CREATING THE FIRST CITY

Eridu is the Mesopotamian Eden, the place of creation. This is the beginning of a story that relates how the Babylonian god Marduk created the world:

This narrative is a myth of origin – how the world as Mesopotamian people knew it came into being, a charter for the notion of the city as a sacred place while, at the same time, the myth refers to a particular city, Eridu. The time before creation is described as an absence of all the characteristic features of civilized life as the Mesopotamians knew it. Amid the primeval sea, the first city, Eridu, and the great Marduk temple of Babylon, Esagila, are ‘made’ – or rather conceived through an act of divine thought that sparks off the process of actual creation. Just like the marsh dwellers of southern Iraq, who still build their huts on floating islands of reed, the god spreads mud upon a reed frame to fashion a platform. From this primordial, rather flimsy basis, the cities and their temples take their beginning. Henceforth the gods take up residence on earth and live in cities. And because the gods have the dwelling of ‘their hearts’ delight' in cities, Mesopotamian cities are always sacred.

Thus the Mesopotamian Eden is not a garden but a city, formed from a piece of dry land surrounded by the waters. The first building is a temple. Then mankind is created to render service to god and temple. This is how Mesopotamian tradition presented the evolution and function of cities, and Eridu provides the mythical paradigm. Contrary to the biblical Eden, from which man was banished for ever after the Fall, Eridu remained a real place, imbued with sacredness but always accessible. The strong local flavour of this mythical narrative, with its references to the particular conditions of the region, is striking and can only be understood if we consider the setting of Eridu.

Eridu is the ancient name of a site now known as Abu Shahrein. The etymology of the word Eridu is unknown and may well belong to a linguistic substratum of an early, pre-Sumerian culture. The Sumerians wrote it with the sign NUN, which looks like some kind of tree or even a reed. Eridu's geographical situation is unique. It is one of the most southerly sites, at the very edge of the alluvial river plain and close to the marshes: the transitional zone between sea and land, with its shifting watercourses, islands and deep reed thickets.² At the same time the western desert, stretching for many hundreds of miles and containing nothing but sand-dunes and rock-strewn wastes, is close enough to threaten the site and engulf it with sand. This placement meant that ancient Eridu had immediate access to the three widely different physical systems: the alluvium, the desert and the marshes, and hence to the different modes of subsistence: farming, nomadic pastoralism and fishing. Most importantly, however, the city commanded its own ecosystem, since it was built upon a hillock within a depression about twenty feet below the level of the surrounding land, which allowed the subterranean waters to collect together. This swampy place can still become a sizeable lake in the months of high water.³ The earliest Mesopotamian texts, from the early third millennium, underline the importance of this lagoon. In Sumerian this was known as the abzu (in Akkadian Apsû). In the almost rainless southern regions, the most obvious and crucial manifestation of water was the abzu. At Eridu, so the texts say, it surrounded the religious centre and became synonymous with it. According to the Mesopotamian notion of the cosmos, the earth was a solid, disc-like expanse within a huge body of water. Below the earth was the abzu, above the earth the sky formed a more or less impermeable vault holding back the upper body of water, which at certain places and times fell as rain through the holes in the sky's ceiling. Eridu was the centre of the cult for the god or goddess of sweet water.

The text quoted at the beginning comes from a cuneiform tablet written in the Neo-Babylonian period, some time in the sixth century BCE. Hormuzd Rassam found it among the ruins of Sippar;⁴ perhaps it belonged to the collection of a learned priest since it was written in both Sumerian and Babylonian. The creation story forms the introduction to a long incantation meant to be recited so as to purify the temple of Nabu at Borsippa.⁵ Although this particular text version is of relatively late date (first millennium), the tradition that Eridu is the oldest city goes back to the very first written texts from the late fourth millennium; Eridu already heads a list of geographical names. The Sumerian King-list⁶ begins with the following lines: ‘After kingship had descended from heaven, Eridu became (the seat) of kingship. Alulim reigned 28,800 years as king; Alalgar reigned 36,000 years. Eridu was abandoned, (and) its kingship carried off to Badtibira.’⁷ The antiquity of Eridu was a matter of traditional knowledge, repeated again and again until it became a fact, or a cliché, so far as the scholars of the Western world were concerned who read these Mesopotamian texts before the site was discovered. The story of the archaeological exploration of Eridu shows how the received wisdom of ancient sources came to be both refuted and confirmed.

DIGGING UP ERIDU

The mount of Eridu, called Tell Abu Shahrein, which lies just some 24 kilometres south of Ur, had been targeted for excavation as early as the mid nineteenth century. It is a typical cone-shaped tell, or ruin-heap, half a kilometre in diameter, rising some 25 metres above the plain. Six smaller mounds were dotted around in the vicinity, indicating that the population centre moved throughout time, perhaps in accordance with the lagoon's shifting shoreline.

First attempts at excavation were made by J. H. Taylor on behalf of the British Museum, in 18 54. The race was then on for all the newly created national museums of the Western world to be filled with artefacts from the most distant places and times for the greater glory of the imperial powers of the nineteenth century, symbolizing their command over time and space. In the mid nineteenth century highly crafted objects of great antiquity were particularly sought after – statues, reliefs and vessels made of precious metals – and any written documents in strange scripts were to be mastered and rendered intelligible through the advanced linguistic skills of Victorian scholars. The exploration of the great northern mounds in Assyria had already yielded carved stone reliefs, colossal bulls with bearded human faces and great quantities of cuneiform tablets.⁸ All that Taylor found at Abu Shahrein was a mound of mud brick, heavily eroded and compacted. Since at the time methods to trace mudbrick walls had not yet been developed, his workmen tunnelled through the solid mass of earthwork in a generally futile search for ‘antiquities’. Even the handsomely carved lion in black granite Taylor unearthed had to be left behind for lack of suitable transport.

The British Museum meanwhile continued to be filled with the popular Assyrian artefacts and Abu Shahrein was again left to the local foxes and jackals. In 1918 Campbell Thompson and in 1919 H. R. H. Hall explored the site for the museum yet again, but these attempts likewise produced few results. By the end of the Second World War in 1945 the colonial era of archaeology was over and the newly created Directorate General of Antiquities in Iraq chose the so far unpromising site as the target for the very first full-scale Iraqi excavation project. The symbolic significance of this site played an important role as it was to bring new evidence ‘for the strong thread of continuity which runs throughout the past of Iraq’.⁹ The new Iraqi nation, the first Mesopotamian state to enjoy political self-determination after centuries of Ottoman and European rule, marked its new beginning with the recovery of the most ancient of cities so as to reaffirm the continuity of the local culture. It was also significant that the secular government chose a pre-Islamic site in an attempt to make its people proud of their ancient history and to help foster a sense of national, non-Muslim identity.

Work at the site began in 1946, with the Iraqi Fuad Safar as field director and the British archaeologist Seton Lloyd acting as technical adviser to the Department of Antiquities. The team spent three seasons until the end of February 1949.¹⁰ From the beginning, the aims of the excavations were to produce a comprehensive and systematic exploration of the site, using the latest archaeological methods. The site was considered ‘representative for the earliest epoch of human habitation in the drying Mesopotamia… as the setting for the evolution of Sumerian civilization…’,¹¹ but the main political concern was to substantiate ancient claims for the site as a place of origin, not just for the Sumerian but also by implication for all of civilization, not least for that of the Middle East.

The Iraqi team began work on the ‘acropolis’, where the remains of a large stepped monument or ziggurat very similar to the one at nearby Ur¹² were clearly visible. This structure could be dated with the help of some inscribed bricks to the kings Ur-Nammu and Amar-Sin, rulers of the Third Dynasty of Ur (c. twenty-first century). Beneath one of the corners of the ziggurat workmen discovered walls of a much earlier edifice, which could on the basis of potsherds be dated to the late Ubaid period (c.3800). The term ‘Ubaid period’ derives from the Tell Ubaid site near Ur excavated by Sir Leonard Woolley in the 1920s. It refers to the Chalcolithic cultural levels of southern Mesopotamia, overlapping with the Halaf levels well known from northern Mesopotamian sites.

The Iraqi archaeologists decided to remove this late Ubaid layer to see if any earlier remains could be found underneath, and ended up with another sequence of eighteen occupational levels. At the lowest level, ‘on a dune of clean sand’, they discovered the very first building, ‘a primitive chapel’ no larger than 3 metres square. It contained a pedestal facing the entrance, and a recessed niche. The proposed date for this ‘chapel’ is 4900 or Ubaid level 1. The whole building, and all those of the subsequent layers, were made of sun-dried brick. This was an interesting feature in light of the fact that the typical Ubaid dwellings in the region, known from Woolley's excavations, were made from reeds, in a form of construction still used by the Marsh Arabs today: tightly wrapped, slim bundles of tall reeds serving as poles to form a framework covered with reed mats. Given that this was the traditional construction technique in the south, it was remarkable to find so much brick architecture at the earliest levels at Eridu, with all the important structures built in brick. The Ubaid culture is generally not well documented in southern Mesopotamia,¹³ so the material recovered at Eridu helped to modify the impression, gained from Woolley's excavations, that the Ubaid in the south was a poorly developed primitive ‘village culture’.¹⁴

The sequence of ‘temples’ showed they had been built on the same sites for hundreds of years. When the first ‘chapel’ of level XVII had fallen into disrepair, a new one was built directly on top of the ruins of the older walls. The new building was practically a reconstruction of the earlier one, with the addition of more sophisticated features, such as interior plastering. At the next level the plan was changed from an approximate square to a rectangle, nearly double in overall size. Then, at level xiv, the whole building had been levelled and all the earlier ruins filled in with sand and enclosed with a brick wall, to provide the basis for a new building elevated from the surrounding areas by 1 metre and accessed by a ramp. This showed features that had much in common with other monumental buildings of the time discovered in the north of Mesopotamia. A spacious central chamber (4.50 metres wide and at least 12.60 metres long) was surrounded by smaller rooms. Also characteristic was a rhythmical articulation of the walls by niches and buttresses, and a clear symmetrical layout.

The same layout was repeated when the building was reconstructed in the next two levels (x and ix). Then came a complete change in plan and general character. The walls became much thicker (70 centimetres) and the building covered much more ground. It had a central nave flanked by subsidiary side rooms. The walls, inside and out, were heavily corrugated, which made the high and dense walls more stable and produced a pleasing aesthetic effect as the play of light and shade created regular vertical patterns. Within the main chamber were several false doorways, one revealing a niche containing a spouted vessel full of fish bones. Buried beneath the pavement of the nave, directly below the ‘altar’, were a number of curious snake-like clay coils some 30 to 40 centimetres long. These may have had something to do with an underworld cult.¹⁵

At level VI the building became truly monumental. The walls of the previous building were once again levelled, to a height of 1.2 metres, and the space between them was packed with mudbricks. This formed another platform, extended considerably, to make room for the new structure. The denudation of the mound meant that not much of the plan could be recovered, but the main sanctuary was a very long room, measuring 14.40 by 3.70 metres. Twin doors were set into the shorter wall at each end. The podium was plastered, its surface burnt dark red. It was covered by a thick deposit of ashes, which also lay scattered all over the floor. Mixed with the ashes were large quantities of fish bones, as well as bones of smaller animals. Such remains extended over the whole of the north-eastern end of the chamber. The state of the bones indicated that the fish had been eaten, and curiously the remains had not been cleared out but were kept within the building. A small subsidiary room completely filled with ashes and debris to the consistency of clinker suggested that the fish remains and other offerings were burnt there.

After the collapse of temple vi preparations were apparently made for building yet another temple on top, at a higher level but more or less to the same plan. Only traces of subsequent buildings (temples v–I) remain because of extensive levelling at the end of the second millennium in preparation for the ziggurat. But since temple VI had been very dilapidated when the restoration attempts were made, it may well have been the last of the Ubaid ‘temples’ at Eridu.

From the time of the humble beginnings of the first ‘chapel’ to the huge and sophisticated temple vi more than a thousand years had passed. With hardly any interruption, one building had followed another on the very same spot, each succeeding one becoming larger than the one before, and as a consequence of carefully levelling the earlier remains the platforms grew higher and higher. A reconstruction by the Iraqi architects showed the last Ubaid ‘temple’ to be raised high above the ground, elevated by the successive layers of previous buildings. The interior furnishings, the platforms, niches and podia, also show a remarkable continuity of form and perhaps of purpose.

The archaeologists have routinely called these structures ‘temples’, a word signifying a place dedicated specifically to sacred ritual. It is not at all clear whether the Ubaid ‘temples’ were exclusively used for worship or seen as dwellings of the gods, as in later periods. What is certain is that they provided at least periodically a setting for particular activities. As has been pointed out, they were all filled with ‘ordinary rubbish’, small bones, unremarkable pottery and so forth. They were not apparently kept spotlessly clean or reserved for one specific purpose. Few objects of value were recovered: some clay figurines of naked male and female humans, small animal figures, a model snake decorated to look like one of the serpents still common around Abu Shahrein and the buried clay snake coils mentioned above. Otherwise there were only some beads of semi-precious stones, stone tools, obsidian blades, spindle whorls of clay and a copper axe.

It is clear that the buildings not only became progressively larger and better built with time, but that the interior arrangements – the pedestals, false doorways, symmetrically arranged niches – seem to have had symbolic as well as functional characteristics which point to ritual rather than mundane purposes. These late Ubaid temples may have anticipated the Mesopotamian temple architecture of the historical periods some 1,500 years later. Their plan, a central chamber flanked by side rooms with a free-standing podium in the middle, became a standard feature, as was the façade decorated with pilasters and niches. The custom of burying votive gifts and ritual implements was also never abandoned. Most significant perhaps was the idea of sealing the remains of earlier structures and their contents (to preserve their sanctity?) and then erecting the new building on top of the levelled ruins. Thus most Mesopotamian temples enshrined the substance of older temples and the very platform they stood on was made venerable by the accumulated sacred debris. As such they became a visible sign of continuity and antiquity.

It must be remembered that, especially in the southern plains, the features of the landscape were not as permanent as in other regions. The rivers would shift their beds, inundations destroyed cultivated areas, sand dunes encroached on deserted villages within weeks – all factors that threatened the desire for permanence and anchorage to a place. Only the cities, and especially the ‘brickwork’ of the temples, persisted throughout the ages and grew like living beings. This process can first be observed in Eridu. In the Babylonian narrative, the platform arising from the Apsu became the first seat of the gods; in the archaeological sequence, the simple hut-like structure on the sand was continually built over to become one of the most venerable sanctuaries in the country. Eridu's reputation in later times as the first shrine was amply justified by the archaeological sequence.¹⁶

THE SÈVRES OF THE CHALCOLITHIC

The archaeologists did not dig only on the main mound at Abu Shahrein, but also at several other, subsidiary sites. They discovered a small part of the residential quarters and an extensive cemetery. In all these areas, as well as in the ‘temples’ on the main mound, they found potsherds. Potsherds are always important in excavations, since they are invaluable for dating purposes. Other than that they are rarely particularly exciting. The pottery of the fifth millennium, however, is not only interesting for chronological reasons but is also very beautiful. Besides this it tells us something about the socio-cultural organization of the age. Both the northern Halaf and Samarra ware and the southern Ubaid and Hajji Muhammed ware – all named after their initial places of discovery – are instantly recognizable: they are thin-walled, mainly creamy in colour and decorated with a simple, striking elegance unparalleled in the whole of the prehistoric Near East. The patterns vary from freely applied dots and strokes to intricate geometrical hatching and zoomorph and human figures. The repertoire of shapes includes beakers, round and oval plates, bowls, spouted vessels, large and small jars, goblets, thin cups and coarser cooking ware.

Such high-quality pottery was made by skilled specialists at different centres of production; the pottery discovered at the early levels of Eridu (from the earliest in level xix to level xiii) was locally manufactured. However, the distribution of the painted pottery went far beyond the areas of production. This pottery was no mundane, all-purpose household ware but an élite product; it needed special places for storage and display, was not easily transported and unsuitable for a nomadic way of life for which containers of unbreakable materials were preferred. But sedentarism and pottery go well together and it is conceivable that elaborately produced and decorated clay vessels may have contributed to popularize and proclaim the values of settled populations. Of course, they also have their practical functions since they were used for eating and drinking and serving food. Ethnographic accounts have demonstrated both the communicative purpose of decorated objects and the importance of collective feasting, sometimes called ‘conspicuous commensuality’ – a precursor to ‘conspicuous consumption’. The term stresses the importance of the group bonding through shared meals, elaborately served and displayed. Such communal feasting strengthened the status of those who were able to procure the wherewithal, from the comestibles to the precious dinner service.

Anthropologists have argued that the emergence of prestige goods and precious tableware is connected to early forms of hierarchy – in which one group has secured access to territory and goods and is responsible for distribution.¹⁷ The culture of the Chalcolithic (or Copper Age) in the Near East, of which the Ubaid is the southern manifestation, was characterized by increasing sedentarism, horticulture and exchange. Painted pottery from the latest phase (c.4000–3500) can be found all over Mesopotamia, Syria, western and southern Iran, Anatolia, and along the Persian Gulf. This points to a lively network of exchange throughout the region, which was to become even more intensified and organized during the subsequent age known as the Uruk period.¹⁸ At Eridu the earliest settlement levels, built directly on the virgin sand, already have the Chalcolithic pottery typical of the southern Ubaid culture. Whoever decided to settle there was familiar with the technology and decorative repertoire.

As demand for such pottery increased, so the professional expertise developed; temperatures could be regulated more evenly and some form of tournette was used to produce even more regular shapes. Locally produced pottery dominates in the earlier levels, enabling archaeologists to speak of ‘Eridu ware’. Later all evidence for local manufacture ceases (after level xiii), for reasons unknown. The various floor levels of the ‘temples’ were strewn with Ubaid potsherds, though occasionally whole vessels and jars were recovered. This may suggest that the ceramics were used for actual meals, not just to bring offerings. If we remember the large quantities of ashes and of fish bones as well as the animal remains so common around the pedestals, it is tempting to see the consumption of food as an important part of the activities taking place. However, some types of pottery were only found in the ‘temples’ and may have served a more clearly defined ritual character. Little vessels, no more than 15 centimetres in diameter, with very thin walls and very elaborate decorative schemes, were sometimes found in clusters, one nestling inside the other (typical of levels x and xi). Also noteworthy are the so-called tortoise vessels, a speciality of Eridu. These had long spouts projecting diagonally from the shoulder and were also abnormally thin and fragile. One of them, filled with fish bones, survived in a niche behind the altar of temple viii.

Particularly striking are the ‘censers’ common among the debris of the later ‘temples’. The pierced triangular window openings could have allowed the smoke of incense to escape. The placement of the openings suggests doorways and the whole decorative scheme makes them look like miniature buildings – though ones more reminiscent of reed than of brick architecture.¹⁹ Figurative terracottas were also found among the ‘temple’ debris, including fragmentary (female?) figurines, with the lower body covered with some sort of dress whose patterns are painted on with wavy stripes in black. It is unclear whether they represent any particular type of person or function or a deity. Other anthropomorph statues from the Ubaid cemetery levels are all nude.