Experiments with double chamber sunken up-draught kilns

Transcription

1 Experiments with double chamber sunken up-draught kilns Dragoş Gheorghiu Department of Master, National University of Arts, 19 Budisteanu, Bucharest, Romania Abstract - Sunken up-draught kilns appeared in the 5 th millennium BC in the context of the South Eastern European Chalcolithic Cucuteni-Tripolye tradition, as a mechanism for the mass production of ceramics fired at temperatures up to 9008C. Compared with the other pyro-instruments of the respective epoch, the sunken up-draught kiln presented a series of advantages, such as high temperatures, protection of vessels made of fine paste, and a higher output of vessels. Experiments allowed identifying and refining the stages of the chaîne opératoire of this type of kiln, as well as the discovery of new features, such as ergonomics or micro-physical processes, which will help archaeologists to understand the complex technology involved in Chalcolithic ceramic making. 1. Introduction Kilns are a subject often encountered in the archaeological and experimental archaeology literature (Rhodes 1981; Dawson and Kent 1984; Kingery 1997), and they were among the first subjects of experimental archaeology in the last decades. Although this might appear to be a wellexhausted subject, experiments with this pyro-instrument have the potential of contributing new data. This paper will discuss one of the types of prehistoric draught kilns, which the author believes to represent one of the most extraordinary prehistoric machines due to its simplicity and efficiency, as demonstrated by experiments (Gheorghiu 2002; 2006; 2007). This proposition is also based on the fact that sunken up-draught kilns did not change from the 6 th millennium BC until post-medieval times in some parts of South Eastern Europe. Sunken up-draught kilns are attested in the archaeological record starting from the 6 th millennium BC in the Near East (see Simpson 1997, 39), but only from the 5 th -4 th millennia BC in South Eastern Europe (see Comşa 1976; Markevic, 1981). Their invention occurred very early, probably before the emergence of ceramics in the Near East, and they could have been the source of the vaiselle blanche/white ware (de Contenson and Courtois, 1979), produced by plastering wickerwork containers with lime plaster (Gourdin and Kingery 1975), the latter material requiring a minimum temperature of 9008C (Rehder 2000, 47-8), which could have been reached in draught kilns. 2. The archaeological background The emergence of efficient sunken up-draught kilns corresponds to a mastery of fire by means of the control of the ascending flow of hot gases that carried and fed the combustion through perforated surfaces conferring an airabsorbing character to the platform dividing the inner space of the kiln (see Rye and Evans 1976, 164; Arnold 1997, 213). This technological innovation allowed the control of the thermal shock, and provided a clean firing for the decorated pottery, which was separated from the fuel, thus creating the premises of an increase in the volume of ceramic production and the emergence of craft specialisation in Chalcolithic South Eastern Europe (Gheorghiu 2001; 2007, 45). Sunken up-draught kilns dating to the 5 th millennium BC were excavated in this region, belonging to the Chalcolithic Cucuteni-Tripolye tradition (Dumitrescu et al. 1954; Markevic, 1981). Additional evidence of the existence of such kilns in the Cucuteni-Tripolye tradition is the vast quantity of fine ceramics, modelled from a high quality paste (resulting from levigated clays), with very thin walls (down to 3 mm), decorated before firing with complex patterns, and fired at high temperatures (see Ellis 1984; Gâţa 2000; Marinescu-Bîlcu, 2000). A large quantity of pottery could be produced with the help of the slow potter s wheel or tournette, identified in the archaeological record of the middle and final phases of this tradition (Ellis 1984). The replicas used for experimentation in this study were reproduced after the sunken up-draught kilns discovered at Costeşti (Markevic, 1981) and Glǎvǎneşti Vechi (Comşa 1976, 25), dated to the final phase of the Cucuteni-Tripolye tradition (5 th -4 th millennia BC). All the theoretical work presented in this paper is based on the experiments carried out with replicas of these two kilns during the last decade. Cite this article as: Gheorghiu D (2014). Experiments with double chamber sunken up-draught kilns. In M Martinón- Torres (Ed.), Craft and science: International perspectives on archaeological ceramics. Doha, Qatar: Bloomsbury Qatar Foundation UCL Qatar Series in Archaeology and Cultural Heritage

2 Gheorghiu Open firings vs. controlled firings While in South Eastern Europe Chalcolithic container-like pyro-instruments (like ovens or pits) were successfully used for firing ceramics (Comşa 1976; Gheorghiu 2002; 2007), the level of a mass production of pottery was reached only after the introduction of sunken up-draught kilns. The differences between the open structures of combustion and the sunken up-draught kilns consist of the following features:. the fuel is separated from the clay objects, which are not in direct contact with the flame;. the inner space of the kiln is divided by a perforated platform into a firebox and a firing chamber;. the thermal shock of the hot gases and flames is moderated by the tubes of the perforated platform;. the amount of fuel in the firing box is regulated in order to control the interior temperature;. the opening or closing of the apertures of the kiln (Rye and Evans 1976, 164) resulted in a control of the oxygen, an operation which could create an oxidising or reduced atmosphere. All these features defined by the notion of control (control of air flow and consequently control of the fire by changing the time and temperature of firing by means of manipulating the fuel, or control of the atmosphere by closing the apertures) render the sunken up-draught kiln a veritable machine, i.e., a complex instrument with inputs and outputs which can be regulated. The term machine indicates in this context that the pyrocontainer was no longer a simple box, but a space where the fire was separated, fed and controlled throughout the firing. The shape of sunken up-draught kilns Depending on the direction of the flow of air, up-draught and downdraught kilns can be distinguished, the former category being encountered in the archaeological record of the late Eastern European Chalcolithic. A sunken updraught kiln was cut into a slopping bank, whose medium height allowed the operator, at the same time, to see the opening of the firebox and the flue of the firing chamber. The functional shape of a sunken up-draught kiln was composed of the following parts: a fire tunnel/firebox situated below a firing chamber, a perforated platform, a support for the platform, and ceramic waste or a movable dome (Dawson and Kent 1985) to cover the flue. The shapes of the Cucuteni sunken up-draught kilns that were replicated for the purposes of this study could be reduced to the following basic volumes: a semi-cylindrical firebox ( m), intersected with an ovoid volume (diameter at the base 1.5 m, depth 2 m) divided by a clay platform (height: cm) with perforations (diameter m) (Fig. 1). It is evident that the small openings of the firebox and of the flue allowed an easy shutting down of the kiln at the end of the firing process to homogenise the temperature inside, a process which could generate a reduced atmosphere inside the vase chamber. This post-firing shuttering of the kiln was the method employed for the production of the light grey ceramics in the final phase Cucuteni B (Gheorghiu 2006, 38). The ovoid shape seems Figure 1. A cross-section through the Cucuteni sunken up-draught kiln (after Markevic 1981). to have been the most efficient for creating a good airdraught and for conserving thermal energy during the combustion process and after; the difficulty with it consists in the loading and unloading of the ceramic objects. Functioning of sunken up-draught kilns There is general agreement that the key operations involved in the functioning of a draught kiln are the following: a gradual evacuation of capillary water (Arnold 1997, 61), heat containment (Shepard 1956, 75; Kingery 1997, 11) and heat transfer to wares, protection of ceramic objects contained by the firing chamber from thermal shocks during the processes of heating or cooling, protection against humidity (rain or moisture) (Arnold 1997, 213), and the production of higher temperatures (Shepard 1956, 75) compared to other pyro-instruments. The perforated platform has two important roles: to control the thermal shock (i.e., to transform the large flames originating from the firebox into a set of small flames which came in contact with the vessels long after the initiation of the fire), and to store heat, being the principal source for transferring heat (Fig. 2). 3. Experiments Chaîne opératoire. Physical determinism and subjective choices of the potter It is not the purpose of this paper to present a detailed account of firing with a sunken up-draught kiln, but merely to emphasise those aspects that have not been mentioned in the existing literature hitherto. Thus, a significant number of details concerning the firing process, such as the time duration of the procedures, the temperatures reached, or the rate of waste in each firing are not mentioned here. A first set of operations entailed the loading of the firing chamber through the flue, followed by the covering of the flue with shards, in order to form a ceramic strainer, and the structured positioning of the fuel in front of the firebox. Both the fuel and shard covering were positioned so as to allow the creation of an airflow between the firebox and the flue. Because the environment associated to the Cucuteni-Trypolye culture was deciduous forest, pinewood was used in these experiments only to initiate the fire, and beech wood to continue the firing. Craft and science: International perspectives on archaeological ceramics 174

3 Experiments with double chamber sunken up-draught kilns Figure 2. The perforated platform of the sunken up-draught kiln with a thermocouple fitted; Vadastra, Photograph by Dragoş Gheorghiu. After initiating the fire, the fired fuel was gradually introduced inside the firebox, the fire being stabilised at a temperature between 208C and 1208C (see Hamer 1975, 22) to allow water smoking, i.e., the slow elimination of capillary water (Arnold 1997, 61) (Fig. 3). There is a relationship between the duration of this stage and the dimensions of the sunken up-draught kiln, as well as of the pieces which form the load. The temperature can be increased to over 3508C, the limit of combustion for the organic material contained into the clay s fabric, but stabilised under 7008C. This represents the second stage of firing, when the process of sintering of the internal particles is initiated (Kingery 1997, 12). At a temperature between 5708C and 6008C, the elimination of water from the crystal structure of the minerals begins (Shepard 1956, 72, 81; Arnold 1997, 610; Gâţa 2000, 124), which is associated with the beginning of the emission of red glow from the incandescent clay objects (Fig. 4). The positioning of the wasters on the flue allows the potter to observe this phenomenon, and consequently to control the firing. The heating of the interior of the sunken up-draught kilns and of the ceramic load over 7008C produces a strong air draught, which allows the flames from the firebox to reach up to the flue. Frequently, when the temperature reaches 9008C at the middle of the load, coals already fill the m firebox and the air draught diminish, consequently impeding the rise of the temperature. One can infer that the volumetric structure and the dimensions of this type of kiln are optimal for temperatures under 9008C (as the analyses of most of the Cucuteni- Tripolye fine ceramics show). In order to raise the temperature over this thermal threshold, one should Figure 3. The water smoking of the sunken up-draught kiln; Vadastra, Photograph by Dragoş Gheorghiu. 175 Ed. Marcos Martinón-Torres

4 Gheorghiu Figure 4. Incandescent clay objects seen through the flue; Vadastra, Photograph by Dragoş Gheorghiu. discharge the firebox and initiate a new fire, which would help increase the temperature to over 10008C. As this is a procedure with high energy expenditure and a high level of difficulty, it is possible that high temperatures were not reached in draught sunken up-draught kilns in the East European prehistory, a hypothesis supported by the analysis of prehistoric ceramics (Gâţa 2000, 125). It is important to mention that a sunken up-draught kiln does not produce a uniform temperature inside the vessel chamber, the difference between the temperatures (measured with the help of Platinum-Rhodium thermocouples placed on the upper side of the platform and at the middle of the load) sometimes being up to 2008C (see Gheorghiu 2002). Except for the stages related to water elimination, which require the observance of a well determined time span, the other stages of the firing process can be carried out using different time spans. Therefore, a complete firing process could last between 5 and 18 hours (see Deca 1982, 214) depending on the type of load, fuel, and temperature of the environment, but also on the style of firing adopted by the potter. 4. Discussion and archaeological implications The experiments carried out with sunken up-draught kilns brought to light a series of problems, not mentioned so far in the specialist literature. Some of these leave visible traces in the archaeological record, while some only help experimenters to refine the chaînes opératoires pertaining Craft and science: International perspectives on archaeological ceramics 176

5 Experiments with double chamber sunken up-draught kilns Figure 5. A child unloading the kiln; Vadastra, Photograph by Dragoş Gheorghiu. to the construction and use of this type of kilns. From the first category, four examples will be discussed. The first example refers to the technology involved in digging the ovoid and cylindrical shapes, which is a difficult operation due to the narrow dimensions of these features. This is the reason why the author assumed that children performed this operation. During the experiments, children were very efficient in loading and unloading the vessels, with the platform supporting their weight. The dimensions of the opening of the sunken up-draught (Fig. 5) kiln only allowed children to excavate its shapes. When the author built the platform, he asked a child to plait a twig structure inside the excavated fire chamber, which he fixed in the walls of the sunken up-draught kiln and on Figure 6. The walls of the kiln after 5 successive firings; Vadastra, Photograph by Dragoş Gheorghiu. 177 Ed. Marcos Martinón-Torres

6 Gheorghiu Figure 7. The partially collapsed platform at 12008C. One can notice the filled perforations near the fracture line; Vadastra, Photograph by Dragoş Gheorghiu. the upper part of a pedestal of clay left at the bottom of the fire chamber. One of the Cucuteni-Tripolye sunken up-draught kilns displays a profile left on the walls of the firing chamber (see Markevic 1981) for a better fixing of the twig structure of the platform. After finishing the plaiting of twigs, the child pressed wet clay on this structure and carried out some of the perforations sitting on the un-perforated area on top of the pedestal. Other perforations were made from outside the sunken up-draught kiln, with the help of a long pointed stick. As the Cucuteni-Tripolye excavated kilns show (see Markevic 1981), the large dimensions of the pedestals left the central part of the platforms without perforations, and so the positioning of the vessels inside the firing chamber should have been done in such a manner as to allow the circulation of the air and flames in this area. One might also assume that a child (entering from the firebox aperture) finished the lower surface of the wet platform and enlarged the perforations. The second example refers to the changes in the dimensions of the sunken up-draught kiln due to repeated firings. In a first stage, a sunken up-draught kiln is a negative shape dug into the ground, but consecutive firings transform into ceramic not only the load of vessels but also the soil, creating the walls of the kiln, which are visible in the archaeological record. One can thus conclude that the kiln is a ceramic construction produced by itself (Fig. 6). The third example refers to the behaviour of the platform at high temperatures. For a better understanding of the functioning of the sunken up-draught kilns, the perforated platform can be approached as a set of tubes for air absorption, with the role of inducing a distribution of the thermal shock. At temperatures over 11508C, one can notice the beginning of a process of vitrification in the filling up of the air-draught tubes with glass and slag, with consequences on the lowering of the temperature in the vessel chamber. It is assumed that the tubes of the platforms were cleaned periodically, because the occurrence of glass is not mentioned in relation to the Cucuteni-Tripolye excavated sunken up-draught kilns. The experiment with the collapsed alluvial clay platform at temperatures over 11508C led to the hypothesis that the respective feature was made from a refractory material, probably clay with a high proportion of sand. The repair of collapsed platforms (Fig. 7) in the Cucuteni-Tripolye tradition is not mentioned either in the existing literature, but only the repeated plastering of the interior of the firing chambers (Markevic 1981). This experiment could thus offer some hints for future analyses of the platform materials and for potential reconstructions. The fourth aspect refers to the micro-processes of evaporation of capillary water. When a wet sunken updraught kiln is fired for the first time, one can observe rhythmic fluctuations of the temperature inside the vessel chamber (Gheorghiu 2002, 92, diagram 3), due to the recurring evaporation of the water contained in the platform and the walls, which decreases the level of heat inside the sunken up-draught kiln on each occasion. One may infer that this phenomenon occurs (but at a microscale) also on the surface of the vessels fired during the water smoking stage. 5. Conclusion Although sunken up-draught kilns seem to have been studied in depth, one can conclude that every time experiments are conducted, they could bring new data and a new perspective that would lead to a better understanding of these complex ancient machines. Experiments Craft and science: International perspectives on archaeological ceramics 178

7 Experiments with double chamber sunken up-draught kilns demonstrate that the Cucuteni-Tripolye sunken up-draught kilns showed good visual ergonomics (since one operator could control the inputs and outputs of the kiln at every stage of the process), a good ergonomic dimension (both for adults and for the children who helped in the process) (Gheorghiu 2012), and that these dimensions, together with the shape of the vessel chamber, conferred to the machine the advantages of thermal efficiency. From the physical processes related to the growing in time of the efficiency of the vessel chamber, one can infer that sunken up-draught kilns were used for long periods of time, thereby constituting specialised places that lasted for years. References Arnold E.D., 1997, Ceramic theory and cultural process, Cambridge University Press, Cambridge. Comşa E., 1976, Caracteristicile şi însemna tatea cuptoarelor de ars din aria culturii Cucuteni-Ariuşd. Studii şi cerceta ri de istorie veche şi arheologie, 27, de Contenson H. and Courtois L.C., 1979, A propos des vases en chaux: Recherches sur leur fabrication et leur origine. Paléorient, 5, Dawson D. and Kent O., 1984, Methods of kiln reconstruction. Bulletin of the Experimental Firing Group, 2, Dawson D. and Kent O., 1985, Kiln superstructures - The Bickley experiment. Bulletin of the Experimental Firing Group, 3, Deca E., 1982, Centrul de ceramica din comuna Lungeşti, Jud. Vâlcea. Buridava - Studii şi materiale, , Râmnicu Vâlcea Museum, Râmnicu Vâlcea. Ellis L., 1984, The Cucuteni-Tripolye culture: A study in technology and the origins of complex society, British Archaeological Reports International Series 217, Oxford. Dumitrescu Vl., Dumitrescu H., Petrescu Dâmboviţa M. and Gostar N., 1954, Ha ba şeşti. Monografie arheologica, Editura Academiei, Bucharest. Gâţa G., 2000, A technological survey of the pottery. In Dra guşeni. A Cucutenian community, (eds. S. Marinescu- Bîlcu and A. Bolomey), , Editura Enciclopedica, Bucharest; Wasmuth Verlag, Tübingen. Gheorghiu D., 2001, South-Eastern European Early and Late Chalcolithic. In Encyclopedia of Prehistory, Vol. IV, (eds. P. Peregrine and M. Ember), , Kluwer Academic and Plenum, New York and London. Gheorghiu D., 2002, Fire and air draught: Experimenting the Chalcolithic pyroinstruments. In Fire in archaeology, (ed. D. Gheorghiu), 83-94, British Archaeological Reports International Series Archaeopress, Oxford. Gheorghiu D., 2006, On Chalcolithic ceramic technology: A study case from the Lower Danube traditions. In Ceramic studies. Papers on social and cultural significance of ceramics in Europe and Eurasia from prehistoric and historic times, (ed. D. Gheorghiu), 29-42, British Archaeological Reports International Series Archaeopress, Oxford. Gheorghiu D., 2007, Chalcolithic pyroinstruments with airdraught - An outline. In Fire as an instrument. The Archaeology of pyrotechnologies, (ed. D. Gheorghiu), 41-51, British Archaeological Reports International Series Archaeopress, Oxford. Gheorghiu D., 2012, The ergonomics of the chainesoperatoires. In Estudos de tecnologia ceramica, (eds. L. Oosterbeek and J.F. Cerezer), 59-62, Arkeos 31, Tomar. Gourdin W.H. and Kingery W.D., 1975, The Beginnings of Pyrotechnology: Neolithic and Egyptian lime plaster. Journal of Field Archaeology, 2, Hamer F., 1975, The Potter s Dictionary of materials ad technologies. Pitman Publishing, London; Watson Guptill Publications, New York. Kingery W.D., 1997, Operational Principles of Ceramic kilns. In The Prehistory and History of Ceramic Kilns, (ed. P.M. Rice), 11-20, Proceedings of the Prehistory and History of Ceramic Kilns, 98 th Annual Meeting of the American Ceramic Society in Indianapolis, American Ceramic Society, Colombus, OH. Marinescu-Bîlcu S., 2000, The pottery. Tradition and innovation. In Dra guşeni. A Cucutenian community, (eds. S. Marinescu-Bîlcu and A. Bolomey), , Editura Enciclopedica, Bucharest; Wasmuth Verlag, Tübingen. Markevic V.I., 1981, Pozne-Tripolskie plemena severnog moldovii. Kishinev. Rehder J.E., 2000, The mastery and uses of fire in Antiquity, McGill-Queen s University Press, Montreal. Rhodes D., 1981, Kilns. Design, construction and operation, Chilton Book Company, Radnor, PA. Rye O.S. and Evans C., 1976, Traditional pottery techniques of Pakistan: field and laboratory studies, Smithsonian Contributions to Anthropology 21, Smithsonian Institution Press, Washington DC. Shepard A.O., 1956, Ceramics for the archaeologist, Carnegie Institution of Washington Publication 609, Washington DC. Simpson J., 1997, Prehistoric ceramics in Mesopotamia. In Pottery in the making. World Ceramic Traditions, (eds. I. Freestone and D. Gaimster), 38-43, British Museum Press, London. ACKNOWLEDGEMENTS I thank Dr. Michela Spataro for support and Professor George Nash for reviewing the English of the manuscript and for useful comments. I would also like to thank the two reviewers who improved the coherence of the text. The pyro-experiments were financed by three CNCSIS grants (Nos. 1612; 112; 945), and my first experiments with sunken up-draught kilns were carried out with Dr. Alex Gibson. 179 Ed. Marcos Martinón-Torres