The archaeological site of Monte Albán in the state of Oaxaca, southern Mexico, was founded as the capital of the Zapotec state ca. 500 BCE. Ceramic production in the immediate vicinity of the site has made continuous use of materials derived from the local feldspathic rocks. Materials analyzed in the present study were collected from potters at Santa María Atzompa, a modern pottery-producing village six kilometers north of the center of the archaeological site. Broad-spectrum chemical analyses using instrumental neutron-activation analysis (INAA) were complemented with x-ray diffraction analysis of the feldspars and petrographic examination under polarized light microscope and an electron microprobe. Comparison of the combined results with previously published studies of Monte Albán ceramics indicate continuity in methods of procuring and preparing ceramic raw materials as well as in general geologic “provenance.” The Atzompa materials thus represent a key contribution to the growing database for sourcing Mesoamerican archaeological ceramics and tracing the evolution of ceramic technology

Keywords: Ceramics, Mesoamerica, provenance, geochemistry, mineralogy, petrography

1. Local Geology and Sources of Ceramic Raw Materials

Monte Albán is the highest ridge of a complex of hills on the western edge of the Valley of Oaxaca (see Figure 1). Hills ascend more than 1800 meters above mean sea level, three hundred meters above the valley floor and the Atoyac River floodplain. Cretaceous limestones outcrop at the surface but are underlain by a Precambrian basement complex of feldspar-rich metamorphic rocks (Fries & Schmitter, 1962; INEGI, 1984; Ortega-Gutiérrez, 1990). These "Oaxaca Complex" rocks are at least a billion years old and are thought to have been attached to both South American and Laurentian terranes during the period of the Grenville orogeny (Keppie & Ortega-Gutiérrez, 1995; Mora & Valley, 1985; Mora et al., 1986). Outcrops in gullies on the valley sides indicate long-term erosion and transport into valley floor deposits. The materials used for pottery production at Atzompa are all derived from erosion and weathering of these rocks but at least five separate sources are mined by the potters we studied (see Figure 2).

The Atzompa potters distinguish two main classes of materials: 1.) smooth alluvial clays which are soaked, and 2.) coarse gritty clays which contain nonplastic, feldspathic “temper” material. Sources of white and black gritty clay are distinguished by the potters and both are generally beaten, then sieved to form a powder. The most widely used source of "white gritty clay" feldspathic temper is 3.5 km west-southwest of Santa María Atzompa on a hillslope which shows evidence of extensive utilization (see Figure 3). Potters have mined poorly sorted deposits of rock and clay which appear to result from a large debris flow, probably in response to an earthquake. All of the mineral grains in the rock fragments examined and most of the clasts used as temper are plagioclase feldspar. The parent rock should therefore be classified as anorthosite as opposed to diorite (Shepard, 1963, 1967) or gneiss (INEGI, 1984; Payne, 1994). Ongoing petrographic analyses have corroborated the previous work on Monte Albán pottery by Shepard (1963, 1967), however, in that the feldspar clasts in both pottery fabrics and the rock fragments from this source exhibit antiperthite twinning and spots of higher birefringence caused by alteration to epidote and mica group minerals. X-ray diffraction spectra exhibit a 3.19-3.21 angstrom peak characteristic of plagioclase feldspars but peak intensities were otherwise highly variable among the grains analyzed.

The Atzompa potters have three sources for alluvial clay. A source in "high alluvium" of a small tributary valley near San Lorenzo Cacaotepec (see Figure 4) is preferred for buff firing (crema) paste. Although an alluvial chronology has yet to be developed in the Valley of Oaxaca, preliminary geomorphological studies suggest that the high alluvium predates the founding of Monte Albán (Kirkby, 1973). The "low alluvium" of the Atoyac River may be much younger, however, and the Laguna source of the Atzompa potters was in a fairly extensive area of these deposits subject to seasonal inundation (see Figure 5). The Laguna clays contain more extraneous material than those from San Lorenzo and require more preparation, including levigation. The third source of alluvial clay is near the town of San Felipe Tejalapan, approximately 10 km east of Atzompa. Formerly a favored source (Hendry, 1992, p. 64), the mine is no longer used and is reported to be deep and dangerous. A gray clay, it fires to a warm slightly reddish color.

2. Production Recipes and Compositional Groups

The Atzompa potters procure the raw materials, combine them, and process them based on criteria of vessel type, size, and finish. Both the petrographic fabric and the chemical composition therefore vary considerably within this modern ceramic production "site." The Laguna clay is always mixed with the "white gritty clay" feldspathic temper, which is particularly valued for making griddles (comales) and other cooking vessels and for making larger vessels. The powder is spread on a mat and the wet clay kneaded into it. Producers of comales prefer to use clay from the upper levels of the Laguna source and form their vessels on a bed of wood ash.

Thin-sections were prepared from an Atzompa comal sherd for petrographic examination under both polarized light microscope and an electron microprobe (Figure 6). Table 1 is a modal analysis of the sherd's inclusions, which are set in a matrix of dehydrated clay minerals, opaque grains, and very fine-grained (<.05 mm) quartz and plagioclase. The latter inclusions probably derive from the Laguna alluvial clay deposits rather than having been deliberately introduced as temper. Antiperthite lamellae were visible in approximately half of the plagioclase grains, and measurement of extinction angles using the Michel-Levy method (Nesse, 1991, p. 272-273) suggests a sodic composition, ranging from An₂₅ to An₃₅.

Table 1: Petrographic Modal Analysis of Comal Sherd from Santa Maria Atzompa

Seventy-nine samples of raw materials and pottery from various stages of the production process were subjected to instrumental neutron activation analysis (INAA) at the University of Missouri Research Reactor (Thieme & Neff, 1993). Compositional groups identified by principal components analysis (Neff, 1994) generally mirror the geography of raw material sources (Figure 7). Non-plastic material (temper) is at the low end of component 1, and Laguna clays at the high end. Laguna pottery, which incorporates temper added by the potters, falls at an intermediate position. San Felipe clays and sherds fall within the range of chemical variation of the Laguna pottery. Also near the center of the plot, but partially distinct from the Laguna ware, are the San Lorenzo pottery, prepared pastes and clays. These form a single, relatively homogeneous group. Colorants, kiln clays and pottery from other towns are clearly separated from the various subgroups of Atzompa pottery clays and pottery.

In Figure 8, the chemical basis of group separation is illustrated by plotting coordinates for the elements on the first two components, in effect providing a two-dimensional representation of the variance-covariance structure of the data (Neff, 1994). As shown, deviations towards the low end of component 1 are created largely by high sodium, which in the case of the tempering materials must come from the feldspar-containing rock fragments, and by dilution of a number of elements, most obviously arsenic, thorium, antimony, and chromium. Laguna clays are depleted in sodium but enriched in the dilution elements while Laguna pottery, being a mixture of Laguna clay and tempering material, occupies an intermediate position on this enrichment-dilution axis. San Lorenzo clays are not as noticeably affected by addition of temper, and the clays and pottery thus form a single group. The colorants and kiln clays are distinguished by anomalously high hafnium and zirconium, probably reflecting high proportions of zircon grains in these relatively silty materials.

3. Implications for "Sourcing" Archaeological Ceramics

The Atzompa ceramic technology is of interest to archaeologists working in Mesoamerica because of its spatial proximity to Monte Albán and apparent continuity in methods of processing, preparation, and vessel manufacture (Shepard, 1963, 1967; Thieme and Thieme, 1997). The compositional profile based on the INAA results summarized above may therefore represent a "fingerprint" for this portion of the Valley of Oaxaca in archaeological ceramics as well. Differences between the Atzompa profile and those for contemporary ethnographic materials from San Marcos Tlapazola, San Bartolo Coyotepec, and Ocotlan de Morelos (Herrera et al., 1999, p. 975; Thieme and Neff, 1993) derive largely from geologic variation within the valley. High concentrations of sodium, strontium, and aluminum differentiate the Atzompa materials and presumably all pottery manufactured in the immediate vicinity of Monte Albán.

Some of the pitfalls in the compositional analysis of archaeological ceramics are also evident, however, from the Atzompa ethnographic results. The ways in which Atzompa potters classify their materials are different from both archaeological and industrial ceramic classification systems, for example, and this is sometimes manifested in the products and byproducts of their technology. All of the materials are referred to as clay (barro) although nonplastic inclusions are introduced into the pottery fabric with the gritty clays and some of the smooth clays. Materials recognized as distinct sources by the potters, such as the white and black gritty clays, cannot be distinguished using the chemistry of the final product.

Although Atzompa potters have been studied by ethnographers since the 1950's (e.g. Hendry, 1992) the present study is uniquely ethnoarchaeological in that production recipes were documented in detail and materials were analyzed using methods applicable to archaeological ceramics. Mixing of different alluvial clays as well as addition of various "temper" materials results in products derived from deposits as much as 10 km away from the site of production. The relative contribution of the various materials does not simply "fall off" with geographical distance (c.f. Arnold et al., 1991) since choices of materials are based on the potters' own knowledge of sources and methods of production. Although many innovations have been documented by the ethnographers, there is also much continuity in knowledge transmitted between generations in an evolving technological tradition.

Some changes in the raw materials available to potters in the Valley of Oaxaca have undoubtedly occurred due to landscape changes since the birth of the Zapotec state ca. 500 BCE (Blanton et al., 1999; Joyce, 1994; Kirkby, 1973). Even over the 50 year period of detailed ethnographic documentation, several sources appear to have been exhausted or become unavailable. More detailed geological mapping and prospection are therefore essential if we are to trace the technology's Prehispanic roots at the numerous ceramic production sites documented by archaeological surveys (Blanton, 1978; Feinman, 1982; Kowalewski, 1976). Compositional profiles for archaeological assemblages have been tentatively correlated with the ethnographic compositional profiles (e.g. Herrera et al., 1999) but archaeological conclusions may require modification once a more comprehensive geochemical database has been developed.

Ongoing study of the materials from Atzompa and nearby ceramic production sites in the Valley of Oaxaca thus parallels a general shift in archaeological ceramic studies from simply assigning provenance towards reconstruction of local scale production and distribution (Day et al., 1997, 1999; Feinman et al., 1992) and technological traditions (Feinman, 1986; Feinman et al., 1989; Wright, 1993; Whitelaw et al., 1997). It is possible to "source" some Oaxacan pottery using geochemical, mineralogical, and petrographic methods, and pottery manufactured in the immediate vicinity of Monte Albán appears to have a unique geochemical fingerprint. The growing database for sourcing archaeological ceramics from Monte Albán and other Prehispanic sites in southern Mexico will enable us to more precisely chart the evolution of Mesoamerican ceramic traditions.

References Cited

Arnold, D. E., Neff, H. & Bishop, R. L. (1991). Compositional analysis and "sources" of pottery: an ethnoarchaeological approach. American Anthropologist 93: 70-90.

Blanton, R. E. (1978). Monte Albán: Settlement Patterns at the Ancient Zapotec Capitol. New York: Academic Press.

Blanton, R. E., Feinman, G. M., Kowalewski, S. A. & Nicholas, L. M. (1999). Ancient Oaxaca: The Monte Albán State. Cambridge: Cambridge University Press.

Day, P. M., Wilson, D. E. & Kiriatzi, E. (1997). Reassessing specialization in prepalatial Cretan ceramic production. In (R. Laffineur & P. P. Betancourt, Eds) TEXNH: Craftsmen, Craftswomen, and Craftsmanship in the Aegean Bronze Age, pp. 275-290. Liege: Aegaeum 16.

Day, P. M., Kiriatzi, E., Tsolakidou, A.. & Kilikoglou, V. (1999). Group therapy in Crete: A comparison between analyses by NAA and thin section petrography of Early Minoan pottery. Journal of Archaeological Science 26, 1025-1036.

Feinman, G. M. (1982). Appendix IX: Production Sites. In (R. E. Blanton, S. A. Kowalewski, G. M.Feinman, & J. Appel) Monte Albán’s hinterland, pt. I: pre-Hispanic settlement patterns of the central and southern parts of the Valley of Oaxaca, Mexico. Museum of Anthropology Memoirs 15. Ann Arbor: University of Michigan.

Feinman, G. M. (1986). The emergence of specialized ceramic production in formative Oaxaca. Research in Economic Anthropology, supplement 2, 347-373.

Feinman, G. M., Cooper, R. F., Cook, G. B. & Nicholas, L. M. (1989). A technological perspective on changes in the ancient Oaxacan grayware ceramic tradition: preliminary results. Journal of Field Archaeology 16, 331-344.

Feinman, G. M., Kowalewski, S. A., Banker, S., and Nicholas, L. M. (1992). Ceramic production and distribution in late postclassic Oaxaca: stylistic and petrographic perspectives. In (G. J. Bey III & C. A. Pool, Eds) Ceramic production and distribution: an integrated approach, pp. 235-259. Boulder: Westview Press.

Fries, C. Jr. & Schmitter, E. (1962). Rocas Precámbricas de Edad Grenviliana de la Parte Central de Oaxaca en el Sur de México. Estudios Geocronológicos de Rocas Mexicanas, Instituto de Geología Boletin 64. Mexico, D.F.: Universidad Nacional Autonóma de México.

Hendry, J. (1992). Atzompa: A Pottery Producing Village of Southern Mexico in the Mid-1950’s. Vanderbilt University Publications in Anthropology 40. Nashville: Vanderbilt University.

Herrera, R. S., Neff, H., Glascock, M. D. & Elam, J. M. (1999). Ceramic patterns, social interaction, and the Olmec: Neutron activation analysis of Early Formative pottery in the Oaxaca Highlands of Mexico. Journal of Archaeological Science 26, 967-987.

Instituto Nacional de Estadistica, Geografia, e Informatica (1984). Geologic Map E14-9, Oaxaca 1:250,000. Mexico City: INEGI.

Joyce, A. A. (1994). Monte Albán en el contexto pan-regional. In (M. Winter, Ed) Monte Albán, Estudios Recientes, pp. 63-76. Oaxaca: Proyecto Especial Monte Albán 1992-1994.

Keppie, J. D. & Ortega-Gutiérrez, F. (1995). Provenance of Mexican terranes: isotopic constraints. International Geological Review 37, 813-824.

Kirkby, A. V. T. (1973). The Use of Land and Water Resources in the Past and Present Valley of Oaxaca, Mexico. Prehistory and Human Ecology of the Valley of Oaxaca 1. Ann Arbor: University of Michigan Museum of Anthropology.

Kowalewski, S. A. (1976). Prehistoric Settlement Patterns in the Central Part of the Valley of Oaxaca. Tucson: University of Arizona, Ph.D. dissertation.

Mora, C. I. & Valley, J. W. (1985). Ternary feldspar thermometry in granulites from the Oaxacan Complex, Mexico. Contributions to Mineralogy and Petrology 89, 215-225.

Mora, C. I., Valley, J. W. & Ortega-Gutiérrez, F. (1986). The temperature and pressure conditions of Grenville-age granulite-facies metamorphism of the Oaxacan Complex, southern Mexico. Instituto de Geología, UNAM, Revísta 6, 222-242.

Neff, H. (1994). RQ-Mode principal component analysis of ceramic compositional data. Archaeometry 36, 115-130.

Nesse, W. D. (1991). Introduction to Optical Mineralogy. New York: Oxford University Press.

Ortega-Gutiérrez, F. (1984). La evolución tectónica premissisipica del sur de México. Instituto de Geología, UNAM, Revísta 2, 112-131.

Ortega-Gutiérrez, F. (1990). Acapulco Trench to the Gulf of Mexico across Southern Mexico, Centennial Continent/ Ocean Transect 13. Boulder: Geological Society of America.

Payne, W. O. (1994). The raw material and pottery-making techniques of Early Formative Oaxaca: an introduction. In (K. V. Flannery & J. Marcus, Eds) Early Formative Pottery of the Valley of Oaxaca, Mexico. Prehistory and Human Ecology of the Valley of Oaxaca 10, pp. 7-16. Ann Arbor: University of Michigan Museum of Anthropology.

Shepard, A. O. (1963). Notes from a Ceramics Laboratory, Beginnings of Ceramic Industrialization, An example from the Oaxaca Valley. Washington, D. C.: Carnegie Institution of Washington.

Shepard, A. O. (1967). Preliminary notes on the paste composition of Monte Albán pottery. In (A. Caso, I. Bernal, and J. R. Acosta), La Cerámica de Monte Albán, pp. 477-484. Mexico City: Instituto de Antropología e Historia.

Thieme, M. S. & Neff, H. (1993). Examination and Analysis of Clay Materials in a Pottery Producing Town in Oaxaca, Mexico. Paper presented in Ceramic Ecology Symposium at the 92nd Annual Meeting of the American Anthropological Association, Washington, D.C.

Thieme, M. S., & Thieme, D. M (1997). Tradition and continuity of ceramic production and utilization of clay sources in the vicinity of Santa María Atzompa from the Late Formative to the present. Paper presented in the Ceramic Ecology Symposium at the 98th Annual Meeting of the American Anthropological Association, Washington, D. C.

Whitelaw, T. M., Day, P. M., Kiriatzi, E., Kilikoglou, V. & Wilson, D. E. (1997). Ceramic traditions at EMIIB Myrtos, Fournou Korifi. In (R. Laffineur & P. P. Betancourt, Eds) TEXNH: Craftsmen, Craftswomen, and Craftsmanship in the Aegean Bronze Age, pp. 265-274. Liege: Aegaeum 16.

Wright, R. (1993). Technological styles: transforming a natural material into a cultural object. In (S. Lubar & W. D. Kingery, Eds), History from Things: Essays in Material Culture, pp. 242-269. Washington D.C.: Smithsonian Institution Press.