by Georges Roux
Apart from a few interesting wall-plaques, some of them inscribed, and pieces such as the celebrated ‘Stele of the Vultures’ from Girsu, the sculpture of that period is represented mainly by statues of worshippers which once stood on the brick benches that ran around the cella of most temples. Usually upright but sometimes seated, their hands folded in front of their chest, these long-haired or bald, shaven or bearded men wearing the traditional Sumerian woollen skirt, and these women wrapped in a kind of saree were staring at some divine statue with their shell-and-lapis eyes set in bitumen – the eyes we have already encountered at Tell es-Sawman two and a half millennia ago. But these statuettes are not all of the same quality.12 Those found at Tell Khueira are rather crude and clumsy, those discovered in the archaic Ishtar temple at Assur are mediocre, and those, widely publicized, which come from the ‘square temple’ of Tell Asmar are stiff, angular and, with their huge, haunting eyes and their corrugated beards, more impressive than beautiful. In contrast, many of the statues unearthed at Mari are marvellous portraits extremely well-carved, but Mari is very far from Sumer and these remarkable sculptures cannot be regarded as being representative of Sumerian art: in all probability, they were the work of local artists drawing their inspiration from Sumer, the ancestors of the great sculptors of the Akkad period. What is strange is that
The oval temple at Khafaje, Early Dynastic III period. The two areas enclosed by walls measure 103 × 74 and 74 × 59 metres respectively. We do not know to which god or goddess this temple was dedicated. From P. Amiet, L‘Art Antique du Proche-Orient, 1977; after P. Delougaz, The Oval Temple at Khafaje, 1940.
the statues of worshippers discovered at Nippur and Girsu, in the Sumerian heartland, give the impression of being mass-produced and cut a sorry figure when compared with the masterpieces of the Uruk period. Were they made in workshops for ‘impoverished pilgrims’,13 or do they reflect the inevitable decline that seems to follow all exceptional periods in the history of art?
To be fair, it must be said that in other fields the art of the Early Dynastic period was far from being decadent. Thus, in a limited area of central Mesopotamia there flourished, in ED I, an attractive polychrome pottery called scarlet ware clearly derived from the Jemdat Nasr ware. On the other hand, many sites of the upper Tigris valley and the Khabur basin have yielded samples of the very elegant ‘Ninevite 5' ceramic, at first painted, then heavily incised, and remarkable for its shapes: tall fruit stands with pedestal bases, high-necked vases with angular shoulders, carinated bowls.14 The scarlet ware was short-lived, the Ninevite 5 ware vanished towards the middle of ED III after a very long existence, and both were replaced, throughout Mesopotamia, by an unpainted pottery with very few artistic qualities.
The art of the stone-cutter followed a contrary course towards improvement.15 The short and narrow cylinder-seals of the ED I period bearing monotonous friezes of schematized animals or geometric designs (the so-called ‘brocade’ style) were replaced, in ED II and III, by longer and wider seals with totally different compositions depicting either ‘banquet scenes’ or ‘animal-contest scenes’. The former showed men and women drinking from cups or from tall jars through a tube. The latter consisted of cattle attacked by lions and defended by naked heroes and bull-men. There were also some religious motifs, such as the sun-god on a boat. As time went by, the compositions remained basically the same, but they were executed with greater skill. Some seals, notably those of kings, were made of lapis-lazuli or other semi-precious stones, or even of gold, and they were sometimes capped with silver at both ends. An important novelty was the appearance, at Ur, Jemdat Nasr and Uruk, of the first short cuneiform inscriptions on cylinder-seals.
However, it is in metal work that the Sumerians made the most striking advances due to a great extent to the introduction of two new techniques: cire perdue (lost wax) for bronze and repoussé for precious metals. As we shall see in going through the marvellous pieces found in the Royal Cemetery of Ur, the Early Dynastic period was the time when the art of the goldsmith reached a degree of proficiency unequalled in any other contemporary civilization. But the raw material had to be imported and paid for with what southern Mesopotamia could offer: cereals, hides, wool, textiles, manufactured objects and bitumen. How, then, were the Sumerians organized to run their economy? What was their social structure? Who were their rulers and what can we know of their political history? To try to answer these questions (and some others) we must leave archaeology and turn to the few texts that are available.
The Sumerian City-states
Because our attention is now focused on Sumer we are tempted to forget what a small country it really was: thirty thousand square kilometres, a little less than the area of Belgium, about the size of four or five English counties. As life remained concentrated along the Euphrates, its branches and irrigation canals, the ‘cradle of civilization’ was in fact a fairly narrow strip of land extending from the latitude of Baghdad to the swamps that bordered the shores of the Arabo-Persian Gulf. In addition, a linguistic barrier, somewhere between Kish and Nippur, separated the Semitic-speaking people of the north (the future Akkad) from the Sumerian-speaking people of the south, making Sumer proper even smaller.
In the third millennium B.C. both Sumer and Akkad were divided into political units which we call ‘city-states’. Each city-state consisted of a city, its suburbs and satellite towns and villages, and of a well-defined territory comprising gardens, palm-groves and fields of barley and wheat. The open steppe between irrigated areas served as pasture land. The average surface of a city-state is unknown, but one of the largest, Lagash, is said to have measured some 2,880 square kilometres and to have numbered 30,000 – 35,000 people.
For the Early Dynastic period our sources do not list more than eighteen major cities in the whole of Sumer and Akkad. These were, from north to south: Sippar, Kish, Akshak, Larak, Isin, Nippur, Adab, Zabalam, Shuruppak, Umma, Girsu, Lagash, Nina, Badtibira, Uruk, Larsa, Ur and Eridu. But many other towns and villages, as yet unlocated, are also mentioned, whilst archaeologists have unearthed settlements – such as al-‘Ubaid and Abu Salabikh – whose ancient names remain unknown.
Each Sumerian city was formed of several districts, and each district had its own god with his temple. The city as a whole and its territory were under the protection of a ‘national’ god who ideally owned the city-state. Lagash, for example, ‘belonged’ to Ningirsu as its rival Umma belonged to Shara and Ur to the moon-god Nanna. This fictitious concept and the fact that the first administrative records available in large numbers came from a temple that of the goddess Baba in Girsu – have led to the hasty conclusion that all the land of the city-state was the property of the temples and that all its inhabitants were temple servants or clients. This might have been true for the Uruk period, but the picture that emerges from other Presargonic tablets now in our possession and from a careful reappraisal of old and new data by modern scholars is very different from the picture presented some years ago.16
It is now estimated that about one-third of the arable land surrounding the city was owned by the temples. This temple-land could neither be sold nor exchanged and was divided into three parts: the ‘land of the Lord’ (gàna-ni-enna), which fed the priests and the numerous persons employed by the temple; the ‘food land’ (gàna-shukura), which was allotted in small parcels to the farmers who worked the ‘land of the Lord’ and to some temple officials for their subsistence, but which did not fully belong to them and could be taken away at any time; and the ‘plough land’ (gàna-uru-lá), which was let out to tenants against one-seventh or one-eighth of the harvest. The temples also exploited or hired out orchards, pastures, fisheries, as well as cattle and flocks of sheep and goats. Taken collectively, the revenues of the temples in cereals, fruit, livestock and by-products were therefore considerable. They were partly used for the maintenance of the priests, scribes and other temple officials, partly stored as provision against drought and partly exchanged for imported goods. Probably th
e largest part, however, was redistributed as wages or gratuities to the thousands of people – mostly women, but also men and slaves of both sexes – who permanently laboured in temple workshops and premises, milling grain, spinning and weaving wool or hair, brewing beer or acting as cooks, gardeners and servants. Also paid in kind (usually barley) were the temple farmers, who could be mobilized by the ruler in case of war or for such large-scale public works as the building of sanctuaries and fortifications and the digging of canals. The scale of wages seems to have varied considerably from place to place and also with time.
All this required continuous planning, control and bookkeeping, but the Sumerians had meticulous minds and were extremely well organized. Not only did their ‘bureaucrats’ leave us thousands of payrolls, vouchers, lists of workers and other similar documents, but we learn from tablets found at Girsu and Shuruppak that members of the same profession were divided into highly specialized groups. For instance, there were separate shepherds for male and female asses and separate fishermen according to whether they fished in fresh, brackish or sea water; even the snake-charmers formed a ‘corporation’, which had its own chief. Artisans and merchants, similarly organized, worked partly for private citizens and partly for the state (temple or palace), though trade with foreign countries as far away as Afghanistan and the Indus valley was largely in the hands of the latter. An army of scribes, controllers, overseers and other officials, directed by chief inspectors (nu-bànda) and by superintendents (agrig) under the leadership of the high priest (sanga) of each temple kept this intricate machinery running.
The other single major economic unit was the palace, although much less is known about its role and administration. Tablets from Shuruppak, c. 2600 B.C., indicate that the ruler maintained six or seven hundred soldiers with their equipment – no doubt his own bodyguards and the regular army of the city-state – and employed people of various professions on his estates. We also learn from a handful of contracts that he occasionally purchased land from wealthy individuals or high officials. We have no means of assessing the total size of the royal domain, but if we include the possessions of the princes and their families, it might have been as large as the temple-land.
Finally, other contracts – more numerous, it is true, in Semitic Akkad than in Sumer proper – tell us that private persons of all ranks could freely sell, exchange, donate or let out houses, fields, gardens, fishery ponds, livestock and slaves belonging to them – or rather, perhaps, to family communities.17 Naturally, the area of the plots of land in private possession varied according to the social status of the owner: it could reach more than 240 hectares for a high official and be as small as 6 hectares or even one hectare for a simple civil-servant, a currier or a stone-cutter, all people attested as vendors in these contracts.
As for the social structure of the city-state, our texts mention only freemen and slaves, but it is clear from the tripartite economic system described that in Early Dynastic times the Sumerian society was divided into three main layers: at the bottom the slaves, usually recruited among prisoners of war or kidnapped in foreign countries but never very numerous; then those peasants and workers who served the temple or the palace, were maintained by them and possessed no land; and then the group of landowners or ‘freemen’, which covers the whole range from artisans to members of the royal family. And above all these, of course, the ruler of the city-state about whom more will now be said.
Early Sumerian Rulers
For the Sumerians, the ruler was the ‘shepherd’ chosen by the gods and responsible to them for the safety and prosperity of the city-state. On the archaic tablets from Uruk the ruler is called en, a title translated by ‘lord’ but implying both secular and religious functions. The en probably resided in the temple precinct, and it is reasonable to assume that he was also the high priest of the ‘national’ god, the head of the temple around which the Sumerian city had grown. This title persisted until the middle of the nineteenth century B.C., but in other Early Dynastic states the ruler was known as either ensi, ‘governor’, or lugal, ‘king’. Ensi is written PA.TE.SI, a compound logogram of uncertain meaning; lugal simply means ‘great man’. Why some rulers called themselves ensi and others lugal, or sometimes took both titles according to circumstances, is by no means clear.18 In some cases at least it seems that the lugal reigned over several city-states and that the ensi was the vassal of a lugal. The ruler's wife, known in any case as nin, ‘lady’, ‘queen’, played an important part in public life. In Girsu, for example, she managed the affairs of the temple of the goddess Baba.
The ruler and his family lived in a palace (é-gal, ‘big house’) distinct from the temple. Three such palaces have been excavated in Mesopotamia: one at Eridu, another at Kish and yet another – or rather two superimposed palaces built in succession – at Mari.19 They differed in a number of details but were strikingly similar in plan. All had a square central courtyard surrounded by chambers on three sides and communicating, on the fourth side, with a long, rectangular room which probably served as an audience hall. Two parallel thick walls separated by a narrow corridor surrounded the building. In Mari, the palace contained numerous ritual installations suggesting royal chapels. In Kish, a second building alongside the palace included a spacious hall with four central mud-brick columns and a pillared loggia.
The ruler governed the city-state on behalf of the gods. As most ancient and modern kings, he led his troops against the enemy, signed peace treaties and saw to it that fair judgements were rendered. One of his most sacred duties was the building, maintenance and restoration of the temples, in keeping with the
Meskalamdug's helmet (or, more exactly, wig) in massive gold, from the Royal Cemetery of Ur.
After Sir Leonard Woolley, Ur Excavations, II, 1934.
belief that humanity had been created for the service of the gods and that he was only the first of their servants. Numerous inscriptions refer to such building activities, and from Ur-Nanshe to Ashurbanipal several Mesopotamian monarchs have been portrayed in stone or bronze with baskets on their heads, carrying bricks for the new sanctuaries. Lugals and ensis also played a leading role in feasts, processions and other religious ceremonies. In Uruk, but also possibly elsewhere, the Early Dynastic ruler acted as the male god in the Sacred Marriage rite and, indeed, there is reason to believe that early in the third millennium B.C., in the days of Lugalbanda, Dumuzi and Gilgamesh – all qualified as ‘divine’ in the Sumerian King List – some royal couples were considered as ‘living gods’ or, more correctly, as human replicas of the divine couple to whom the city-state belonged. This, might be one of the answers to the many questions raised by the most startling discovery ever made in the course of Mesopotamian excavations: the Royal Cemetery of Ur.
A detailed description of the Royal Cemetery cannot be given here; it should be read in the excellent articles and books written by Sir Leonard Woolley on this fascinating subject.20 None but the discoverer himself could effectively convey the feeling of excitement that seized him and his team as gold literally oozed from the earth under their picks and as marvel after marvel was brought to light. None but this outstanding archaeologist could describe the delicate and painstaking removal, the patient and skilled restoration of the magnificent objects, ornaments and weapons that accompanied the dead: the golden vessels and daggers, the gold and lapis-lazuli statuettes of a ram ‘caught in a thicket’, the golden and silver bulls' heads which decorated the harps, the gold head-dress of ‘Queen Puabi’ formerly known as Shubad, and, above all perhaps, the splendid golden helmet of Meskalamdug – to quote only the main pieces. Woolley's dramatic evocation of these strange funerals where musicians with their harps, soldiers with their weapons and court ladies in gorgeous attire willingly followed their masters into the awesome pits where they were drugged to a painless death never fails to leave the reader with a poignant, unforgettable feeling of horror, mingled with wonder and admiration.
But the Royal Cemetery of Ur presents the histo
rian with very difficult problems. There is no doubt that it belongs to the dawn of history, to the period immediately preceding the First Dynasty of Ur (c. 2600 B.C.). It would seem at first sight that the people so lavishly buried could be no other than kings, queens and princes, but in the seventeen royal tombs where several inscriptions were found, most of them on cylinder-seals, two names only, Meskalamdug (‘The hero of the good land’) and Akalamdug (‘The son of the good land’), are followed by the title lugal, ‘king’, and two other names, those of Ninbanda, wife of Meskalamdug, and of Puabi, spouse of an unknown monarch, are qualified by the title nin, ‘queen’; and while the fact that all but two tombs had been plundered in antiquity might account for the absence of other royal inscriptions, this absence is nevertheless disconcerting. Even more puzzling is the practice of collective burials involving from three to seventy-four attendants, mostly female here – practically a whole royal household. It is attested on a smaller scale and mostly with male servants in other countries and in other times – in Egypt during the First Dynasty, among the Scythians and the Mongols, in Assam, and even among the Comans of southern Russia as late as the thirteenth century A.D.21 – but nowhere in Mesopotamia outside Ur and possibly Kish. Again, it can be argued that practically all the royal tombs in ancient Iraq were found plundered and that we have no written description of a royal funeral. Yet this silence about a ceremony which must have been of paramount importance is surprising and can only be explained by assuming that royal burials with human sacrifices fell into disuse at a very early date, probably during the Early Dynastic period. But why this sacrifice? The only text in our possession alluding to a king going to the grave with his retinue is, significantly, a Sumerian epic tale known as ‘the death of Gilgamesh’.22 Now, we know that Gilgamesh and Meskalamdug were but a few generations apart, and we also know from other sources that Gilgamesh was considered to be a god of the Netherworld. This would tend to confirm the theory first propounded by Woolley that Meskalamdug, Akalamdug, Puabi and the other anonymous kings and queens of the Royal Cemetery were more than monarchs: they were gods, or at least they represented the gods on earth and, as such, were entitled to take their court with them into another life, a life no doubt incomparably more enjoyable than that of the human being. However, this theory, as all others, is open to criticism, and the drama of the Cemetery of Ur remains a mystery.23
