A late classic lime-plaster kiln from the Maya center of Copan, Honduras.

Abrams, Elliott M. ; Freter, AnnCorinne

Introduction

At the ancient Maya kingdom of Copan, Honduras [ILLUSTRATION FOR FIGURE 1 OMITTED], a multi-year programme of settlement archaeology has addressed broad-ranging goals with concomitant results. Settlement distribution and concentration, defined for the Late Classic (AD 600-900) and early Post-Classic periods (AD 900-1200), show that the fragmentation of the Classic kingdom was not a sudden catastrophe but a more complex and protracted process (Freter 1988; 1992; 1994; Webster & Freter 1990a). A detailed methodology for generating population size was developed (Webster & Freter 1990b), and the extensive regional data provided the opportunity to refine obsidian hydration dating as a chronometric technique in the Maya region (Freter 1993).

Economic loci of production are often evident only through systematic random sampling within the context of regional settlement research. Too often, sites with sizable and conspicuous architectural remains are the exclusive subject of excavation. Yet production areas, especially in societies with relatively limited economic specialization such as the Classic Maya, are often located away from buildings and are thus overlooked (Freter 1991).

Among results from the large-scale settlement research at Copan is a lime-plaster kiln, published in this note (referred to previously in Abrams & Freter 1988; Abrams 1994). Although the presence of a lime kiln at the Maya centre of Tikal, Guatemala, has been supposed (Coe 1990: 877), the kilns themselves at Tikal have proved elusive. To our knowledge, the lime kiln described here is the only one discovered at a Classic Maya site.

The kiln

During the 1984 archaeological survey of the Copan Valley, a small mound - Site 70, Mound 32 - was test-pitted in the 15% stratified random sampling of the 1425 sites within the Copan Valley (Webster & Freter 1985; Freter 1988; 1992). It is located about 1 km east of the Sepulturas barrio, on the second terrace above the Copan River at approximately 596 m ASL [ILLUSTRATION FOR FIGURES 2 & 3 OMITTED]. At the time, the mound was in a ploughed field planted in tobacco, elevated only about 60-70 cm above the level of the field. Its area was about 16 sq. m. Late Classic period ceramic sherds and obsidian tools were collected from the surface. A 2-x-2-m excavation unit, in the southwest corner of the mound and straddling the rise and the level surface, extended to the centre of the mound. (This placement was dictated by the rural test excavation permit from the Instituto Hondureno de Antropologia e Historia, which required that no exposed architecture be trenched or removed unless it could be restored). Thus the goal of the test excavation unit was to uncover midden or floor contexts associated with the mound to reconstruct basic site function and chronology; it exposed about 20% of the total surface area and a sample of the mound's floor deposits.

Level 1, the top 20 cm, was the plough zone. Very few artefacts were recovered. However, towards the bottom of this level in the northeast section, burnt earth and charcoal were encountered; at this soil change, level 1 was terminated. In level 2, 20-40 cm below the ground surface on the exterior of the burnt earth, was an in-curving wall about 25 cm thick constructed of burnt earth (not adobe brick) tempered with grass, and pedologically distinct (10 YR 5/3) from the surrounding soil matrix. Pieces of burnt limestone were encountered in this level. The final excavation level 3 reached the exterior base of the kiln at a depth of 75 cm, at which point a burnt clay floor (7.5 YR 6/8) was encountered. Here the soil was very hard and reddened. In the wall of the kiln, 50 cm thick and standing about 60 cm high, was a small opening 20 cm in height. A reconstruction drawing [ILLUSTRATION FOR FIGURE 4 OMITTED] sketches a domed or semi-enclosed kiln, about 4 m in diameter with a projected interior volume of perhaps 17 cu. m.

Function of the kiln

This feature is identified as a kiln for the burning, or calcining, of limestone, a necessary process in producing lime, a key ingredient in lime plaster. Whereas construction plaster in the Old World was based on burnt gypsum or limestone (Kingery et al. 1988), plaster in Mesoamerica was exclusively lime-based. Without lime, a simple mixture of crushed limestone and water is too weak to protect masonry walls and roofs from the destructive force of torrential tropical rains in the southern Maya lowlands (Hyman 1970). The need for a strong and durable lime plaster was especially important at Copan, where only mud was used as mortar in masonry (Pollock 1965).

This functional identification is based on positive as well as negative evidence. The interior contents included only materials which directly indicate the burning of wood and of limestone; and the only feature that would have exclusively contained those materials is a lime-processing kiln. The firing of ceramics in the rural zones of Copan did not involve a kiln (Freter 1991), eliminating this possible alternative. Similarly, sweatbaths at Copan (Cheek & Spink 1986: 59) and elsewhere (Rivero Torres 1987: 16; Sheets 1992: 98) bear significant differences from the lime kiln: sweatbaths are

1 rectangular rather than circular,

2 bear straight rather than in-curving walls,

3 have localized rather than scattered interior burning and

4 are adjacent to residential structures rather than distanced from them.

The amount of lime generated from this type of kiln would have been considerable. Its interior volume of 17 cu. m would have contained 6-10 cu. m of limestone, capable of generating an equivalent amount of lime cement. Since lime plaster is a mixture of approximately one part lime to two parts aggregate (generally crushed stone or sascab, a naturally decomposed limestone), the kiln could have generated material for about 24 cu. m of lime plaster at each burning. This amount could coat the average masonry structure at Copan; several kilns fired simultaneously, enough to surface the largest (Abrams 1994). These volumes suggest the lime's use in construction rather than for soaking maize, a domestic use of lime. And lime used to make limewater was probably produced on a smaller scale, in smaller features, and on a household or courtyard basis of production and consumption.

Chronology of the kiln

The sherds from the plough zone overlying the kiln feature date to the Coner period, c. AD 650-1150 (Webster & Freter 1990a). Two obsidian blades from the plough zone were dated as part of the Copan Obsidian Hydration Dating Project, and yielded dates of AD 901 and AD 952[+ or -]140 years (2[Sigma] error range; from Freter 1988). These overlying samples, coupled with the ceramic data, indicate the kiln was in operation between AD 750 and 900, or towards the end of the Late Classic period.

Discussion

For over 100 years, Classic Maya architecture has been a central focus of archaeological investigation, yet it has only been of late that the social, political and economic contexts of architecture have been analytically considered (summarized in Abrams 1994). This lime-plaster kiln broadens our ability to consider some of these dimensions of ancient Maya culture.

The kiln tells us directly about the technology of plaster production. The technology assumed for the Late Classic Maya, based on ethnographic descriptions of limestone burning by early 20th-century Yucatecan Maya, is the open-air wooden pyre, or calera; described by Earl Morris and his colleagues (1931) during their excavation of the Temple of the Warriors at Chichen Itza, it is assumed to be prehispanic. Morris et al. (1931: 220) state:

The Yucatecan method of producing lime . . . is an ancestral heritage that has come down through the centuries with practically no change.

The Copan lime kiln indicates that technological variability may exist among Maya centres. Its operation during the Late Classic period suggests that kilns perhaps replaced caleras to increase fuelwood efficiency under deteriorating environmental conditions (Abrams & Rue 1988; Wingard 1992).

Lime plaster is the most complex manufactured building component in Maya architecture. To calcine limestone, a temperature of about 900 [degrees] C is needed; too low or too high a heat fails to produce the required qualities in the lime (Eckel 1928). The mixing of lime with its various aggregates also required technical skills, as did several other aspects of the manufacturing process. Certainly a specialist was required to operate the kiln.

Several traits of this specialized economic position can be inferred. Since plaster (as well as masonry architecture) could only be produced in the dry season, these specialists necessarily were seasonal rather than full-time. Further, the mounds associated with Site 70, Mound 32 [ILLUSTRATION FOR FIGURE 3 OMITTED] represented commoner structures; this sets the socio-economic status of plaster specialists as not significantly (if at all) above that of the majority commoner segment within Maya society. This inference supports a similar conclusion concerning the socio-economic status of some ceramic producers (Freter 1991), wood-workers and obsidian-workers (Mallory 1984) and ground-stone artisans (Spink 1983).

The producers of plaster were not the consumers of their product; rather, lime plaster was consumed by the members of the elite to protect and decorate masonry structures, the majority of which were used or occupied by the elite. Thus plaster produced by commoners had to be moved up, in a sense, to elite consumers within a redistributive system: there certainly was not a 'market' for lime plaster. Located about 200 m northeast of the kiln is a larger courtyard (Mounds 7-10; [ILLUSTRATION FOR FIGURE 3 OMITTED]), representing the lowest level of ranked lineage elite within the five-tiered hierarchy of elite at Copan (Abrams 1994). Within the model of the Classic Maya kingdom structured as a segmentary state of clustered maximal lineages (Sanders 1989), commoner plaster specialists were seasonally required to manufacture plaster for the various elite members of their lineage and, on occasion, for the royal elite. The management and movement of that plaster would have been most directly orchestrated by low-ranking elite members such as those living in one of the structures of Mounds 7-10. This illustrates the varied responsibilities of ranked members of the elite and supports the model of a redistributive economy characterizing the ancient Maya state.

The fact that one kiln firing could generate enough plaster to coat an average-sized masonry structure (Abrams 1994) suggests that each of the maximal lineages during the Classic period in Copan's history operated perhaps only one or two kilns, depending on the population size and wealth of lineage units. If plaster for some large structures in the Main Centre was obtained from several lineages as their obligation to the state, this would account for the variations in plaster quality and composition on individual buildings. Three samples of plaster from the Casa de Monjas at Chichen Itza revealed considerable variation in aggregate proportions (Rays 1934), as did samples analysed from the Main Centre of Copan (Hyman 1970). This variation supports the concept of independent lineage-based producers; further, the low number of kilns required by the Classic Maya may also account for the uniqueness of its discovery.

Conclusion

This note has described a lime kiln at the Classic Maya centre of Copan, Honduras, the first discovered at a Maya site. Its location and size respectively suggest that plaster manufacture was conducted by commoner seasonal specialists and that few kilns were needed to generate the plaster placed on Maya masonry architecture. The low number of kilns and their location outside courtyards may account for their presumed absence at other Maya centres.

Acknowledgements. The field and laboratory research for these data were funded by the National Science Foundation (BNS-8419933) and supported by the Instituto Hondureno de Antropologia e Historia. We are indebted to Nancy Gonlin and Mike Davis for their assistance in the field excavations, David Webster and William T. Sanders, the directors of the PAC II project, and Steve Houston for critical and editorial suggestions. Sam Girton of Instructional Media at Ohio University assisted with the preparation of the final figures. As always, we bear full responsibility for all errors of fact, logic or interpretation.

References

ABRAMS, E. 1994. How the Maya built their world: energetics and ancient architecture. Austin (TX): University of Texas Press.

ABRAMS, E. & A. FRETER. 1988. Intra-polity economics at the Maya center of Copan, Honduras. Paper presented at the 53rd Annual Meeting of the Society for American Archaeology, Phoenix.

ABRAMS, E. & D. RUE. 1988. The causes and consequences of deforestation among the Prehistoric Maya, Human Ecology 16(4): 377-95.

CHEEK, C. & M. SPINK. 1986. Excavaciones en el grupo 3, estructura 223 (operacion VII), in W. Sanders (ed.), Excavaciones en el area urbana de Copan I: 27-154. Tegucigalpa: Instituto Hondureno de Antropologia e Historia.

COE, M. 1990. Excavation in the Great Plaza, North Terrace and North Acropolis of Tikal. Philadelphia (PA): University of Pennsylvania Press. Tikal Report 14(III), University Museum Monograph 61.

ECKEL, E. 1928. Cements, limes, and plasters: their materials, manufacture and properties. New York (NY): John Wiley.

FASH, W. & K. LONG. 1983. Mapa arqueologica del Valle de Copan, in C. Baudez (ed.), Introduccion de la Arqueologia de Copan, Honduras III: 3-48. Tegucigalpa: Instituto Hondureno de Antropologia e Historia.

FRETER, A. 1988. The Classic Maya collapse at Copan, Honduras: a regional settlement perspective. Unpublished Ph.D thesis, Department of Anthropology, Pennsylvania State University.

1991. A reconstruction of the Late Classic rural ceramic production system in the Copan Valley, Honduras. Paper presented at the 56th Annual Meeting of the Society for American Archaeology, New Orleans.

1992. Chronological research at Copan: methods and implications, Ancient Mesoamerica 3: 117-34.

1993. Obsidian-Hydration dating: its past, present, and future application in Mesoamerica, Ancient Mesoamerica 4: 285-303.

1994. The Classic Maya collapse at Copan, Honduras: an analysis of Maya rural settlement trends, in G. Schwartz & S. Falconer (ed.), Archaeological views from the countryside: village communities in early complex societies: 160-76. Washington (DC): Smithsonian Institution Press.

HYMAN, D. 1970. Precolumbian cements: a study of the calcareous cements in Prehispanic Mesoamerican building construction. Baltimore (MD): Johns Hopkins University Press.

KINGERY, D., P. VANDIVER & M. PRICKETT. 1988. The beginnings of pyrotechnology, part II: production and use of lime and gypsum plaster in the Pre-Pottery Neolithic Near East, Journal of Field Archaeology 15: 219-44.

MALLORY, J. 1984. Late Classic Maya economic specialization: evidence from the Copan obsidian assemblage. Unpublished Ph.D thesis, Department of Anthropology, Pennsylvania State University.

MORRIS, E., J. CHARLOT & A. MORRIS. 1931. The Temple of the Warriors at Chichen Itza, Yucatan 2. Washington (DC): Carnegie Institution of Washington. Publication 406.

POLLOCK, H.E.D. 1965. Architecture of the Maya Lowlands, in G. Willey (ed.), Handbook of Middle American Indians 2: 378-440. Austin (TX): University of Texas Press.

RIVERO TORRES, S. 1987. Los Cimientos, Chiapas, Mexico: a Late Classic Maya community. Provo (UT): Brigham Young University. Papers of the New World Archaeological Foundation 51.

ROYS, L. 1934. The engineering knowledge of the Maya, Carnegie Institution of Washington 436(6): 27-105. Washington (DC).

SANDERS, W.T. 1989. Household, lineage and the state in 8th-century Copan, in D. Webster (ed.), House of the Bacabs, Copan: a study of the iconography, epigraphy and social context of a Maya elite structure: 89-105. Washington (DC): Dumbarton Oaks.

SHEETS,P. 1992. The Ceren site. Fort Worth (TX): Holt, Rinehart & Winston.

SPINK, M. 1983. Metates as socioeconomic indicators during the Classic Period at Copan, Honduras. Unpublished Ph.D thesis, Department of Anthropology, Pennsylvania State University.

WEBSTER, D. & A. FRETER. 1990a. Settlement history and the Classic collapse at Copan, Honduras: a redefined chronological perspective, Latin American Antiquity 1: 66-85.

1990b. The demography of Late Classic Copan, in T.P. Culbert & D. Rice (ed.), Precolumbian population history in the Maya lowlands: 37-61. Albuquerque (NM): University of New Mexico Press.

WINGARD, J. 1992. The role of soils in the development and collapse of Classic Maya civilization at Copan, Honduras. Unpublished Ph.D thesis, Department of Anthropology, Pennsylvania State University.