TT
tr

ArticleEditorial draft

Feasting, beer and labour at Göbeklitepe

Were the builders of Göbeklitepe paid in meat and beer? The work-feast model rests on bones, stone troughs and thousands of grinding stones. Here is what each line of evidence can carry, and what it cannot.

TT Living History Editorial — draft for author reviewPublished 9 min read

Every account of Göbeklitepe runs sooner or later into arithmetic. The pillars of the large enclosures stand up to 5.5 m tall. They were cut from the surrounding limestone plateau with stone tools, dragged to the mound and set upright by people who had no draught animals, no wheels and no chiefs that anyone can identify. How many hands that took is debated, but nobody thinks one small band managed it alone. So the question follows: why did they come, and what kept them there?

The answer proposed by the excavation team in 2012 has become one of the best-known ideas about the site. People came because there was a party. The work was organised as a feast, with meat in quantity and, quite possibly, beer.1 This article looks at that idea piece by piece.

The problem of labour

Anthropologists have a name for the arrangement the excavators had in mind. In a work feast, a host who needs many hands for a short time, to raise a house or bring in a harvest, pays not in wages but in hospitality. The workers eat and drink well, the host gains a building and a reputation, and everyone expects the favour to be returned. Michael Dietler and Brian Hayden's comparative work on feasting documented the practice across small-scale farming societies on several continents.2

The model is attractive for Göbeklitepe because it explains cooperation without requiring coercion, and because it links two things that the site has in abundance: monumental building and food refuse. It is also the starting point for the question of how the clans were persuaded and for the game's treatment of motivation. But an ethnographic analogy is a guide to what to look for, not evidence that it happened. What did the excavations actually produce?

The 2012 paper: bones, basins and a spot test

In Antiquity in 2012, Oliver Dietrich, Manfred Heun, Jens Notroff, Klaus Schmidt and the brewing scientist Martin Zarnkow set out the case. Three observations carried it.

First, the fill. The enclosures were found packed with rubble mixed with enormous quantities of broken animal bone, all from wild species, with gazelle and aurochs prominent. The team read this as the debris of large-scale meat consumption.

Second, the vessels. Six very large limestone containers, barrel- or trough-shaped, were recorded in place in rooms of the site's later, rectangular architecture. The largest could hold about 160 litres. They were far too heavy to move and were evidently fixtures; fragments of similar vessels turned up in every layer.1

Third, a chemical hint. Greyish-black residues on the walls of some vessels were tested by Zarnkow at the Technical University of Munich for oxalate, a compound that settles out when cereals are steeped, mashed and fermented, and which brewers know as beerstone. Some samples gave what the authors called probable evidence of oxalate.

Oxalate has a respectable history as a marker. In the early 1990s it was used to identify beer in a jar from Godin Tepe in Iran. But that was a pottery jar of the late fourth millennium BCE, from a world that already had written signs for beer. A film on an open limestone trough some five thousand years older is a weaker witness. Oxalate occurs in many plants, and a team including several of the same authors acknowledged in 2020 that cereals can leave it by routes other than brewing.3

Seven thousand grinding stones

The stronger evidence came later, and it concerned not drink but grain. In 2019 Laura Dietrich and colleagues published in PLOS ONE the first study of the site's ground stone tools: 7268 objects, among them almost 3400 handstones, together with grinding slabs, bowls, mortars and pestles, nearly all of basalt from an outcrop about 2 km away.4

The team combined several methods. They classified the shapes and found the handstones strikingly standardised. They made replicas and ground einkorn and other materials with them to build a reference collection of wear patterns, then compared those patterns under the microscope with the ancient surfaces. Most of the archaeological tools matched cereal processing. Sediment samples from inside and outside buildings were rich in phytoliths, the microscopic silica bodies that plants leave behind, and grasses dominated them.

In the experiments, 500 g of einkorn took 40 to 60 minutes to grind. The authors judged the quantity of equipment too great for the everyday needs of a small resident group. They added one more strand: the animal bones suggest that gazelle, a migratory animal, was hunted in large numbers between midsummer and autumn. Put together, they argued, the evidence shows seasonal peaks of activity, with crowds, meat and cereal food prepared on the spot. They interpreted these peaks as large work feasts.4

This is a good deal more solid than the oxalate. It is also, strictly, evidence for cereal processing on a scale that needs explaining, and for seasonal hunting. "Feast" is the explanation offered, not the thing observed.

Beer or porridge?

What, then, was in the troughs? A follow-up study in 2020 examined the stone vessels as a group: around 600 pieces, mostly fragments, the largest assemblage known from the Pre-Pottery Neolithic of the Near East, with the big troughs reaching 165 litres. Bringing together find context, wear traces, residue chemistry and experiment, the authors concluded that the limestone troughs had been used for cooking cereals.3 Porridge or gruel is the simplest reading. Beer remains possible, and the two are not far apart: a cereal mash left standing will ferment whether or not anyone intends it.

That is where the matter stands. A visitor will still hear that Göbeklitepe has the world's oldest brewery. What it has is a set of large cooking troughs, a great deal of evidence for ground cereal, and a chemical trace that its own investigators call tentative. The game's brewer is built on that hypothesis and is labelled as such.

The critiques

Objections to the work-feast model come from several directions, and some of them have reshaped it.

Who lived there? The 2012 argument assumed a sanctuary without residents, to which hunter-gatherers travelled. The Canadian archaeologist E. B. Banning challenged this in 2011, arguing that the buildings could be richly decorated houses and that large quantities of food waste and tools are what a settlement looks like.5 Since then the excavators themselves have moved. The 2019 study does not rule out a small permanent population, notes rock-cut cisterns holding some 153 cubic metres of rainwater, and associates the grinding equipment with the rectangular buildings. If people lived on the mound, at least part of the grinding was simply dinner. The dispute is followed in temple or settlement?.

Is the rubbish from feasts? The bone-rich fill is not a floor deposit left where people ate. How the enclosures came to be filled, whether deliberately by people, by slope collapse, or both, is under revision in the current project (see why were they backfilled?). Material that slid in from elsewhere on the mound may mix refuse of many kinds and periods. That weakens any direct line from a given bone to a given banquet.

How fine is the dating? The troughs come from later building phases; the great pillars were raised earlier. Thousands of grinding stones accumulated over many centuries. The authors of the 2019 paper concede that the number of tools in use at any one time cannot be established. A large total is compatible with occasional huge gatherings and with modest, steady use.

How much labour was really needed? Estimates of the workforce depend on assumptions about how fast the enclosures were built and whether they stood complete at the same time. If construction was piecemeal and spread over generations, the labour problem shrinks, and with it the need for a special mechanism to solve it.

What survives the criticism

A careful summary would run like this. People at Göbeklitepe processed wild cereals intensively, with standardised tools, and cooked them in very large stone vessels. They consumed great quantities of game, some of it hunted seasonally. They did this at a place where they also invested extraordinary effort in building. Communal eating on a scale beyond the household is a reasonable inference from that combination, and a familiar one from the ethnographic record.

That the meals were organised specifically to recruit builders, that guests arrived from far-off groups, that beer flowed: these are interpretations of decreasing security. They may be right. The honest position is that the work feast is the best-developed model for how the labour was mobilised, not a demonstrated fact, and that the link to the origins of agriculture is more speculative still. For the climatic and botanical background, see the end of the Younger Dryas; for the practical side of feeding a workforce, the wiki pages on food and the division of labour.

Footnotes

  1. Dietrich, O., Heun, M., Notroff, J., Schmidt, K. and Zarnkow, M. 2012. The role of cult and feasting in the emergence of Neolithic communities. New evidence from Göbekli Tepe, south-eastern Turkey. Antiquity 86(333): 674–695. ↩ ↩2

  2. Dietler, M. and Hayden, B. (eds) 2001. Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power. Washington: Smithsonian Institution Press. ↩

  3. Dietrich, L. and colleagues 2020. Investigating the function of Pre-Pottery Neolithic stone troughs from Göbekli Tepe – An integrated approach. Journal of Archaeological Science: Reports 34: 102618. ↩ ↩2

  4. Dietrich, L., Meister, J., Dietrich, O., Notroff, J., Kiep, J., Heeb, J. and colleagues 2019. Cereal processing at Early Neolithic Göbekli Tepe, southeastern Turkey. PLOS ONE 14(5): e0215214. The figures for tool counts, grinding experiments, phytoliths and cistern capacity are taken from this open-access paper. ↩ ↩2

  5. Banning, E. B. 2011. So Fair a House: Göbekli Tepe and the Identification of Temples in the Pre-Pottery Neolithic of the Near East. Current Anthropology 52(5): 619–660. ↩

Sources

  1. [1]Oliver Dietrich, Manfred Heun, Jens Notroff, Klaus Schmidt, Martin Zarnkow. The role of cult and feasting in the emergence of Neolithic communities. New evidence from Göbekli Tepe, south-eastern Turkey. Antiquity 86(333): 674–695. Antiquity / Cambridge University Press (2012)
  2. [2]Laura Dietrich, Julia Meister, Oliver Dietrich, Jens Notroff, Janika Kiep, Julia Heeb and colleagues. Cereal processing at Early Neolithic Göbekli Tepe, southeastern Turkey. PLOS ONE 14(5): e0215214. PLOS ONE (2019)
  3. [3]Laura Dietrich and colleagues. Investigating the function of Pre-Pottery Neolithic stone troughs from Göbekli Tepe – An integrated approach. Journal of Archaeological Science: Reports 34: 102618. Elsevier (2020)
  4. [4]E. B. Banning. So Fair a House: Göbekli Tepe and the Identification of Temples in the Pre-Pottery Neolithic of the Near East. Current Anthropology 52(5): 619–660. University of Chicago Press (2011)
  5. [5]Brian Hayden, Neil Canuel, Jennifer Shanse. What Was Brewing in the Natufian? An Archaeological Assessment of Brewing Technology in the Epipaleolithic. Journal of Archaeological Method and Theory 20: 102–150. Springer (2013)
  6. [6]Li Liu and colleagues. Fermented beverage and food storage in 13,000 y-old stone mortars at Raqefet Cave, Israel: Investigating Natufian ritual feasting. Journal of Archaeological Science: Reports 21: 783–793. Elsevier (2018)
  7. [7]Amaia Arranz-Otaegui and colleagues. Archaeobotanical evidence reveals the origins of bread 14,400 years ago in northeastern Jordan. PNAS 115(31): 7925–7930. PNAS (2018)
  8. [8]Michael Dietler, Brian Hayden (eds). Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power. Smithsonian Institution Press (2001)
ShareXFacebookWhatsApp

Related

No ratings yet
Rate this

Comments (0)

**bold**, *italic*, [link](https://…), > quote

Loading…