Underwater Archaeology MARS 3360-001 Professor William C. Phoel April 29, 2010 Chryso Tsikkou Title: An Early Helladic II Period Submerged Cultural Resource Found in Dokos, Greece. 1. ABSTRACT The Hellenic Institute of Marine Archaeology (HIMA) has undertaken several significant projects. Among them, the most important is the one at Dokos (1988-1992). The Early Helladic shipwreck of Dokos, sometimes cited as the earliest known wreck, was the first systematic and efficient full-scale investigation of an ancient shipwreck to be conducted in Greece.
The rich ceramic finds (raised among 10,000 objects) dating to the Early Helladic II period are judged to be particularly important both for their large size and type variety, and also in that they represent the largest closed group of Early Helladic ceramic ware found to date in the Aegean. 2. A BRIEF HISTORY OF UNDERWATER ARCHAEOLOGY IN GREECE. Figure [ 1 ]: Ancient remains found in underwater excavations of various submerged cultural resources. The possibilities of underwater archaeology in Greece were appreciated relatively early by Greek archaeologists.
In 1884, Keeper of Antiquities Christos Tsoudas, with the help of sponge divers, concluded the first methodical underwater survey in the channel between the island of Salamis and Attica. Yet, it was not archaeologists that raised some of the Ancient World’s few surviving masterpieces; it was ordinary fishermen and sponge divers. Some of these works of art include the Poseidon of Kreusis found in the Gulf of Corinth (1889), the Boy or Ephebe of Marathon (1925) extricated by Evangelos Leonidas from his fishing nets, the Poseidon (or Zeus), and the Jockey of Cape Artemision (1928, see Fig. ). Nevertheless, the field of underwater archaeology continued to develop in Greece over the past century and finally reached the point in 1996 where Nikos Tsouchlos, then director of the Hellenic Institute of Marine Archaeology (HIMA), could state that “in Greece underwater archaeology has now finally established its place in the archaeological field” (1). 3. BACKGROUND INFORMATION ABOUT THE ISLAND AND THE SHIPWRECK IN DOKOS, GREECE.
According to the Hellenic History of Marine Archaeology, “the underwater archaeological excavation at Dokos, carried out by the Hellenic Institute of Marine Archaeology from 1989 to 1992 under the direction of the archaeologist George Papathanasopoulos, was the first full scale excavation of an ancient wreck in Greece which also employed the most up-to-date technological methods of the time. ” The underwater site of Dokos is the most valuable concrete evidence for navigation, sea trade, technology, and the economy in the Aegean during the late 3rd millennium BC (3).
The website of Hellenic Institute of Marine Archaeology also states that, “the island of Dokos owes its name to the Dokos family of Hydriot ship owners, to whom it belonged at the end of the 18th century. In antiquity it was called Aperopia, a name perhaps meaning “the mountain island. ” It is in fact craggy and precipitous, with few natural springs. Its highly important geographical location, however, at the entrance to the Argolic Gulf and on the sea route to and from the coasts of Argolida and Lakonia attracted the interest of mariners from early on” (3).
The Hellenic Institute of Marine Archaeology continues by stating that, “[Dokos] was colonized from the end of the Neolithic period (4th millennium BC), but human presence on the island augmented during the Early Helladic period (2500-2300/2200 BC), when sea trade developed. After that time it vanished from the record for a long period of history, becoming the territory of fishermen and shepherds. In times of trouble, however, its role was upgraded because of its position and fortified character. During the 13th century BC, the powerful settlements of Myti Kommeni and Ledeza grew up.
In the middle of the 7th century AD an actual castle town was formed in the district of the Kastro, and at the time of the national uprising in 1821 the island was used as a naval station for the Hydriot fleet” (3). After Independence, Dokos passed into the possession of different Hydriot families. 4. LOCATION AND SIGNIFICANCE OF THE WRECK – RECONNAISSANCE SURVEYS. In the summer of 1975, Peter Throckmorton, a diver and an investigator of sea bottoms detected a great number of pottery at a depth of 20 meters at the seabed around the island of Dokos (3, see Fig. 9 and 10). Soon afterwards,
Throckmorton, having communicated the information to the Greek Archaeological Service, returned to the island accompanied by the Greek archaeologist Georgios Papathanasopoulos, who examined the pottery, dated it in the Early Helladic period and recommended that it was a shipwreck probably around 2200 BC. Soon after, in 1975 and 1977, they carried out two investigating surveys, which resulted to a more precise dating: according to the Hellenic Institute of Marine Archaeology, “the shipwreck could be located to the Early Helladic II period, which means around 2700-2200/2100, BC” (3).
During the same survey, the two archaeologists had defined the area of the shipwreck at a depth of 15 to 30 meters and found out that it was the most ancient wreck discovered worldwide ever since (3). Figure [ 2 ]: Map displaying where Dokos is located. A methodical excavation of the site started in the summer of 1989 by the Institute of Marine Archaeological Surveys of Greece when government restrictions were lifted and adequate funds were accessible.
The excavation was under the direction of the archaeologist Georgios Papathanasopoulos. The entire process of excavation was completed in 1992, and the finds were kept in a special space of the Museum of Spetses (3). The shipwreck of Dokos signifies one of the most essential verifications for prehistoric navigation and the growth and expansion of marine trade. It also offered tremendously significant information about the level of technology during the Bronze Age and about the economy and exchange in the Aegean in late 3rd millennium BC.
The pottery exposed from the sea bottom represented the shipload of a merchant ship and would be transported to an unknown destination. The pottery of the ship substantiates for the high level of technology during that period of time. There have been found vessels of diverse kinds, along with objects of everyday use, which bear a resemblance to those already found in the excavations of the coastal Early Helladic colonies of Lerna and of Tiryns, as well as in other settlements of the general Argolida region.
The survey of the shipwreck has supplemented a great deal of information to the land surveys of the Early Helladic civilization and made scholars appreciate and recognize better the marine transportation and communication processes in the Myrtoan Sea and in the Saronicos Bay (6, see Fig. 2). 5. Figure [ 3 ]: Amphorae found on board. STUDY AND RECOVERY OF THE ARTIFACTS. The online database of Underwater Archaeology had published a journal article regarding the cargo found in Dokos, Greece.
In this section of the paper, I will summarize what the article acknowledged. The pottery from the Dokos cargo site, which all emerged to belong to a late phase of Early Helladic II period, includes all the fundamental Early Helladic II period types and styles identified from several land sites as well as various kinds of household utensils. The commonest forms of artifacts found were bowls, sauceboats, basins, wide-mouthed jars with plastic bands, and amphoras (see Fig. 3); utensils included “spit-supports”, braziers, and baking trays.
An appealing feature of the pottery studied from the 1989 and 1990 campaigns was the presence of Cycladic elements; this validates the primary impression created of the pottery at the time of the discovery of the cargo in 1975. The three commonest shapes are bowls, sauceboats and amphoras, which are characterized by many dozens of sherds; the first two are the commonest Early Helladic II period types in southern and central Greece and on the islands of the Saronic and Argolic gulfs (3).
In the course of reconnaissance dives made in 1989 west of the main site, two stone slabs were found, each with a perforation near its edge. The first, which is roughly triangular in shape, was at a depth of 34 meters and some 33 meters west of point 8 on the perimeter of the wreck. It was lying rather close to the shore on a rocky bottom with patches of sand. The second, nearly circular in shape, was found at a depth of 38 meters, 46 meters west of point 8 on the wreck perimeter and 14 meters southwest of the first slab and on a similar bottom.
When the two pieces were found, they were lodged among the rocks with their perforations topmost and indicating eastward in the direction of the wreck. Both slabs were photographed in situ and their positions were set by calculating their distances from different points on the perimeter of the main site and from each other. Before they were elevated up to the surface, archaeologists conducted an experiment: ropes were secured to the holes and two people standing on the diving raft positioned over the middle of the site attempted to haul in the slabs.
They found that it took a large effort by the two people just to shift the slabs from their location. Subsequently, they tried to pull them up, and the two team members stationed a boat directly over the two slabs and hauled first on one and then on the other one. That time required less exertion to move the slabs, but still called for a considerable effort. The initial observations made at the time the slabs were found and the consequent examination of them after they were raised both propose that they were in all probability Early Age Bronze stone anchors and directly related with the Dokos underwater site (7).
The shapes of the stone slabs, their weight, the location of the holes, and the method used to drill them were all distinctive attributes of Bronze Age anchors. The fact that they each have only one hole for the anchor rope, as well as their somewhat slight weight, the nature of the bottom where they were found and their positions on the seabed were all indications that they were anchors planned for use on rocky bottoms; they were in fact characteristic Early Bronze Age anchors.
All these facts, in addition to the breaks and edge damage which were apparent evidence of their frequent use, exclude the likelihood that they were weights used by divers. Stones used as diving weights were usually left on the bottom after the dive, and in any case did not undergo the kind of wear and tear that would cause that kind of damage (7). Bearing in mind their distances from the site of the cargo and their locations to the west of it, it was not irrational to recommend that they probably belonged to the Early Helladic II period shipwreck discovered in Dokos.
The rationale was that if the vessel was anchored in approximately the position of the wreck site and were caught by a strong westerly wind, which is the only thing that seriously agitates the sea in the little bay of Skindos, and if it then broke up on the rocky shore and sank, its anchors would have been in just about the same positions as those two: that is, approximately 40 meters west of the bottom site and with the holes for the anchor ropes orientated in its direction.
The number, size, and particularly the weight of the anchors suggest that they belonged to a moderately small vessel of perhaps 5 to 10 tons and 12 to 15 meters in length. This surpasses an earlier estimate of the vessel’s size, which was based on the relatively large cargo it seemed to have been carrying. 6. EQUIPMENT USED, MAPPING DESCRIPTIONS, AND CONSERVATION-PRESERVATION OF THE ARTIFACTS. The Hellenic Institute of Marine Archaeology engaged in an intensive conservation campaign in their lab constructed at the Museum of Spetses (see Fig. 8).
Due to the amount of finds in need of treatment, with the addition of the lack of funds and manpower, the work had been slow, but consistent. To increase productivity, all the finds were logged into a computer for easy referencing and research (8). Due to the impracticability of using the conventional grid and survey method, mainly because of the steep gradients and abnormality of the seabed in the area of the excavation, HIMA turned to a new system, the Sonic High Accuracy Ranging and Positioning System, for mapping and plotting the positions of finds underwater.
Known as SHARPS, the system was designed by scientists of the Institute of Nautical Archaeology (INA) for use in underwater excavations, mapping, and plotting positions of finds on the seabed by employing an echo system for fixing the three-dimensional position of a point. The system consists basically of four transmitters-receivers, a control unit, and a computer with at least 640k of RAM, a hard disk, the 8087 mathematics processor, and a graphic screen. The transmitter-receivers are linked to the control unit and then to the computer by cables of up to 300 meters in length.
Three of the transmitter-receivers are situated firmly on the seabed at a maximum distance of 100 meters from each other so as to form a triangle. They act as receivers for the sound pulses transmitted at a permanent rate by the process of a trigger; the latter allows the diver to decide and manage the frequency and duration of the transmission. This transmitter is moved by the diver to the different points with a measured precision in the order of two centimeters at 100 meters.
The triangular grid employed is defined by measuring the distances between the three receivers A, B, and C on the inclined plane bounded by the lines joining each pair of vertexes, namely AB, BC, CA, and the depths hA, hB, and hC. The position of the transmitter (trigger) in relation to the three receivers is established in the same way both horizontally and vertically: the receivers gather the transmitted signals and send them to the control unit, and the distance of the transmitter from each receiver is calculated from the time taken for the signal to be received.
The three distances and the three depths are used to calculate the coordinates of the transmitter in the three dimensions x, y, and z in an arbitrary system of reference defined by the receivers, the calculations are based on the mathematical relationships resulting from the triangular grid and on the speed of the sound pulse through the water (2). Figure [ 4 ]: Divers working on the Dokos shipwreck trying to retrieve artifacts. In addition to the SHARPS system, a widespread photographic documentation was compiled.
All phases of the excavation were photographed, from the set-up of the installations on land and in the sea to marked feature finds while still in place, after being raised, and before and after conservation work. To make a photographic plan of the site and the main bulk of the finds on the bottom, a stereophotographic frame was used. It enabled the archaeologists, after the photographs had been correctly connected, to assemble a simple photomosaic of the underwater site, and the same photographs were used for photogrammetry, to make a precise scale plan of the site.
The photomosaic is made up of partially overlapping photographs taken with a succession of exposures from the frame’s first position. The stereographic frame was designed and used mainly by the president of the Hellenic Institute of Marine Archaeology, Nikos Tsouchlos, while the underwater photography of the work on the bottom and individual finds in situ were photographed by Kyle Jachney (2). In general, pottery survives well in marine environments and requires only minimal handling after recovery.
It is necessary, however, that the conservator be able to distinguish earthenware, stoneware, and porcelain, and to be familiar with the alternative treatments for conserving them (9). Stoneware and porcelain are fired at such high temperatures that they are impervious to liquids and thus do not take up soluble salts from their archaeological environment; therefore, it is not necessary to take them through long rinses to remove soluble salts. However, with certain kinds of stoneware and porcelain, glazes are applied in consequent firings, and sometimes salts may be accumulated between the glaze and the body.
If these salts are not removed, the glaze may flake off. Consequently, a gigantic caution must be exercised with stoneware and porcelain. Well-fired pottery require only to be rinsed in a mild detergent and the edges and surfaces scrubbed with a soft brush. Care should be taken not to remove traces of food, paint, pigments, or soot that is left on the interior or exterior surfaces. The conservator must be careful not to mark the pottery surface when using a brush or any other object throughout preservation and conservation (see Fig. ). Fragile, poorly fired pottery require more care, but the procedure is the same. Fragile pieces, pottery with friable surfaces, flaking surfaces, or fugitive paints may necessitate consolidation with a resin (8). Earthenware excavated from marine sites become saturated with soluble salts, and/or the surfaces often become covered with insoluble salts, such as calcium carbonate and calcium sulfate. In many instances, pottery adjacent to metal objects, mostly iron objects, will be enclosed by the encrustation forming around the metal.
Soluble salts (chlorides, phosphates, and nitrates) are potentially most hazardous to the integrity of pottery, and they must be removed in order for the object to be stable. The soluble salts are hygroscopic, and as the relative humidity rises and falls, the salts constantly dissolve and crystallize. These salts ultimately reach the surface of the pot, where extensive crystallization takes place causing exfoliation of the surface of the pot. Eventually, the pot will break as a result of internal stresses. At times, masses of needle-like crystals may cover the surface, hiding all details.
Soluble salts can be removed by repeated rinsing in water (a running bath is the quickest and most effective method but is very wasteful). There are many ways of setting up a series of vats so that water runs into one vat and cascades into a series of other vats. This minimizes water waste, particularly if using de-ionized water. Very simple rinsing procedures exist, such as putting soluble, salt-laden sherds in a mesh bag and placing the bag in the reservoir of a toilet. Countless volunteers help out every day in altering the water and the salt content in the sherds equalize with that contained in the supply water.
Then, if necessary, the rinsing can be continued in several baths of de-ionized water to lower the salt content even further. This is a simple trick that is very efficient. Monitor the rinsing progress with a conductivity meter. If sherds or pottery are too fragile to withstand the rinsing process, surfaces may be consolidated first with Acryloid B-72 then rinsed. Since Acryloid B-72 is somewhat water-permeable, it will permit the salts to diffuse out, although significantly more gradually than in non-consolidated material (7).
In most cases, the safest and most satisfactory technique of removing insoluble salts from the surface of pottery is by hand. Most calcareous concretions can be removed without difficulty when wet by scraping with a scalpel, dental tool, or an analogous appliance. Dental burrs and pneumatic air chisels are also quite useful. The insoluble salts may also be removed chemically, but it is important to pre-wet the sherd. Nitric acid, hydrochloric acid, and oxalic acid are most commonly used.
Before using any acid on pottery, however, make sure that the paste is thoroughly wetted so that the acid will not be absorbed. Although 10 to 20 percent of nitric acid can be used to remove calcareous concretions, it is potentially the most damaging acid of the three. More care should be implemented in its use, as dilute nitric acid will dissolve lead glazes. In most cases, 10 to 20 percent hydrochloric acid is safer than nitric acid to clean glazed pottery. The sherds are left immersed in the acid until all gas evolution comes to an end (usually less than an hour); this procedure may be repeated if required.
Care must be exercised, since hydrochloric acid can discolor glazes, especially lead glazes, which will turn milky. The samples are then washed thoroughly in tap water and, if needed, immersed in a 10 percent oxalic acid for 10 to 20 hours to eliminate iron stains. A systematic rinsing should follow, and the sherds should be dried. It is crucial that pottery with a carbonate temper (shell, calcium carbonate) not to be submerged in hydrochloric or nitric acid because the tempering material will be removed from the paste, resulting in the weakening of the pottery (2).
While nitric, oxalic, and hydrochloric acid treatments will remove calcareous deposits (especially hydrochloric), they are inclined to dissolve the iron oxides from pottery containing iron oxides in the paste or in the glazes (many stoneware glazes contain iron oxides). The use of these acids on glazes containing iron oxides intensifies their tendency to exfoliate, in particular if the glazes are friable. To avoid over-cleaning, the sherds should always be pre-wetted by soaking in water and then by applying the acid locally on the surfaces with a cotton swab or by drops.
The excess acid is directly removed when the effervescing action stops, either by wiping the area or rinsing the object(s) under running water to remove the acid. Earthenware and terra cotta often contain iron oxides, are more porous, and thus more prone to deteriorate when treated with these acids; acid treatments should be used on such materials with good judgment and carefulness (9). A helpful chemical for eliminating calcareous deposits from ceramics is ethylene-diaminetetraacetic acid (EDTA). A 5 percent solution of the tetra-sodium salts of EDTA (pH 11. 5) works best for removing calcareous aterial without critically affecting the iron content of the pottery. Iron is more soluble at pH 4, while calcareous deposits are more soluble at pH 13. In this treatment, the sherds are immersed in the solution and left until the deposits are removed. Sporadically, the solution may have to be replenished. During the process, the iron stains that are usually bound in with the calcium salts are removed along with the calcium. It is a slow but successful treatment. Soaking calcareous-encrusted sherds in a 5 percent aqueous solution of sodium hexametaphosphate has been used to remove calcareous deposits.
Care must be taken, however, since a solution of sodium hexametaphosphate has a tendency to soften the paste of the sherd more readily than the calcareous encrustation. Silicates on the surface of pottery can be removed with hydrofluoric acid, but this acid is extremely hazardous, and it is not recommended to be used by amateurs (2). The conservation of ceramics recovered from a marine site is not complex. When pieces are found encrusted, the most complicated part of the conservation procedure is the removal of the adhering material without any possible destruction of the paste or glaze.
For this motive, mechanical cleaning techniques are favored, but hydrochloric acid is used with some reliability to remove calcareous encrustations. The soluble salts that are perpetually present in any porous material recovered from a marine site are removed by rinsing in water. In most instances, tap water is all that is needed, but to the use of de-ionized water in the final baths will remove more soluble salts. Sulfide staining is easily removed with hydrogen peroxide, but other stains, such as iron stains, are more difficult to remove without negatively affecting the material.
If the decision is made to remove the more complicated stains, the material should be methodically wetted with water before immersing or applying the proper chemicals. Supervising the progression of the process cautiously and careful washing of the hands in running water after using any chemicals are recommended (2). After treatment, permit the pottery to air dry. Solvent drying is not obligatory, but it may be used if preferred. After drying, consolidate by entirely immersing the material in a dilute solution of PVA or Acryloid.
Pottery vessels can be reconstructed after the consolidated sherds have dried. Apparatus required to preserve and conserve ceramics includes properly sized vats, tap water, de-ionized water, acetone, ethanol, PVA, Acryloid, hydrogen peroxide, hydrochloric acid, EDTA, dental picks, and pneumatic chisels (2). Figure [ 5 ]: Archaeologists working on the artifacts found on the wreck. Little items made of stone can be handled in basically the same approach as described for pottery (once pottery has been fired, it is actually a form of stone).
Numerous sedimentary rocks can absorb soluble salts and be stained. The same treatments and chemicals described under pottery were used, but the acids should be no stronger than 5 percent. The usage of some acids on any of the sedimentary rocks (e. g. , limestone, marble, sandstone, etc. ) will be tremendously unsafe as these can be quickly destroyed by acid treatments. The acids can be used effectively on metamorphic and igneous rocks (7). Figure [ 6 ]: Pithoi found on the shipwreck.
Upon early assessment of pottery and other surface finds, the site was dated to circa 2200 BC, or the Early Helladic II period, comprising the largest closed body of materials from the Early Helladic period (4). The clay vases were estimated to number more than 500, including most known Early Helladic pottery types. They consist of many of the deep spouted sauceboats in a mixture of shapes and sizes, as well as cutaway jugs, shallow and deep bowls. Furthermore, in a variety of shapes and sizes there were amphorae, plates, cups, jars, askoi (type of pottery) and pithoi (large storage jars of a certain type, see Fig. ), and household utensils and grinders. An examination of the pottery and an inspection of the many sauceboats propose that the types are similar to those from Askitario in Attica, and the sauceboats compare to those from Lerna in the Argolid, Lithares in Boeotia, and from Cyclades (8). After a thorough examination of the origin of the pottery, archaeologists concluded that the shipwreck lied directly on the maritime trade route from South Euboea to the Saronic and Argolid gulfs, ending at the Early Helladic center of Lerna (9).
The significance of the wreck is enhanced by other finds: two fragments of the same lead bar pointed to an Attic provenance for the pottery for it is recognized that the ore was being mined at Lavrion in the Early Helladic period. Additionally, the two primitive stone anchors that were found at a distance of 40 meters from the wreck and in deeper water, were accurately on the line the ship would have taken after entering the bay. The anchors may have been dropped by the Early Helladic ship just before it sank (5).
The thousands of obsidian blades, essentially from Milos, that have been found at every Neolithic and Early Helladic site in Greece, on the mainland and the islands, are clear verifications of trade by sea in the distance past (7000-2000 BC). As a consequence, the cargo of the Early Helladic ship that was wrecked by the entrance to the small bay on the island of Dokos is considered tangible evidence of sea trade in Greece, and one of the oldest seafaring documents so far to come into sight from the bottom of the Aegean (3). 7.
EXHIBITION, PUBLICATION, AND PRESENTATION. Figure [ 7 ]: The sign of ENALIA, a specialized journal. Since 1989, the Hellenic Institute of Marine Archaeology publishes ENALIA, the only specialized journal in underwater and nautical archaeology in Greece (see Fig. 7), which is widely appreciated both in Greece and abroad. The shipwreck of Dokos is incorporated in the publications, and HIMA’s work at the Dokos wreck made it all too clear that there was virtually no infrastructure in Greece for anyone wishing to carry out marine archaeological research.
This prompted archaeologists to carry out an architectural study at the Athens Polytechnic School to explore the possibilities of building a marine archaeology center. This proposal had the advantage of being based on very real coordinates with very specific functions to fulfill. Nevertheless, the proposed site of Dokos (specifically, at Myti Kommeni in the Bay of Skindos), with its wealthy contribution of marine and land archaeological sites round about, portrays the architect with considerable challenges: the land is rough and inaccessible, and there are no modern facilities close at hand (3).
Additionally, The Danish Institute at Athens offered a presentation of The work of the Hellenic Institute of Marine Archaeology, 2005 – 2008. Presentations on underwater surveys and excavations included the following projects: 1. Southern Salamis by Yannos G. Lolos, in English 2. Pagasetikos Gulf by Elias Spondylis, in Greek 3. Argosaronic Gulf by Christos S. Agouridis, in English 4. Southern Euboean Gulf by George Koutsouflakis, in English Integrated in the presentation was the ancient shipwreck found in Dokos, Greece.
It fell in the category of Argosaronic Gulf by Christos S. Agouridis (3). 8. GENERAL CONCLUSION The vast amount of Early Helladic II pottery from the Dokos cargo site is imperative not only for the diversity of its types and sizes, but chiefly because it comprises the only possible closed find of this nature with Early Helladic II pottery thus far known from the Aegean (4). The Late Helladic sherds, the teeth and bone fragments were primarily found together with the Early Helladic sherds in the lower levels of the second trial channel.
They cannot, however, be considered as a part of the large closed find of pottery, which for the most part came from the surface level of the site, but rather as rejected pieces of pottery from the land; this is only to be anticipated in a place that has been used as a natural harbor from prehistoric times to the present day (4 and 5). On the basis of the conclusions achieved so far it appears that the great mass of Early Helladic II pottery recovered from the underwater site at Myti Kommeni on Dokos was part of the cargo of one or more Early Helladic ships that foundered, capsized or abandoned their load in that bay (6).
The bay shaped a natural harbor, which would no doubt have always been the scene of much action and movement due to the settlement, probably a trading station, on the Myti Kommeni peninsula during the Early Helladic period (3rd millennium BC).