Playing with fire: Exploring ceramic pyrotechnology in the Late Neolithic Balkans through an archaeometric and experimental approach

Highlights

• Neolithic pottery is investigated via experimental archaeology and archaeometry.

• Pyrotechnology is studied from a technological and a social point of view.

• The paper addresses pyrotechnology and cultural choices in the Neolithic Balkans.

• We formulate a methodological framework for the study of ancient pyrotechnology.

Abstract

Addressing ceramic pyrotechnology plays a key role in understanding a wide range of cultural and social behaviours associated to pottery production. Firing is the process which transforms clay into ceramic, which is one of the most frequently preserved materials in the majority of Neolithic and later archaeological sites.

Though firing temperatures and the functions of various pyrotechnological installations have been extensively investigated in archaeology, both have often been addressed separately. Most of our knowledge on firing structures and procedures in the Neolithic are still largely based on ethnoarchaeological evidence. To move forward, we need to consider all aspects involved in ancient pyrotechnology, together with use of additional investigative tools. This study aims to address Neolithic pottery firing from a diverse perspective that merges archaeometric analyses and experimental archaeology. To demonstrate the potential of this approach, we combined an archaeometric case study of pottery from the late Neolithic (5200–4800 BCE) from the site of Gradište-Iđjoš (Serbia) with experimental pit firings, likely one of the mostly frequently employed firing techniques used in prehistoric periods.

Scientific analyses include X-ray powder diffraction (XRPD), scanning electron microscopy (SEM), and ceramic petrography. These methods were run on both archaeological materials and experimental reproductions. Additionally, a detailed program of firing temperature monitoring, integrated observations on atmospheric conditions, soaking time, and duration were recorded to contribute to the study. The experiments enabled us to collect results useful for our understanding of the pyrotechnological knowledge of Neolithic potters from a technological and social point of view. In addition, they demonstrated the potential of a dedicated methodological framework for studying pottery firing that can be applied to other chronological and cultural contexts.

Introduction

The study of ancient pyrotechnology is certainly one of the crucial themes in anthropological and archaeological research (e.g. Gibbs, 2015, McDonnell, 2001) as it enables scholars to explore topics such as invention, innovation, and technological advancement to build narratives of large-scale interactions in global archaeology (e.g. Roberts and Radivojević, 2015, Roberts and Vander Linden, 2011). Ceramics, as one of the most abundant materials preserved in the archaeological record, are at the focus of several pyrotechnological studies. These works illustrate how pottery firing is a complex procedure, due to the large number of variables that are involved in this process (e.g. Gosselain, 1992, Livingstone Smith, 2001, Rice, 2015).

Among the different approaches that have been used to reconstruct ancient ceramic pyrotechnology, archaeometric analyses and experimental archaeology have played a critical role. On the one hand, scientific analyses allow a degree of resolution that cannot be obtained solely with macroscopic investigations (Tite, 1995, 37–38). Archaeometric studies that focus on the reconstruction of pottery pyrotechnology employ a variety of methods that aim especially at the estimation of firing temperatures. This is done through the identification of relationships between firing temperatures and changes in the pottery microstructure (e.g. porosity, clay matrix, progressive sintering, vitrification and mineralogy: Gliozzo, 2020, Maniatis and Tite, 1981, Rice, 2015). On the other hand, the employment of experimental archaeology (Coles, 1979, Godino et al., 2020, Outram, 2008, Reynolds, 1999) not only helps to test hypotheses developed on the basis of the archaeometric results, but, most importantly, gives insightful information on different aspects of firing procedures. This knowledge helps to have a more nuanced understanding of ancient ceramic pytorechnology and the complex social behaviour behind this practice (e.g. Gheorghiu, 2019), that goes beyond the mere estimation of firing temperatures.

Despite the clear advantages that both approaches contribute, they are only rarely systematically combined (e.g. Aloupi-Siotis, 2020, Kudelić, 2017, Thér et al., 2019). In this work, using the case study of the Late Neolithic Vinča settlement of Gradište near Iđjoš in the Serbian Banat (hereafter Gradište-Iđjoš), we show that the combination of archaeometric analysis of materials deriving from both archaeological contexts and our experiments is the key to a better understanding of ancient pyrotechnology.

Such an approach gives us important information on different aspects of firing procedures and how these are reflected in the microstructural and compositional characteristics of archaeological ceramics. These data then aid a better interpretation of archaeometric results and help us developing a well-rounded technological and social reconstruction of ancient pyrotechnology.

The settlement of Gradište-Iđjoš (Fig. 1) is situated in the north-centre Banat, approximately 3 km east of the Tisza river. This Neolithic and Chalcolithic settlement was excavated before and after the Second World War and is currently investigated by the Bordeland: ARISE project (Mirković-Marić and Marić, 2017). The excavations carried out at this site gave evidence of a Starčevo-Körös culture phase (second half of the 6th millennium BCE) and a Vinča and Tisza occupation (5200–4900 BCE). Late Neolithic mixed assemblages are typical for the area of northern and central Serbian Banat and are found in many other sites in this region (Brukner, 1968). Two other examples for this are the sites of Kremenjak-Čoka and Akača-Novo Miloševo, both situated close to Gradište-Iđjoš (Fig. 2).

The Vinča phenomenon, whose pottery is at the centre of this investigation, is a Neolithic/Chalcolithic material culture that developed in a vast area in the northern and central Balkans. In terms of absolute dates, the estimated duration of the Vinča phenomenon spans from c. 5350 to c. 4600 BCE (Whittle et al., 2016 and literature therein).

The Tisza material culture (Korek, 1989, Raczky, 1987) spread during the Late Neolithic (c. 5000 to c. 4600 BCE) in an area spanning from Slovakia and Ukraine to the north, up to the Körös river on the east. The Serbian Banat represents the southern part of the territory of Tisza material culture, reaching the confluences of the Aranka and the Zlatica rivers into the Tisza.

The geology (Fig. 2) of the north-centre Serbian Banat (close to the location of Gradište-Iđjoš) is marked by several Pleistocene and Holocene alluvial sediments containing gravel, sand, and clay layers (Koprivica and Strajin, 1994). In a previous work, geological samples near Gradište-Iđjoš were selected to study the nature and distribution of these alluvial sediments (Amicone et al., 2020a), thereby demonstrating that two main clay sources mark this area: very fine sandy-clay sources deposited during the Holocene and available in the proximity of the site, and sandier Pleistocene sources that outcrop c. 10 km from Gradište-Iđjoš.

Several researchers have focused on the study of Late Neolithic and Chalcolithic pottery pyrotechnology from the Balkans (e.g. Gardner, 1978, Gardner, 2003, Goleanu et al., 2005, Kaiser et al., 1986, Linda, 1984, Maniatis and Tite, 1981, Perišić et al., 2016, Spataro, 2017, Szakmány et al., 2019, Yiouni, 2000). While targeted studies on ceramic pyrotechnology of Tisza material culture are missing, pottery produced by the Late Neolithic and Chalcolithic communities labelled as Vinča received particular attention for the purported link between pottery firing technology and the origins of metallurgy in the Vinča phenomenon (Amicone et al., forthcoming).

By applying a vast range of archaeometric techniques, these studies (e.g. Kaiser et al., 1986) were especially focused on the estimation of firing temperatures in the attempt to understand if Vinča pottery was fired to temperatures comparable to those necessary to smelt copper (c. 1083 °C, Pollard et al., 1991) and if this pyrotechnology knowledge could have been transferred from ceramic manufacture to metallurgy. Nevertheless, to have a more comprehensive understanding of the pyrotechnological processes, more attention must be paid to other parameters of ceramic manufacture, such as how firing atmosphere was controlled to create redox conditions.

A more recent study (Amicone et al., 2020b) utilised a multi-pronged scientific approach to investigate pottery from Belovode and Pločnik (Serbia), home of the world’s earliest metallurgy. This work illustrates that potters fired ceramics at highly variable temperatures, which did not appear to have exceeded 900 °C and employed either oxidising or reducing conditions. This study also proposed a model of production for dark-burnished pottery, a tradition widespread throughout the Balkans in the Late Neolithic and typical feature of Vinča material culture (Chapman, 2006, Chapman, 2007). This model consists of a two-step firing procedure that involves an oxidising firing followed by a reducing phase during cooling obtained through smudging of the vessels. This work concluded that potters at these sites were certainly able to manipulate the amount of oxygen in their firings and that this knowledge could have been important for the development of early metallurgy pyrotechnology.

Vinča pottery has often been regarded as the outcome of specialised and skilled productions (e.g. Kaiser, 1984, Spataro, 2018) and therefore it was often assumed that potters were certainly employing kilns rather than open or pit firing installations where the firing process is less controlled (Rice, 2015, 172–181). However, there is no conclusive evidence for pottery kilns in Vinča culture settlements (Amicone et al., forthcoming). Recent experiments (Svoboda et al., 2005, Vuković, 2018) suggested that the complete range of pottery manufactured by Vinča potters could have been produced using pit firings. The use of this technique could have even been preferred, despite the lack of control over different variables of the firing procedure, as it allows for a relatively fast and fuel-efficient process (Rice, 2015, 172–181). Traces of pit firings are not always easy to be identified in the archaeological record (Costa, 2017). If pit firings were indeed the main type of firing technique employed at Vinča sites, this would explain the general lack of corroborated evidence for pottery firing installations in the archaeological record of these settlements.

On this basis, we set up an experimental framework (Table 1) to test the efficiency in terms of temperatures and atmosphere of pit firing, one of the most likely diffused firing structures in prehistory. We combined this approach with laboratory investigations that allowed us to give particular attention to the observation of microstructural and mineralogical changes taking place in the clay objects fired in this type of installation. Therefore, our experiments helped us to create a reference collection to compare archaeological materials to and furthermore provided us a baseline to better understand how ancient firing processes might have worked.