Version classiqueVersion mobile

Multidisciplinary approaches to food and foodways in the medieval Eastern Mediterranean

Sylvie Yona Waksman

Trading goods, trading tastes

The transportation of amphorae, tableware and foodstuffs in the Middle and Late Byzantine period

The evidence from Aegean shipwrecks

George Koutsouflakis


Collective data show a high increase in the number of Late Roman/Early Byzantine shipwrecks in the Aegean, but a dramatic decline in the rest of the Mediterranean. A second even greater increase is reported for the 12th-13th centuries AD. In absolute numbers, Byzantine shipwrecks rival in the Aegean those of the Hellenistic and Roman periods.
The intention of this paper is to examine Byzantine shipwrecks with cargoes of amphorae, focusing on their chronological and spatial distribution and their contents, as well as to put a spotlight on the consignments of glazed tableware favored by markets from the middle of the 12th century onwards.

Texte intégral


  • 1 In writing this essay I benefited greatly from the contribution of several colleagues, to whom I o (...)
  • 2 Van Doorninck (2002, p. 899) reported the existence of roughly 80 known Byzantine wrecks throughou (...)
  • 3 Out of a total of 1,259 shipwrecks, Parker’s catalogue contains 95 shipwrecks dated to after 600 A (...)
  • 4 A more recent statistical analysis, still unpublished, reduces this percentage to the still genero (...)

1“Byzantium, mistress of the sea”:1 Fr. van Doorninck, when he thus entitled his chapter on Byzantine seafaring in one of the early synthesizing works concerning underwater archaeology, well captured the spirit of his subject (van Doorninck 1972, p. 134). The title itself, in its very essence, implies a very close relation of the empire with the many seas surrounding its vital territories and farther away. Almost half a century later, it is true that in discussion of Byzantine maritime trade in the Aegean it is difficult to avoid understatement. The nature of underwater evidence and the widespread presence of Byzantine ceramics scattered over the coastlines demonstrate the intensity of maritime trade, counterbalancing to a certain extent the absence of solid, published data. In an ever-growing catalogue of shipwrecks in which the exact number of “known” Byzantine wrecks remains elusive,2 statistics clearly show an unprecedented increase in the number of shipwrecks in relation to what appears to have occurred in the rest of the Mediterranean. “Medieval” shipwrecks dated to between 600 and 1500 AD occupied only 7.5% of the total amount of Mediterranean shipwrecks referred to by Parker.3 For the same time period, the number of amphorae-laden Byzantine wrecks in the Aegean rises to 26%, according to some of the latest statistics published on the chronological range of wrecks in Hellenic territorial waters, more than the number of wrecks of the Hellenistic and Roman periods (Micha 2005-2006, pp. 84, 88‑90).4

  • 5 The term “time capsule”, adopted in order to express this disconnected reasoning, has become so ol (...)

2As noted elsewhere in the past, shipwrecks provide evidence for archaeological studies that other terrestrial or underwater sites seldom provide (Parker 1992, pp. 2‑4; Adams 2001; Gibbins & Adams 2001, pp. 279‑280). The wide range and quality of intact artifacts is one element of research on shipwrecks, which also offer the opportunity to trace the connections between trade, trade routes, commercial tactics, shipping and ship management. But if we are to comment on the value of a shipwreck in a few words, its most recognized quality is its existence per se in space and time. Ships are self-contained and self-organized entities, containing a contemporary assemblage of artifacts lost accidentally while in transit and deposited in a more or less neutral environment.5 This transitional context, in which goods can be studied in their stacked positions, not affected by secondary human action, can provide unexpected evidence for commercial strategies and vivid insight into seaborne trade.

3However, this theoretically clear context in which we seek to define the contributions of wrecks to Byzantine studies can be affected by a number of disadvantages. Wrecks may vary tremendously in their states of preservation, and underwater investigation may not be evenly applied to them. Thus the evidence provided by Byzantine wrecks is not uniform in scale and value. Only a handful of Byzantine wrecks have been fully excavated, studied and published. A far larger group consists of wrecks that have been surveyed and recorded, resulting in publications that vary from a few sentences in annual reports to fully documented archaeological studies. The information gained from such publications differs greatly and basically lacks the kind of internal evidence that is provided by excavated wrecks. Another fairly large category of wrecks comprises unsurveyed wrecks of unknown or hypothetical location, whose existence is indicated only by material handed on to archaeologists by fishermen or amateur divers, or confiscated by the port police. Since the nature of these sites remains unknown and unverified, they should not be listed as shipwrecks, however much their material may have been studied. Byzantine studies are inundated with artifacts retrieved from unknown underwater sites. Some of these almost certainly come from wrecks. But for others the evidence is more ambiguous, as they have been taken over borders illegally and are now found in collections, some of which are in well-respected institutions.

  • 6 The term “Aegean” is used here in a somewhat formulaic way without strict, absolute geographical l (...)

4This study focuses on the shipwrecks of the Middle and Late Byzantine periods in the Aegean6 and especially on wrecks which mark the resurgence of Byzantine trading ventures, dating roughly from the 10th up to the 13th century. I consider that the wrecks of the earlier period, especially the large number of wrecks of between the 4th and the 7th century AD, are related too strongly to the ancient world to be considered Byzantine or simply medieval and should be treated in a separate study.

  • 7 It should be noted that the four basic amphora types proposed by Günsenin are considered here as m (...)

5Byzantine wrecks undergo the same deterioration processes as wrecks of any other period; the remains correspond mainly to the remains of their cargoes, and pottery demonstrates a hardy resistance to underwater disintegration. Organic materials may be preserved in anoxic sediments, but do not leave discernible traces or survive when exposed on the sea floor. Consequently, the evidence gained from Byzantine wrecks reveals only certain aspects of maritime trade: the trade in ceramics, the trade in building materials and mainly the trade in commodities stored and carried in ceramic containers (amphorae or pithoi). Accordingly, the study of cargoes may easily be relegated to a study of amphorae, especially when the bulk of the evidence originates from poorly recorded sites. Amphorae are indeed by far the most common items retrieved from Byzantine wrecks, and appear also in large numbers as isolated, jettisoned material. For the purpose of this study and for the sake of simplicity, the basic Günsenin typology will be followed here, as it is the most accepted (Günsenin 1989).7

  • 8 My present position, as director of the Department of Underwater Archaeological Sites, Monuments a (...)

6The catalogue attached at the end of this study, to which discoveries of recent years have been added, lists the existing evidence from the Aegean, according to published information and the personal experience of the author.8 It contains basic information concerning geographical position, depth, circumstances of discovery (when available), a concise description of the site and the remains and a selected bibliography. The dates given are those provided by the researchers/authors.

Chronological distribution

  • 9 The evolution of many different subtypes is an almost standard phenomenon in most LR2 amphorae car (...)

7While the very Late Roman period, from the 5th century up to the first half of the 7th century, appears to have been an extremely busy period in the Aegean, almost exclusively dominated by cargoes of (Benghazi) LR1 and LR2 amphorae, this evidence itself contains the seeds of a forthcoming decline. Ships that navigated in this transitional period between the ancient and the medieval worlds were smaller in size and indicative of attention to the development of faster shipbuilding techniques. Economy of time, wood and labor took precedence over the durability, longevity and tonnage of vessels (Steffy 1982, pp. 84‑85; 1994, pp. 83‑85; van Doorninck 2002, pp. 900‑901). At the same time, the sense of a controlled and well-travelled sea under Roman rule was gradually lost. The loss of this sense of security would not have encouraged investment in large cargoes. Apart from a few middle-sized vessels, the bulk of Late Roman/Early Byzantine cargoes discovered in the Aegean belong to smaller vessels, measuring less than 15 meters and often containing cargoes of reused amphorae belonging to several different subtypes and owners (van Doorninck 1989, pp. 248‑253; 2015b, pp. 206‑208; van Alfen 2015, pp. 18‑30).9 The three tumultuous centuries that followed present a serious gap in the evidence, unlike any other period in the post-Archaic Aegean. Apart from a well-documented wreck found at Bozburun on the southwest coast of Asia Minor (s/w 51), almost nothing else has come to light in the Aegean.

8A resurgence of commercial activities appears towards the 10th century, strongly indicated by several shipwrecks carrying homogenous cargoes of Günsenin I amphorae, some of them quite large. No doubt mixed consignments containing reused amphorae, glazed bowls, jugs, pitchers and recycled glass, such as the 11th century Serçe Limanı wreck (s/w 52), would not have been uncommon. The intensity of maritime commercial activities escalates during the 12th and the 13th centuries. Since all the wrecks of this period remain unexcavated, they cannot provide a more precise dating than the one offered by the circulation time-range of the amphorae on board. In cases of more precisely dated material, such as ceramic glazed tableware, the evidence converges mainly in the second half of the 12th century. This period, before the seizure of the capital city during the 4th Crusade and the establishment of the new Frankish administration, may represent the peak of Byzantine trading ventures. The term “Byzantine” is here used in a more chronological and geographical sense than a cultural one.

9The chronological distribution of the wrecks dovetails well with historical sources and archaeological data. The 12th century is considered a period of significant growth in the Byzantine economy with rising population levels, considerable increase in the size of urban settlements and extensive tracts of new agricultural land being brought into production (Hendy 1985, pp. 85‑90, 106‑107; Harvey 1989, chap. 4; Magdalino 1993, pp. 140‑142; Laiou & Morrisson 2007, pp. 92‑93, 101‑105, 130‑133; Brandejs 2013, pp. 16‑18).

Spatial distribution

10General charts demonstrating the distribution of known wrecks in wide areas which have not yet been subject to an integrated survey policy do not necessarily reflect historical reality. The only thing they do reflect is a present state of knowledge. They should not be read as incontestable evidence for trade patterns or commercial routes, since the sea has not been equally treated by archaeological research; a handful of exhaustively investigated areas cannot be considered to be representative of the vast extent of archaeologically uncharted seabed in the Aegean. Presence cannot equalize absence, and statistic analysis, even when based on solid numbers, can be very misleading.

11The existing evidence, in regard to spatial distribution, indicates that Byzantine shipwrecks are not scattered equally. While some areas that occupy relatively small tracts of sea can be crowded with large numbers of wrecks, others appear to be empty, although they may have played an important role in Byzantine commercial trade and navigation (fig. 1‑2). These inconsistencies are the outcome of two decisive factors: survey policy and environmental conditions. The large number of wrecks located in areas such as the Pagasetic Gulf, the northern Sporades, and the Sea of Marmara are the result of focused underwater surveys. Survey policy tends to focus on areas where the yields are expected to be higher according to studies of historical and geopolitical data, or on evaluation of existing evidence and information. Shipwrecks engender surveys and surveys produce shipwrecks; the understanding of this interrelationship is crucial for interpreting maps presenting the distribution of wrecks. The second factor consists of the peculiarities of the physical environment. The northern Aegean, and especially the coasts of Thessaly above Mount Pelion, Macedonia and Thrace, although active in the Byzantine period, have failed significantly to provide remains of shipwrecks, in spite of efforts to search in these areas. The many rivers that annually receive heavy precipitation from the southern Balkans and flow into the sea along the coastline of northern mainland Greece create a very dynamic and heavily silted littoral environment that is aggressive to the remains of wrecks. Traditional means of survey are much less effective if the underwater environment does not facilitate the recognition of shipwreck remains.

Fig. 1 – Distribution of cargoes with Günsenin I‑IV amphora types in the Aegean.

Fig. 1 – Distribution of cargoes with Günsenin I‑IV amphora types in the Aegean.

Fig. 2 – Distribution of cargoes of glazed tableware in the Aegean, 12th-14th century.

Fig. 2 – Distribution of cargoes of glazed tableware in the Aegean, 12th-14th century.

12Considering all these limitations, a first reading of the distribution maps shows that the northern half of the Aegean is considerably richer in Byzantine shipwrecks than the southern half (fig. 1). The low numbers of identified wrecks along the coastlines of Macedonia and Thrace are counterbalanced by the considerable number of amphorae brought in every year to the EUA, the local port police stations and local museums by owners of deep-trawling fishing boats (Micha 2005-2006, p. 90; Bakirtzis 2003, p. 77). The fact should not be ignored that a large number of Günsenin I and III type amphorae are steadily brought to the Department, trawled from an area that extends from Lemnos up to Kavala and the Chalkidiki peninsula. This evidence, although not evaluated in any great detail, is highly indicative for the existence of a large number of wrecks in more open and less silted waters.

13Wrecks scattered around the channel of Euboea, the Pagasetic Gulf and the northern Sporades provide the most solid evidence for Byzantine seafaring and are inevitably linked with centers of major administrative and military importance such as Thebes and its port Chalcis (Kontogiannis 2012; Koilakou 2013), Corinth (Sanders 2003, pp. 395‑397), Athens to a lesser extent (Kazanaki-Lappa 2002), and the two capitals of the empire, Thessaloniki and Constantinople. The abundance of wrecks in these waters clearly implies itineraries that were regularly followed and close trade relations. Accordingly, workshops that produced amphorae to supply containers for the maritime transportation of agricultural products or prestigious ceramic tableware should be sought within the limits of this area. Beyond the seas that lie between these centers, the evidence is more scarce. However, long-distance trade is also indicated by a concentration of wrecks along the Lycian coast of southern Asia Minor. The evidence for cargoes transported along the coast of Asia Minor and to the Dodecanese islands is clearly present for the 12th and early 13th centuries, when control of the hinterland had been irrevocably lost, reflecting a struggle to maintain trading contacts with the traditionally lucrative markets of Cyprus and the Syro-Palestinian coast.

The nature of the cargoes

14For the purpose of this study, a basic separation of cargoes into homogenous and heterogenous (mixed) has been adopted. This separation is made according to the visible physical remains of wrecks on the sea floor and does not deal with the possible existence of perishable commodities that left no traceable remains behind.

  • 10 The same amphorae are known also under the names Bakirtzis class 2 (Bakirtzis 2003, pp. 76‑77, pl. (...)
  • 11 On the standardization of Late Roman 1 and 2 and Günsenin I type amphorae see van Alfen 1996; 2015 (...)
  • 12 As stated above, the evidence comes mainly from amphorae regularly found by deep trawlers over a p (...)

15The 9th and 10th centuries were dominated by the existence of seemingly homogenous cargoes of Günsenin type I amphorae which seemed to enter the market in large numbers, indicating a first boom in maritime trade after a long pause (for Günsenin type I see Günsenin 1989, pp. 269‑271; 1990, I, pp. 21‑24; 2018, pp. 92‑96).10 Distribution patterns present a high concentration of shipwrecks in the area of the Marmara islands – Propontis (Byzantine Ganos), with which the largest part if not the entire production of these amphorae should be linked (fig. 1). In accordance with that which occurred at the beginning of the very Late Roman period, amphora shapes and volumes appear to adhere to certain volumetric systems11 and cargoes may contain a high number of units having the same technical characteristics. Reused amphorae are present to a lesser extent, and are mostly seen in mixed consignments such as in the Serçe Limanı wreck. Günsenin I amphorae are the forerunners of a resurgence in the trade of liquids, which was to increase in the following three centuries. The concentration of wrecks in the Sea of Marmara (s/w 42‑49) could well reflect consignments heading to Constantinople. Another large part of the production was probably intended for transport to the Aegean. The presence of wrecks near the Chalkidiki peninsula (s/w 1‑2), in the area around Lemnos (unknown number),12 Kythera (s/w 29), Crete (s/w 38) and along the coasts of Asia Minor and the eastern Aegean islands (s/w 13, 35, 41, 50, 52‑55) correspond to trade networks that reached far. As mentioned, the evidence does not illuminate a more dense chronology. The Serçe Limanı wreck (s/w 52) provides a fixed date, ca 1025, but this may not be the most representative wreck of the period. The journey of this ship began on the Fatimid Syrian coast and continued northward in the Aegean with a mixed cargo of ceramic wares, glassware and glass-cullet; the 89 Günsenin I type amphorae were clearly an assortment of worn and reused vessels, scratched with initials signifying ownership and apparently heading in the wrong direction. They would thus be less associated with their original production sites (van Doorninck 1989, pp. 253‑256).

  • 13 Also known under the names Bakirtzis class 4 (Bakirtzis 2003, pp. 78‑80, pl. 19), Saraçhane type 6 (...)

16When Günsenin type I, newly-produced or reused, was still predominant for the transportation of wine, type II entered the market, and both types enjoyed a long period of concurrent circulation. Günsenin type II (Günsenin 1989, p. 271; 1990, pp. 26‑28; 2018, pp. 98‑100)13 was an advanced shape for the time, more user-friendly, and over a period of about 2 centuries spawned several variations. However, the shipwreck evidence indicates that it never achieved the commercial success of its predecessor although its production was long-standing. Two type II cargoes are known from the Gulf of Euboea (s/w 22) and the island of Pelagonisos (s/w 3, fig. 3), the second also containing variations of type IV. Furthermore, type II appears less commonly among jettisoned material scattered along the coast of mainland Greece. Again, accurate numeric data are lacking.

Fig. 3 – The Pelagonisos shipwreck: view of the amphorae cargo (s/w 4).

Fig. 3 – The Pelagonisos shipwreck: view of the amphorae cargo (s/w 4).
  • 14 Also known as Bakirtzis class 5 (Bakirtzis 2003, p. 80, pl. 20), Saraçhane type 61 (Hayes 1992, p. (...)

17Günsenin III type amphorae – with their many variations in shape – represent the late Middle and Late Byzantine maritime trade at its highest volume (Günsenin 1989, pp. 271‑274; 1990, pp. 28‑30; 2018, pp. 100‑102).14 Type III is by far the most common Byzantine amphora in the western Aegean, present in large numbers also as isolated find. The coasts of mainland Greece, Euboea and the northern Sporades are interspersed with such evidence. Cargoes of type III are as a general rule homogenous, without the inclusion of other amphora types on board. They shared the market with type II for almost a century, before completely dominating and overshadowing any other ceramic container. Distribution patterns show an immensely increased presence along the coasts of the middle and the northern Greek mainland (s/w 15‑18, 20, 24, 28), and the northern Sporades (s/w 4‑7, 9, 11[?]): thirteen out of the fifteen recorded wrecks are to be found in these locations, while the remaining two are scattered in the Dodecanese (s/w 34) and the southern Peloponnese (s/w 30).

18Three of the most interesting of these cargoes, preserved in excellent condition at depths inaccessible for regular divers, clearly demonstrate the volume that this trade could reach (fig. 4‑6). The three wrecks (s/w 5, 6 and 7) lie only some tens of meters apart and carry large numbers of the same Günsenin type III amphorae. All the examples studied have the very same shape, dimensions and technical characteristics, making it highly possible that they originated from the same production site, which is likely to be Chalcis according to recent archaeometric research (Waksman et al. 2018a). Two of the wrecks (s/w 5, 6) are apparently representative of the middle-sized vessels that sailed the Aegean and present visible traces of the cargo, around 15 m in length. The third wreck (s/w 7) obviously belongs to a larger and wider type of vessel and presents an 8-pattern figure on the sea floor (fig. 6). In all three wrecks the amphorae are piled in at least three different layers (fig. 7), and shipwreck 7 strongly indicated the existence of a fourth one. There is no reliable estimation of the size of the cargo. The largest may have well have carried a total number of more than 3,000 units, the two smaller wrecks between 1,000 and 1,500 units.

Fig. 4 – Photomosaic of the Günsenin III cargo in the Galyfas Bay, Peristera – Alonessos (s/w 5).

Fig. 4 – Photomosaic of the Günsenin III cargo in the Galyfas Bay, Peristera – Alonessos (s/w 5).

Fig. 5 – Photomosaic of the Günsenin III cargo in the Galyfas Bay, Peristera – Alonessos (s/w 6).

Fig. 5 – Photomosaic of the Günsenin III cargo in the Galyfas Bay, Peristera – Alonessos (s/w 6).

Fig. 6 – Photomosaic of the Günsenin III cargo in the Galyfas Bay, Peristera – Alonessos (s/w 7).

Fig. 6 – Photomosaic of the Günsenin III cargo in the Galyfas Bay, Peristera – Alonessos (s/w 7).

Fig. 7 – Layers of Günsenin III amphorae in the cargo of shipwreck 5, Peristera – Alonessos.

Fig. 7 – Layers of Günsenin III amphorae in the cargo of shipwreck 5, Peristera – Alonessos.

19The presence of three wrecks with identical cargoes in the same area, at the same depth and only a few meters apart strongly suggests a simultaneous accident resulting in a total loss of ships and cargoes. It is highly unlikely that they represent individual occurrences on different occasions over a longer period of time. These three vessels were probably sailing together following the same itinerary and were lost simultaneously due to the same cause. The exact cause is not easily understood. All three wrecks have retained their coherence surprisingly well and were located at some distance from the shore. Whether or not this group of three ships represents an exceptional case or an ordinary situation is not yet known. A state-generated function may have operated here, reminiscent of the muda system of convoys practiced by the Venetians (Lane 1973, pp. 123‑128; Thiriet 1959, pp. 337‑338, 343‑345, 421‑423).

20The term muda was used for the Venetian galley convoys that were sponsored and regulated by the state for the time period that they sailed together with their cargoes. The muda system implied that merchant caravans or convoys ran on a set schedule along prescribed routes. These state-supervised convoys reached Constantinople following a sea lane that passed along the western Aegean coast and usually stopped at Chalcis, a centre of Venetian interest, on their way to the capital or Thessaloniki. The muda system as a practice became generalized in the 13th and 14th centuries, although it pre-existed the Latin conquest of the Empire. Byzantine sources are literally devoid of obvious references to tactical commercial maritime convoys. Certainly there is some evidence to suggest that during the 12th century the coast of mainland Greece was unsafe for commercial enterprises; however, closer evaluation of the existing evidence reveals that for most of the century the Byzantine administration maintained a firm hold on the coasts, if not the seas (Magdalino 1993, pp. 139‑140).

21An even larger cargo, from a ship measuring well over 30 meters in length, was partly excavated in 2013 inside the modern commercial harbor of Rhodes (s/w 34). The ship sank due to an extensive fire that burned all wooden structures down to the waterline and caused the collapse and complete fragmentation of the cargo, which was found in a 1‑meter-thick stratum containing only heavily fragmented Günsenin III amphorae (fig. 8). The size of the cargo suggests that massive shipments to Rhodes or even farther east took place.

22However, much of the other evidence from type III cargoes indicates significantly smaller consignments, possibly distributed to regional sites. Again, data can be qualified but not quantified in any detail. Taking into account the distribution of wrecks and isolated finds of Günsenin III amphorae, it could be said that the route from Chalcis to Thessaloniki and the Propontis is interspersed with these amphorae, which is consistent with the hypothesis of a Chalcis origin.

Fig. 8 – Fragmented cargo of Günsenin III amphorae in the hold of shipwreck 34 in the port of Rhodes.

Fig. 8 – Fragmented cargo of Günsenin III amphorae in the hold of shipwreck 34 in the port of Rhodes.
  • 15 Also known as Bakirtzis class 6 (Bakirtzis 2003, pp. 80‑81, pl. 21), Bjelajac type III (Bjelajac 1 (...)

23Günsenin IV amphorae (Günsenin 1989, p. 275; 1990, pp. 31‑33; 2018, pp. 102‑108)15 appeared on the market at some point during the course of the 13th century and are best represented by a cargo off Çamaltı Burnu, in the Marmara islands, which is the only excavated homogenous cargo of this type (s/w 44). There are some indications that the production of the type is related to the area surrounding Byzantine Ganos (Günsenin 1990, p. 50; 2001, pp. 118, 121, but see Günsenin 2018, p. 95). The presence of type IV is much more limited in the Aegean, with only one homogenous cargo reported off the island of Skopelos (s/w 8), which is understudied and unpublished. Type IV appears occasionally in limited numbers as supplementary cargo in type III consignments and vice versa.

24Epitomizing the evidence of shipwrecks, amphorae types I and III are the types that dominated the Aegean market in contiguous periods extending from the 10th to the 13th century. Their spatial distribution appears to differ (fig. 1). While type I is more common in the north – northeastern Aegean and along the coast of Asia Minor, type III appears mostly along the coasts of mainland Greece, the Sporades and the Peloponnese. This probably reflects the geographical locations of two different workshops, but could also be an indication of the loss of control of Byzantine territories in Asia and the shift of the economic interests of the empire to mainland Greece during the Komnenian dynasty.

Glazed tableware

  • 16 Glazed tableware appears in most Byzantine shipwrecks of the 12th and 13th centuries; however, the (...)

25Consignments of glazed tableware, mainly deep bowls, shallow bowls and plates, appear as supplementary cargoes in amphora wrecks around the middle of the 12th century and enable a more detailed chronology of these wrecks than do amphorae alone.16 Such shipwrecks offer a unique opportunity to study chronological overlap in detail and to evaluate the synchronicity of circulation of different decorative styles.

  • 17 Only a few details of the excavation and the layout of the cargo are published. Kritzas (1971) ref (...)
  • 18 For a synopsis of the decorative styles of the period see Sanders 2003, pp. 388‑389, and Papanikol (...)

26Aside from shipwrecks containing tableware of other cultures, as is the case in the Serçe Limanı 11th century wreck, the earliest example of such a mixed cargo is the Pelagonisos wreck in the northern Sporades (s/w 4). It is the earliest wreck that provides a consignment of fine incised sgraffito ware, present in large numbers of shallow bowls and dishes (Kritzas 1971; Throckmorton 1971; Ioannidi-Dostoglou 1989; Dellaporta 1999a, pp. 122‑142). The site was never properly excavated, and much ink has been wasted in describing and analyzing styles that appear to correspond to a date in the middle of the 12th century. However, the main cargo of this wreck in terms of bulk, contrary to what has been repeatedly stated, consisted of Günsenin type II‑III amphorae and millstones (fig. 3) (Papanikola-Bakirtzi 2012, p. 200).17 It is the first example of a wreck to provide evidence that high-quality tableware was successful in the market. The ceramic vessels raised from the Pelagonisos wreck are more or less homogeneous in fabric, technical characteristics, form and shape and should be attributed to a single workshop (Papanikola-Bakirtzi 2012, p. 200; Megaw & Jones 1983, pp. 251‑252): large shallow dishes with vertical walls, as well as less common deep hemispherical bowls, decorated with a fine sgraffito and occasionally with painted decoration (Painted Fine Sgraffito Ware).18 The craftsmanship demonstrates particular care and a preponderance of detailed work. The intention of imitating metal prototypes is clear in the shapes, firing, decoration and styles, probably indicative of customers and consumers who wished to achieve higher social standards.

27A chance find from the same area of the northern Aegean indicates the existence of a second wreck in deep water. A small number of shallow and deep plates, vases and an amphora were brought to the authorities by a fisherman in 2002, apparently obtained from a wreck site that was swept by dragged nets, in the area of the underwater bank known as Glafki, somewhere between Lemnos and the northern Sporades (s/w 11, Tsanana & Amprazogoula 2015). The details of this recovery are still unknown, but the raised examples show a striking resemblance to the material raised from the Pelagonisos, indicating an origin from the same workshop and a similar date, somewhere in the middle of the 12th century. The amphora, a typical Günsenin III example, if it was lifted during the same fishing operation and recovered from the same wreck, should indicate a mixed cargo, containing both amphorae and tableware in the way of the Pelagonisos wreck.

28A substantial change in trading strategies appears towards the second half of the 12th century, with the presence of several cargoes containing exclusively glazed tableware. It appears that glazed pottery artefacts were no longer just objects to fill the gaps between the amphorae stowed in the hold, they had become a main cargo. Accordingly, the transport of these items would not have necessarily followed the itineraries of liquids and semi-liquids. This change would appear to indicate that these products became much more valued and appreciated as articles of trade and that commercial enterprises based on transport by sea could have come to be based just on the profit from table wares.

29One of the earliest examples of such a cargo comes from a wreck on the east coast of Attica, close to Thorikos, found accidentally during construction work in the shallows (s/w 23, fig. 9) (Dimitriadou 2002, p. 577; Koutsouflakis 2013, pp. 255‑261). The site was never properly excavated and remains unpublished. Not a single fragment of amphora is reported and the samples of tableware collected and brought to the EUA should be considered to be indicative of the nature of the cargo. As in the case of the Pelagonisos wreck, the Thorikos shipwreck contained mainly shallow plates with vertical walls presenting a more developed fine sgraffito style, less detailed but still close to the examples of the Pelagonisos island decorative repertoire. However about 50% of that very same consignment consisted of deep bowls with pronounced ring bases, a shape that was to dominate the market for the next half-century to the detriment of the more shallow vessels. These large bowls are painted with white slip motifs, presenting patterns of spirals, triangles and rhombs that continued well into the second half of the century. The detailed incisions seen on the pottery of the Pelagonisos wreck gave way to more hasty execution, while slip-painted wares, both shallow plates and large deep bowls, came to be successful as articles of trade, perhaps as a reaction to a need for mass production in order to satisfy the demand of a rapidly expanding market. If we accept a line of development in the tableware produced by Aegean workshops, the Thorikos wreck should be dated somewhere in the third quarter of the 12th century.

Fig. 9 – Exposed glazed tableware from the Thorikos shipwreck (s/w 23).

Fig. 9 – Exposed glazed tableware from the Thorikos shipwreck (s/w 23).
  • 19 Another “plate wreck” of the same period, still unpublished, is located in the area of Cape Gelido (...)

30A recently located and excavated wreck, in Adrasan Bay (s/w 56), on the southwest coast of Anatolia, is another example of a homogenous cargo of tableware, dated to about the same period (Özdaş & Öniz 2014; Doğer & Özdaş 2016).19 The cargo consists of beige-slipped, monochrome-glazed, open vessels, mainly plates and bowls decorated with fine sgraffito medallions, palmettes and a stylized floral spiraling branch motif. Some of the samples resemble those of the Pelagonisos wreck, others however present a style that is more advanced chronologically. The cargo also contained some examples of brown and green-brown painted bowls, seen also in the Thorikos wreck.

  • 20 It is tempting, for example, to attribute several of the bowls of underwater origin now in possess (...)

31A further step in the evolution of styles may be observed in the Kavalliani shipwreck (s/w 25) (Koutsouflakis et al. 2012, pp. 57‑58; Koutsouflakis & Argyri 2012, p. 881; Koutsouflakis 2013, pp. 85‑104; Koutsouflakis & Tsompanidis 2018; Waksman et al. 2018b) which follows chronologically and is named after the desolate island at the southern entrance of the Gulf of Euboea, where it was discovered in 2008. The Kavalliani wreck is another typical Byzantine plate shipwreck, where the archaeologist came too late, obviously in the wake of looters. The remains of the cargo were raised quickly before any further acts of looting, but a number of ceramic vessels taken long ago from the site are to be found in anonymous private collections or even in collections of well-respected institutions.20 The pottery from this wreck presents a series of subsequent changes in decoration and shapes. Shallow bowls and plates are almost absent and the dominant shape is that of large, base-ringed deep bowls, conical, hemispherical or footed, sometimes with an intense carination under the lips and a diameter measuring between 23 and 27 cm. A second and third series of bowls present the same shapes with lesser dimensions, but they form only a small proportion of the total cargo.

32The decoration of fine sgraffito ware had changed to one style which employed thicker engravings and made extensive use of a new cut-slip technique, a sort of “negative champlevé technique”, in which large areas of slip were removed to create a silhouette effect. The styles of decoration can differ to a great extent. A large number of samples are decorated with free-style birds (fig. 10), fishes or animals at the centre of the vessel interior surrounded by wavy lines or stylized spikes depicted in a rotating manner (Koutsouflakis & Tsompanidis 2018, pp. 41‑44). These new motifs share very little, if anything, with metal vessels, which had been the prototypes for much of the previous production. Other decoration presents spears or curvilinear motifs in a radial composition and central medallions framed by jagged zones in the periphery, occasionally interrupted by smaller medallions. Some unique samples, only partly preserved, carry a unique depiction of a galley – extremely rare for this period –, a fully equipped marching foot-soldier and a fierce lion-head surrounded by sword-like jewels. There is an unusual use of spears and banners as supplementary decorative motives not seen elsewhere. Thin incision is used only to render certain minor details. The decorative impression is achieved by the contrast of the deep brown incisions and cut-slips on the pale slipped surfaces. The lines are skillful, succinct and confident, creating a new style that is in contrast to the somewhat elaborate techniques of the previous fine sgraffito style.

Fig. 10 – Glazed tableware from the cargo of the Kavalliani shipwreck (s/w 25).

Fig. 10 – Glazed tableware from the cargo of the Kavalliani shipwreck (s/w 25).

33The detailed study of this material revealed the hands of several different craftsmen, varying in dexterity. White slip-painted decoration is still present, but it corresponds to a percentage of less than 15% of the samples raised from the sea and presents a more conservative style that is very similar to that seen in examples from the Thorikos wreck. The styles of both sgraffito and slip-painted wares strongly imply a connection with material found on land sites around the town of Thebes; the study of the remains of the wrecks clearly demonstrated that the ship was sailing out of the Gulf of Euboea at its southern end and had probably departed earlier from the town of Chalcis. This connection between the cargo and workshops in the area of Chalcis and material distributed in Thebes was shown by chemical analysis of bowls found in the shipwreck (Waksman et al. 2018b). The Kavalliani wreck should be dated to towards the end of the 12th century.

34The well-studied cargo of the Kastellorizo wreck (s/w 36) (Philotheou & Michailidou 1986; Michailidou & Philotheou 1989; Loucas 1989; Michailidou 1999; Dellaporta 2002, p. 570), by the southern coast of Asia Minor and on the seaway to Cilicia and Cyprus, shows more or less the decline of the style presented by the pottery from the Kavalliani wreck. The site itself was never properly investigated; what we know about the cargo comes mainly from material confiscated by the port police and a countless number of ceramics interspersed in museums and private collections in Greece, Turkey and elsewhere. However, nothing in the present state of knowledge implies the presence of commodities on board other than tableware.

35The Kastellorizo material is characterized by a consistent standardization in forms and sizes and a hasty rendering of decoration. The large, ring-based bowls are again predominant in the cargo, with very few examples of other shapes. Fabric and technical characteristics imply a production originating from the same workshop, or different workshops operating in the same area and using similar clay (Papanikola-Bakirtzi 2012, p. 203; Philotheou & Michailidou 1986, p. 308). Vessels decorated with incised sgraffito formed the majority of the consignment and were accompanied by a lesser number of vessels of brown and green painted style, slip-painted style and champlevé style. The latter is absent from all previous wrecks and should indicate a more advanced chronology. Free-style motifs (birds and fishes) in the centre of the interior, surrounded by rotating fronds or stylized spikes follow the style found in the Kavalliani wreck; however, the rendering demonstrates a degenerated, careless production with clear signs of mannerism. The lines are still confident, but artistic creativity, care and attention to the final result appear to have been considered less important. The success of the more accomplished decorative style as seen in the pottery from the Kavalliani wreck was not able to be maintained for long: it became extremely popular in a large part of the Eastern Mediterranean and beyond, reaching regions as far as the Syro-Palestinian coast, Italy and France. The workshop(s) would have found it difficult to meet the demands of mass production without making serious compromises in the quality of incised decoration, which was perhaps the most time-consuming stage of production. If we accept a chronological evolution in decorative styles, the Kastellorizo wreck could be dated to the beginning of the 13th century.

36The so-called “Skopelos wreck” (s/w 10) (Armstrong 1991) is only poorly represented by a collection of 30 fragmentary ceramics, raised under unknown circumstances from the sea around Skopelos (northern Sporades) and now in possession of the Ashmolean Museum in Oxford. Altogether the group comprises twenty-four glazed bowls and six amphora sherds, which leaves open the hypothesis that they probably originate from a mixed cargo, containing both amphorae and tableware. The amphorae fragments have not been typologically identified or published yet. The remaining tableware does not show consistency in its shapes and presents many variations and sizes. Unlike the Kavalliani and the Kastellorizo wrecks, shallow bowls appear again, making up a considerable part of the consignment. The decorative styles are very similar to what we have seen in the Kastellorizo material: nine samples are decorated in the same sgraffito fashion with thick incision and present the same mannerism in the style. Five of these are decorated with central silhouetted animals in a free field using the cut-slip technique more common in the Kavalliani wreck. The most coherent group of the collection consists of seven examples decorated with chubby hares in the champlevé technique. Plain slip-painted ware is missing with the exception of an odd example decorated with pairs of thick, white slip-painted lines converging from the rim to the centre and covered with green glaze. The production from the Skopelos wreck, first assigned to the end of the 12th century (Armstrong 1991, p. 335), should be very close to that of Kastellorizo and may correspond to a slightly later date, considering the high percentage of the champlevé samples.

  • 21 For the “Main Middle Byzantine Production” (MBP), see Waksman et al. 2014.

37The evidence from shipwrecks, according to our present state of knowledge, appears to indicate a steady flow in mass maritime transportation of tableware from the middle of the 12th century to the first years or decades of the 13th. Additional evidence confirms that the main volume of the Aegean workshops’ maritime enterprises was constricted to this time span. A large number of plates and bowls of underwater provenance reported in museums at Istanbul, Izmir, Bodrum, Antalya and Alanya is evidence for the possible existence of three more wrecks along the Asia Minor coast, at Izmir, Bodrum and Beşadalar, dated to this period (Doğer & Özdaş 2016, pp. 446‑448). No matter how formulaic the dates given to Byzantine “plate-wrecks” are, it remains a fact that the Zeuxippus ware that succeeded in the markets is definitely absent from any Aegean shipwreck. Scars from tripod stilts, a characteristic trait of the more advanced 13th century production, are missing from any bowl recovered from Aegean wrecks. This would appear to imply a pause in the mass maritime transportation of tableware. If this reflects a temporary dysfunction in the production of Chalcis’ MBP,21 possibly caused by factors related to the change of administration, or a subsequent change in commercial strategies, it is not well understood. The tableware of the more advanced 13th century production may have again become supplementary cargo, undetectable under consignments of amphorae and leaving no visible traces on the sea floor. Evidence from a plate-wreck located off the island of Skyros (s/w 12) and hypothetically ascribed to the 14th century remains poorly published and cannot be evaluated here (Karambinis 2015, pp. 257‑258, 267‑268).

Amphorae content

38Shipwrecks are by definition unexpected events, deposited in an environment subjected to a minimum of secondary human activity; however, the amphorae that make up their cargos can hardly be considered “undisturbed”. The vast majority of amphorae recovered from the sea lack their stoppers and are filled with solid and liquid debris, the result of post-depositional contamination. The liquid contents, probably the most common, escape the vessels without leaving any visible traces. Semi-liquids, according to their solubility in sea water, may also disappear or solidify, depending on their nature. Solids may disintegrate, leaving tiny remains, easily removable by marine action. As a result, the content of any amphora recovered from a wreck site represents mainly the environment of the surrounding sea floor. Remains of foodstuffs such as animal or fish bones, pits, seeds and shells found in amphorae may well not belong to their original content. Amphorae deposited underwater tend to become colonized by shells, cephallopods, fish and crustaceans, which leave their own remains or collect inside objects.

39During the early decades of systematic archaeological research on shipwrecks, defining the content of amphorae was not considered an important issue. Contents were assumed to be either self-evident from the shape of the vessel itself, or were somewhat arbitrarily determined from excerpts of literary sources or pictorial art. The notion that amphorae were a) specifically designed according to criteria of density, weight, physical and/or chemical properties and domestic use of the commodity they carried, and b) thus recognized in markets as typical containers of that particular commodity, dominated amphorae studies from the beginning; this view is still maintained, and fervently supported by a number of researchers (Opaiţ 2007, pp. 101‑122; 2010, pp. 154‑155).

40A more systematic approach was adopted from the late 1970s onwards, when the study of amphorae was extended to include the description and analysis of residues to determine content, coating or preservatives. The introduction of better sampling methods by G. Bass (1982, pp. 163‑165) resulted in a series of new data, related mostly to the presence of botanical remains. The betterment of sampling methods and the introduction of multidisciplinary studies applied to this new field created a new skepticism towards long-standing common fallacies and raised new questions. Were amphorae containers for one commercial use? Were they primarily designed for a specific commodity? Were they traded as single empty containers?

  • 22 The examples from the Aegean are numerous, although mostly understudied or unpublished. For some p (...)

41Indeed, it is largely thanks to wrecks that we acquired the first certainties on the content of some forms of amphorae. Grape seeds, for example, are among the botanical remains, most commonly reported inside intact amphorae.22 However, the information they provide is not necessarily unequivocal. Grape seeds have been found together with olive pits and pine seed hulls in the same amphorae from the Yassi Ada 7th century AD shipwreck (Bryant & Robert 1982, pp. 328‑329; van Doorninck 1989, pp. 252‑253). Although grape seeds outnumbered the other botanical remains and were considered to be evidence for the transportation of a kind of cheap wine whose impurities had not been carefully removed, the question of an earlier use of these vessels remained open. The same amphorae might have carried olive oil, olives or olives preserved in wine during an earlier trip, remnants of which were the olive pits adhering to the coating of the amphora (van Doorninck 2005, p. 96). Moreover, grape seeds present in the coating pitch may be remnants of an earlier use that does not necessarily correspond with the final use of the vessels (Ward 2015, p. 60). Clearly in many instances, botanical remains fail to provide definite answers to long-standing questions such as “wine or oil?”.

  • 23 Pigments and iron nails are reported to have been found inside Late Roman amphorae in the vaulted (...)

42Were Byzantine amphorae advertising their contents and place of origin? While this might have been the case for the most part in the Greek and Roman periods, Byzantine merchants were actually far more practical in their use of food containers. The evidence from the Late Roman period suggests that after their first use amphorae could lose their commercial affiliations. An undefined number of them may have been returned or sent to production centers for refilling, while others were discarded or channeled to urban and agricultural sites for secondary uses. And of course, a large number of them would have remained present in markets, workshops, ships, ports and households, becoming multifunctional, and in some cases used for purposes other than carrying or storing foodstuffs.23

  • 24 The evidence of vitis species seeds from the excavation of the Serçe Limanı wreck, in light of the (...)

43The Ganos amphorae production, concerning mainly the Günsenin I type, has been convincingly linked to the production of wine, promoted by a thriving monastic community from the 10th century onward. Workshops and kilns have been located along the northern shores of the Sea of Marmara and on Marmara Island (Günsenin 2009, pp. 147‑148; 1995, pp. 165‑178; 1989, p. 253) while the presence of a number of wrecks in the area appears to verify that this was a main area of production, although perhaps not the only one. Botanical remains that would corroborate this link between type I and wine are still lacking,24 but a metrological approach would appear to confirm this argument (van Doorninck 2015a).

  • 25 DNA analysis was carried out by Maria Hansson, Lund University, Sweden, GC/MS analysis by R. Nelso (...)

44The production sites for Günsenin type II are unknown except for Chalcis, which was one among several workshops manufacturing this type (Waksman et al. 2018a). The distribution of the type shows a dense concentration of material around the Black Sea (Samsun, Sinop, Constantinople, Pliska, Dinogetia, Mangalia, Capidava, Aegyssus, Chersonesos, Sarkel, Dniepr, Taman, Kiev, Sudak, Ivan) and in lesser numbers in the Aegean – Izmir, Thessaloniki and Athens (Todorova 2011, p. 135). It has been suggested that another main production center may have been an important commercial hub in the empire such as Trebizond, but so far there is no archaeological evidence for workshops to support this point of view. The liquid primarily contained in type II is also a subject of speculation. The contemporaneous circulation with other wine amphorae, the absence of visible resinous residues within the vessels and the much more careful design around the neck and rim may suggest a commodity more valuable than wine, perhaps olive-oil, that could be widely distributed to places of consumption where there was little agricultural production. An intact sample raised from the Katergon wreck at Pelagonisos island (s/w 3), when cleaned and placed in a desalination tank released an rancid, oily substance indicative of an oily content (fig. 11). DNA analysis of samples taken from several points on the interior of that amphora did not provide a clear answer: traces of grape and olive DNA were recognized in almost equal amounts, while GC/MS chemical analysis indicated that the content “could have been” olive oil.25 Contaminated interiors, evidence of reuse or indication of a product containing both grapes and olives – the question remains open.

Fig. 11 – Oily substance pouring from the mouth of a Günsenin IIb amphora from the Katergon shipwreck at Pelagonisos – Kyra Panagia (s/w 3).

Fig. 11 – Oily substance pouring from the mouth of a Günsenin IIb amphora from the Katergon shipwreck at Pelagonisos – Kyra Panagia (s/w 3).

45Günsenin III proved to be a much more popular and durable shape. It presents characteristics never before seen on amphorae: heavy wear, overhead handles that protect the neck and the rim and make carrying easier (but do not facilitate pouring the content) and thick walls of well-fired coarse clay, which is rare in most Byzantine amphorae. These characteristics appear to indicate a deliberate intention to produce durable vessels able to stand up to continuous use. The question of which agricultural production primarily employed these vessels remains obscure. Solidified resin was observed on almost every broken base raised from the mid‑12th century wreck in the port of Rhodes (s/w 34). This could be considered as strong evidence for wine, but it should be kept in mind that resinous coatings are not necessarily linked only to wine content. Dozens of other type III complete amphorae examined in the storerooms of the Hellenic Ephorate of Underwater Antiquities did not present similar visible remains. Residue analysis of examples found at Thebes and Chalcis demonstrates the use of these amphorae for the transport of fermented substances, probably (red) wine or its derivatives, and possibly plant oils, which are sometimes detected in the same containers (Pecci et al., in this volume).

46The distribution patterns of Günsenin type III imply that these were amphorae produced in the Aegean area. Chemical analysis of the clay showed the existence of workshops in Chalcis, whose output was found in central Greece, as well as in the 13th century Novy Svet shipwreck in the Black Sea (Waksman et al. 2018a; Morozova et al. and Kontogiannis et al., in this volume). The abundance of variations, sizes and volumes could suggest other areas of production as well. The possibility that these amphorae were massively produced to be exported as empty containers to other agricultural sites that lacked local productions should be also considered.

47The Günsenin IV type was probably initially intended for the transport of wine. Grape seeds were collected from the interiors of intact type IV amphorae lined with pine pitch, as those from the wreck of Çamaltı Burnu (Günsenin 2005, p. 120). The fact that type IV shares strong similarities of shape with type I, along with the existence of an intermediate and perhaps transitional type I/IV, may indicate that it was produced in the same areas. Evidence for type IV from shipwrecks is relatively rare in the Aegean, whereas distribution patterns on land sites show a greater density on sites around the Black Sea (Todorova 2011, pp. 136‑137; Günsenin 1989, pp. 274‑276, fig. 12‑14). Variants of the type are dated as early as the 11th century; the type reached its peak in the 13th century and continued to be produced in the 14th and the early 15th century, with geographical distribution influenced by the irrevocable loss of crucial territories of the empire during the late centuries. Residue analysis of examples of Günsenin IV found in the 13th century Novy Svet shipwreck revealed fermented substances, probably wine and its derivatives, and possibly plant oils as in the case of Günsenin III amphorae (Pecci et al., in this volume).

48Evidence for the existence of more elaborate tastes may be seen in amphorae originating from the periphery of the Byzantine world. The Crimean amphorae that composed the main cargo in the Bozburun wreck contained, according to sieved material from intact samples, a cheap red wine flavored with spices and fish paste (Hocker 2005, p. 104). The Byzantine populations inherited much of the Roman tastes in food, and fish sauces are frequently reported in sources recounting royal dinners by visitors to the Byzantine court. The Black Sea area had a long tradition of processing fish products and probably remained the main supplier of the shrinking Byzantine world. However, we have no evidence so far of special containers for the maritime transport of such products.


49Medieval wrecks (8th-15th centuries), depending upon their state of preservation and the type of research applied, can enlighten many aspects of maritime commercial transportation and reveal the vigor and extent of Byzantine trade. Underwater surveys usually exploit only a tiny fraction of the potential that newly discovered wrecks have to offer, and in many cases result only in short reports that reveal little. Even when the reports are poor, they can still provide information on geographical distribution, synchronicity in the circulation of ceramic vessels, freightage strategies and volume of consignments, but they rarely provide a deeper insight into commercial structures and practices. From this point of view the scientific investigation of this cultural capital is still in its infancy, only a handful of sites having been properly excavated and fully published. Although some finds may be rich, surveys provide limited information on Byzantine wrecks and cannot answer crucial questions regarding the ethnic affiliations of traders, the national identities of ships, the institutions involved in maritime trade and the historical circumstances of this activity. These are questions that can only be answered by fully excavated wrecks.

50With these aspects in mind, not every 12th century wreck found in the Aegean should be automatically thought to be Byzantine. The 12th century was a time of intensive activity for the new commercial powers in the Aegean, with Venice taking full advantage of the privileges accorded by the court of Constantinople (Thiriet 1959, pp. 40‑49). Byzantine elites in the same period were much engaged in clearing new agricultural lands and did not perceive trade to be an activity that enhanced social prestige (Brandejs 2013, p. 17). The percentage the Venetians held in the maritime transportation of goods around the Aegean and beyond cannot be estimated in any definite way. Venetian tradesmen and agents had already become settled in crucial production and distribution centers; it is doubtful, for example, that Romano Mairano, a Venetian agent who dominated the oil market of Corinth, Sparta and Thebes between the years 1165 and 1171 would have relied on ships other than Venetian (Sanders 2003, p. 396; Thiriet 1959, pp. 39, 47‑48). The provenance of a cargo, when defined, may link a trading venture to a certain artisanship, workshop or agricultural production, but it does not necessarily reflect the nationality of the carrier or the ethnic affiliations of the trader.

51On the contrary, well-excavated wrecks may answer a wide range of questions and have a long-reaching effect on the dialogue between archaeological and historical research. Shipwrecks such as the 7th century Yassi Ada wreck or the 11th century Serçe Limanı wreck, although excavated in the 1960s and 70s, are good examples of the impact excavated wrecks can exert and are predominant in any historical account concerning trade in those periods.

52Once again, it should be underscored that surveyed wrecks offer only a limited view of the commodities transported. Perishable materials do not generally survive in underwater environments. Byzantine wrecks provide little information on a wide range of food products, textiles, wood and livestock that are listed in Byzantine and Venetian sources. Some of those commodities might be traced in the future into the anoxic layers that occur within several shipwrecks. The recovery of a goat pelt in the form of an utricle, full of cheese, is one of the latest astonishing finds from a 12th century shipwreck excavated by EUA in the port of Rhodes (s/w 34, fig. 12). The pelt was found astern, carefully stored between two frames in the lower part of the hull. Surprisingly, the hairy part of the skin was turned inwards, in contact with the cheese, which might be considered distasteful today. However, historical references and folklore reveal that this was a standard method of making, maintaining and transporting cheese during the Ottoman period that survived well into the 20th century and still exists on some islands of the northern Aegean and in Turkey. Living evidence of this method was even discovered in a little shop in an alley off Athenas Avenue, in the traditional market of modern Athens. Such accidental present-day discoveries remind us of how little we know when our studies rely only on the macroscopic recognition of amphorae and ceramic vessels.

Fig. 12 – Goat pelt containing cheese from the 12th century shipwreck excavated in the port of Rhodes (s/w 34).

Fig. 12 – Goat pelt containing cheese from the 12th century shipwreck excavated in the port of Rhodes (s/w 34).

Fig. 13 – Concreted cargo of the Katergon shipwreck at Pelagonisos – Kyra Panagia (s/w 3).

Fig. 13 – Concreted cargo of the Katergon shipwreck at Pelagonisos – Kyra Panagia (s/w 3).

Catalogue of Byzantine shipwrecks in the Aegean


1. Porto Koupho, Chalkidiki. Depth: 105 m

53Spotted by sound scan sonar during a joint 2003 survey of EUA and HCMR and recorded by the submarine “Thetis”, the wreck site consists of two large concentrations of amphorae and two groups of Y-shaped anchors, partially retaining their order, resting on a sandy seabed. The visible remains are reported to cover an area of about 40×50 m. The cargo appears to be homogeneous, of Günsenin I type amphorae. Date: 9th-11th century.

Bibl.: Mela 2003, pp. 587‑588.

2. Porto Koupho, Chalkidiki. Depth: 110‑115 m

54A second wreck, spotted by the same mission, was recorded as heavily silted with visible remains consisting of 20‑25 scattered amphorae along a straight line, resting on a sandy seabed. Based on the single sample raised, the cargo consisted of Günsenin I type amphorae, of a different variation than the above. Date: 9th-11th century.

Bibl.: Mela 2003, p. 588.

3. Kyra Panagia (Pelagonisos), cape Katergo, northern Sporades. Depth: 17‑32 m (fig. 13)

55The wreck, discovered by the EUA in 1994, rests on a mixed rocky and sandy seabed on the west side of a cliff ending in a sandy plateau. Visible remains consist of an extended, dense mass of concreted half-broken amphorae, deeply embedded in the terrain. Many of the visible amphorae were identified as Günsenin type IIb. Several other parts of small-sized amphorae were not identified due to their fragmentary state. The assemblage also contains pieces of several Y-shaped iron anchors. Date: 12th century.

Bibl.: Haniotes 1994, p. 864; Koutsouflakis et al. 2017.

4. Kyra Panagia (Pelagonisos), Hagh. Petros bay, northern Sporades. Depth: 34‑40 m (fig. 3)

56The wreck rests on a sandy seabed surrounded by rocks; severely looted during the 1960s, it was partly excavated by the Archaeological Service. The ship had pine planking fastened to frames with iron nails and was estimated to have measured at least 25 m. Traces of the wooden structure of the ship are still visible on the site. The cargo consisted of Günsenin III type amphorae in two variations, pithoi, glazed tableware belonging to fine sgraffito and slip-painted wares and six huge granite millstones. Date: middle 12th century.

Bibl.: Kritzas 1971; Throckmorton 1971; Ioannidi-Dostoglou 1989; Parker 1992, no. 796.

5. Peristera-Alonessos, Galyfas bay, northern Sporades. Depth: 48‑55 m (fig. 4, 7)

57A large pile of amphorae resting on a sandy, slightly inclined seabed was located by the EUA in 1994. The concentration measured 15×6 m and contained several hundred intact, concreted amphorae of the Günsenin III type. The assemblage contained also a Y-shaped iron anchor. Date: 10th-12th century.

Bibl.: Haniotes 1994, pp. 862‑863; Kazianes 1994, p. 854; Preka et al. 2008, p. 1396.

6. Peristera-Alonessos, Galyfas bay, northern Sporades. Depth: 48‑55 m (fig. 5)

58Six meters distant from the latter wreck there is a second pile of amphorae of similar type (Günsenin III), measuring 15.5×5 m. The layout of the material appears to indicate a separate shipwreck, but this is not certain. Date: 10th-12th century.

Bibl.: Haniotes 1994, pp. 862‑863; Kazianes 1994, p. 854; Preka et al. 2008, p. 1396.

7. Peristera-Alonessos, Galyfas bay, northern Sporades. Depth: 56 m (fig. 6)

59A large 8-shaped pile of amphorae resting on a sandy seabed was located inside Galyfas bay on the NW coast of Peristera, not far from the two latter wrecks, during the joint survey of the EUA and the University of Trodheim. The cargo appears to be homogeneous, consisting of the same Günsenin III amphorae. Date: 10th-12th century.

Bibl.: Dellaporta 1999b, p. 1022; Dellaporta et al. 2006, pp. 82‑84.

8. Skopelos, Cape Mytikas, northern Sporades. Depth: 12‑22 m

60Several fragmented amphorae of Günsenin type IV were spotted in small concentrations along the rocky side of a cliff and are interpreted as remains of a severely looted cargo. Date: 12th-13th century.

Bibl.: Spondylis 2002, p. 30.

9. Skopelos, Dhasia islet, northern Sporades. Depth: 13‑33 m

61The wreck site consist of two large concentrations of amphorae, most of them in fragmentary condition, resting on a steep, rocky seabed, half covered by sediments and recognized as Günsenin III type. Date: 11th-12th century.

Bibl.: unpublished, H. Spondylis, personal communication.

10. Skopelos, unknown location, northern Sporades. Depth: not reported

62A collection of “Aegean Ware” tableware and amphorae in Oxford is said to have been raised under unknown circumstances from a wreck site off Skopelos. The assemblage contains 24 glazed bowls, mainly sgraffito and champlevé wares and 6 amphorae fragments, the latter as yet unpublished. Date: late 12th century.

Bibl.: Armstrong 1991.

11. Glafki bank, north Aegean, between Lemnos and northern Sporades. Depth: not reported

63An assemblage of several fine incised sgraffito bowls and plates, and an almost intact Günsenin III type amphora were raised by trawling nets and handed to the Archaeological Service by a fisherman in 2002. The evidence probably indicates the presence of a wreck in the area. Date: 12th century.

Bibl.: Tsanana & Amprazogoula 2015.

12. Skyros, Achileion. Depth: 6‑8 m

64A heavily silted shipwreck was located in the shallows during construction works of a local port installation. A large part of the ship’s hull remains in situ, covered by sediments. The site has not yet been investigated in any detail. From the cargo, consisting of fine glazed tableware, several deep bowls were brought to the local museum. Date: 14th century.

Bibl.: Karambinis 2015, pp. 257‑258, 267‑268.

13. Chios, Melanios bay. Depth: 2 m

65The wreck site consists of a number of concreted blocks of amphora sherds and scattered material over an extended area 100 m in length. The cargo consists of amphorae, Günsenin I type. Date: 9th-11th century.

Bibl.: Theodoulou 2002, p. 573.

14. Chios, Cape Pachi-Tholos. Depth: 25‑35 m

66A number of intact and fragmented amphorae, much scattered over the NW side of the promontory, are probably remains of a cargo. The amphorae are referred to as Byzantine without distinguishing a particular type. Date: unknown.

Bibl.: Dellaporta 2001, p. 553.

15. Cape Glaros, Nies, west coast of southern Pagasetic Gulf, Magnesia. Depth: 7‑10 m

67Surveyed by the HIMA between the years 2000-2010, the site extends over a length of 250 m, presenting multiple small heaps of pottery and a large number of Byzantine iron anchors of the Y and T types. The exact number of shipwrecks and of deposits resulting from jettison remains uncertain. The presence of at least one shipwreck close to the tip of the cape, containing a cargo of Günsenin III type amphorae is clearly reported. Date: 12th-13th century.

Bibl.: Spondylis 2002, p. 25.

16. Cape Glaros, Nies, west coast of southern Pagasetic Gulf, Magnesia. Depth: 7‑15 m

68Found further to the south in the same area as the latter, a second site produced a number of Günsenin III type amphorae, probably remains of a separate shipwreck. Both sites are poorly preserved and looted. Date: 12th-13th century.

Bibl.: Spondylis 2002, p. 25.

17. Steni Angali Bay, Nies, west coast of southern Pagasetic Gulf, Magnesia. Depth: 5‑8 m

69The site was surveyed by the HIMA during 2000. A concentration of fragmented Günsenin III type amphorae and a partially preserved glazed bowl are estimated to be remains of a looted cargo. Date 12th-13th century.

Bibl.: Spondylis 2002, p. 25.

18. Cape Telegraphos, west coast of southern Pagasetic Gulf, Magnesia. Depth: 10‑20 m

70Site surveyed by the HIMA during 2000. A number of broken Günsenin III type amphorae appear to indicate the presence of another poorly preserved shipwreck. Date: 12th-13th century.

Bibl.: Spondylis 2002, p. 25.

19. Cape Telegraphos, west coast of southern Pagasetic Gulf, Magnesia. Depth: 19‑25 m

71A poorly preserved cargo containing Günsenin IV type amphorae and flat-based amphorae of an unidentified type is reported. Surveyed by the HIMA during 2000. Date: 12th-13th century.

Bibl.: Spondylis 2002, pp. 25, 28‑29.

20. Amaliapoli, Cape Perikles, Pagasetic Gulf, Magnesia. Depth: 19‑25 m

72A cargo of half-broken amphorae of Günsenin III type is reported on the west side of the cape, without further information. Surveyed by the HIMA during 2000. Date: 12th-13th century.

Bibl.: Spondylis 2002, p. 26.

21. Amaliapoli, Kikynthos island, Pagasetic Gulf, Magnesia. Depth: 4‑9 m

73The wreck site, surveyed by the HIMA, consists of a large accumulation of broken pithoi sherds on a mixed, rocky and sandy seabed. The assemblage also contained fragments of two different types of amphorae, recognized as belonging to the Günsenin I and III types. Date: 12th-13th century.

Bibl.: Spondylis 2012, pp. 32‑34.

22. Agh. Ioannis Theologos, Phthiotis, northern Euboean Gulf. Depth: 20 m

74The wreck, traced und surveyed by the EUA, rests on a sandy sea floor about 750 m off a rocky coast. The cargo consists of amphorae of Günsenin II type. On the site are also reported two Y-shaped anchors. Date: 11th century.

Bibl.: Kazianes et al. 1990, pp. 228‑231; Parker 1992, no. 70.

23. Thorikos, cape Vrysaki, east Attica. Depth: 5‑10 m (fig. 9)

75The shipwreck was buried under silt deposits and was accidently discovered during construction works which revealed part of a cargo, consisting of glazed plates and deep bowls of “Aegean Ware”. Several samples were brought to the EUA and exposed material was also collected during 2012. Date: Second half of 12th century.

Bibl.: Dimitriadou 2002, p. 577; Koutsouflakis 2013, pp. 255‑261.

24. Cape Sesi, Portolafia Bay, Euboea. Depth: 16‑35 m

76A number of 20 intact and fragmentary amphorae of identical type were raised from a small area in the southern part of the cape, along its sloping sides. Several others were found isolated in the deeper zone of 35 m, half buried in the sediments. All recognized amphorae were of Günsenin III type in two variations. Date: 12th-13th century.

Bibl.: Koutsouflakis et al. 2012, pp. 53‑54; Koutsouflakis 2013, pp. 167‑169; Koutsouflakis & Argyri 2012, pp. 879‑880.

25. Cape Mavros Kavos, Kavalliani, southern Euboean Gulf. Depth: 9‑18 m (fig. 10)

77Remains of a severely looted wreck were discovered and surveyed by the HIMA and the EUA during 2008 on the northern part of a reef, east of the island of Kavalliani. Remains of the cargo, consisting of glazed tableware, mainly incised or painted deep bowls, were found scattered over an extensive area covered by shallow deposits. Although seriously disturbed, much of the cargo remains buried under the sediments. Date: late 12th century.

Bibl.: Koutsouflakis et al. 2012, pp. 57‑58; Koutsouflakis & Argyri 2012, p. 881; Koutsouflakis 2013, pp. 85‑104; Koutsouflakis & Tsompanidis 2018.

26. Aigina, Ypsili islet, Saronic Gulf. Depth: 42 m

78A cargo of Byzantine amphorae, some of them intact, was located by the EUA during 1997, on a flat, sandy seabed. The amphorae were recognized as belonging to the magarika category, 1st group of Bakirtzis classification (Bakirtzis 2003). Date: 8th-9th century.

Bibl.: Kazianes 1997, p. 1186.

27. Aigina, Diaporia islets, Soupia islet. Depth: not reported

79A cargo of roof tiles resting on a sandy seabed, discovered during 1997 by the EUA, is only briefly reported. An almost intact amphora from the same assemblage was identified as a magarikon of the 1st group of Bakirtzis classification and provides a date somewhere between the 9th and the 11th century.

Bibl.: Kazianes 1997, p. 1186.

28. Petrokaravo islet, south of Aigina, Saronic Gulf. Depth: 20‑37 m

80A number of about fifteen scattered, intact and broken amphorae were located on the side of an underwater cliff. The main wreck may lie in deeper waters further down the cliff. They were identified as belonging to the Günsenin III type. Date: 12th-13th century.

Bibl.: unpublished.

29. Kythera, Lydia islet, near Cape Gaidaros. Depth: 17‑25 m

81A large concentration of broken amphorae and a few intact samples have been located along the east side of the islet, exposed on the rocky, stepped sides of a descent with several sandy pockets. The amphorae have been recognized as belonging to Günsenin type I. Date: 10th-11th century.

Bibl.: unpublished.

30. Methone, SW Peloponnese. Depth: 5‑6 m

82A small accumulation of Byzantine amphorae and a Y-shaped iron anchor probably belong to the remains of a wreck. The amphorae are of the Günsenin III type. Date: 11th-12th century.

Bibl.: unpublished, H. Spondylis, personal communication.

31. Syrna, Kassiotis-bay, Dodecanese. Depth: 45‑52 m

83The wreck was located in summer 2002 during a joint research project of the EUA and the HCMR. A large cargo of amphorae still in their shipping order lies half-embedded on a sandy seabed. The cargo consists of Byzantine amphorae, most possibly corresponding to Otranto 1 and 2 types. Date: 10th-12th century.

Bibl.: Dellaporta 2002, p. 571; Micha 2005‑2006, p. 89, fig. 10.

32. Arkioi, Cape Koumaros, Dodecanese. Depth: 10‑35 m

84The wreck is located on the east side of the promontory, resting on a rocky sea floor, and consists of several blocks of concreted amphorae sherds and scattered material in the deeper zone. The cargo comprises amphorae of the Bakirtzis type 1 form. Date: 9th-11th century or slightly earlier.

Bibl.: Dellaporta 2001, p. 554.

33. Rhodes, commercial harbor. Depth: 16 m

  • 26 On the specific shipbuilding terminology see e.g. Steffy 1994.

85The remains of a long hull were unearthed near the entrance of the commercial port of Rhodes during 2007. The hull, in quite good condition, was half-buried in the sediments of the port and was preserved in a length of 20 m and a width of 3.40 m. The original length of the vessel was estimated to exceed 30 m. Built with a “skeleton first” technique, the hull was flat-bottomed, without projecting keel but preserved parts of a keelson.26 These characteristics, unique so far in the Eastern Mediterranean, may indicate a galley used for transportation. The interior was found to be empty, without cargo or ballast stones. Date: 13th century (dated after C‑14).

Bibl.: Koutsouflakis 2017.

34. Rhodes, commercial harbor. Depth: 12‑14 m (fig. 8)

86The remains of an extensively burned shipwreck were partly excavated by the EUA inside the commercial port of Rhodes during 2013. The wreck was heavily buried under 1 m of sediment. Excavation focused at one extremity of the vessel, revealing extended parts of the shell, ceiling, frames, keel, keelson and bulwark. The ship was fully loaded with a cargo of Günsenin III type amphorae, badly affected by fire. Among the raised items were also glazed pottery, foodstuffs and organic material (ropes, animal skins, wooden objects). Based on the pottery, the ship was dated to the middle of the 12th century.

Bibl.: Koutsouflakis 2017.

35. Rhodes, commercial harbor. Depth: 10‑15 m

87Several dozen intact Günsenin type I amphorae were brought to the authorities and the local museum by a commercial diver during the years 2002‑2008, all originating from the same area, near the two latter wrecks. Some of the samples raised are still sealed with pitch. Although not yet located, the evidence supports the existence of a third, heavily silted wreck in the area, loaded with a cargo of Günsenin type I amphorae. Date: 10th-11th century.

Bibl.: Koutsouflakis 2017.

36. Kastellorizo, Cape Zephyrion, Dodecanese. Depth: 10‑50 m

88The wreck is mostly known from material confiscated by the port police and recovered by fishermen and divers, today in many museums and private collections around the world. The site has been severely looted in the past. Although surveyed by the EUA during 2002 and 2005, it was never investigated properly. From the existing rather poor records, it seems that the main cargo consisted of glazed tableware (plates and deep bowls), mainly belonging to incised sgraffito and slip-painted wares. Date: late 12th-early 13th century.

Bibl.: Philotheou & Michailidou 1986; Michailidou & Philotheou 1989; Loucas 1989; Parker 1992, no. 538; Dellaporta 2002, p. 570.

37. Kastellorizo, Cape Hagh. Stephanos, Dodecanese. Depth: 15‑20 m

89Two concentrations of fragmented Byzantine amphorae are reported as remains of a cargo in a bay south of Cape Hagh. Stephanos. The exact type of amphorae is not mentioned. Date: not reported.

Bibl.: Dellaporta 2002, p. 569.

38. Crete, Dhia island, Cape Anginara, Herakleion. Depth: 10‑28‑20 m

90A huge concentration of pottery, broken and half-concreted on the rocky sea floor, located during a 2011 EUA survey, should correspond to the remains of at least one shipwreck. The site is severely looted and disturbed. The cargo included many different types of small amphorae of the magarika family, but also much intruding pottery from other periods. One of the main types has been recognized as Günsenin I type. Date: 9th-12th century.

Bibl.: Theodoulou et al. 2015, pp. 1092‑1093.

39. Ithaca, Cape Hagh. Andreas, Ionian Sea. Depth: 17‑25 m

91A much looted and destroyed accumulation of broken Byzantine amphorae has been located on a sloping seabed. The type of amphora is not specified beyond the fact that they belong to the magarika category. Date: unknown.

Bibl.: Εvangelistis & Dellaporta 2005, p. 1256.

40. Fournoi, Katergakia islets, north side of Thymaina island. Depth: ca 50 m

92A pile of about 20 amphorae of the Günsenin XVI type were located during an EUA-RPMNF survey in Fournoi in 2015. The cargo is resting on a sandy seafloor at the end of a steep rocky downfall in the depth zone of ‑50 meters and contains several intruders from an early Hellenistic cargo that rests in the shallows. Date: 12th century.

Bibl.: Viglaki Sofianou et al. 2019, pp. 210‑211.

41. Fournoi, Kisiria island. Depth: 15‑35 m

93A much scattered cargo of Günsenin I amphorae, containing though some intact specimens, was located during an EUA-RPMNF survey off the northeastern side of the island of Kisiria in the depth zone between 15 and 35 meters. Date: 9th-11th century.

Bibl.: Viglaki Sofianou et al. 2019, pp. 209‑211.


42. Tekmezar I, Marmara islands, Propontis. Depth: 35‑45 m

94A huge cargo, covering an area of about 800 square meters and containing well over 20,000 amphorae in three visible layers was located fifty meters from Cape Tekmezar. The cargo was homogenous, consisting of Günsenin I type amphorae. The assemblage contains at least five Y-shaped iron anchors. Date: 11th century.

Bibl.: Günsenin 1999, pp. 20‑21; 1998, pp. 311‑312.

43. Tekmezar II, Marmara islands, Propontis. Depth: 35‑45 m

95A second cargo of Günsenin I amphorae was discovered twelve meters to the west of Tekmezar I wreck, dispersed over an area of 180 square meters. The assemblage contains again at least five Y-shaped iron anchors. Date: 11th century.

Bibl.: Günsenin 1999, pp. 20‑21; 1998, pp. 311‑312.

44. Çamaltı Burnu, Marmara islands, Propontis. Depth: 20‑32 m

96The wreck was widely distributed in three pockets on the sea floor and consisted of about 800 amphorae of Günsenin IV type, piled according to size. The consignment contains also a large number of flat-bottomed storage jars and several Günsenin III examples. Several anchors in the same area are related to the wreck. Date: 13th century.

Bibl.: Günsenin 1999, p. 21; 1998, pp. 310‑311.

45. Ocaklar Burnu, Marmara islands, Propontis. Depth: 25‑41 m

97The wreck is located to the west of the large Marmara Island and contains a cargo of Günsenin I amphorae. Date: 11th century.

Bibl.: Günsenin 1999, p. 19; 1998, p. 310.

46. Kocayemişlik, Marmara islands, Propontis. Depth: 25‑35 m

98A cargo of several dozen amphorae of Günsenin I type and two Y-shaped anchors. Date: 11th century.

Bibl.: Günsenin 1998, p. 312.

47. Anataş adacık, Marmara islands, Propontis. Depth: 34 m

99The cargo consists of Günsenin I amphorae of large dimensions, lying on a sandy sea floor. A Y-shaped iron anchor is reported within the cargo. Date: 11th century.

Bibl.: Günsenin 1998, p. 312.

48. Taşada (Viranköy), Marmara islands, Propontis. Depth: 22 m

100Cargo of Günsenin I type amphorae. Date: 11th century.

Bibl.: Günsenin 1998, p. 315.

49. Eşek adaları (Saraylar), Marmara islands, Propontis. Depth: 22 m

101The wreck is located at the sandy end of a steep decline and exposes about 150‑160 amphorae of the Günsenin I type. Slightly to the north lies a Y-shaped anchor. Date: 11th century.

Bibl.: Günsenin 1998, p. 315.

50. Tenedos – Bozcaada, unknown location. Depth: not reported

102A large number of amphorae of Günsenin I type, confiscated in Tenedos and now on display in Çanakkale Museum, indicate the existence of a wreck there. Date: 11th century.

Bibl.: Günsenin 1990, I, p. 68; Parker 1992, no. 1139.

51. Bozburun. Depth: 30‑35 m

103Lying on a clean sandy slope, the wreck was excavated in the years 1995‑1998 by INA under the direction of F. Hocker. The cargo consisted of about 1,200‑1,500 amphorae originating from the Crimea region, northern Black Sea, and containing a kind of pulped red wine, flavored with spices and fish paste. The wreck is dated towards the end of the 9th century.

Bibl.: Hocker 2005.

52. Serçe Limanı. Depth: 33 m

104Excavated by G. Bass and Fr. Van Doorninck in the years 1977-1979, the wreck known as “the glass wreck” produced 3 tons of broken glass vessels, Islamic glazed pottery bowls and Günsenin I type amphorae lined with pitch and bearing Greek graffiti. Date: ca 1025.

Bibl.: van Doorninck 1991; Bass et al. 2004.

53. Kötü Burun. Depth: 36‑42 m

105A cargo including over 60 amphorae of the Günsenin I type in two sizes and 10 large globular jars lies scattered on a rocky slope. Date: 10th-11th century.

Bibl.: Pulak 1985; Günsenin 1990, p. 22, fig. 7‑8.

54. Çömlek Burun, Bozburun peninsula. Depth: 65 m

106Located during the 2005 survey of the RPM Nautical Foundation, the wreck site consists of about 60 amphorae piled together in three levels and covering an area of 10×3 m. There is evidence of two forms of amphorae that appear to be variations of a single type, parallel to Günsenin I type. A grapnel anchor near one end of the concentration should be related to the same wreck. Date: 11th-12th century.

Bibl.: Royal 2006, p. 8.

55. Kızılburun. Depth: 28‑37 m

107Located during the 1993 annual survey of the INA. The site consists of hundreds of amphorae, intact and broken, lying on a rocky slope at the extreme southern point of Kızılburun. The amphorae are all of the same type, undetermined. Date: 10th-12th century.

Bibl.: Pulak & Rogers 1994, p. 19.

56. Adrasan Bay, southern Turkey. Depth: 15‑18 m

108The wreck was identified during a survey conducted by the Marine Science and Technology Institute of Dokuz Eylül University and partly excavated during 2014. It rests on a mixed sandy and rocky seafloor and covers an area of about 150 m2. The cargo consisted of glazed tableware, mainly deep bowls with plain or upright rims and low ring bases, decorated in fine sgraffito, brown painted and green-brown painted wares. In addition, three Y‑shaped anchors, marble architectural pieces and ballast-stones were also observed. Date: mid 12th-early 13th century.

Bibl.: Özdaş & Öniz 2014; Doğer & Özdaş 2016.


Adams 2001: Adams J., “Ships and boats as archaeological source material”, World archaeology 32, 2001, pp. 292‑310.

Antonova et al. 1971: Антонова И.А., Даниленко В.Н., Ивашута Л.П., Кадеев В.И., Романчук А.И. [= Antonova I.A., Danilenko V.I., Ivashuta L.P., Kadeev V.I., Romanchuk A.I.], “Средневековые амфоры Херсонеса” [= “Medieval amphorae of Chersonesos”], Античная древность и средние века [= Ancient Antiquity and Middle Ages] 7, 1971, pp. 81‑101.

Armstrong 1991: Armstrong P., “A group of Byzantine bowls from Skopelos”, Oxford journal of archaeology 10/3, 1991, pp. 335‑347.

Bakirtzis 2003: Μπακιρτζής X. [= Bakirtzis Ch.], Βυζαντινά Τσουκαλολάγηνα [= Byzantine pottery] (2nd ed.), Athens, 2003.

Bass 1982: Bass G.F., “The pottery”, in Bass & van Doorninck 1982, pp. 155‑188.

Bass 2005: Bass G.F. (ed.), Beneath the seven seas, London, 2005.

Bass & van Doorninck 1982: Bass G.F., van Doorninck F.H. Jr. (ed.), Yassı Ada, vol. I, A seventh century Byzantine shipwreck, College Station, 1982.

Bass et al. 2004: Bass G.F., Matthews Sh.D., Steffy J.R., van Doorninck F.H. Jr. (ed.), Serçe Limanı: An 11th-century shipwreck, vol. 1, The ship and its anchorage, crew, and passengers, College Station, 2004.

Bjelajac 1989: Bjelajac L., “Byzantine amphorae in the Serbian Danubian area in the 11th-12th centuries”, in Déroche & Spieser 1989, pp. 109‑118.

Brandejs 2013: Brandejs J., “Some remarks on the economic development of the Komnenian Byzantium”, West Bohemian historical review 1, 2013, pp. 13‑22.

Cameron 2016: Cameron A., “Late Antiquity and Byzantium: An identity problem”, Byzantine and modern Greek studies 40/1, 2016, pp. 27‑37.

Carlson 2003: Carlson D.N., “The classical Greek shipwreck at Tektaş Burnu, Turkey”, American journal of archaeology 107, 2003, pp. 581‑600.

Carlson et al. 2015: Carlson D.N., Leidwanger J., Kampbell S.M. (ed.), Maritime studies in the wake of the Byzantine shipwreck at Yassiada, Turkey, College Station, 2015.

Chaniotis 1994: Chaniotis F.K., Archaeological deltion 49, Chronicles, B’2, 1994, pp. 862‑864.

Dellaporta 1999a: Dellaporta Aik., “Byzantine shipwrecks and underwater evidence for Byzantine pottery”, in Papanikola-Bakirtzi 1999, pp. 118‑144.

Dellaporta 1999b: Dellaporta Aik., Archaeological deltion 54, Chronicles, B’2, 1999, pp. 1019, 1028‑1030.

Dellaporta 2001: Dellaporta K.P., Archaeological deltion 56, Chronicles, B’6, 2001, pp. 553‑558.

Dellaporta 2002: Dellaporta K.P., Archaeological deltion 57, Chronicles, B’6, 2002, pp. 569‑572.

Dellaporta et al. 2006: Dellaporta K.P., Jasinski M.E., Søreide Fr., “The Greek-Norvegian deep-water archaeological survey”, International journal of nautical archaeology 35/1, 2006, pp. 79‑87.

Déroche & Spieser 1989: Déroche V., Spieser J.‑M. (ed.), Recherches sur la céramique byzantine, Bulletin de correspondance hellénique supplément 18, Athens, 1989.

Dimitriadou 2002: Dimitriadou Th., Archaeological deltion 57, Chronicles, B’6, 2002, pp. 576‑577.

Doğer & Özdaş 2016: Doğer L., Özdaş H., “Adrasan: Ceramic finds from a Byzantine shipwreck”, in P. Magdalino, N. Necipoğlu (ed.), Trade in Byzantium: Papers from the Third International Sevgi Gönül Byzantine Studies Symposium (Istanbul, 24‑27 June 2013), Istanbul, 2016, pp. 445‑463.

Doncheva-Petkova 1977: Дончева-Петкова Л. [= Doncheva-Petkova L.], Българска битова керамика през ранното Средновековие (втората половина на VI – края на X в.) [= Early medieval Bulgarian coarse ware (second half of the 6th-end of the 10th century)], Sofia, 1977.

Evangelistis & Dellaporta 2005: Evangelistis D., Dellaporta K.P., Archaeological deltion 60, Chronicles, B’2, 2005, pp. 1254‑1256.

Gibbins & Adams 2001: Gibbins D., Adams J., “Shipwrecks and maritime archaeology”, World archaeology 32, 2001, pp. 279‑291.

Günsenin 1989: Günsenin N., “Recherches sur les amphores byzantines dans les musées turcs”, in Déroche & Spieser 1989, pp. 267‑276.

Günsenin 1990: Günsenin N., Les amphores byzantines (xe-xiiie siècle). Typologie, production, circulation d’après les collections turques, PhD, University Paris 1 Panthéon-Sorbonne, 1990 (unpublished).

Günsenin 1998: Günsenin N., “Récentes découvertes sur l’île de Marmara (Proconnèse) à l’époque byzantine. Épaves et lieux de chargement”, Archaeonautica 14, 1998, pp. 309‑316.

Günsenin 1999: Günsenin N., “From Ganos to Serçe Limani: Social and economic activities in the Propontis during medieval times illuminated by recent archaeological and historical discoveries”, The INA quarterly 26/3, 1999, pp. 18‑23.

Günsenin 2001: Günsenin N., “L’épave de Çamaltı Burnu I (île de Marmara, Proconnèse). Résultats des campagnes 1998‑2000”, Anatolia Antiqua 9, 2001, pp. 117‑133.

Günsenin 2005: Günsenin N., “A 13th-century wine carrier: Çamalti Burnu, Turkey”, in Bass 2005, pp. 118‑123.

Günsenin 2009: Günsenin N., “Ganos wine and its circulation in the 11th century”, in M. Mundell Mango (ed.), Byzantine trade (4th-12th centuries): The archaeology of local, regional and international exchange. Papers of the Thirty-eighth Spring Symposium of Byzantine Studies (St John’s College, University of Oxford, March 2004), Aldershot, 2009, pp. 145‑153.

Günsenin 2018: Günsenin N., “La typologie des amphores Günsenin. Une mise au point nouvelle”, Anatolia Antiqua 26, 2018, pp. 89‑124.

Haniotes 1994: Haniotes F.K., Archaeological deltion 49, Chronicles, B’2, 1994, pp. 862‑864.

Harvey 1989: Harvey A., Economic expansion in the Byzantine Empire (900-1200), Cambridge, 1989.

Hayes 1992: Hayes J.W., Excavations at Saraçhane in Istanbul, vol. 2, The pottery, Princeton, 1992.

Hendy 1985: Hendy M.F., Studies in the Byzantine monetary economy (c. 300-1450), Cambridge, 1985.

Hocker 2005: Hocker Fr., “Sampling a Byzantine vintage: Bozburun, Turkey”, in Bass 2005, pp. 100‑105.

Ioannidaki-Dostoglou 1989: Ioannidaki-Dostoglou E., “Les vases de l’épave byzantine de Pélagonèse-Halonnèse”, in Déroche & Spieser 1989, pp. 157‑171.

Karambinis 2015: Karambinis M., The island of Skyros from the Late Roman to Early Modern times: An archaeological survey, Archaeological studies Leiden University 28, Leiden, 2015.

Kazanaki-Lappa 2002: Kazanaki-Lappa M., “Medieval Athens”, in A.E. Laiou (ed.), The economic history of Byzantium: From the seventh through the fifteenth century, vol. 2, Washington DC, 2002, pp. 639‑646.

Kazianes 1994: Kazianes D., Archaeological deltion 49, Chronicles, B’2, 1994, pp. 853‑858.

Kazianes 1997: Kazianes D., Archaeological deltion 52, Chronicles, B’3, 1997, pp. 1185‑1187.

Kazianes et al. 1990: Kazianes D., Simossi A., Haniotes F.K., “Three amphorae wrecks from the Greek world”, International journal of nautical archaeology 19, 1990, pp. 225‑232.

Koilakou 2013: Koilakou Ch., “Thebes”, in J. Albani, E. Chalkia (ed.), Heaven and earth: Cities and countryside in Byzantine Greece, Athens, 2013, pp. 180‑191.

Kontogiannis 2012: Kontogiannis N.D., “Euripos – Negroponte – Eğriboz: Material culture and historical topography of Chalcis from Byzantium to the end of the Ottoman rule”, Jahrbuch der Österreichischen Byzantinistik 62, 2012, pp. 29‑56.

Kontogiannis et al., in this volume: Kontogiannis N.D., Skartsis S.S., Vaxevanis G., Waksman S.Y., “Ceramic vessels and food consumption: Chalcis as a major production and distribution center in the Byzantine and Frankish periods”, in Waksman 2020, pp. 239‑254.

Koutsouflakis 2013: Κουτσουφλακης Γ. [= Koutsouflakis G.], Ναυσιπλοΐα και Θαλάσσια Εμπορευματική Διακίνηση στο Νότιο Ευβοϊκό (6ος αι. π.Χ.-14ος αι. μ.Χ.) [= Νavigation and commercial transportation in southern Euboean Gulf (6th century BC-14th century AD)], PhD, National Kapodistrian University of Athens, 2013 (unpublished).

Koutsouflakis 2017: Κουτσουφλακης Γ. [= Koutsouflakis G.], “Τρία μεσαιωνικά ναυάγια στον Εμπορικό Λιμένα της Ρόδου” [= “Three shipwrecks of the medieval era in the commercial port of Rhodes”], in Το Αρχαιολογικό Έργο στα Νησιά του Αιγαίου [= The archaeological work in the Aegean Islands], vol. C, Lesbos, 2017, pp. 477‑500.

Koutsouflakis, forthcoming: Koutsouflakis G., “‘The unharvested sea’: A century of underwater exploration in the Hellenic Archipelagos. The evidence of shipwrecks”, in The archaeology of the sea, Brussels, forthcoming.

Koutsouflakis & Argyri 2012: Κουτσουφλακης Γ., Αργύρη Ξ. [= Koutsouflakis G., Argyri X.], “Αρχαία Ναυάγια και φορτία στο Θαλάσσιο Χώρο του Νότιου Ευβοϊκού” [= “Ancient shipwrecks and cargos in the south Euboean Gulf”], in Αρχαιολογικό Έργο Θεσσαλίας και Στερεάς Ελλάδας 3, Πρακτικά Επιστημονικής Συνάντησης, Βόλος 12.3‑15.3.2009 [= The archaeological work in Thessaly and Sterea Hellas 3, Proceedings of scientific meeting, Volos 12.3‑15.3.2009], Volos, 2012, pp. 875‑888.

Koutsouflakis & Tsompanidis 2018: Koutsouflakis G., Tsompanidis A., “The Kavalliani shipwreck: A new cargo of Byzantine glazed tableware from the south Euboean Gulf, Aegean”, in F. Yenişehirlioğlu (ed.), Proceedings of the XIth Congress AIECM3 on Medieval and Modern Period Mediterranean Ceramics (Antalya, 19‑24 October 2015), vol. 1, Ankara, 2018, pp. 39‑48.

Koutsouflakis et al. 2012: Κουτσουφλακης Γ., Αργυρη Ξ., Παπαδοπουλου ΧΡ., Σαπουντζης Ι. [= Koutsouflakis G., Argyri X., Papadopoulou Ch., Sapountzis J.], “Υποβρύχια Αναγνωριστική Έρευνα στο Νότιο Ευβοϊκό (2006-2008)” [= “Underwater survey in the south Euboean Gulf (2006-2008)”], Εnalia XI, 2012, pp. 40‑69.

Koutsouflakis et al. 2017: Κουτσουφλακης Γ., Θεοδουλου Θ., Foley B. [= Koutsouflakis G., Theodoulou Th., Foley B.], “Αυτοψία σε ένα άγνωστο ναυάγιο στη νήσο Κυρα-Παναγιά” [= “An autopsy on an unknown shipwreck in the island of Kyra-Panagia”], Αρχαιολογία [= Archaiologia] 123, 2017, pp. 48‑57.

Kritzas 1971: Κριτζας Χ. [= Kritzas Ch.], “Το βυζαντινόν ναυάγιον Πελαγοννήσου Αλοννήσου” [= “The byzantine shipwreck in Pelagonisos – Alonessos”], Athens annals of archaeology 4, 1971, pp. 176‑182.

Laiou & Morrisson 2007: Laiou A.E., Morrisson C., The Byzantine economy, Cambridge, 2007.

Lane 1973: Lane F.C., Venice: A maritime republic, Baltimore, 1973.

Loucas 1989: Loucas I., “Les plats byzantins à glaçure inédits d’une collection privée de Bruxelles”, in Déroche & Spieser 1989, pp. 177‑183.

Magdalino 1993: Magdalino P., The empire of Manuel I Komnenos (1143-1180), Cambridge, 1993.

Megaw & Jones 1983: Megaw A.H.S., Jones R.E., “Byzantine and allied pottery: A contribution by chemical analyses to problems of origin and distribution”, Annual of the British School at Athens 78, 1983, pp. 235‑263, pl. 24‑30.

Mela 2003: Mela T.‑P., Archaeological deltion 58, Chronicles, B’6, 2003, pp. 585‑589.

Micha 2005-2006: Micha P., “Amphora shipwrecks in the Aegean: A database of the Ephorate of underwater antiquities”, Skyllis 7, 2005-2006, pp. 82‑93.

Michailidou & Philotheou 1989: Michailidou M., Philotheou G., “Plats byzantins provenant d’une épave près de Castellorizo”, in Déroche & Spieser 1989, pp. 173‑176.

Morozova et al., in this volume: Morozova Y., Waksman S.Y., Zelenko S., “Byzantine amphorae of the 10th-13th centuries from the Novy Svet shipwrecks, Crimea, the Black Sea: Preliminary typology and archaeometric studies”, in Waksman 2020, pp. 429‑446.

Opaiţ 2007: Opaiţ A., “A weighty matter: Pontic fish amphorae”, in V. Gabrielsen, J. Lund (ed.), The Black Sea in Antiquity: Regional and interregional economic exchanges, Black Sea studies 6, Aarhus, 2007, pp. 101‑122.

Özdaş & Öniz 2014: Özdaş H., Öniz H., “Underwater excavation of Adrasan plate wreck”, TINA maritime archaeology periodical 2, 2014, pp. 32‑33.

Papanikola-Bakirtzi 1999: Papanikola-Bakirtzi D. (ed.), Byzantine glazed ceramics: The art of sgraffito, Athens, 1999.

Papanikola-Bakirtzi 2012: Papanikola-Bakirtzi D., “Byzantine glazed ceramics on the market”, in C. Morrisson (ed.), Trade and markets in Byzantium, Washington DC, 2012, pp. 193‑216.

Parker 1992: Parker A.J., Ancient shipwrecks of the Mediterranean and the Roman provinces, BAR International series 580, Oxford, 1992.

Pecci et al., in this volume: Pecci A., Garnier N., Waksman S.Y., “Residue analysis of medieval amphorae from the Eastern Mediterranean”, in Waksman 2020, pp. 417‑428.

Philotheou & Michailidou 1986: Φιλοθεου Γ., Μιχαηλιδου Μ. [= Philotheou G., Michailidou M.], “Βυζαντινά πινάκια από το φορτίο ναυαγισμένου πλοίου κοντά στο Καστελλόριζο” [= “Byzantine plates from a shipwreck’s cargo off Kastellorizo”], Archaeological deltion 41, Studies, 1986, pp. 271‑329.

Preka et al. 2008: Preka K.A., Theodoulou Th., Kourkoumelis D., Foley B., Archaeological deltion 63, Chronicles, B’2, 2008, pp. 1394‑1405.

Pulak 1985: Pulak C., “Turkish coastal survey yields wrecksite inventory”, INA newsletter 12/2, 1985, pp. 2‑3.

Pulak & Rogers 1994: Pulak C., Rogers E., “The 1993-1994 Turkish shipwreck surveys”, The INA quarterly 21/4, 1994, pp. 17‑21.

Pulak & Townsend 1987: Pulak C.R., Townsend R.F., “The Hellenistic shipwreck at Serçe Limanı, Turkey: Preliminary report”, American journal of archaeology 91, 1987, pp. 31‑57.

Radulescu 1973: Radulescu A.V., “Amfore cu inscriptii de la edificiul roman cu mosaic”, Pontica 6, 1973, pp. 193‑207.

Romanchuk et al. 1995: Романчук А.И., Сазанов А.В., Седикова Л.В. [= Romanchuk A.I., Sazanov A.I., Sedikova L.V.], Амфоры из комплексов византийского Херсона [Amphorae from the complexes of Byzantine Chersonesos], Ekaterinburg, 1995.

Royal 2006: Royal G.J., “Beyond the shallows: Shipwreck discoveries from the 2005 Bozburun Peninsula survey, Turkey”, The INA quarterly 33/3, 2006, pp. 3‑11.

Sanders 2003: Sanders G.D.R., “Recent developments in the chronology of Byzantine Corinth”, in C.K. Williams II, N. Bookidis (ed.), Corinth, vol. XX, Corinth: The centenary (1896-1996), Princeton, 2003, pp. 385‑399.

Sazanov 1997: Sazanov A., “Les amphores de l’Antiquité tardive et du Moyen Âge: continuité ou rupture? Le cas de la mer Noire”, in G. Démians d’Archimbaud (ed.), La céramique médiévale en Méditerranée. Actes du VIe congrès de l’AIECM2 (Aix-en-Provence, 13‑18 novembre 1995), Aix-en-Provence, 1997, pp. 87‑102.

Spondylis 2002: Σπονδυλησ Η. [= Spondylis H.], “Αναγνωριστική έρευνα ΙΕΝΑΕ Δυτικών Ακτών Νοτίου Παγασητικού, Έτους 2000” [= “Underwater Survey of the western coast in the south part of the Golf of Pagasai (2000)”], Enalia VI, 2002, pp. 24‑31.

Spondylis 2012: Σπονδυλησ Η. [= Spondylis H.], “Υποβρύχια έρευνα ΙΕΝΑΕ στον Παγασητικό Κόλπο. Ερευνητική περίοδος 2005” [= “Underwater archaeological research in the Gulf of Pagasai, 2005 campaign”], Εnalia ΧΙ, 2012, pp. 16‑39.

Steffy 1982: Steffy J.R., “Reconstructing the hull”, in Bass & van Doorninck 1982, pp. 65‑86.

Steffy 1994: Steffy J.R., Wooden ship building and the interpretation of shipwrecks, College Station, 1994.

Talbot-Rice 1966: Talbot-Rice D., “Late Byzantine pottery at Dumbarton Oaks”, Dumbarton Oaks papers 20, 1966, pp. 209‑219.

Theodoulou 2002: Theodoulou Th., Archaeological deltion 57, Chronicles, B’6, 2002, pp. 572‑573.

Theodoulou et al. 2015: Theodoulou Th., Foley Br., Evaggelistis D., Koutsouflakis G., Sakellariou D., Tourtas A., “Crete project 2011: Underwater archaeological survey at the area of Herakleion and Dia Island. A preliminary report”, in Archaeological work in Crete 3 (Proceedings of the 3rd Meeting, Rethymno, 5‑8/12/2013), 2015, pp. 1091‑1101.

Thiriet 1959: Thiriet F., La Romanie vénitienne au Moyen Âge. Le développement et l’exploitation du domaine colonial vénitien (xiie-xve siècle), Paris, 1959.

Throckmorton 1971: Throckmorton P., “Exploration of a Byzantine wreck at Pelagos Island near Alonessos”, Athens annals of archaeology 4, 1971, pp. 183‑185.

Todorova 2011: Todorova E., “The medieval amphorae (9th to 14th centuries AD) from excavation at Silistra in 2007 (preliminary report)”, in Ch. Tzochev, T. Stoyanov, A. Bozkova (ed.), PATABS II. Production and trade of amphorae in the Black Sea: Acts of the International Round Table held in Kiten, Nessebar and Sredetz, September 26‑30, 2007, Sofia, 2011, pp. 131‑140.

Todorova, in this volume: Todorova E., “One amphora, different contents: The multiple purposes of Byzantine amphorae according to written and archaeological data”, in Waksman 2020, pp. 403‑416.

Tsanana & Amprazogoula 2015: Tsanana Aik., Amprazogoula K., “Céramique d’une épave méso-byzantine de la mer Égée du Nord”, communication presented at the XIth Congress AIECM3 on Medieval and Modern Period Mediterranean Ceramics (Antalya, 19‑24 October 2015).

Van Alfen 1996: van Alfen P., “New light on the 7th-c. Yassi Ada shipwreck: Capacities and standard sizes of LRA1 amphoras”, Journal of Roman archaeology 9, 1996, pp. 189‑213.

Van Alfen 2015: van Alfen P.G., “The restudy of the LR2 amphoras from the 7th-century Yassıada shipwreck”, in Carlson et al. 2015, pp. 17‑34.

Van Doorninck 1972: van Doorninck F.H. Jr., “Byzantium, mistress of the sea (330‑641)”, in G.F. Bass (ed.), A history of seafaring based on underwater archaeology, London, 1972, pp. 134‑158.

Van Doorninck 1989: van Doorninck F.H. Jr., “The cargo amphoras on the 7th century Yassi Ada and 11th century Serçe Limani shipwrecks: Two examples of a reuse of Byzantine amphoras as transport jars”, in Déroche & Spieser 1989, pp. 247‑257.

Van Doorninck 1991: van Doorninck F.H. Jr., “The medieval shipwreck at Serçe Limani: An early 11th-century Fatimid-Byzantine commercial voyage”, Graeco-Arabica 4, 1991, pp. 45‑52.

Van Doorninck 2002: van Doorninck F.H. Jr., “Byzantine shipwrecks”, in A.E. Laiou (ed.), The economic history of Byzantium: From the seventh through the fifteenth century, Washington DC, 2002, pp. 899‑905.

Van Doorninck 2005: van Doorninck F.H. Jr., “The ship of Georgios, priest and sea captain: Yassiada, Turkey”, in Bass 2005, pp. 92‑97.

Van Doorninck 2015a: van Doorninck F.H. Jr., “The metrology of the piriform amphoras from the 11th-century Byzantine ship at Serçe Limani: New designs but an old system”, in Carlson et al. 2015, pp. 35‑54.

Van Doorninck 2015b: van Doorninck F.H. Jr., “The 7th-century ship at Yassıada and her final voyage: Present thoughts”, in Carlson et al. 2015, pp. 205‑216.

Viglaki Sofianou et al. 2019: Βιγλακη Σοφιανου Μ., Κουτσουφλακησ Γ., Campbell P. [= Viglaki Sofianou M., Koutsouflakis G., Campbell P.], Κορσηιτών νήσοι. Αρχαιολογικά ευρήματα και μία προσέγγιση της ιστορίας των Φούρνων Κορσεών [= The islands of Korseai. Archaeological finds and an approach to the history of Fournoi Korseon], Athens, 2019.

Waksman et al. 2014: Waksman S.Y., Kontogiannis Ν.D., Skartsis S.S., Vaxevanis G., “The main ‘Middle Byzantine Production’ and pottery manufacture in Thebes and Chalcis”, Annual of the British School at Athens 109, 2014, pp. 379‑422.

Waksman et al. 2018a: Waksman S.Y., Skartsis S.S., Kontogiannis N.D., Todorova E.P., Vaxevanis G., “Investigating the origins of two main types of Middle and Late Byzantine amphorae”, Journal of archaeological science: Reports 21, 2018, pp. 1111‑1121 (, accessed 11/04/2019).

Waksman et al. 2018b: Waksman S.Y., Koutsouflakis G., Burlot J., Courbe L., “Archaeometric investigations of the tableware cargo of the Kavalliani shipwreck (Greece) and into the role of the harbour of Chalcis in the Byzantine and Frankish periods”, Journal of archaeological science: Reports 21, 2018, pp. 1122‑1129 (, accessed 11/04/2019).

Ward 2004: Ward Ch., “Plant remains”, in Bass et al. 2004, pp. 495‑511.

Ward 2015: Ward Ch.A., “Plant remains from the old jars on the Byzantine ship at Yassiada”, in Carlson et al. 2015, pp. 55‑62.


1 In writing this essay I benefited greatly from the contribution of several colleagues, to whom I owe my sincere thanks: Y. Waksman, N. Günsenin, Fr. van Doorninck, D. Bakirtzi, L. Doğer, N. Kontogiannis and S. Skartsis were keen to share information, bibliography and to discuss the topic in detail. It goes without saying that they should not be considered responsible for any errors and lapses appearing in this contribution, which remain fully my own. The following abbreviations are used:
– EUA: (Hellenic) Ephorate of Underwater Antiquities;
– HCMR: Hellenic Centre for Marine Research;
– HIMA: Hellenic Institute of Maritime Archaeology;
– INA: Institute of Nautical Archaeology;
– RPMNF: RPM Nautical Foundation;
– s/w: shipwreck (catalogue entry).

2 Van Doorninck (2002, p. 899) reported the existence of roughly 80 known Byzantine wrecks throughout the Mediterranean and about the same number are depicted on a distribution map published by Dellaporta (1999a, p. 119, fig. 2). The catalogue has grown much larger over the past 15 years as a result of several surveys conducted. But it is impossible at any given time to deal with a fixed number of wrecks. Part of the problem lies in the disagreement between scholars in defining clearly the chronological and cultural limits of the term “Byzantine” and the establishment of a clear chronological barrier between the Late Roman and Early Byzantine periods (Cameron 2016). Both counts cited above contain a number of wrecks that date to as early as the 4th century and would otherwise be considered to be Late Roman.

3 Out of a total of 1,259 shipwrecks, Parker’s catalogue contains 95 shipwrecks dated to after 600 AD (Parker 1992, pp. 14‑15, fig. 3).

4 A more recent statistical analysis, still unpublished, reduces this percentage to the still generous 21% (Koutsouflakis, forthcoming). It should be noted, however, that these high numbers may be influenced by two factors. The first concerns a lesser time period of exposure to the destructive marine environment: as historically more recent wrecks, they are more likely to survive and leave distinctive traces on the seafloor. The second factor concerns the ease with which Byzantine cargoes of amphorae may be recognized, thus rendering them less likely to be included in the large reservoir of the unidentified shipwrecks.

5 The term “time capsule”, adopted in order to express this disconnected reasoning, has become so old that no one remembers who introduced it into archaeological discussion. First articulated in the sixties in order to further this common perception concerning shipwrecks, it was much used as an underwater equivalent to the terms “closed find” and “associated group”, popular in terrestrial archaeology. Although not exactly right – the term is rather simplistic and theoretical, too narrow to fit the complexity and variety of ancient shipwrecks – it has survived four decades of criticism and is still largely in use today. For a short discussion of the term, its elusive content and the intricacies it could introduce into ancient shipwreck studies see Adams 2001, pp. 296‑297.

6 The term “Aegean” is used here in a somewhat formulaic way without strict, absolute geographical limits. Under the term are included such interconnected waterways as the Bosporus, the Ionian Sea and the southern coast of modern Turkey.

7 It should be noted that the four basic amphora types proposed by Günsenin are considered here as more or less generic types, since they include a large number of subtypes that vary in shape and volume. For the full version of this typology see Günsenin (1990, I, pp. 20‑45; 2018). A separate typology, also containing amphorae from the earlier Byzantine period, is provided by Bakirtzis (2003, pp. 70‑88, pl. 16‑22).

8 My present position, as director of the Department of Underwater Archaeological Sites, Monuments and Research in the EUA, and my involvement in underwater research over the past 25 years, has enabled me to dive and to personally examine a large number of wrecks. It should be noted that with newly discovered wrecks being published every year and the rapid advance in studies of Byzantine ceramics, the present paper may quickly become outdated. Furthermore, certain fundamental questions, such as “what do we mean exactly by the term shipwreck?” or “when should a site with scattered remains be considered a shipwreck?” are not to be answered here.

9 The evolution of many different subtypes is an almost standard phenomenon in most LR2 amphorae cargoes in the Aegean (Koutsouflakis 2013, pp. 108‑112).

10 The same amphorae are known also under the names Bakirtzis class 2 (Bakirtzis 2003, pp. 76‑77, pl. 17), Saraçhane type 54 (Hayes 1992, p. 75), Bjelajac type II (Bjelajac 1989, pp. 111‑113), Dončeva-Petkova type 1 (Dončeva-Petkova 1977, p. 100), Romanchuk et al. class 42 (Romanchuk et al. 1995) and Sazanov type 45 (Sazanov 1997, p. 97).

11 On the standardization of Late Roman 1 and 2 and Günsenin I type amphorae see van Alfen 1996; 2015; van Doorninck 2015a.

12 As stated above, the evidence comes mainly from amphorae regularly found by deep trawlers over a period of more than 20 years. This material suffered much handling (fishermen, port police, local museums) before it came to the EUA Department, months after being raised. Information on the locations of spots or track-lines of trawling is rare and difficult to crosscheck. More than 40 Günsenin I amphorae of various sizes have been reported from this area and according to their shapes and sizes and the type of calcareous concretion present on them they may be attributed to between four and nine different wrecks.

13 Also known under the names Bakirtzis class 4 (Bakirtzis 2003, pp. 78‑80, pl. 19), Saraçhane type 60 (Hayes 1992) and Sazanov type 52 (Sazanov 1997, p. 98).

14 Also known as Bakirtzis class 5 (Bakirtzis 2003, p. 80, pl. 20), Saraçhane type 61 (Hayes 1992, p. 76), Romanchuk et al. class 48 (Romanchuk et al. 1995), Sazanov type 53 (Sazanov 1997, p. 98) and MBYZ 15 (Vroom 2005, pp. 97‑99).

15 Also known as Bakirtzis class 6 (Bakirtzis 2003, pp. 80‑81, pl. 21), Bjelajac type III (Bjelajac 1989, p. 115), Romanchuk et al. classes 44‑46 (Romanchuk et al. 1995), Sazanov types 54‑55 (Sazanov 1997, p. 98, fig. 4) and Antonova et al. types 23‑24 (Antonova et al. 1971).

16 Glazed tableware appears in most Byzantine shipwrecks of the 12th and 13th centuries; however, their inclusion in the cargo is doubtful when it is present in small quantities. A small number of glazed dishes could have been personal belongings of the captain, the passengers or the crew, or part of the kitchenware of the ship. These cases (Günsenin 2005, pp. 119, 122) are not discussed in detail in this study.

17 Only a few details of the excavation and the layout of the cargo are published. Kritzas (1971) refers to 768 complete or almost complete plates and bowls and 54 intact amphorae, and this was about the number of amphorae observed on the site by the author in 2009. It must be remembered that this was a wreck that had been disturbed and looted before the 1970 excavation.

18 For a synopsis of the decorative styles of the period see Sanders 2003, pp. 388‑389, and Papanikola-Bakirtzi 1999, pp. 17‑24.

19 Another “plate wreck” of the same period, still unpublished, is located in the area of Cape Gelidonya, a few miles to the west of the Adrasan wreck (Doğer & Özdaş 2016, p. 450). The wrecks of Kastellorizo, Adrasan and Cape Gelidonya present the most dense nexus of such homogenous cargoes and are evidence for Byzantine trading ventures towards the Eastern Mediterranean.

20 It is tempting, for example, to attribute several of the bowls of underwater origin now in possession of the Dumbarton Oaks collection (Talbot-Rice 1966) to the Kavalliani shipwreck, given the similarities in shape, dimensions and decoration.

21 For the “Main Middle Byzantine Production” (MBP), see Waksman et al. 2014.

22 The examples from the Aegean are numerous, although mostly understudied or unpublished. For some published examples see Carlson 2003, p. 583; Pulak & Townsend 1987, p. 91; Hocker 2005, p. 104; Günsenin 2005, p. 120.

23 Pigments and iron nails are reported to have been found inside Late Roman amphorae in the vaulted rooms of the mosaic-floored Edifice of the ancient port of Tomis (modern Constanţa, Radulescu 1973, pp. 197‑198).

24 The evidence of vitis species seeds from the excavation of the Serçe Limanı wreck, in light of the belated use of systematic sampling methods, is too limited and obscure to support any such argument (Ward 2004). Sumac was recognized as one of the commodities contained in several amphorae; it should be remembered however that most amphorae found on board were reused.

25 DNA analysis was carried out by Maria Hansson, Lund University, Sweden, GC/MS analysis by R. Nelson and C. Reddy (Woods Hole Oceanographic Institution).

26 On the specific shipbuilding terminology see e.g. Steffy 1994.

Table des illustrations

Titre Fig. 1 – Distribution of cargoes with Günsenin I‑IV amphora types in the Aegean.
Fichier image/jpeg, 693k
Titre Fig. 2 – Distribution of cargoes of glazed tableware in the Aegean, 12th-14th century.
Fichier image/jpeg, 633k
Titre Fig. 3 – The Pelagonisos shipwreck: view of the amphorae cargo (s/w 4).
Fichier image/jpeg, 1,0M
Titre Fig. 4 – Photomosaic of the Günsenin III cargo in the Galyfas Bay, Peristera – Alonessos (s/w 5).
Fichier image/jpeg, 651k
Titre Fig. 5 – Photomosaic of the Günsenin III cargo in the Galyfas Bay, Peristera – Alonessos (s/w 6).
Fichier image/jpeg, 760k
Titre Fig. 6 – Photomosaic of the Günsenin III cargo in the Galyfas Bay, Peristera – Alonessos (s/w 7).
Fichier image/jpeg, 670k
Titre Fig. 7 – Layers of Günsenin III amphorae in the cargo of shipwreck 5, Peristera – Alonessos.
Fichier image/jpeg, 522k
Titre Fig. 8 – Fragmented cargo of Günsenin III amphorae in the hold of shipwreck 34 in the port of Rhodes.
Fichier image/jpeg, 1,6M
Titre Fig. 9 – Exposed glazed tableware from the Thorikos shipwreck (s/w 23).
Fichier image/jpeg, 1011k
Titre Fig. 10 – Glazed tableware from the cargo of the Kavalliani shipwreck (s/w 25).
Fichier image/jpeg, 1,0M
Titre Fig. 11 – Oily substance pouring from the mouth of a Günsenin IIb amphora from the Katergon shipwreck at Pelagonisos – Kyra Panagia (s/w 3).
Fichier image/jpeg, 356k
Titre Fig. 12 – Goat pelt containing cheese from the 12th century shipwreck excavated in the port of Rhodes (s/w 34).
Fichier image/jpeg, 817k
Titre Fig. 13 – Concreted cargo of the Katergon shipwreck at Pelagonisos – Kyra Panagia (s/w 3).
Fichier image/jpeg, 1,4M


Hellenic Ministry of Culture and Sports, Ephorate of Underwater Antiquities, Erechtheiou St. 59, Athens 11742, Greece,


Volume papier
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search