Biblical sites were highly sought after by some of our earliest and greatest archaeologists. One such site, Carchemish, was the famed city of the Hittite Empire. It attracted the attention of T.E. Lawrence and Woolley, pioneers of British Near Eastern Archaeology, who excavated there just before the First World War. Then came the crashing calamity of the Great War, and after it came new political borders…
Carchemish then found itself on the border between Syria and Turkey, which rendered the site inaccessible for further investigation. The region soon fell into neglect for some 70 years. All changed in the late 1980s when the construction of dams along the Euphrates River brought archaeologists back in major local and international rescue projects. But what have the archaeologists found? Edgar Peltenburg and T.J. Wilkinson were among the archaeologists to return. Here they tell of mega-floods linked to climate change, of champagne cup graves and of the changing fortunes of the land of Carchemish.
When we arrived in the land of Carchemish we arrived with a sense of great urgency. Dams were being built and the land was to be flooded. We needed to rescue excavate and we needed to survey. This is a key area of the Ancient Near East since Carchemish was the Hittite Empire’s capital of its Syrian provinces, and thereafter the centre of a paramount kingdom with lively relief sculptures and inscriptions lining processional ways from the Euphrates to the Inner City and Acropolis.
We were coming in the footsteps of T.E. Lawrence and Woolley. While Lawrence is best known for being Lawrence of Arabia, Sir Leonard Woolley (1880-1960) is considered to be one of the first 'modern' archaeologists. Thus, while at Carchemish he determined to set the city in a wider context and began exploring throughout the area, albeit in an unsystematic way.
For our landscape survey, we wanted to get a long-term history of settlement and landscape from the earliest prehistoric settlement of the 'agricultural revolution' through to the expansive growth of population that coincided with the incorporation of the area into the territorial empires of the Assyrians, Seleucids, Romans and Byzantines. Since Woolley's day, the nature of archaeological surveys has changed: they are now more systematic and can give a more rigorous view of patterns of settlement to show how cities, towns and villages expand and decline over the centuries. They have also benefited from developments in satellite imagery and GIS.
However, we have new challenges: in the 19th and early 20th centuries, archaeologists witnessed a landscape that was much less inhabited than that of today, making survey far easier. Moreover, with the current huge investments in dams, earth-moving machines and industrialised agriculture, the landscape is being transformed to such a degree that we are rapidly losing the archaeological record. Modern surveys are therefore trying to record as much as possible of the archaeological record before it is lost for ever. Such work is therefore more pressing than ever before.
In addition to our landscape survey, we focussed on the tell site of Jerablus Tahtani at the north of the flood zone and adjacent to the British Museum-sponsored work at Carchemish. Carchemish and Jerablus Tahtani are tell sites. Tells are the emblematic sites of the Near East, the multi-period layer-cake mounds that accumulated as a result of the superimposition of occupational strata. However, though the history of tell development is frequently taken for granted, careful surveys demonstrate that their history varies in different parts of the Near East. Thus, over much of the Levant, tells were occupied for great lengths of time – in some cases from the Neolithic until the Byzantine or Islamic periods. Further east though, in the semi-arid steppe between the Tigris and Euphrates Rivers, tell development usually finished at some time during the 2nd millennium BC. The area around Carchemish, situated between the Levant to the west and the Jazira to the east, could belong to either tradition.
Through our survey, wediscovered that in this area, tells began to appear following the Neolithic when agricultural settlements developed on the Euphrates River terraces as well as on hill tops (starting from around the 8th-7th millennia). By the important Uruk period (4th millennium BC) tells were, literally, beginning to grow up. This is not surprising: the Uruk period is synonymous with the growth of early civilisation. Its name comes from the eponymous southern Mesopotamian city most closely identified with the invention of writing and other characteristics of Sumerian civilisation. When working at Carchemish, despite the appeal of the tell's later occupation – including a series of stunning sculptures dating from the 8th century BC – Lawrence and Woolley had also uncovered tantalising finds from this Uruk period. The finds came from deep inside the site’s acropolis mound and included Uruk pottery with its early Sumerian connections; but at that time little could be stated about its significance. In post-Uruk levels, they also found a number of richer 'champagne graves', named after the abundance of tall stemmed cups found within metal-rich cist graves. These graves revealed something of the site’s earlier importance, but again little was known of the period at that time. It was thus with great excitement that in our preliminary survey of the 3ha mound of Tell Jerablus Tahtani we found not only Uruk sherds littering the mound but also fragments of the enigmatic champagne cups. We were poised to add information about these ill-defined early periods, and to provide new insights on Carchemish itself. With the support of the Council for British Research in the Levant and other agencies, including the British Museum where some of the earlier Carchemish discoveries are kept, we embarked on rescue excavations at Jerablus Tahtani in autumn of 1991.
We identified five major periods of occupation, from pre-Uruk times in the mid 4th millennium BC when local peoples founded a settlement on a low Pleistocene terrace beside the Euphrates, up to the 12th century AD when a grand building was placed on the top of a mound that then stood some 16m above the flood plain. From the earliest levels we discovered that the first inhabitants did not settle just in the area that would become the tell of Jerablus Tahtani. Rather they were also based on the active floodplain immediately beside the Euphrates, one of the world's formidable high-energy rivers, presumably for direct access to riverine trade and communications. This lower settlement, which has been granted the title of ‘lower town’, was occupied for part of the 4th millennium BC, and apparently the very beginning of the 3rd millennium. It appears to have taken the form of a straggle of pits, middens (waste dumps), and other functional areas to the west and north of the mound Soon after the site was settled, a number of new features more at home in southern Mesopotamia some 900km downstream began to appear in the settlement. These include changes in agriculture, stone bowl-making, metalworking, architecture with highly distinctive multicoloured bricks, and especially in the pottery repertoire with shapes indicating new methods of transport, storage, cuisine, and beliefs.
This pottery includes the proliferation of coarse bowls, many of which were found in dense scatters just like Woolley had observed at Carchemish. Hundreds of these have viscous black bitumen adhering to their sides and rims. Inhabitants w
ere apparently engaged in processing this material which is known to be used for caulking boats. According to chemical analyses, it was brought to Jerablus Tahtani from as far away as southern Iran. Sealings, such as one bearing an impression of spirals, are another new feature suggestive of the administration of commodities.
Debate continues about the meaning of these intrusive features that occur widely along the great bend of the Euphrates. Who left them? Some say they point to the establishment of new colonies and enclaves in order to procure desirable raw materials lacking in the lowlands. They are therefore referred to as 'Uruk', after the great city in the south. Other researchers suggest the situation was more mixed, with southerners having to accommodate to developed political and economic infrastructures in the highland zone, and indigenous peoples adopting Uruk features.
The changes we noted at Jerablus Tahtani were abrupt but there was no destruction separating indigenous deposits from the overlying Uruk deposits. So how to explain this? One possibility is that Jerablus Tahtani was a small outpost of Carchemish, located where there was easier access to the river. After the Uruk phase, Jerablus Tahtani’s open settlement still retained long-distance contacts. For example, the link with South West Iran, and its great cities, may have continued since one sealing on a jar is closely similar to another from Susa in Iran.
However, during the early 3rd millennium BC, the outer ‘lower town’ then disappeared, and settlement was confined within the walls of the site itself. This settlement 'implosion' is known at a number of other Near Eastern sites, notably Tell Brak. This time, for Jerablus Tahtani, the change was not a peaceful process since the village was destroyed and a compact but massively walled fort was erected over its ruins.
The fort had a stone-based defensive wall enclosing an area of about 300 m2. It still survives some 3m in height. It perched on a 7m high mound, so it was an imposing brilliant white-plastered monument in the midst of the flat green valley beside the dark blue of the Euphrates. Abutting the interior were well-preserved rooms with buttressed walls, hearths, pot stands, plastered bins and stone paved entrances with stairs leading out to passages that sloped up to the unexcavated central core of monument. These are overtly domestic, rather than military, arrangements.
The whole fort was a pre-planned edifice that probably involved master engineers. Sloping down from the centre of the fort beneath passageways was a large, carefully built main drain. Its exit through the stone base of the fort wall was built as an integral component of the wall, so the entire arrangement of houses and communications inside the fort, together with its defensive wall, must have been conceived as an integrated unit. The unified concept demanded forward planning and a well-organised administration capable of mobilising considerable labour, one that was probably beyond the unaided resources of the previous inhabitants. Who then was responsible?
The most likely author of such concerted intervention was Carchemish, a city that is unlikely to have countenanced an independent, heavily fortified site on its doorstep. But who was it defending against? Walled towns and forts proliferated up and down the Euphrates, and although Jerablus Tahtani's walls are a little earlier than most, it would seem that ompetition within an increasing population played a part. Yet there was enough land available for agriculture. Territorial disputes, if they existed, must have been for different reasons than for arable.
One source of friction may have been prized access to riverine traffic. Another concerns the role of pastoral nomads, that is mobile tribal groups usually affiliated to valley settlers. Pastoralists are traditionally associated with this area, and with their immense flocks all targeting the valley at the same time after the barley harvest, tribal gatherings would have considerably swelled the population.
Soon after the fort was built it was decided to surround the hitherto free-standing wall with a massive 12m wide rampart or glacis (see plan). Since it blocked the mouth of the drain, a new drainage system had to be created. The whole interior of the fort was artificially raised with a fill some 2m thick over the first rooms. New buildings and rooms were then placed on the fill, and they in turn were levelled for installing ever higher structures. As a result, the raised fort now became a towering landmark in the valley.
When we know most about the fort, in its later stages, its interior was divided into official and industrial sectors. The official sector contained granaries. Barley was fed into rectangular silos along angled shoots cut through their walls. Lacking entrances, access into these silos was through the roof, as shown in granaries on seals from Mesopotamia.
In the manufacturing sector, concentrations of bobbins attest to textile working, crucibles and moulds to metalworking, and numerous querns, rubbers, pestles and spreads of charred grain to large-scale crop processing. The fort was situated in the highly fertile floodplain so it is not surprising that it became an agricultural focus. One of the seal impressions found at the site has a lozenge design almost identical with well known examples from Ebla at the time when that royal city claimed control of Carchemish. Around 2500 BC, cities like Ebla and Mari began to dominate large areas by diplomacy and warfare. Regional states emerged in Syria. It was a time of major political, social and economic change characterised by intensification of production as more attached specialists were needed to implement the legitimacy of these dynamic new states. Textiles and metal became especially important since the Ebla texts show that kings deployed them in quantity for diplomatic initiatives and expanding armies. The small site of Jerablus Tahtani was not immune from these inter-regional developments and the introduction of a manufacturing zone is probably best understood in terms of macro-political developments.
Deceased workers of the fort were now buried under floors and in abandoned passageways between houses. A variety of burial types existed: single and multiple inhumations in pots, cists, pits and chambers. Many are reburials but we have no idea where the initial burials were placed. Adults were accompanied by quantities of mass-produced pottery, necklaces and metalwork, and if these goods are a reflection of lifestyles, such workers had access to varied material possessions. In death, they were clothed in the manner of courtesans, with bead-festooned crossed pins that fastened cloaks, as depicted on shell inlays from the great city of Mari lower down the Euphrates. However, these burial practices did not prepare us for the grandeur of a communal tomb outside the gate of the fort.
The discovery of one of the largest tombs of 3rd millennium Syria at Jerablus Tahtani was surprising since the site was not a great urban centre but a small fort. It raises many questions about how we interpret the social and political roles of small sites at a time of state formation.
Situated beside a road and staircase leading into the fort, the monumental Tomb 302 consisted of a two-roomed rectangular stone structure covered by a mound c.15m by 7m and 2.4m high. This tumulus contained over 30 adults and children buried with approximately 100 tall pedestalled cups exactly like the 'champagne cups' Woolley found at Carchemish. As shown on a plaque from Mari, they are ornate ceremonial vessels
probably used in mortuary feasts. Ancient looters had overlooked pieces of gold, silver, rock crystal, ostrich egg shell, and ivory dagger pommels in crevices between the corbel-walled sides. In a final stage of remembrance, Tomb 302 became a memorial where offerings were made to the ancestors. On the fill that concealed the numerous bodies below mourners or worshippers had placed shaft-hole axes like those known from the Royal Cemetery of Ur, poker-butted spearheads, thin-bladed daggers, terracotta bull figurines, juglets laid in 'frying pans' and hundreds of bowls. Some 100 years separated the two phases of use of the monument, during which time satellite burials in jars were placed around it. Rarely do archaeologists find such dedications clearly distinguished from the burials themselves.
Much debate exists about world-wide climate change in the later 3rd millennium BC, one that may have led to aridification in the east of Syria. In the west, the Euphrates regime had become more unstable even before that date, exacerbated by soil erosion caused by over-exploitation of the land. The results in the valley bottom around Jerablus Tahtani were devastating. We found evidence in the much refurbished mound of Tomb 302 for exceptionally high and recurrent Euphrates flood waters. When the mound was damaged, caretakers patched it up again, giving us a precious record of inundations. Because of the height of these Euphrates intrusions in the mound, we know that flood waters must have washed several metres over the adjacent fields. Since floods most likely occurred during spring snow melts just before harvest, they must have devastated surrounding crops and perhaps rendered the site uninhabitable.
In any case, the fort was suddenly abandoned about 2300 BC, and Jerablus Tahtani lay deserted for some 1700 years. Perhaps the fort's inhabitants moved to better protected Carchemish, so helping to lay the very foundations of its greatness. If this was part of a more general pattern, we may infer that Carchemish expanded appreciably at the end of the 3rd millennium. Thereafter, Carchemish grew in importance, becoming one of the most vital centres in the Hittite Empire, and reaching its apogee around the 9th century BC. However, when Carchemish was ttacked and fell in the 7th century BC, Jerablus Tahtani was once again fitfully re-occupied for
more than a millennium.
After the initial abandonment of Jerablus Tahtani at around 2300 BC, other tells in the area continued to be occupied into the Hellenistic, Roman and Byzantine periods. At such sites it appears that the nucleated and apparently walled settlement mound remained as the main focus of occupation and daily life. However, the role of the tell appears to have diminished significantly during the 2nd and 1st millennia BC, and by the time the Seleucids were in power (from the 3rd century BC until the Roman conquest) the countryside underwent a progressive transformation.
Gradually tells were abandoned, to be replaced by lower towns, as well as the dispersal of small settlements and farmsteads across the landscape. At the same time building works including stone-lined water channels, large earthen canals, quarries, and roads also appeared in the landscape beyond the settlements. During the period of these later empires, when the administration of the area changed towards more remote administration from imperial capitals, it is likely that the rights to hold lands were fundamentally changed or re-negotiated. As a result, the countryside itself became more populated and 'busy' with a wide range of activities reflecting the new administrative order.
Overall, some of these changes must relate to the differing administrative role of Carchemish itself, thus despite its current isolation within the nomans- land along the Syrian-Turkish border, it is possible to infer the history of the site from the halo of activity around it.
Today, much of this landscape has been lost beneath the dam waters, but Tell Jerablus Tahtani has been spared. It stands precariously at the northern edge of the lake and, barring unpredictable high water rises, should remain, as Woolley described it in 1921 as the little tell… on the river front…
How grateful we are to this little tell for revealing its secrets to us.
Authors: Edgar Peltenburg is Professor of Archaeology at the University of Edinburgh. His research interests include small-scale society dynamics, archaic states and early technology. T. J. Wilkinson is Professor of Archaeology at Durham University. He is particularly interested in landscape archaeology and geoarchaeology. Both scholars have excavated throughout the Middle East.