The recipe for animal sacrifice in ancient Greece

By Flint Dibble

We’re so used to modern, twenty-first century recipes. Everything is spelled out to a tee: ingredients, amounts, instructions. But, even if you look at earlier 20th century recipes, the detail is sparser. Techniques and amounts could be optional or elided over since certain knowledge was assumed. Ancient recipes, like those by the Roman chef Apicius, are even worse. There’s so much assumed knowledge, and we’re at such a cultural distance that it’s difficult to know exactly how a meal was prepared (though that doesn’t stop us from trying).

The recent online trend in recipes is recipe-blogging. For these, the detail can be excruciating. You need to read (or scroll) through a personal story about the recipe to get to the ingredients, amounts, and process. Understanding ancient Greek food from literary sources is like having access to the story part of a recipe-blog, without the recipe itself.

The recipe for animal sacrifice in ancient Greece is deceivingly simple: wrap a few bones in fat and burn them to a crisp. But, like many ancient recipes, the closer you examine it, the more questions you have. When you put together all our sources of evidence for ancient sacrifice – literary texts, artistic depictions, and ancient animal bones – the variability in ancient practice refreshes old research angles.

The epic poet Hesiod first described the reasons behind ancient animal sacrifice in a story about Prometheus tricking Zeus. Having chopped up an ox, he arranged two portions: 1) for mortals, Prometheus assigned the meat stuffed in the unappetizing stomach, and 2) while Zeus chose the white bones hidden beneath delicious, glistening fat. Hesiod ends this tale of deception with the offhand comment that ever since then, people have burned white bones for the gods (Hesiod Theogony 557).

But animal sacrifice was everywhere in ancient Greece. At every twist and turn of the story, Homeric heroes sacrificed mythical herds of big, beautiful bulls. In Book 3 of the Odyssey, the sacrificial recipe for a cow at the Palace of Nestor at Pylos is described in epic detail. After it was struck with an ax, and its blood collected in a bowl:

They butchered her, cut out the thighs, all in the proper place, and covered them with double fat and placed raw flesh upon them. The old king burned the pieces on the logs, and poured the bright red wine. The young men came to stand beside him holding five-pronged forks. They burned the thigh-bones thoroughly and tasted the entrails, then carved up the rest and skewered the meat on pointed spits, and roasted it (translation from Wilson 2017).

This scene is an exception. Most of the time, the gory details of sacrifice were assumed knowledge. That said, when mentioned, most often, the thigh-bones were mentioned as burned for the gods.

There are only a few exceptions to this pattern of thigh-bone burning in ancient Greek literature. In Aristophanes’ comedic play Peace (1055), the protagonist notes, after burning the thigh-bones of a sacrificed sheep, that the tail is curling. Scholars have connected this line with the many scenes of sacrificial curling tails painted on Athenian pots. Without this iconographic evidence, we wouldn’t have a clear context for this enigmatic mention in literature.

Pottery: red-figured stamnos depicting a an altar on a double plinth, on which two rows of sticks set crosswise, are burning, with a large hook or the horn of an ox and a square object in the midst of the flames; beside it stands a bearded, wreathed man, in a mantle, inscribed.
Athenian red-figure stamnos depicting a tail curling on a flaming altar (British Museum 1839,0214.68 CC BY-NC-SA 4.0).

A second exception is in the epic myth, the Homeric Hymn to Hermes (115-137). Hermes steals a herd of cattle from Apollo, and – to avoid being tracked – he marches them backwards to a cave. Wanting to make it up to the other gods, Hermes slaughters two of the cattle and distributes the meat into 12 portions. Cleaning up the mess, he burns the feet and heads of the cattle in the fire.

This scene is troublesome. There are no parallels in the artistic or literary record. To some scholars of ancient Greek religion, this is an inversion of a typical sacrifice. A literary construct of sorts. After all, the gods here were assigned the meaty human portion. Everything’s mixed up.

But this whole picture is turned on its head when you look at ancient trash: the food remains found in archaeological sites. For a long time, bits of animal bone were ignored in favor of the study of monumental temples or beautiful art. As a zooarchaeologist, I can tell you that when you start looking at ancient trash, the whole picture of ancient Greek animal sacrifice gets messy.

On the one hand, animal bone evidence does somewhat match patterns from literature and art. Most of our burned bones at sanctuaries and temples were thigh-bones. At a few temples, we even have examples of burned tails. More surprisingly, recent evidence shows several sites where the feet (and sometimes heads) of animals were burned. Maybe that scene in the Hymn to Hermes reveals actual ritual practice and not a literary inversion. 

Burned ankle joint
Burned ankle joint from Azoria, Crete. Photography by Jonida Martini.
Feet bones of sheep and goat.
Feet bones of sheep and goats from Azoria, Crete. Photograph by Jonida Martini.

On the other hand, the evidence is also more complicated. Most of the animal bones from ancient Greek sites aren’t burned, including many unburned thigh-bones found in many settlements. Whether this means most animals weren’t sacrificed or that some sacrifices didn’t involve bone burning is unclear.

Plus, even among a pit of burned bones, most of which match one of the patterns above, there are large numbers of exceptions: other anatomical parts that were burned. For example, bones in archaeological deposits from the Palace of Nestor at Pylos do provide evidence for the burning of cattle thigh-bones in feasting contexts, but burned in equal numbers were the jaws and upper-forelimbs (humerus).

The variability presented by this new source of evidence alongside the ambiguities of assumed knowledge means that we need to re-evaluate our evidence. While the burgeoning study of food trash won’t let us recreate all the details of a recipe, it’s an opportunity for us to look upon recipes in old texts with fresh eyes.


Flint Dibble is a Marie Skłodowska-Curie Research Fellow in the School of History, Archaeology, and Religion at Cardiff University. His project ZOOCRETE will be examining the role of animals and foodways in ancient Greece, with a specific focus on Crete. His research touches on topics of urbanism, climate change, religious ritual, and everyday life. Flint is also a public scholar with a strong commitment to sharing knowledge widely. He is active on Twitter (@FlintDibble) where he regularly writes Twitter threads with footnotes that present archaeology to the broader public.  

Cherries Galore in a Cesspit

By Merit Hondelink

As an archaeobotanist, an archaeologist specialised in studying plant remains found in archaeological excavations, I aim to reconstruct and interpret the relationships between humans and plants in the past. Archaeological plant remains, also known as subfossil plant remains, help us to reconstruct the former landscape and inform us how humans exploited it and even transformed the vegetation. Archaeobotanists do not necessarily study one time period, nor a specific region or topic. They can study plant remains from the Palaeolithic or the 20th century, and everything in between. They can focus on one specific site, work across the country or continent, and even work worldwide. They can delve into topics such as natural vegetation, forestation, domestication, trade, food consumption and much more. The one thing that all of this has in common is the link between humans and plants. But most archaeobotanists do specialize, most notably in the plant parts they study, such as fruits, seeds, pollen, wood or phytoliths. And most archaeobotanists have a beloved time period, favourite region or topic that they find most intriguing. In my case my research focuses on early modern Dutch urban food consumption.

I study what people ate in early modern Dutch cities, and how this changed through time. The best way to study what people ate in the past, is to look at their excrement and kitchen refuse, both of which can be found in the archaeologists treasure trove: the cesspit. These latrines were used to empty one’s bowels, but also served as a place to discard kitchen refuse and household waste. The content of a cesspit consists of organic remains from plants and animals, inorganic (culinary) material culture such as earthenware, glassware and ceramics, but also wooden cups and plates, as well as (decorative) objects, personal belongings and much, much more.

Figure 1: A selection of faunal and floral items found in a late medieval cesspit sample from Groningen. Photo: Dirk Fennema.

The content of an archaeobotanical cesspit sample consists of, among others, floral remains in different shapes and sizes (Figure 1). The items are sorted with the use of a microscope (Figure 2) and identified on a species level (and sometimes even on the level of species variety) by using a reference collection (Figure 3). The Groningen Institute of Archaeology offers a wonderful digital, open access, reference collection, see

Figure 2: A peek through the microscope. Visible is a fragment of text and different seeds and fruits, taken from an early modern Delft cesspit sample. Photo: Merit Hondelink.

When the content of a cesspit sample is analysed, sorted and identified, the interpretation begins. What can these plant remains tell us about past human-plant relationships? Most plant species are interpreted in a standardized way: wild plants inform us about the vegetation composition, make-up of soils and hydrology, whilst agricultural weeds in particular inform us about the crops grown and their local, regional, international or even global provenance. Wild but poisonous or toxic plants inform us about potential medicinal applications. A majority of plant species found in cesspits are classified as economic plants, grown as a food crop or cultivated for other useful purposes, such as fibres for textiles or seeds for oil. Identifying edible plants helps us better understand what plants people used for food and which parts people consumed. It also helps us better understand how food was prepared in the past, as preparation marks can be left behind on seeds and fruits.

Some preparation marks are easier to identify than others: nuts need to be cracked to get to the seed; apple seeds may be sliced when cutting up an apple, cereals can be ground, resulting into fragmented bran. But sometimes the archaeobotanist finds fragmented plant parts that, at a first glance, do not make sense.

Figure 3: A small selection of the tubes from the archaeobotany reference collection housed at the Groningen Institute of Archaeology (GIA) at the University of Groningen. Photo via GIA.

I have come across dozens and sometimes hundreds (or even more) cherry stones and plum stones in a single cesspit sample. No surprise there, cherries and plums were grown in local orchards, sold in the market and consumed with gusto. Most of these stones will have been discarded in the cesspit as a result from eating the fruits and spitting out the stones, or after de-pitting the fruits for dinner preparation. Only a small percentage is assumed to have been accidentally swallowed and secreted as excrement. Still, archaeobotanists find many fragments of cherry and plum stones (Figure 4). This is something that raises questions when you think about it. Why would these sturdy fruit stones be fragmented? A more pressing question when you are aware that the Rosaceae family, among others also including almond, peach, and even apple, contains – to varying degrees – hydrocyanic acid, also known as hydrogen cyanide and sometimes called prussic acid. The seed coat and fruit wall protects the consumer from digesting this acid, which can be poisonous when consumed. So why would someone break the stones of these fruits?

Figure 4: Two fragments of cherry stones found in an early modern cesspit in Vlissingen. Photo: Merit Hondelink.

To test the assumption that cherry stones were fragmented intentionally, and not through, for instance, pressure, an experiment was devised. Cherries were bought at the farmer’s market and taken to a physics lab to measure the pressure required to fragment the stones. After a number of tests, the calculated force to fragment a cherry stone averaged 23,9 kg or 239 Newton (Graph 1). This makes it more plausible that the stones were intentionally fragmented, as opposed to – for instance – fragmentation due to soil pressure.

Graph 1: Force needed to fragment a cherry stone. On the vertical axis the force (N), on the horizontal axis the elongation (μm). The point where the line falls is the moment the cherry stone breaks (max. force – max. elongation).

Consulting early modern cookbooks provided me with a list of recipes requiring the cook to de-stone cherries for the preparation of jams, sauces, syrups and tarts. Delicious experiments ensued, but I did not manage to fragment cherry stones whilst cutting and de-stoning, pressing through a cloth or colander, or by just baking the fruit with stones in a tart in the oven. Working a batch of cherries with a mortar and pestle did the job, though. But than you would have to pick the fragmented stones from the mushy cherries: not ideal at all. Picking up the eighteenth century encyclopaedia compiled by Noël Chomel gave me the hint I needed. In the Dutch version of his Dictionnaire œconomique (Algemeen huishoudelijk-, natuur-, zedekundig- en konst- woordenboek), he mentions different recipes for preparing cherries. Two recipes for cherry liquor instruct the reader to fragment the cherry stones by using a mortar and pestle (Figure 5). The fragmented fruits, including the stones and (I assume) the seeds are added to the brandy (Dutch: brandewijn) and, after closing the bottle, the mixture is put in the sun to infuse. Adding spices such as cinnamon, cloves and sugar is optional, according to the author.

Figure 5: How to make a pleasant cherry liquor (Noël Chomel, 1778).

So, it is plausible that the fragmented cherry stones found in early modern cesspits are the result of the domestic production of cherry liquor. Other fruits, such as plums and peaches, are also used to make a fruity liquor according to Chomel’s encyclopedia. However, what happens to the acid contained in the seeds? That requires further research. It might be that the prescribed infusing in the sunlight helps denature the acid into harmless molecules, leaving only the (bitter) taste behind. This line of research will be undertaken come summer with the aid of a brewer and some chemical analysis. In the meantime, a cherry and cinnamon flavoured lemonade is my poison of choice. Bottoms up!