In 1964, the noted American Assyriologist, A. Leo Oppenheim, published his influential Ancient Mesopotamia: portrait of a dead civilisation. ‘Nah ist – und schwer zu fassen der Gott’, the title of Chapter 4, was taken from the well-known verse in Friedrich Hölderlin’s hymn, Patmos, written in 1803. The first section of this chapter was provocatively entitled, ‘Why a ‘Mesopotamian Religion’ should not be written’ and, depending on one’s point of view, it is either extremely dreary reading, 50 years after it was first published, or a call to arms to try to wrest some semblance of meaning from an intractable body of evidence that Oppenheim clearly found very frustrating.

Oppenheim began his defence of this very negative position by insisting upon the impossibility of deriving insights into what people actually thought and felt from the physical arrangements of walls, courtyards, passageways, gateways and rooms ‘built to display the power and wealth of the deity and to harbour and protect its staff and its treasures’. As he noted, by way of comparison, ‘if the monuments of Western Christianity were preserved for some distant and alien generation, or a visitor from outer space, what could they possibly reveal of the essential tenets of their faith?’ Oppenheim was equally dismissive of the utility of studying iconography for, he felt, ‘Not even a perfectly preserved image could indicate to us what it meant for the priest and the pious, how it functioned as the centre of the cult, what its Sitz im Leben was for the community.’

Next Oppenheim turned to three types of cuneiform sources that he felt were relevant: ‘prayers, mythological texts and ritual texts’. Prayers, Oppenheim felt, could only be understood in connection with the rituals, during the performance of which they were recited. Yet while he admitted that prayers ‘convey something of the mood and the emotional climate of Mesopotamian religion’, they did ‘not contain much information’ on important topics, such as ‘the individual in relation to spiritual or moral contexts of universal reach, the problem of death and survival’, or ‘the problem of immediate contact with the divine’.

Oppenheim then turned to the subject of myth, which he characterised as ‘stories about the gods and their doings, about this world of ours and how it came into being, these moralising as well as entertaining stories geared to emotional responses’. These sources were dismissed as little more than ‘a fantastic screen, enticing as they are in their immediate appeal […] but still a screen which one must penetrate to reach the hard core of evidence that bears directly on the forms of religious experience of Mesopotamian man’. And he continued, ‘All these works which we are wont to call mythological should be studied by the literary critic rather than by the historian of religion.’

Finally Oppenheim turned to ritual texts, namely ‘descriptions of specific rituals to be performed by priests and priestly technicians in the sanctuary’. Of the insights to be gained from these sources, Oppenheim was unreservedly pessimistic. ‘Consider what kind of information the codifications of the church rituals […] would impart two or more millennia from now to scholars from a completely different culture, who would be able to understand them linguistically only in the very imperfect way in which we understand cuneiform texts.’

Oppenheim’s views typify what I have often encountered when studying the work of archaeologists concerned with the mental, ideological landscapes of antiquity, rather than the physical ones. To them, I have often said, if not in so many words, that archaeology can tell us many things, but there are entire domains on which it simply cannot shed any light, no matter how earnest or enthusiastic its practitioners and no matter how sophisticated the scientific techniques employed by them may be. To borrow Wittgenstein’s phrase, ‘Wovon man nicht sprechen kann, darüber muss man schweigen.’ (Whereof one cannot speak, thereof one must be silent.) In Oppenheim’s case, I would simply suggest that he was looking for the sort of insights that the available evidence is incapable of providing. However, rather than lamenting the opacity of our sources, I would rather focus on what the documentation at our disposal does reveal.

By way of example, I have chosen to concentrate on the site known to the Sumerians as Uruk (biblical Erech, Greek Orchoë, Arabic Warka), in what is today southern Iraq. One of the first great cities in Mesopotamia and the locus of writing’s invention around 3400 BC, Uruk at its height covered an area in excess of four square kilometres. According to the Sumerian King List, Uruk’s nine-km-long, mudbrick city wall, built around 2900 BC, was the work of Gilgamesh, the fifth king of the so-called First Dynasty of Uruk. Following two exploratory seasons of excavation at Uruk in 1850 and 1854 by the English geologist William Kennett Loftus, German archaeologists conducted a further 39 campaigns between 1912 and 2003, concentrating on a large complex of monumental buildings (including the so-called Limestone Temple, the Mosaic Court, the Square Building, the Stone Cone Mosaic Temple, Temple C, Temple D, the Riemchengebäude, the Pillard Hall and the Great Court) known as Eanna – literally, the ‘house of heaven’ – to the exclusion of all but one other area (the White Temple) of early 4th millennium BC occupation. Collectively, the buildings of Eanna constituted the domain, or better yet, the demesne, of the city-goddess of Uruk, Inanna.

In addition to yielding an enormous amount of information on early architecture, the Eanna excavations resulted in the discovery of nearly 4,000 Archaic texts, the earliest examples of proto-cuneiform known from Mesopotamia and the earliest examples of writing found anywhere in the world. Even though these tablets were recovered from trash deposits, dumped like archived tax documents we might dispose of when it is no longer necessary to retain copies, their importance cannot be overstated. The Archaic texts were first studied by the great German Sumerologist Adam Falkenstein, who published a preliminary sign-list in 1936. Subsequently, work was resumed by Falkenstein’s student, Hans J. Nissen, in 1964 and a major research effort based at the Free University (Berlin), the Max Planck Institute for the History of Science (Berlin) and, later, UCLA (Los Angeles), involving Nissen, Margaret Green, Robert K. Eng- lund and the late Peter Damerow, has made enormous progress in understanding these texts.

What was writing used for and why did it appear in the middle of the 4th millennium BC, within the great architectural complex of Eanna, the goddess Inanna’s household, at Uruk? The earliest texts from Uruk were purely numerical; but they soon became more complex, combining nouns with numbers and compound signs, such as a bowl and a stylised human head, meaning ‘ration’. Complex systems of counting and metrology, as well as time-reckoning, were in use throughout these texts, some of which chart the growth of herds, from sheep and goat to swine and cattle. Many texts keep track of rations given out to labourers for different kinds of services performed, while a sizeable number are lexical texts. These are lists of words grouped by semantic category, such as names of professions, trees (and things made of wood), birds, cattle, fish, metals, containers and so forth. Although in no sense meant to serve as a dictionary or thesaurus, such texts are important witnesses to the regime of scribal education practised from the outset by the inventors of writing and the first generations of Uruk’s scribes.

What do these texts tell us? First and foremost, the Archaic texts confirm that writing was invented to facilitate accounting and record-keeping. It was an almost inevitable outcome of an economic system that was becoming increasingly complex as time wore on, one that involved far more variables than any scribe could possibly keep track of. Writing was not simply helpful, it was indispensable. It rapidly became a tool that was vital for administrators seeking to track income, expenditure, loss and productivity over the course of days, months and sometimes years. This was the context in which writing was invented.

Some of the physical elements of Eanna’s economic system – sheep, goats, cattle, barns, weaving scenes, construction, shrines marked by divine symbols, prisoners of war and ‘rulers’ in action – are illustrated on numerous cylinder seals and, most famously, on a 1.05m tall limestone vessel discovered at Uruk in 1929, which dates to precisely the same period as our earliest texts. The exterior surface of the Warka or Uruk Vase is sub-divided into registers, showing the agricultural and pastoral products that were brought to Eanna and registered. Those involved were paid, i.e. they received rations, in return for their labour and the produce delivered, by the chief administrator of the entire operation, who was, in effect, the steward of the estates of Inanna, goddess of Uruk. In other words, the elaboration of the accounting procedures which we see in action on the Archaic texts was the work of those administrators – some might call them priests – who managed Inanna’s estates, Eanna and all of the associated farms, vineyards, cattle yards, milking sheds, fisheries, etc., as well as the dependent labourers employed there, from reed-cutters, fishermen, potters and cooks, to barbers, ploughmen, stockmen and weavers.

The notion, therefore, that economy, society, religion and politics were somehow discrete domains in antiquity is false. These divisions are artificial. The economic effort attested in the Archaic texts from Uruk – the raw labour, the herding of animals, the building of boats, the casting of metals, the fashioning of wood – was dedicated in toto to Inanna and managed by her worldly overseers. It was this activity that flipped the mental switch, so to speak, leading to the invention of accounting and, very soon thereafter, to a precocious form of writing. The fiction that the awe-inspiring, divine being Inanna was, in fact, the owner of all land and fruits of human toil in and around Uruk permitted the goddess’s stewards to devise a system of resource management and compensation for human effort that persisted, little altered, for the next 3,500 years. This is not to say that it excluded, in later periods, either private or royal ownership of land and production, or that early Mesopotamian society was purely theocratic, but there is no contravening the evidence of Uruk’s Archaic texts in this phase of Mesopotamia’s social development, evidence that is as overwhelming as it is unambiguous. This is an essential, elemental truth about early Mesopotamian religion. It may not have been the truth that Leo Oppenheim was seeking, but it is the truth imparted to us by the sources that have survived from the 4th millennium BC.

In recent years much has been written on early religion in the Near East. The striking animal imagery preserved on carved stone stelae and plastered wall reliefs at sites like Göbekli Tepe and Çatal Höyük in Turkey, both of which predate writing by many thousands of years, have, in part, been responsible for many articles and entire books dedicated to the subject of Near Eastern religion 8,000–11,000 years ago. If Oppenheim was pessimistic about our chances of understanding Mesopotamian religion, even when faced with a wealth of detail from the cuneiform record, what would he think if he were alive today and could witness the often extravagant and hopelessly unverifiable speculations of writers on pre-historic manifestations of religious behaviour in Turkey? How would we ever be in a position to distinguish those religious rituals presumed to have taken place at Göbekli Tepe and Çatal Höyuk, or the spiritual associations of the animals depicted in low-relief on stone stelae and plastered walls and a host of other possible explanations? Why should one believe that powerful zoomorphic symbols necessarily have religious connotations? Who is to say they didn’t have a purely social meaning, as totemic representations held dear by particular kinship groups, or as specially charged images of hunted prey of special significance to groups of hunters?

Just like prehistoric belief systems in Turkey, Mesopotamian religion, I would argue, cannot be reconstructed for the period prior to the invention of writing. Any speculations on belief, the nature of divinity, or the scope of spiritual action, let alone the character of the pantheon and its regional differences, are simply beyond the realm of what archaeological sources can ever hope to provide for this or any other preliterate or aliterate society, for which we lack either written or oral traditions. And while one can agree with much if not all of what Leo Oppenheim argued fifty years ago, I would suggest that he neglected to see what the sources do tell us about early Mesopotamian religion. As noted above, the Archaic texts show us that Inanna’s demesne, Eanna, was characterised by a communal effort of labour with a very particular purpose. That purpose, simply put, was the service of a deity. The labour required to satisfy the deity’s requirements was managed by a bureaucracy of specialists, whom we would today identify as a priesthood. This was a system that sustained its adherents. It was probably less about doctrine and practice, prayer and philosophy, than it was about sublimation to the overarching power of the deep-rooted, elemental forces that supported and protected a community.

Divine protection, as we know from later sources, was all-important. Later, as we know from literary sources dating to between the late 3rd and the late 1st millennium BC, nothing was considered worse than the rejection of a people by its chief god or goddess. Enemies might overrun a city, carrying off the cult statue of the chief deity, but this had more to do with the abandonment of the victim population by its erstwhile divine protector, than the military prowess of its worldly adversary. A deity only abandoned a city or land when its people had sinned. The fault lay not in their stars but in themselves. Inanna was Uruk, its people were fed and housed in her care, through her benevolence, under her watchful eye and through the prudential care of her stewards.

The fact that no ritual texts, hymns, prayers, creation myths or anything that could be associated with the later manifestations of liturgical literature were to be codified or written down until over 1000 years after the invention of writing is surely telling. Religion in early Mesopotamia, so far as the Archaic texts are concerned, was not an amalgam of mythology, ritual and spiritual belief. It was not an abstract set of ideals about life as it should be lived, or divine injunctions. Rather, it was a system of ordered existence, one that was eminently successful in this phase of human existence, at least as manifested in Mesopotamia. Supremely polytheistic, even though each city had its chief god or goddess, early Mesopotamian religion was more about the tending of a deity’s estates, than it was about being tended for by a god’s love. Death was final and in no way to be yearned for. There was no concept of resurrection but, while alive, all who worked for a particular god or goddess were enmeshed in a system of support that sustained thousands and merged those domains we, in the 21st century, are apt to fragment under rubrics like society, economy, politics and religion. This was the essence of early Mesopotamian religion. We cannot account for the first steps that led to this point, though we can trace the superimposed plans of structures, eventually identifiable as ‘temples’ in the 3rd millennium BC, one on top of the other, through the preceding two or three millennia at sites like Eridu, to the south of Uruk. If the later associations of such buildings with individual deities can be used to infer earlier patterns of behaviour, then devotion to a specific deity had a pedigree extending back thousands of years before it first appeared, in crystallised form, in the Eanna precinct. But as Oppenheim would be quick to point out, if he were alive today, this tells us little about belief. What it does tell us, however, is that early Mesopotamian religion was grounded in the earth, in human toil, in the fiction of divine sovereignty and in an all-encompassing system of material production and redistribution.

Early Mesopotamian religion resembles less Chartres, Lourdes, or the Vatican than it does an integrated system of social, economic and religious effort strong enough to withstand the harsh environment and fickle climate of the Tigris-Euphrates river basin and resilient enough to sustain its adherents throughout many millennia of military and political upheaval. Ironically, as Iraq begins to fracture, this seems to be exactly what is missing at the present time.

This essay originally appeared in ‘Knowledge and Information – Perspectives from Engelsberg Seminar, 2018’, Bokförlaget Stolpe, in collaboration with the Axel and Margaret Ax:son Johnson Foundation.