), whose presence on islands can only be explained by sea-crossings. The presence of marine faunal remains in prehistoric settlements often implies the use of boats for inshore or offshore fishing.

Direct evidence

4Direct evidence of boats dated to the Pleistocene is completely absent from the entire Mediterranean region. The oldest boat remains in the Central (Italy) and Eastern Mediterranean (Greece) are dated to the Neolithic (Fugazzola Delpino and Mineo, 1995; Chourmouziadis, 1996), while no shipwreck recovered from the Aegean basin predates the Bronze Age. In particular, two preserved outlines of elongated vessels, i.e. dugouts, and an outline of a wider vessel (either a dugout or a hide boat), were found at the Neolithic lakeside settlement of Dispilio, Kastoria, NW Greece (ca. 5400-3500 cal. BC; Figure 1B.1; Chourmouziadis, 1996, Marangou, 2003). Although it remains the only example from the Aegean, a similar and better-preserved find comes from Lake Bracciano, Lazio, Italy (Figure 1A). An extended dugout produced by the addition of external fittings to the simple logboat was constructed from an oak trunk and was found underwater at the Neolithic lakeside settlement of La Marmota (ca. 5700-5200 cal. BC; Fugazzola Delpino and Mineo, 1995).

Figure 1: Location map of the archaeological sites and islands mentioned in the text and tables.
Figure 1A: Sites with direct (La Marmota - Italy) and indirect (Grabak Cave - Croatia, Akrotiri Aetokremnos - Cyprus) evidence for aquatic navigation throughout the Central and Eastern Mediterranean Basin.
Figure 1B: Archaeological sites and islands in the Aegean. Yellow circle stands for Dispilio, the only site with direct evidence of water transport and red circles stand for the only obsidian raw material sources in the Aegean. (1: Dispilio, 2: Diaplo islet, 3: Kerkyra sites, 4: Kefalonia sites, 5: Zakynthos sites, 6: Gavdos sites, 7: Plakias sites- Crete, 8: Knossos - Crete, 9: Mochlos - Crete, 10: Milos 11: Antiparos, 12: Naxos sites, 13: Yali, 14: Kerame - Ikaria, 15: Strofi - Andros, 16: Maroulas - Kythnos, 17: Franchthi Cave, Argolid, 18: Schisto Cave - Attica, 19: Alonnissos sites, 20: Kyra Panayia, 21: Cave of Cyclops - Youra, 22: Ai Stratis, 23: Tsangli, 24: Otzaki Magoula, 25: Dikili Tash, 26: Ouriakos - Lemnos, 27: Rodafnidia - Lesvos, 28: Skyros, 29: Chalki, 30: Maliq - Albania).

Indirect evidence

5Arguments for the earliest marine crossings in the Aegean are based exclusively on the discovery of lithic assemblages at insular loci. The insularity of the present-day islands of the Aegean can, to a certain degree, be assessed from the available reconstructions of the Pleistocene sea-level stands (Ferentinos et al., 2012; Kapsimalis et al., 2009; Lambeck, 1996; Lykousis, 2009; Table 2). The lithic assemblages, reco- vered from islands of the Ionian and the Aegean Sea, are either composed of exotic or of local raw materials (Chelidonio, 2001; Cubuk, 1976; Kavvadias, 1984; Kopaka and Matzanas, 2011 and 2009; Kourtessi-Philippakis, 1999; Laskaris et al., 2011; Mortensen, 2009; Randsborg, 2002; Sampson et al., 2012; Strasser et al., 2010). In the first case, the best-known argument is related to obsidian procurement. Obsidian raw material sources in the Aegean are restricted to the islands of Milos, Antiparos and Yali (Figure 1B, 10-11 and 13), and thus direct or indirect marine transportation from these islands to the sites of discovery is irrefutable (Laskaris et al., 2011). However, in the second case the situation is more complex. Artefacts with “archaic” morphological attributes made of local raw materials, such as different varieties of chert and quartz, have instigated a discussion of the possibility of Palaeolithic sea-crossings to particular islands of the southern Ionian and southern Aegean Sea (Chelidonio, 2001; Cubuk, 1976; Kavvadias, 1984; Kourtessi-Philippakis, 1999; Randsborg, 2002). The proposed dates for the crossings to Crete and Gavdos (Figure 1B, 6-7 and 9) are ca.130 000-70 000 BP (Kopaka and Matzanas, 2011 and 2009; Strasser et al., 2010 and 2011). All artefacts come from coastal or near-coastal openair sites, which during the Pleistocene would have been situated further inland due to the lower sealevel stands (Lykousis, 2009). The assemblages from Kefalonia (Figure 1B, 4) and Zakynthos (Figure 1B, 5) are composed of local cherts (Cubuk, 1976; Kavvadias, 1984; Kourtessi-Philippakis, 1999; Randsborg, 2002); that from Milos is composed of the local coarse-grained volcanic rocks, including one core made from the local obsidian (Chelidonio, 2001); those from Crete are composed of quartz (Strasser et al., 2010); and those from Gavdos are composed of local fine- and coarse- grained materials such as cherts, limestone, sandstone and others (Kopaka and Matzanas, 2009 and 2011).

6The literature on these early journeys is a growing area in prehistoric archaeology and new finds are constantly being announced (Carter et al., 2014; Runnels et al., 2014); nevertheless the archaeological community is still far from reaching a consensus regarding the optimal interpretation of the aforementioned assemblages (Ammerman, 2013 and 2014; Broodbank et al., 2014). The main reasons for this are (a) the weak geological associations of the finds, and thus the tentative dating; (b) the lack of direct associations with specific hominin or faunal species; and (c) the need for refined palaeogeographic reconstructions which pay special attention to the tectonic parameters for each part of the Aegean. Two additional reasons, both associated with epistemological aspects of the subject matter, are: (d) The notion that our early ancestors lacked the technical and/or cognitive skills to make use of any kind of artefact which would allow them to cross the sea; and (e) the difficultly in challenging established narratives regarding the initial occupation of the NE Mediterranean islands. Interestingly, the islands of the Central and Eastern Mediterranean share common research biographies in the sense that (a) definite Middle and Late Pleistocene sites are situated on present-day islands which were once connected to the mainland, e.g. Capri, Sicily (Italy; Figure 1A), Kerkyra and Lesvos (Greece; Figure 1B, 3 and 27); whereas (b) sites on present-day islands which on the timescales of interest were never connected to the mainland and are usually poorly dated, e.g. Corsica (France), Sardinia (Italy; Figure 1A), Crete and Gavdos (Greece) (Ammerman, 2013 and 2014; Broodbank et al., 2014; Mussi, 2001; Papoulia, 2016). On the other hand, there is a consensus regarding the oldest evidence of obsidian transportation in the Aegean. The Upper Palaeolithic and Mesolithic layers of Franchthi Cave (Argolid, Greece; Figure 1B, 17), and possibly also the Upper Palaeolithic layers of Schisto Cave (Attica, Greece; Figure 1B, 18), have yielded obsidian specimens originating from Milos and dated to the final part of the Late Pleistocene (ca. 14000-11000 BP; Laskaris et al., 2011). Further to the east, similar patterns are evident in Cyprus, where the oldest accepted dates (ca. 12000 BP) come from the coastal site of Akrotiri Aetokremnos (Figure 1A; Simmons, 1999). Towards the onset of the Holocene, an increased frequency and sophistication of maritime networks seems to have occurred, since larger amounts of obsidian artefacts come from a cave and several open-air Mesolithic sites situated on islands of the Aegean Sea, e.g. Youra (Northern Sporades; Figure 1B, 21), Kythnos (Cyclades; Figure 1B, 16) and Ikaria (SE Aegean Sea; Figure 1B, 14; Laskaris et al., 2011; Sampson et al., 2012 and 2010; Sampson 2010). Obsidian, together with non-obsidian artefacts from open-air sites on several other islands, have been attributed to the Mesolithic (Table 2); however, absolute dates are pending. Additionally, the presence of bluefin tuna bones from the Upper Mesolithic layers of Franchthi Cave, and other aquatic faunal resources, as well as fishing equipment, in Mesolithic settlements of the Aegean suggest a high degree of familiarization with marine environments and sophisticated fishing techniques (Mylona, 2014).

Table 2: Association of cultural remains (pre-LGM= before the Last Glacial Maximun, ca. 20,000BP, LUP= Late Upper Palaeolithic, Meso = Mesolithic) and insularity for the islands of the Aegean with indirect evidence for sea-crossings [×✓ = archaeological remains on insular sites, × = archaeological remains on non-insular sites, ×? = archaeological remains on sites whose insularity is not confi med].
Based on the sea level reconstructions provided by Lykousis (2009); Kapsimalis et al. (2009) and Ferentinos et al. (2012).

7With regard to the types of craft used, the only available evidence from the broader Aegean region post-dates the Mesolithic. Th majority of examples are indirect evidence in the form of clay models. Models of dugouts have been found at several sites on the Balkan Peninsula and dated to the Middle and Late Neolithic. A clay model from Tsangli (Thessaly, Greece; Figure 1B, 23), dated to the Middle Neolithic, appears to represent a seagoing vessel (Marangou, 1990). According to Marangou (2001) such a vessel, due to its hydrodynamic prow, would have been capable of transporting large quantities of obsidian and heavy items such as marble or andesite millstones. This particular example exhibits different compartments, as does a decorated clay artefact from the Middle Neolithic layers of Knossos, Crete (Figure 1B, 8; Marangou, 2001). Clay models of double logboats, i.e. vessels produced by the joining together of two logboats of similar size, from Maliq (Albania; Figure 1B, 30), date from the Late Neolithic onwards. Several clay models belonging to the Vinča Culture (ca. 5600-4500 cal. BC) come in a variety of types, including symmetrical or asymmetrical, with ellipsoidal or quadrangular transversal and longitudinal sections; and some comprise fitted transoms (Marangou, 2001). A Late Neolithic incised ceramic bowl, with a series of dots around the rim, from Dikili Tash (Thrace, Greece; Figure 1B, 25) and a clay model fragment from Otzaki Magoula (Thessaly, Greece;; Figure 1B, 24), both with zoomorphic head reliefs, may represent simple hide vessels (Marangou, 1991; Theocharis, 1973); as may the incised sherd from Grabak Cave on the Dalmatian island of Hvar (Croatia; Figure 1A), which seems to depict a rounded and relatively short marine vessel (Novak, 1955; Radmilli, 1963). Finally, the large number of rock carvings of boats, together with boat models, testify to the strong maritime character of Final Neolithic Strofilas at the Cycladic island of Andros (Figure 1B, 15; Televantou, 2008).

The ethnographic record

8A rich ethnographic record provides information on the sea-worthiness of several boat types and their diachronic use worldwide. Simple hide crafts such as curraghs and coracles, made of sewn animal hides covering a simple wooden or bone frame, were used in NW Europe until recently; while the more specialised kayaks and umiaqs were used by the Eskimos (Greenhill and Morrison, 1995; McGrail, 1987). Reeds are still being used for the production of simple rafts, especially in SE Asia. Specifically in the Aegean, a type of simple reed bundle boat, known as papyrella, was used for fishing and lobster trapping until the 20th century in the shallow waters around the island of Kerkyra, Ionian Sea (Figure 1B, 3; Sordinas, 1969). Dugouts similar to the Neolithic example from Dispilio were until recently in use at the lakes of Kastoria (Marangou, 2001). It has been argued that due to the low gunwale, the simple logboat lacks stability and is difficult to control, attributes which meant that it was not safe for use in the open sea. However, the addition of external fittings increased the stability of the original vessel (McGrail, 1987). Propelled by paddles or poles, extended dugouts were used until the 20th century in inland waters of Albania but were also used in the North Aegean Sea during the 16th century (Marangou, 1991). Thus, in contrast to the simple dugouts, which may have been more appropriate for inland waters (McGrail, 2010), paired logboats have also proven to be seaworthy (Figure 2).

Figure 2: Schematic representation of particular sea-going vessel types. A: hide and basket boats, B: raft, C: papyrella-type reed-bundle boat, D: dugout, logboat, monoxylon, E: double logboat.

Experimental voyages

9Based on the ethnographic use of papyrella, in 1988 Tzalas and collaborators tested the potential use of reed- bundle boats to navigate the Aegean between Milos and Franchthi Cave (Tzalas, 1995). Their experimental voyage proved that a combination of land (Argolid to Attica) and sea (Lavrion to Milos) voyages was faster and safer than a direct crossing from the coasts of Argolid to Milos via a circumnavigation of the Saronic Gulf (Figure 3). Heavy winds of 7-8 on the Beaufort Scale were challenging and delayed their arrival at Milos; however, the reed craft proved to be adequate for mild weather conditions and shallow anchorages such as those usually found on the Cycladic islands.

10Although all of the archaeological boat remains from the central and eastern Mediterranean pre-dating the Bronze Age were found in inland waters, this does not preclude their use at sea. The seaworthiness of the long and relatively narrow boats, such as the logboat, was tested by two expeditions in 1995 and 1998 (Tichý, 1999 and 2002). Two oak dugouts (Monoxylon I and II) were constructed based on the archaeological find at La Marmota, in Central West Italy (Fugazzola Delpino and Mineo, 1995). The first dugout travelled the Aegean Sea between Samos, Ikaria, Mykonos, Tinos, Andros and Evia (Figure 3), covering a distance of 300 km. The second travelled approximately 800 km along the coasts of the Central and Western Mediterranean. The vessel was constructed using both stone and metal tools; however, the team estimated that the construction of such a boat solely using stone tools would require about 300 hours. Both trips were made at an average speed of 4 km per hour and proved that the wind was the main constraint. Sea currents mainly affected the Cycladic voyage, whereas the Central Mediterranean voyage was little affected (with the exception of the crossing from Sicily to SW mainland Italy). The team managed to paddle the boat through up to 2-m-high waves and in winds of 9 on the Beaufort Scale. A cargo (consisting of obsidian, wheat, water and the crew) of more than one tonne in weight was regarded as both feasible and safe. The second trip followed parts of the western Mediterranean coasts (including the southern Atlantic facade of Portugal), without having to cross major straits or archipelagos. The maximum distance covered was 290 km along the coast between San Remo (Italy) and Portiragnes (France; Figure 3).

Figure 3: The routes of the experimental trips in the Western and Eastern Mediterranean.

11Regarding the usefulness of the information gained from these ethnoarchaeological experiments, caution is needed both in terms of the accuracy of the reconstructions (Cherry and Leppard, 2015) and because of the sea level fluctuations and the subsequent alterations in the palaeolandscapes, which had a different effect on the coastlines of the Central and Eastern Mediterranean (Lambeck, 1996; Lambeck and Purcell, 2005; Lykousis, 2009; Shackleton, Van Andel and Runnels, 1984). In other words, although the current sea-corridors between the Cycladic islands have not changed signifiantly since the Mesolithic (Kapsimalis et al., 2009), the seascape was completely different prior to the LGM (Lykousis, 2009). Thus, while general inferences regarding the seaworthiness of the particular types of vessels may be made, it is impossible to simulate the exact routes.

12Another experimental trip was recently made between Kythera Island and NW Crete (Figure 3). The crew paddled a raft made of locally sourced reeds (Arundo donax) with the aid of a sail (First Mariners, 2014). The team mainly used metal tools to construct the craft and the paddles, and commercial sisal was used to bind the reeds. They needed three days to reach Chania and proved that a trip with such a vessel was possible under calm weather conditions. Although the present consensus is that the use of a sail post-dates the Neolithic, it is important to note that the distance covered in 2014 was longer than that which would have been confronted during the Pleistocene, when looking out from the now-submerged southern parts of the Peloponnese towards the NW coasts of Crete (Lykousis, 2009); it is certain that land at both ends would have been visible at that time. Overcoming the local climatic conditions would have been a significant challenge for the inexperienced, yet on the other hand climate may have been one of the factors triggering chance dispersals to Crete.


Prehistoric boat construction: social and technical prerequisites

13Any object able to float can potentially serve as a flotation aid (McGrail, 2010; Ruxton and Wilkinson, 2012) and hence becomes an artefact. Simple floats, as well as rafts, can in some cases also carry objects and animals. Yet the use of such vessels is dependent on water temperature and wind conditions (McGrail, 2010), and thus they may have been used for short voyages and only under favourable weather conditions. Given the availability of suitable trees, hollowed-out tree trunks, i.e. simple dugouts, were probably an innovation developed independently in different parts of the world, as were a number of other innovations such as the production of stone tools, the controlled use of fire and the utilisation of natural materials as “ready-made” artefacts.

14Apart from dugouts, hide and basket boats made of a light frame covered by hides or fabrics provide superior control and safety than the simple raft. Both ethnographic (Greenhill and Morrison, 1995) and archaeological (Carter, 2006) examples suggest that organic residues such as bitumen or birch bark tar, and in some cases also clay, were used in order to render the vessel waterproof. Since the Upper Palaeolithic, sewing implements such as bone needles were being used extensively, while spun, dyed, and knotted flax fibres date back to 30 000 BP (Kvavadze et al., 2009). Bitumen and tar were used as adhesives for the production of composite tools since the Middle Palaeolithic (Wragg-Sykes, 2015), whereas stone tools suitable for wood- and hide- working were produced by the genus Homo since the earliest of times. Strictly from a technical perspective, the construction of a dugout could be achieved by any hominin species occupying the NE Mediterranean during the Late Pleistocene, i.e. Homo sapiens, Homo neanderthalensis or Homo heidelbergensis. Yet unfortunately only via proxies are we able to hypothesise how likely it may have been for any of these groups to produce such an idea, have the motivation, and make the effort to attempt a sea-crossing. On social grounds, the controlled use of fire and the subsequent communal gatherings would allow for the transmission of knowledge and interchange of new ideas. High-risk activities, such as hunting large mammals or evading predators and natural hazards, require sufficient communication skills and a certain degree of vigilance to avoid detrimental outcomes – characteristics potentially useful for undertaking sea-crossings. They may not, however, have been sufficient for the strategic organisation and successful implementation of a sea voyage. Per contra, if the initial crossings are perceived as ad hoc responses to natural hazards or external threats, then any kind of expedient vessel may be interpreted as an artefact, a tool used as part of a subsistence strategy. With this in mind it is possible to envisage that species other than our own witnessed some of the earliest serendipitous sea-crossings, which would then gradually have given way to more organised voyages.

Initial (pre-LGM) sea-crossings

15With regard to the arguments for the pre-LGM human presence on islands of the Aegean, it is important to recognise that there is a significant difference between colonisation, conducted by a large number of individuals with permanent occupations on islands; and sea-crossings, which perhaps resulted in “ephemeral colonisation” (Leppard, 2014), conducted by a few individuals or small groups, with non-permanent occupations or rapid abandonments. In such cases the evidence to be found by archaeologists would be very rare, or even non-existent. Impromptu sea-crossings, even due to natural hazards, are unlikely to allow a population to become established in an insular landscape. Consequently, the material culture would be restricted to just a few artefacts, which either travelled along with their artisans or were produced and subsequently discarded shortly after their arrival in the new territory. Thus, the scarcity of the available evidence may either be explained as a result of a flawed interpretation of the archaeological record or as a result of the limited and serendipitous nature of the crossings, premeditated or not, which indeed occurred during the Pleistocene. The fact that almost all of the lithic arte- facts come from the surface, reduces the potential for a future profound appraisal of the available assemblages. On the other hand, overlooking the evidence, no matter how controversial, is certainly not the way forward. The aim should now be to seek evidence within a secure stratigraphic context. Accepting that at least some of the lithics are correctly attributed to before the LGM (Papoulia, 2016), it would be interesting to consider the feasibility of such trips.

16A raft with – or more probably without – a sail could be one way of reaching Crete from the southern ends of the Peloponnese, since the distance between Kythera, Antikythera and Crete was markedly less than today (< 30 km for a non-stop trip; Figure 4C). The same route is suggested by the dispersal patterns of large mammals to Crete, such as elephants and rhinos, which are able to cross small aquatic distances. Such a trip would certainly be contingent upon weather conditions, which might often be lethal for the inexperienced mariner. On the other hand, a direct trip from Africa to Crete (Mortensen, 2009; Strasser et al., 2010) appears unlikely, since a non-stop trip three times greater than that from the Peloponnese would be required (< 300 km); moreover, there would be no visibility of the opposite coasts and thus no land to aim for (Schüle, 1993; Papoulia, 2014). Further to the north, a much smaller distance would need to be crossed at the Kafireas Strait (ca. 6 km; Figure 4D) in order to reach the Cycladic mega-island from Evia during MIS 2.

Figure 4: The Aegean during the Pleistocene low sea-level stands. Potential sea routes are annotated in red.
Modifi after Lykousis, 2009 and Tourloukis, 2010.

17Such a crossing is thought to have been rather challenging in terms of sea currents (Kapsimalis et al., 2009). In contrast, coastal or inshore waters would have been the perfect settings for marine adaptations. Such an “inter-visible” and “inter-accessible” set of islands was certainly found at the Cyclades from the LGM onwards (Kapsimalis et al., 2009), and may also have occurred diachronically in the southern Ionian Sea (Figure 4D). There, an enclosed aquatic milieu exists today, formed by the arrangement of the large islands of Lefkas, Kefalonia, Zakynthos, and the smaller isles and islets of the Inner Ionian Sea Archipelago (IISA). The available palaeogeographic reconstructions indicate that while the majority of these smaller islets were connected to the mainland, Kefalonia and Zakynthos were insular and several “stepping stone” islets emerged during low sea-level stands (Ferentinos et al., 2012; Zavitsanou et al., 2015). Therefore, based both on the available archaeological (Papoulia, 2016) for a critical review of the lithic finds) and geological evidence (Ferentinos et al., 2012; Zavitsanou et al., 2015), arguments for small-scale (< 7 km) seacrossings in the region during the Late Pleistocene seem credible. However, refined palaeogeographic reconstructions and chronological determinations are needed in order to define whether the proposed pre-MIS 2 assemblages from islands such as Milos, Naxos (Figure 4D) and Alonnissos (Figure 1B, 19) are indications of marine crossings, terrestrial crossings, or both (Papoulia, 2016).

Post-LGM marine voyages

18The post-LGM Aegean was significantly altered in its coastal palaeogeography (Kapsimalis et al., 2009). Many of the previously non-insular islands begin to approximate their present-day forms (Figure 5). Thus although boat remains are absent, the material culture found on these islands may provide proof of marine crossings. The obsidian artefacts, in particular, constitute unambiguous evidence of marine crossings from ca. 14000 BP onwards. With regard to the nature of the craft used, although archaeological remains of hide boats are rare, their potential use during the Upper Palaeolithic has been suggested (Marangou, 2001). A possible hypothesis is that such vessels were used for small-scale sea-crossings between the Cycladic islands, since they can be easily controlled, especially in small anchorages. Other types may include simple log rafts and basket boats and, while the use of the simple logboat is generally thought to have been restricted to the more protected inland waters (McGrail, 2010), the Monoxylon trips proved that sea voyages with extended logboats were also technically possible. According to the experimental papyrella voyage, a journey between Milos and Attica was necessary in order to reach Franchthi Cave in the Argolid. A preference for this particular route may be indicated by the absence of finds on the islets of Falconera and Parapola (Sampson, 2006). Schisto Cave, situated in SW Attica, could be approached via the same route, as could the Mesolithic sites on Kythnos since the island is situated on the route between Milos and Attica.

Figure 5: The Cyclades since the LGM. Potential sea routes are annotated in red. Modified after Kapsimalis et al., 2009.

19A larger radius needs to be covered for the transport of obsidian from Milos to Youra in the Northern Sporades, as well as to Crete and Gavdos in the Southern Aegean. Especially in the first case, a combination of land and sea voyages seems probable. Obsidian both from Milos and Yali has been found at the island of Ikaria. The Yali raw material sources are a distinctive type of obsidian easily recognisable due to its macroscopically-visible white spherulites. Due to its poor suitability for knapping, the exchange network of Yali obsidian is mainly restricted to the neighbouring areas, with the earliest use at Mesolithic Kerame (Sampson et al., 2012) – although subsequently it was preferred for the production and trade of stone vessels such as the ones found at Knossos, Crete (Figure 1B 8; Georgiades, 2008).

20While the vessels used for the initial sea- crossings would need to carry the minimum number of people and perhaps some personal objects, at least since 9000 cal. BP boats travelling to islands such as Cyprus and Crete had to be large and stable enough in order to also carry small mammals. It has been argued that the transportation of ruminants in particular would have influenced both the type of vessel and the distance and the speed of the journey, since due to the “downer cow syndrome” the animals would be unable to remain motionless for more that 3-4 hours and yet their movement had to be controlled in order to prevent overturning the boat (Vigne, 2013). Subsequently, the transition from small anchorages to proper harbours (Broodbank, 2000) would also have been related to the types of vessels used. Hence the maritime character of the Mesolithic, which was primarily focused on the acquisition and consumption of marine resources, was gradually transformed into a sedentary lifestyle sustained by sophisticated marine networks, not only in order to procure and trade exotic materials (lithics, pottery and metals) but also for permanent colonization.


21It is increasingly appreciated that the prehistory of marine crossings is much older than was previously thought. However, based on the available geoarchaeological record, it is clear that while the maritime character of Neolithic societies can be studied in a considerable amount of detail, and corroborated by a small amount of direct and much indirect evidence for seafaring activities such as fishing, trading, or colonization, there is still a high degree of uncertainty regarding the oldest sea-crossings. Although the technological capacity for constructing simple vessels, such as dugouts, hide vessels and reed-bundle rafts, was present at least since the Middle Palaeolithic, the types of vessels used, the frequency of the crossings and the exact routes are among the main questions regarding the Mesolithic and Late Upper Palaeolithic voyagers. As for the pre-LGM sea-crossings, a number of questions remain to be answered by future geoarchaeological investigations, since the controversy regarding the interpretation of the available indirect evidence lies at the heart of the issue regarding the intentionality of the crossings, if not the feasibility of the crossings themselves. It is highly likely that events such as sea-crossings made by a very small number of individuals have left only minor if any archaeological evidence at all. Thus in order to resolve this issue, meticulous investigation at sites where stratigraphic sequences permit absolute dating of the archaeological context is required. Furthermore, it is possible that evidence for the occupation of the Pleistocene Aegean coasts remains hidden at depths well below current sea level (Sakellariou and Galanidou, 2015; Papoulia, 2013; Tourloukis, 2010). Thus the submerged palaeolandscapes between the Aegean islands and the Eurasian mainland can no longer be neglected, since these were either the coastal parts of the present-day islands or the connecting terrestrial bridges between the landmasses. Therefore, it is clear that future surveys seeking to answer questions regarding the earliest attempts of hominins to confront and navigate the sea need to be pursued in the sea.

22Evidently, the choice of particular terrestrial and marine migration routes, subsistence strategies, exploitation of marine resources, settlement patterns and the origins of seafaring, would all have been shaped by the morphology of the palaeolandscapes. Therefore, detailed palaeogeographic reconstructions are the foundation for this research, since while for the Mesolithic onwards we may have doubts about the exact bathymetry of the anchorages, for the Pleistocene a key issue is the insularity of a number of present-day islands. A combination of an unequivocal temporal context for the archaeological finds, corroborated by detailed regional sea level reconstructions, may be able to test the Pleistocene seaward or terrestrial dispersal hypothesis. Finally, it must be stressed that discovery of direct evidence of the earliest Pleistocene sea-crossings in the Aegean is highly unlikely. This can be better appreciated when we consider the evidence from the Holocene record. Even though seafaring formed an integral part of both Mesolithic and Neolithic lifestyles, only two archaeological vessels have been recovered from the entire Mediterranean region and both date to the Neolithic. The development of a marine lifestyle should not be interpreted as an instant “revolution” but rather as a result of gradual and progressive familiarisation with the sea and development of the required skills and techniques. By accepting that proxies are the only way to understand the diverse mechanisms involved in this long-term process, a thorough study of the few pieces of indirect, yet unequivocal, evidence for the initial crossings will permit a conclusive interpretation of this controversial yet potentially highly significant record.


23Dr Margarita Nazou and an anonymous reviewer are acknowledged for reading and commenting on a previous draft. The editor, Dr Matthieu Ghilardi, is warmly thanked for his useful comments and support during the editing process and for organizing a most enjoyable conference in Corsica.



AMMERMAN A.J., «Introduction», in Ammerman A. J., Davis T.W., Island Archaeology and the Origins of Seafaring in the Eastern Mediterranean. Eurasian Prehistory, 10 (1-2), 2013, p.9-30.

AMMERMAN A.J., «Setting our sights on the distant horizon», in Ammerman A.J., Davis T.W., Island Archaeology and the Origins of Seafaring in the Eastern Mediterranean. Eurasian Prehistory, 11, 2014, p.203-236.

ANDERSON, A., BARRETT, J.H., BOYLE, K. V. (eds.), The Global Origins and Development of Seafaring, Cambridge, McDonald Institute for Archaeological Research, 2010.


Pleistocene road signs of human dispersals across Eurasia»,

Quaternary International, 285, 2013, p.30-43.

BROODBANK C., An Island Archaeology of the Early Cyclades, Cambridge, Cambridge University Press, 2000.

BROODBANK C., GALANIDOU N., LEPPARD T.P., RABETT R.J., RUNNELS C., PHOCA-COSMETATOU N., «Discussion and Debate», Journal of Mediterranean Archaeology, 27, 2014, p.255-278.

CARTER R., «Boat remains and maritime traide in the Persia Gulf during the sixth and fifth millenia BC», Antiquity, 80, 307, 2006, p.52-63.


Archaeological Project: new data on the Middle Palaeolithic and Mesolithic Cyclades», Antiquity Project Gallery 341, 2014, URL:

CHELIDONIO G., «Manufatti Litici su Ciottolo da Milos (Isole Cicladi)», Pegaso, Rivista Annuale Di Cultura Mediterranea, 1, 2001, p.117-148.

CHERRY J., LEPPARD T., «Experimental archaeology and the earliest seagoing: the limitations of inference», World Archaeology, 47(5), 2015, p.740-755.

CHOURMOUZIADIS G., Dispilio (Kastoria). A prehistoric lake settlement, Thessaloniki, Kodikas, 1996, (in Greek).

CUBUK G. A., «Altpaläolitische Funde von den Mittelmeerterrassen bei Nea Skala auf Kephallinia», Archäologisches Korrespondenzblatt, 6, 1976, p.175-181.

DERRICOURT R., «Getting “Out of Africa”: Sea Crossings, Land Crossings and Culture in the Hominin Migrations», Journal of World Prehistory, 19, 2006 , p.119-132.

EFSTRATIOU N., Agios Petros. A Neolithic site in Northern Sporades. Archaeopress, Oxford, British Archaeological Reports, International Series 245, 1985.


E., «A late Palaeolithic Site at Ouriakos (Island of Limnos, Greece) in the north-eastern Aegean Sea», Antiquity Project Gallery 335. URL: efstratiou335/

FARR H., «Seafaring as social action», Journal of Maritime Archaeology, 1(1), 2006, p.85-99.

FIRST MARINERS. A Million Years of Seafaring: Scientific Re-enactment of Man’s First Encounters with the Sea, 2014. URL:


PAPATHEODOROU G., «Early seafaring activity in the southern Ionian Islands, Mediterranean Sea», Journal of Archaeological Science, 39, 2012, p.2167-2176.


C., «Coastal and Marine Palaeo-Environments and Human Dispersal Points Across the Africa-Eurasia Boundary», in Brebbia C.A, Gambin T., The Maritime and Underwater Heritage, Southampton, Wessex Institute of Technology Press, 2003, p.61-74.

FUGAZZOLA DELPINO M.A., MINEO M., «La piroga Neolitica del Lago Bracciano, La Marmota 1», Bullettino di Paletnologia Italiano, 86, 1995, p.197-266.

GALANIDOU N., «Mesolithic Cave Use in Greece and the Mosaic of Human Communities», Journal of Mediterranean Archaeology, 24(2), 2011, p.219-242.

GALANIDOU N., COLE J., ILIOPOULOS G., McNABB J., «East meets West: the Middle Pleistocene site of Rodafnidia on Lesvos, Greece», Antiquity Project Gallery 336. URL:

GEORGIADES M., «The obsidian in the Aegean beyond Melos: An outlook from Yali», Oxford Journal of Archaeology, 27, 2008, p.101-117.

GREENHILL B., MORRISON J., The archaeology of boats and ships. An introduction, London, Conway Maritime Press, 1995.

KACZANOWSKA M., KOZLOWSKI J., «The Aegean Mesolithic: material culture, chronology, networks of contacts», in Ammermann A.J., Davis, T.W., Island Archaeology and the Origins of Seafaring in the Eastern Mediterranean. Eurasian Prehistory, 11, 2014, p.31-62.

KAPSIMALIS V. , PAVLOPOULOS K., PANAGIOTOPOULOS I., DRAKOPOULOS P., VANDARAKIS D., SAKELLARIOU D., ANAGNOSTOU C., «Geoarchaeological challenges in the Cyclades continental shelf (Aegean Sea)», Zeitschrift für Geomorphologie, 53, 2009, p.169-190.

KAVVADIAS G., Palaeolithic Kefalonia: The Fiskardo Culture, Athens, Fitrakis, 1984, (in Greek).

KOPAKA K., MATZANAS C., «Early Sea Travels to the Aegean and Crete? Thoughts based upon the chipped stone industries from the island of Gavdos», Acts of the 10th International Cretological Conference A1 (Chania 2006), Chania, 2011, p.43-82, (in Greek).

KOPAKA K., MATZANAS C., «Palaeolithic industries from the island of Gavdos, near neighbour to Crete in Greece», Antiquity Project Gallery, 321, 2009. URL: http://antiquity/

KOURTESSI-PHILIPPAKIS G., «The Lower and Middle Palaeolithic in the Ionian Islands», in Bailey, G.N., Adam, E., Panagopoulou, E., Perles, C., Zachos, K., The Palaeolithic Archaeology of Greece and Adjacent Areas. Proceedings of the ICOPAG Conference, Ioannina, September 1994. London, British School at Athens, 1999, p.282-287.

KVAVADZE E., BAR-YOSEF O., BELFER-COHEN A., BOARETTO E., JAKELI N., MATSKEVICH Z., MESHVELIANI T., «30,000-Year-Old Wild Flax Fibers», Science 325, 2009, p.1359.

LAMBECK K., «Sea-level change and shore-line evolution in Aegean Greece since Upper Palaeolithic time», Antiquity, 70, 1996, p.588-611.

LAMBECK K., PURCELL A., «Sea-level change in the Mediterranean Sea since the LGM: model predictions for tectonically stable areas», Quaternary Science Reviews, 24, 2005, p.1969-1988.

LASKARIS N., SAMPSON A., MAVRIDIS F., LIRITZIS I., «Late Pleistocene/Early Holocene seafaring in the Aegean: New obsidian hydration dates with the SIMS-SS method», Journal of Archaeological Science, 38, 2011, p.2475-2479.

LEPPARD T.P., «Passive dispersal versus strategic dispersal in island colonisation by hominins» Current Anthropology, 56(4), 2015, p.590-595.

LEPPARD T.P., «Modeling the Impacts of Mediterranean Island Colonization by Archaic Hominins: The Likelihood of an Insular Lower Palaeolithic», Journal of Mediterranean Archaeology, 27, 2014, p.231-253.

LYKOUSSIS V., «Sea-level changes and shelf break prograding sequences during the last 400ka in the Aegean margins: Subsidence rates and palaeogeographic implications», Continental Shelf Research, 29, 2009, p.2037-2044.

MARANGOU C., «Neolithic Watercraft in Greece: Circustantial Evidence and Serious Guesses», in Beltrame C. (Ed.), Boats, Ships and Shipyards. Proceedings of the Ninth International Symposium on Boat and Ship Archaeology Venice 2000. Oxbow Books, Oxford, 2003, p.14-18.

MARANGOU, C., «Neolithic Craft: Evidence about boat types and uses», in Bassiakos Y., Aloupi E., Facorellis Y., Archaeometry Issues in Greek Prehistory and Antiquity, Athens, Hellenic Society of Archaeometry, 2001, p.737-752.

MARANGOU C., «Maquettes d’embarcations: les débuts», in Laffineur, R., Thalassa. L’Égée Préhistorique et La Mer. Actes de La 3e Rencontre Internationale de l’Université de Liège, Calvi (Corse) 1990, Aegeum 7, Liège, Peeters, 1991, p.21-41.

McGRAIL S., «The Global Origins of Seagoing Water Transport», in Anderson, A., Barrett, J.H., Boyle, K. V., The Global Origins and Development of Seafaring. Cambridge, McDonald Institute for Archaeological Research, 2010, p.95-107.

McGRAIL S., Ancient boats in North Western Europe. The archaeology of water transport to AD 1500, London, New York, Longman, 1987.

MORTENSEN P., «Lower to Middle Palaeolithic Artefacts from Loutro on the South Coast of Crete», Antiquity Project Gallery, 317, 2008, URL: ProjGall/mortensen/index.html

MUSSI M., Earliest Italy: An Overview of the Italian Paleolithic and Mesolithic, New York, Kluwer / Plenum Publishers, 2001.

MYLONA D., «Aquatic animal resources in Prehistoric Aegean, Greece», Journal of Biological Research-Thessaloniki 21, 2, 2014.

NOVAK G., Prethistorijski Hvar. Grapceva Spilja, Zagreb, Jugoslavenska Akademija Znanosti i Umjetnosti, 1955.

PAPOULIA C., «Confronting the Sea: Navigation Skills in pre-Modern Human Societies», in Touchais, G., Laffineur, R., Rougement, F., Physis: L’environnement Naturelet La Relation Homme-Milieu Dans Le Monde Égéen Protohistorique. Actes de La 14e Recontre Égéenne Internationale, Paris, Institut National d’Histoire de l’Art (INHA), 11-14 Décembre 2012, Aegeum 37, Leuven - Liege, Peeters, 2014, p.521-523.

PAPOULIA C., «Below sea-level: Combining Palaeolithic and Underwater Archaeologyinthe Eastern Mediterranean Sea», in Breen C., Forsythe W. ACUA Underwater Archaeology Proceedings 2013, Advisory Council for Underwater Archaeology 2013, p.11-17.

PAPOULIA C., «Seaward dispersals to the NE Mediterranean Islands in the Pleistocene. The lithic evidence in retros- pect», Quaternary International, 2016. doi:10.1016/j. quaint.2016.02.019

RADMILLI A.M., La Preistoria d’Italia alla Luce delle Ultime Scoperte, Florence, Instituto Geografico Militare, 1963.

RANDSBORG K., Kephallenia: Archaeology & History. Acta Archaeologica 73, 2002.

ROBERTS P., PETRAGLIA M.D., «Pleistocene rainforests: barriers or attractive environments for early human foragers?», World Archaeology, 47(5), 2015, p.718-739.

ROCKMAN M., STEELE J. (eds.), Colonization of Unfamiliar Landscapes. The Archaeology of Adaptations, London, Routledge, 2003.

RUNNELS C., McCOY F., BAUSLAUGH R., MURRAY P., «Palaeolithic research at Mochlos, Crete: new evidence for Pleistocene maritime activity in the Aegean», Antiquity Project Gallery, 2014, URL: runnels342

RUXTON G.D., WILKINSON D.M., «Population trajecto- ries for accidental versus planned colonisation of islands», Journal of Human Evolution, 63, 2012, p.507-511.

SAKELLARIOU D., GALANIDOU N., «Pleistocene submerged landscapes and Palaeolithic archaeology in the tectonically active Aegean region», Geological Society, London, Special Publications, 411, 2016, p.145-178.

SAMPSON A., Μesolithic Greece. Palaeoenvironment, Palaeoeconomy, Technology, Athens, Ion, 2010, (in Greek).

SAMPSON A., The Prehistory of the Aegean: Palaeolithic, Mesolithic, Neolithic, Athens, Atrapos, 2006, (in Greek).

SAMPSON A., KACZANOWSKA M., KOZLOWSKI J.K., «Mesolithic occupations and environments on the Island of Ikaria, Aegean, Greece», Folia Quaternaria, 80, 2012, p.5-40.

SAMPSON A., KACZANOWSKA M., KOZLOWSKI J.K., The Prehistory of the Island of Kythnos (Cyclades, Greece) and the Mesolithic settlement at Maroulas, Krakow, PAU, 2010.

SCHÜLE W., «Mammals, vegetation and the initial human settlement of the Mediterranean islands: a palaeoecological approach», Journal of Biogeography, 20, 1993, p.399-411.

SERANGELI J., BOLUS M., «Out of Europe - The dispersal of a successful European hominin form», Quartär, 55, 2008, p.83-98.

SHACKLETON J.C., Van ANDEL, T.H., RUNNELS C.N., «Coastal Paleogeography of the Central and Western Mediterranean during the Last 125,000 Years and Its Archaeological Implications», Journal of Field Archaeology, 11(3), 1984, p.307-314.

SIMMONS A.H., Faunal Extinction in an Island Society. Pygmy Hippopotamus Hunters of Cyprus, New York, Kluwer / Plenum Publishers, 1999.

SORDINAS A., «Investigations of the prehistory of Corfu during 1964-1965», Balkan Studies, 10, 1969, p.393-424.

STRASSER T.F., PANAGOPOULOU E., RUNNELS C.N., MURRAY P.M., THOMPSON N., KARKANAS P., McCOY F.W., WEGMANN K.W., «Stone Age Seafaring in the Mediterranean: Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Crete», Hesperia, 79, 2010, p.145-190.

STRASSER T.F., RUNNELS C., WEGMANN K., PANAGOPOULOU E., McCOY F., DiGREGORIO C., KARKANAS P., THOMPSON N., «Dating Palaeolithic sites in southwestern Crete, Greece», Journal of Quaternary Science, 26, 2011, p.553-560.

TELEVANTOU C., «Strofilas: A Neolithic Settlement on Andros», in Brodie N., Doole J., Gavalas G., Renfrew C., Horizon. A Colloquium on the Prehistory of the Cyclades, Cambridge, McDonald Institute for Archaeological Research, 2008, p.43-53.

THEOCHARIS D., Neolithic Greece, Athens, 1973. (in Greek)

TICHY R., «Monoxylon Expeditions 1995 and 1998. Facts about the oldest Sea Navigation», Experimentelle Archäologie in Europa 1, 2002, p.189-197.

TICHY R., Monoxylon II: Plavba po 8000 Letech. Dobrodružství Experimentální Archeologie, Náchod, JB Production, 1999.

TOURLOUKIS V., The Lower and Middle Pleistocene Archaeological Record of Greece, Leiden, Leiden University Press, 2010.

TZALAS H., «On the Obsidian Trail with a Papyrus Craft in the Cyclades», Tropis III. Third International Symposium on Ship Construction in Antiquity: Proceedings. Hellenic Institute for the Preservation of Nautical Tradition in Athens, Athens, 1995, p.441-468.

VIGNE J.-D., «The origins of mammals on the Mediterranean islands as an indicator of early voyaging», in Ammermann A.J, Davis T.W., Island Archaeology and the Origins of Seafaring in the Eastern Mediterranean. Eurasian Prehistory, 10, 2013, p.45-56.

WRAGG-SYKES R., «To see a world in a hafted tool: Birch pitch composite technology, cognition and memory in Neanderthals», in Coward, F., Hosfield, R., Pope, M., Wenban-Smith, F., Settlement, Society and Cognition in Human Evolution. Landscapes in Mind, New York, Cambridge University Press, 2015, p.117-137.

WURSTER C.M., BIRD M.I., «Barriers and bridges: early human dispersals in equatorial SE Asia», Geological Society, London, Special Publications 411, 2014, URL:

ZAVITSANOU A., SAKELLARIOU D., ROUSAKIS G., GEORGIOU P., GALANIDOU N., «Paleogeographic reconstruction of the Inner Ionian Sea during Late Pleistocene low sea level stands: Preliminary results», Proceedings of the 11th Panhellenic Symposium on Oceanography and Fisheries, Mytilene, Greece, 2015, p.997-1000.

Table des illustrations

Légende Table 1: Types of evidence (indirect and direct) for sea-crossings in the Aegean for each cultural period (pre-LGM = before the Last Glacial Maximum, ca. 20,000BP, LUP = Late Upper Palaeolithic, Meso = Mesolithic, Neo = Neolithic).
Fichier image/jpeg, 88k
Légende Figure 1: Location map of the archaeological sites and islands mentioned in the text and tables.Figure 1A: Sites with direct (La Marmota - Italy) and indirect (Grabak Cave - Croatia, Akrotiri Aetokremnos - Cyprus) evidence for aquatic navigation throughout the Central and Eastern Mediterranean Basin.Figure 1B: Archaeological sites and islands in the Aegean. Yellow circle stands for Dispilio, the only site with direct evidence of water transport and red circles stand for the only obsidian raw material sources in the Aegean. (1: Dispilio, 2: Diaplo islet, 3: Kerkyra sites, 4: Kefalonia sites, 5: Zakynthos sites, 6: Gavdos sites, 7: Plakias sites- Crete, 8: Knossos - Crete, 9: Mochlos - Crete, 10: Milos 11: Antiparos, 12: Naxos sites, 13: Yali, 14: Kerame - Ikaria, 15: Strofi - Andros, 16: Maroulas - Kythnos, 17: Franchthi Cave, Argolid, 18: Schisto Cave - Attica, 19: Alonnissos sites, 20: Kyra Panayia, 21: Cave of Cyclops - Youra, 22: Ai Stratis, 23: Tsangli, 24: Otzaki Magoula, 25: Dikili Tash, 26: Ouriakos - Lemnos, 27: Rodafnidia - Lesvos, 28: Skyros, 29: Chalki, 30: Maliq - Albania).
Fichier image/jpeg, 528k
Légende Table 2: Association of cultural remains (pre-LGM= before the Last Glacial Maximun, ca. 20,000BP, LUP= Late Upper Palaeolithic, Meso = Mesolithic) and insularity for the islands of the Aegean with indirect evidence for sea-crossings [×✓ = archaeological remains on insular sites, × = archaeological remains on non-insular sites, ×? = archaeological remains on sites whose insularity is not confi med].Based on the sea level reconstructions provided by Lykousis (2009); Kapsimalis et al. (2009) and Ferentinos et al. (2012).
Fichier image/jpeg, 219k
Légende Figure 2: Schematic representation of particular sea-going vessel types. A: hide and basket boats, B: raft, C: papyrella-type reed-bundle boat, D: dugout, logboat, monoxylon, E: double logboat.
Fichier image/jpeg, 126k
Légende Figure 3: The routes of the experimental trips in the Western and Eastern Mediterranean.
Fichier image/jpeg, 216k
Légende Figure 4: The Aegean during the Pleistocene low sea-level stands. Potential sea routes are annotated in red.Modifi after Lykousis, 2009 and Tourloukis, 2010.
Fichier image/jpeg, 390k
Légende Figure 5: The Cyclades since the LGM. Potential sea routes are annotated in red. Modified after Kapsimalis et al., 2009.
Fichier image/jpeg, 544k


University of Crete, Department of History and Archaeology, Gallos University Campus, 74100 Rethymnon, Crete, Greece Email

© CNRS Éditions, 2016

Conditions d’utilisation :