Technology and Cuisine. Experimenting with Chalcolithic obsidian-tempered pottery from the Southern Caucasus

Raw materials such as obsidian and clay, like the lithic tools and ceramic vessels made from them, have been traditionally employed in archaeological research as two parallel and often complementary sources of information to reconstruct past cultures and societies, modalities of production, and regional interaction patterns.
Obsidian-tempered ceramics like those found in the Southern Caucasus during the Chalcolithic period (beginning of the fifth – mid fourth millennium BC) have enormous potential for understanding production, interaction and mobility patterns because they contain a unique combination of two materials that can be sourced and studied by combining different disciplines and methodological approaches.

This contribution summarises the main results (Palumbi et al. 2014) obtained during the last two years by a team composed of C. Chataigner, B. Gratuze, A. Harutyunyan, G. Pulitani and myself. Using different approaches, this multidisciplinary team has investigated obsidian-tempered ceramics with the aim of shedding light on their origins, circulation, function and cultural meanings.

Fig.1: Map of the obsidian sources of Eastern Anatolia and the Southern Caucasus (Drawn by C.Chataigner).

Fig.1: Map of the obsidian sources of Eastern Anatolia and the Southern Caucasus (Drawn by C.Chataigner).

The selective use of obsidian intentionally added as a temper to ceramic fabrics is sporadically documented in the ancient world, for example in Yemen (Lewis et al. 2010) and in the Aegean (Katsarou et al. 2002). However, it represents a potting tradition that is largely and systematically attested in the Southern Caucasus during the Chalcolithic period. The large and easy availability of obsidian in the Southern Caucasus (fig. 1) may partially explain the existence of this ceramic tradition in the region, however it has to be pointed out that obsidian-tempered ceramics are completely absent in other regions equally rich in obsidian, such as Central Anatolia.

The archaeology of the Chalcolithic in the Southern Caucasus is still in its early stages. While an absolute chronological framework has now been provided for this period (Lyonnet 2007), it is far from being understood when it comes to its social, economic and cultural developments.

Fig. 2 : Vegetal-tempered ceramics from Aratashen.

Fig. 2 : Vegetal-tempered ceramics from Aratashen.

 It seems that the Chalcolithic period in the Southern Caucasus saw the emergence of productive, organisational and social complexity. This process is witnessed by an enlargement of the territories occupied (from lowlands to highlands), an increase in the range of raw materials exploited (copper minerals) and the development of specialised “primary” production (e.g. pastoralism, wine making) and handicrafts (metallurgy).

As for ceramic production, the most distinguishable marker of this period is the vegetal-tempered ceramics (fig. 2) (Sagona 2014; Marro 2010). This tradition had already emerged in the Neolithic but came to represent the main ceramic horizon throughout the Southern Caucasus during the Chalcolithic. However, obsidian-tempered ceramics were also a distinctive potting tradition in this period.

Fig. 3 : Cross-sections of obsidian-tempered ceramics from Aratashen (Photo by B. Gratuze).

Fig. 3 : Cross-sections of obsidian-tempered ceramics from Aratashen (Photo by B. Gratuze).

Often named “Sioni Ware” after the eponymous site of Sioni in Eastern Georgia (Kiguradze and Sagona 2003, Sagona 2014), obsidian-tempered ceramics are an easily identifiable ware-group which stand out from the vegetal-tempered ceramics for the following reasons:

1 – The dense presence of small and angular-shaped obsidian inclusions (fig. 3).

2 – The surface colours which are usually darker than those of the vegetal-tempered ceramics.

3 – The forms mainly consisting of small to medium jars.

4 – Incisions and circular or comb impression decorations always applied on the rims (fig. 4).

Fig. 4 : Vessels from Sioni (Georgia) (Kiguradze and Sagona 2003)

Fig. 4 : Vessels from Sioni (Georgia) (Kiguradze and Sagona 2003)

Since their discovery, several interpretations of this ceramic production have been proposed:
  • a tradition attributed to indigenous Caucasian communities in contrast to the vegetal-tempered ceramics attributed to “Mesopotamian-oriented” populations (Marro 2008);
  • a chronological marker of the Early Chalcolithic period (Palumbi 2011).
  • a production typical of a single site or region (like for instance Sioni in Georgia) that spread throughout the Southern Caucasus as a result of exchange or population movements.
However, new absolute dates from the site of Sioni (Sagona 2104) and recent excavations in several Chalcolithic sites of the region (Lyonnet and Guliyev 2012) have demonstrated that obsidian- and vegetal-tempered ceramics were chronologically and geographically coterminous, coexisting in the same sites and levels. Thus the hypotheses put forth on obsidian-tempered ceramics thus far, need to be revised.
In order to shed light on these questions the above mentioned research group decided to undertake experimental research on the fabrication of obsidian-tempered ceramics from the site of Aratashen in Armenia.
Located in the Ararat plain (fig. 5), near the Kasakh river, Aratashen is a small artificial mound excavated by a joint French-Armenian expedition within the framework of the “Mission Caucase” directed by Christine Chataigner (Badalyan et al. 2004).
Fig. 5 : View of the Ararat plain from Aratashen, with Mount Ararat in the background (Photo by C. Chataigner).

Fig. 5 : View of the Ararat plain from Aratashen, with Mount Ararat in the background (Photo by C. Chataigner).

The Chalcolithic levels at Aratashen yielded large amounts of pottery, the majority of which was vegetal-tempered and some of which was obsidian-tempered.

The low occurrence of obsidian-tempered ceramics at Aratashen and their close morphological and decorative similarities with the analogous productions from other Chalcolithic sites in the Southern Caucasus prompted the hypothesis that they were not locally produced.

By considering the potential traceability of the obsidian contained in the ceramic paste, we decided to verify this hypothesis by investigating the provenance of the obsidian-tempered vessels as if they were obsidian artefacts.

This approach had never been attempted prior to our analyses and we did not even know whether the chemical fingerprint of the small particles of obsidian could have been affected or distorted by the firing of the vessels. If these radical modifications had taken place, it would not have been possible to match the obsidian temper to existing obsidian sources thus hampering any possibility of identifying the provenance of these vessels.

Thanks to the collaboration of Armine Harutyunyan, 13 obsidian-tempered potsherds from Aratashen were carefully selected (fig. 6).

Fig. 6 : Obsidian tempered pottery from Aratashen (Photo by A. Harutyunyan).

Fig. 6 : Obsidian tempered pottery from Aratashen (Photo by A. Harutyunyan).

The analyses on the chemical composition of the obsidian temper were carried out by Bernard Gratuze by means of Laser Ablation High-Resolution Inductively Coupled Plasma Mass-Spectrometry (LA-HR-ICP-MS), which at the moment is the only technique allowing the sourcing analysis of obsidian particles of such small dimensions.

The following section of this article briefly summarises our work, which has been published in full detail (Palumbi et al. 2014).

The most important result obtained was that the obsidian used as a temper for ceramic production can be analysed for provenance studies, as the firing of the vessels does not affect the chemical fingerprint of the rare earth elements present in the obsidian inclusions.

The analyses of the obsidian inclusions in the ceramics from the site of Aratashen have shown that most of them originate from the outcrops of Arteni and Tsaghkunyats (Armenia) and that the obsidian from Meydan Dağ (Eastern Turkey) was also sporadically used.

The ubiquitous presence of obsidian from Arteni in these ceramics reflects the same results provided by obsidian sourcing of the lithic industry (Badalyan 2010), suggesting that the obsidian used in the ceramics was “recycled” from the products of obsidian knapping that took place on sites where ceramic production took place.

Conversely, the obsidian from Tsaghkunyats, also largely used in the ceramic temper, was not used for knapping at Aratashen. This inconsistency can be explained by the fact that small pebbles of obsidian from Tsaghkunyats are easily available in the nearby Kasakh riverbed. While their dimensions were of no use as a raw material for knapping, these pebbles could have been intentionally acquired from the river to be crushed and transformed into ceramic-temper.

The intentional selection of the obsidian from Tsaghkunyats for ceramic tempering suggests that obsidian procurement for ceramic production was partially independent from the obsidian networks supplying the lithic industry. This sheds light on the co-existence of various obsidian acquisition strategies at the site of Aratashen.
As for the origin of obsidian-tempered ceramics at Aratahsen, we were able to conclude that the potsherds containing obsidian from Tsaghkunyats and those containing a mix of different types of obsidian were locally produced.

This conclusion begged the following question: if the obsidian-tempered ceramics were produced and used at Aratashen in coexistence with the vegetal-tempered ceramics, was there any functional difference between these two productions?

Fig. 7: G. Pulitani checks on the size and shape of the four vessels.

Fig. 7: G. Pulitani checks on the size and shape of the four vessels.

In order to answer this question, we set up an experimental session with Giuseppe Pulitani, who is a professional potter, to test the cooking properties of the obsidian-tempered vessels. To do so, we fashioned 2 vegetal-tempered and 2 obsidian-tempered pots of the same size and shape (fig. 7).

The vegetal-tempered and obsidian-tempered fabrics were prepared respectively by finely chopping wild grass and by crushing three different types of obsidian (those from Arteni were kindly provided by the GéObs Project) which were to be added to the clay (fig. 8).

Fig. 8 : Chopping grass and crushing obsidian for tempering; mixing clay and crushed obsidian

Fig. 8 : Chopping grass and crushing obsidian for tempering; mixing clay and crushed obsidian

The vessels were built by means of slabs of clay and, after they reached a leather-hard state, their surfaces were smoothed with a pebble (fig. 9). The vessels were then fired in a gas kiln at 800° C for about 12 hours.
Fig. 9 : Slab-building, finishing and smoothing of the vessels

Fig. 9 : Slab-building, finishing and smoothing of the vessels

Once the vessels were fired, we set up a cooking session in uncontrolled firing conditions (fig. 10) by boiling three litres of water and two litres of milk to test and compare the cooking “performances” of the vegetal-tempered and obsidian-tempered vessels against different quantities of liquids. The results of this experiment were astonishing.

Fig. 10 : The cooking session

Fig. 10 : The cooking session

Milk in the obsidian-tempered vessel boiled in 18 minutes while it took 28 minutes to boil the same amount of milk in the vegetal-tempered vessel (play video).

By increasing the amount of liquid (water) the difference was even more blatant: it took 20 minutes to boil 3 litres of water in an obsidian-tempered vessel, while it took 35 minutes to boil the same amount of water in the vegetal-tempered vessel.

The temperature curves of these performances show that the rise in temperature of the contents of the obsidian-tempered vessels was more rapid and abrupt than in the vegetal-tempered vessels where it was slower and more gradual (fig. 11).

Fig.11 : Temperature curves of the two cooking sessions in uncontrolled firing conditions.

Fig.11 : Temperature curves of the two cooking sessions in uncontrolled firing conditions.

How do we explain these results?

The differences in the physical properties of the two ceramic pastes are the key factor for understanding the difference in the cooking performance of the obsidian and vegetal-tempered vessels.

The high silica content of obsidian is probably an important factor determining the high thermal conductivity of the obsidian-tempered ceramics as is the compactness of the obsidian-tempered paste which improves heat-conductivity.

Conversely, the porosity characterising the vegetal-tempered pastes which is the result of the voids left by the burnt vegetal inclusions, acts more like an insulator than as a heat conductor.

These experiments have shown that vegetal-tempered and obsidian-tempered vessels are characterised by radically different properties in terms of heat conductivity and insulation.

These different properties could have been motivated by specific needs and uses, which obsidian-tempered and vegetal-tempered ceramics could fulfil.

Vessels with different physical properties could have allowed the storage and processing of different types of liquids or foodstuff, and possibly also the use of different cooking techniques for serving and eating different kinds of meals.

A first hypothesis for the “invention” of the obsidian-tempered ceramics is that they could be linked to the emergence of a “Chalcolithic cuisine” made of new alimentary practices, cooking techniques and culinary tastes that developed among the fifth-millennium BC communities of the Southern Caucasus.

More experiments and new analyses are needed to verify this hypothesis.

If the Chalcolithic was the period that saw the development of productive, organizational and cultural complexity in the Southern Caucasus, our research on the obsidian-tempered ceramics are a further step towards the investigation of these developments as studies on food processing and consumption are fundamental to understanding any society and its traditions.


Badalyan R. 2010. Obsidian in the Southern Caucasus: The use of raw materials in the Neolithic to Early Iron Ages, in: Hansen S., Hauptmann A., Motzenbacker I., Pernicka E. (eds), Von Majkop bis Trialet. Gewinnung und Verbreitung von Metallen und Obsidian in Kaukasien im 4.-2. Jt.v.Chr., Bonn, Rudolf Habelt, 27-38.

 Badalyan R., Lombard P., Chataigner C., Avetisyan P. 2004. The Neolithic and Chalcolithic phases in the Ararat plain (Armenia): the view from Aratashen, in: Sagona, A. (eds), A View from the Highlands. Archaeological Studies in Honour of Charles Burney, Herent, Peeters, 399-420.

 Katsarou, S., Sampson, A., Dimou, E., 2002. Obsidian as temper in the Neolithic pottery from Yali, Greece, in: Kilikoglu, V., Hein, A., Maniatis, Y. (eds), Modern Trends in Scientific Studies on Ancient Ceramics. Papers Presented at the 5th European Meeting on Ancient Ceramics, Athens 1999, Oxford, BAR International Series 1011, pp. 111-120.

 Kiguradze T., Sagona A. 2003. On the origins of the Kura-Araks cultural complex, in: Smith A., Rubinson K. (eds), Archaeology in the Borderlands: Investigations in Caucasia and beyond, Los Angeles, The Cotsen Institute of Archaeology, 38-94.

 Lewis, K., Khalidi, L., Eisenberger, B., Sanabani, A. 2010. Mapping Masna’at Maryah: Using GIS to reconstruct the development of a multi-period site in the highlands of Yemen, in: Proceedings of the Seminar for Arabian Studies 40, Oxford, Archaeopress, 213-226.

 Lyonnet B. 2007. Introduction, in: Lyonnet B. (eds), Les cultures du Caucase (VIe-IIIe millénaires avant notre ère). Leurs relations avec le Proche-Orient, Paris, CNRS Editions, 11-20.

 Lyonnet B., Guliyev F. 2012. Mentesh Tepe, Archäologische Mitteilungen aus Iran und Turan 44, 86-108.

 Marro C. 2010. Where did Late-Chalcolithic chaff-faced ware originate? Cultural dynamics in Anatolia and Transcaucasia at the dawn of urban civilization (ca 4500-3500 BC), Paléorient 36 (2), 35-55.

 Marro C. 2008. The Late Chalcolithic ceramic cultures in the Anatolian Highlands, in: K. Rubinson K., Sagona A. (eds), Ceramic in Transitions. Chalcolithic through Iron Age in the Highlands of the Southern Caucasus and Anatolia, Leuven, Peeters, 9-37.

 Palumbi G. 2011. The Chalcolithic of Eastern Anatolia, in: Steadman S., Mc Mahon G. (eds) Oxford Handbook of Ancient Anatolia, Oxford, Oxford University Press, 205-226.

 Palumbi G., Gratuze B., Harutyunyan A., Chataigner C. 2014. Obsidian-tempered pottery in the Southern Caucasus: a new approach to obsidian as a ceramic-temper, Journal of Archaeological Science 44, 43-54.

Sagona A. 2014. Rethinking the Kura-Araxes genesis, Paléorient 40 (2), 23-46.


Giulio Palumbi est chargé de Recherche au CNRS. Il est spécialiste de la préhistoire d’Anatolie orientale et du Caucase. UMR 5133 – Archéorient, MSH Maison de l’Orient et de la Méditerranée, Lyon

En collaboration avec :

Christine Chataigner est Ingénieur de Recherche au CNRS, laboratoire Archéorient/UMR 5133. Responsable de la mission “Caucase” du MAE (Arménie, Géorgie), elle étudie les périodes néolithique et chalcolithique dans le Caucase et la circulation de l’obsidienne dans cette région et au Proche Orient.

Bernard Gratuze est directeur de recherche au CNRS, laboratoire IRAMAT-CEB, UMR 5060, CNRS/Université d’Orléans. Il développe et applique des méthodes d’analyse pour les objets archéologiques. Ses domaines de recherche concernent principalement les technologies de fabrication du verre et la diffusion des productions de cette industrie de l’Age du Bronze à la période moderne, ainsi que la circulation de l’obsidienne.

Armine Harutyunyan est archéologue de l’Institut d’Archéologie et Ethnographie de l’Arménie. Elle étudie les céramiques des périodes néolithique et chalcolithique du Caucase.

Giuseppe Pulitani dont l’atelier Vasellarius Tusculanus est à Colonna dans la province de Rome, est un artisan potier, expert de technologies anciennes et de techniques traditionnelles de manufacture de la céramique.

Pour citer ce billet : Palumbi G. 2015. Technology and Cuisine. Experimenting with Chalcolithic obsidian-tempered pottery from the Southern Caucasus, ArchéOrient – Le Blog (, 3 juillet 2015. [En ligne]

Laisser un commentaire

Votre adresse de messagerie ne sera pas publiée. Les champs obligatoires sont indiqués avec *