Notes on the Social Structure of Third Millenium BC Sumer
March 10, 2015 – 9:37 pm
The People (sag-giga)
The principal identity of the people seems to have been with the city itself. One sign of this is the tendency to take names incorporating the name of the city.[1] There is no evidence that alternative identities actually opposed to the city played any significant role in Sumerian history. The idea that there might have been a ‘racial’ distinction felt between semitic-speaking Akkadians and their Sumerian compatriots was once conjectured, but is not now credited. There is no evidence at all of such super-city identities as the oligarchs and democrats in the Greek cities, with the dire effects that that often had. On the other hand, they were always keen to distinguish themselves from the (semi-)nomads that surrounded the civilised areas and frequently appeared within it. (The myth of the Marriage of Martu gives evidence of their attitude.) There were, however, many other normal divisions that might appear within the city population itself. These divisions do not seem, for the most part, to imply inimical relations more than usual.
Kin
Tribe
If the Sumerians had ever been organised tribally those structures had disappeared over the millennia of their urban existence leaving no trace. We can be fairly confident, however, that nomadic peoples who settled in the cities would retain some degree of tribal identity at least. This would especially be the case in the north of the land where the Semitic and recently nomadic element was significant and where the pressure to conform to Sumerian norms would have been less.
Clan
The evidence of land sale practices from this period, and also as seen on the Akkadian ‘Obelisk of Maništusu’, suggests that the extended and augmented family was an important economic unit. This implies, of course, that it was also an important social division. The evaluation of the precise degree of its importance in this period depends largely upon whether we accept that housing enclosures seen in the north were standard in the south. The evidence for this is not very strong, and it does not improve as we look into later periods. Records of house sales in Shuruppak may suggest that neighbours were relatives, while studies of documents found in adjoining houses in OB Ur suggest quite the contrary.[2] If it is accepted then it is likely that the population of the city was sharply divided into family communes or hierarchical extended families. This has been argued for by Diakonoff[3], and Jacobsen has even suggested that the institution of lu-gal is a development of the rôle of the clan head and that the é-gal is a bloated version of his extended household.[4]
Associated with the family core in such an organisation of society would also be a more or less extensive periphery of dependants. This would include cadet members of the family, slaves, retainers, persons of no family who would seek incorporation as a form of social insurance, and many others. Not all of these would need to be living in the actual enclosure.
Family
As with us, the nuclear family appears to be the basic social unit. Men and women at this time were more approximately equal than at almost any other time in the history of this area. Men were the heads of their families but women had property and business rights, and could qualify as witnesses.[5] Marriage was theoretically a business arrangement by the parents involving a money gift, the ni-mi-usa, to the bride’s father to cover the cost of the wedding.[6] It could also involve a written contract. This aspect of marriage may lie behind the proverb ‘A joyful heart – the bride; a sorrowful heart – the groom.’[7] Nevertheless, the tone of the literature in general suggests that marriages were also contracted for reasons of the heart. The tales of Inanna show two lovers determining their own marriage and a proverb also states ‘For his pleasure – marriage; on thinking it over – divorce.’[8] Divorce seems to have been relatively easy and equitable. Children are, of course, supposed to be absolutely submissive, but isn’t this always the case – and literature indicates that it was no more realistic an ideal then than in most other times.[9] Wet nursing and adoption were common.[10]
Imrua
There is an institution of im-ru-a which we hear mentioned but not explained.[11] A tablet of ED Shuruppak, for example, speaks of 539 boys from 7 imru, and Gudea of Lagash began the E-ninnu temple with labourers from 3 imrua. In the Lagash case the imru were all represented by a standard (šu-nir) named after a major deity: Ningirsu – ‘the king smiting the foreign land’, Nanshe – ‘the pure bow of the ship’, Inanna – ‘the rosette’. Such a scheme of names may indicate that these groups were associated with the temples of those gods, and organised through them; but the term seems to mean ‘family,’ so it may have referred to a clan organization within the city.
Classes
The population of the city seems to have become much more stratified in this period. We have seen this reflected in house sizes, records of indebtedness, and, most strikingly, in the Royal cemetery at Ur. The researches of Diakonoff[12] indicate that the population was divided into four categories; nobility, commoners, clients, and slaves. Others are inclined to see the distinction as a gradation of liberties and to use the vocabulary of free, semi-free and slave. In any case, these distinctions were not those drawn by the Sumerians themselves, and we sometimes have difficulty matching their vocabulary to our functional vocabulary. Partly this is because the meanings of the relevant Sumerian terms changed over time, and partly because our main clue to class differences below the nobility is in the different kinds of ration schedules that are applied by the temple or palace to their workers according to their status.
First Class
The nobles and commoners of the first division are the free people of the second. At most times the greater part of the population of the Land were formally free. There seems to have been no other distinction between nobility and commoner except wealth. Kramer suggests that the two groups were the source of members for the assemblies of ‘old men’ and ‘young men’ respectively, but this is merely a guess. In any case, to be free is not to be without formalised obligations to the society: taxes had to be paid and there were doubtless other duties required. Amongst the classes that we hear of as free are the officials and the engar (peasants.)
Most free citizens were eligible for corvée for the temple or palace, for which they received rations and sometime special supplements according to the standard ration system originally established for the temple. But these rations, even when supplemented, could surely not have sufficed for a healthy diet, since, as we have seen, the food rations were mostly cereals. We must assume that the recipients of these rations, and the citizenry in general, also had other sources of food and goods. Unfortunately we know almost nothing of this domestic economy.
We do know that essential to the status of freedom was the ability to provide for oneself (and family.) Thus for most of the period, free citizens must have had access to the usufruct of the land, whether as a rent or as a freehold farmer. Other possibilities are, of course, craft industries or trade or employment as an official with whatever recompense that brought. It was not until Ur III that we begin to see hired labour, lú-hun-ga, working for wages, á.
Second Class
The clients were of three sorts:
- Higher dependents of the temple, like administrators and some craftsmen.
- Lower dependents of the temple who were much more numerous.
- Dependents of the nobility.
The typical terms at this time for adult serfs were guruš for the males and gemé for the females. They were generally occupied in the agricultural or manufacturing sectors. They formed the personnel (gìr-s?-ga) of households – be they temple, palace, or private – which might specialise in certain industries, such as weaving, milling, etc. Those who specialised in certain crafts were the giš-kin-ti of that household, working in special workshops, and we know that they could follow various trades, such as smith, carpenter, leather-worker, reed-mat maker, upholsterer, mason, potter, fuller, shipwright, and so on.[13]
In ED Lagash a dependent class existed called the lu-kur-dab-ba. Of this class a subset were eligible for corvée or military service. These were called the šub-lugal, ‘royal subjects’. This class seems to have been very extensive and included most of the able-bodied population. State service was required in monthly turns, though in times of stress the working period scould be extended. It was performed under institutional control in groups of 10 (or multiples thereof.) Workers were then said to be ‘serving their turn’ (erín bala guba,) and while they were so employed they would receive a ration. When the workers of this class were ‘sitting out their turn’ (bala tuša) they received á, a wage.[14]
In the early period the term erín was applied to generally foreign male prisoners of war who survived their capture and were then set to work alongside the native guruš in the agricultural sector.[15] By the time of Ur III the term was used more generally to refer to the class formerly known as šub-lugal. In references to the erín class, their dual soldier/worker role is emphasised by the coincidence of terms for their organization: for example, nubanda as captain or overseer, and ugula as sergeant or foreman.[16]
Others sections of the dependent class worked for rations for greater parts of the year. The distinctions here are as yet unclear, but we see evidence that there were two kinds of personnel working for any household: those like the male igi-nu-du8 or the íl, who were rationed each month of the year, and those who were not.[17]
Third Class
Slavery existed but seems to have had no great effect upon the economy or political structure.[18] In fact, the prevalence of slavery is much debated. Male slaves were arád and females were gemé, but female serfs were also called gemé. Thus many apparent references to female slaves may only refer to serfs.[19] The inscriptional etymology of gemé (from ‘mountain’ + ‘woman’) suggests that they were very often prisoners of war. The slaves arád and sag also seem to have been booty of war whose status, however, seems often to have been merely temporary until they could be properly integrated into their new homeland.[20] In other cases we know that slavery could be the result of one’s own indebtedness or of having been sold off by one’s guardian to repay their debts: children could be sold into slavery by their parents, or mothers could be sold by their sons. This seems to account for about 2/5 of the slaves known, and would become more prevalent in the age that followed Ur III. Slaves might be marked by a special hair style – perhaps a tonsure – or by a fetter if working outside. [21]
Slaves are known to have been held in private houses where they did domestic chores for the family, but they played a more economically significant role in the larger institutions attached to the temple or palace where more tractable females were employed in quite large numbers. At many periods it was possible for slaves to accumulate the price of their release, so some personal space was allowed to them. Indeed, the fact that the relationship of the ruler to the god, and of the people to the god, and so on was taken to be a trope expressing their mutual obligations, indicates that the life prospects of a slave were not as dire as the term suggests to us today.[22] Slaves had no actual rights to good treatment by their owners, but nothing is said of brutality being used on them. Most male prisoners of war were, unfortunately, either killed on the battlefield or blinded to make them harmless. Such men could still be useful as drawers of water or as workers in orchards apparently, but they could hardly aspire to an independent life. With that exception then, slavery for most seems to have been merely an unpleasant stage in one’s life with independence by manumission or by self-purchase at the end of it. Perhaps for this reason, we hear of few cases of slaves running away.
[1] vdMieroop:C, p. 43.
[2] Postgate, p. 91; vdMieroop:C, p. 105.
[3] Maisels, p. 191 (ref. to Diakonoff:RDS, p. 179.)
[4] Maisels, p. 158 (ref. to Jacobsen:EPD.)
[5] Kramer:S, p. 78.
[6] Hawkes:FGC, p. 207.
[7] Kramer:HBS, p. 124.
[8] Kramer:HBS, p. 124.
[9] Kramer:HBS, ch. 3.
[10] Hawkes:FGC, p. 212.
[11] Postgate, p. 80; vdMieroop:C, p. 105.
[12] Kramer:S, pp. 78 f. (ref. to Diakonoff:SSS.)
[13] Gelb:RS, p. 242.
[14] Maisels, p. 146 (ref. to Maekawa.)
[15] Maisels, p. 150 (ref. to Gelb:PW, 94-6, 83.)
[16] Maisels, p. 146 n. 4 (ref. to Gelb:PW, 84.)
[17] Gelb:RS, p. 241.
[18] Redman, p. 303; Oppenheim, p. 74.
[19] Maisels, p. 149 (ref. to Gelb:TS, p. 92.)
[20] Maisels, p. 149 (ref. to Gelb:PW, p. 91.)
[21] Saggs, pp. 169f.
[22] Oppenheim, p. 75.