Inferring provenance, manufacturing technique, and firing temperatures of the Monagrillo ware (3520–1300 cal BC), Panama’s first pottery

Monagrillo (3520–1300 cal BC) is Panama’s oldest pottery. Archaeologists assumed it was a low-fired expedient ware made with any available clay. We studied 1) clay sources (thin sections; DTA; shrinkage, porosity, and plasticity tests), 2) manufacturing techniques (xeroradiography; thin sections; visual inspection), and 3) firing temperature (SEM-EDS; porosity tests). We identified two clay types, one restricted to the Pacific coast, one widely distributed. Vessels were made by layering slabs and occasionally lumps. Rimand lip-finishing is variable. Firing temperature (>800–950°C) is relatively high for open firing. Porosity is quite low. These aspects indicate that Monagrillo is not an experimental or expedient ware. 1. Monagrillo, Panama's first pottery In the Americas, the earliest known pottery becomes progressively younger along the Central American isthmus and into Mexico. Therefore, some archaeologists have reasonably assumed that diffusion was the primary mechanism for the dispersal of ceramics northwards from South America (Ford 1969; Fonseca Zamora 1997;Meggers 1997). Others have argued that independent invention is as plausible an explanation, as there are geographical gaps in the distribution of the earliest known wares in Central and Mesoamerica which differ stylistically and technologically from each other and from contemporary South American wares (Cooke 1995; 2005; Hoopes 1995). Clark and Gosser (1995) propose that invention was ‘dependent’ – i.e., that people adopted the technology from other groups but manipulated style and function for their own purposes. The Monagrillo ware (Willey and McGimsey 1954) is the earliest known pottery in Panama. It was produced between 2600 and 1200 BC (3520–1300 cal BC) over an area of 5600 km in the central part of the country, between Parita Bay on the Pacific coast and the Coclé del Norte drainage in the Caribbean foothills (Fig. 1; Cooke 1995, Fig. 14.1; Griggs 2005). It has not been reported elsewhere in Panama. It appears stratified directly above Preceramic layers at two rock-shelters (Cueva de los Ladrones (Cl1) and the Aguadulce Shelter (Ag13)). This fact supports the hypothesis that it was the first pottery made in central Panama. Current evaluations suggest that the Monagrillo pottery is typologically coherent, showing little evidence for diachronic change during its long period of manufacture. Vessel shapes are limited to bowls and restricted collarless vessels (Cooke 1995, Fig. 14.2). No handles, lugs, or feet have been found. Decoration consists of redpainted bands and daubs, and rare incised decoration using lines and volutes. Towards the end of the tradition, short collars are added to jars, and the variety of incised, impressed and modelled motifs increases. These developments, however, are poorly dated (Cooke 1995, Fig. 14.3; Cooke and Sánchez-Herrera 2004). Previous assessments of Monagrillo pottery have relied on comparing data obtained from visual (usually noninstrumental) inspection of manufacturing processes, intuitive typological studies and regional site surveys supported by radiocarbon chronologies. Our study of the Monagrillo ware uses rigorous analytical methods derived from materials science and geology, focusing on the production process, with particular attention to raw material selection and firing temperatures. 2. Subsistence economy, settlement patterns and social interactions of the Monagrillo potters Data about the subsistence economy, settlement patterns, and social interactionsof theMonagrillopottershavebeenprovided by (1) archaeological surveys (Cooke and Ranere 1992), Cite this article as: Iizuka F, Cooke R, Frame L , Vandiver P (2014). Inferring provenance, manufacturing technique, and firing temperatures of the Monagrillo ware (3520–1300 cal BC), Panama’s first pottery. In M Martinón-Torres (Ed.), Craft and science: International perspectives on archaeological ceramics. Doha, Qatar: Bloomsbury Qatar Foundation http://dx.doi.org/10.5339/uclq.2014.cas.ch3 UCL Qatar Series in Archaeology and Cultural Heritage Figure 1. Geological map of Panama with Monagrillo sites The map, modified from del Giudice and Recchi (1969), was produced with the assistance of Natalia Hoyos. Iizuka et al. Craft and science: International perspectives on archaeological ceramics 20 (2) excavations at stratified sites (Willey and McGimsey 1954; Bird and Cooke 1978; Ranere and Hansell 1978; Cooke 1984), (3) archaeobotany (Dickau 2005, 2010; Piperno and Pearsall 1998; Perry et al. 2007), (4) zooarchaeology (Hansell 1979; Carvajal Contreras and Hansell 2008; Cooke and Jiménez 2008; Cooke et al. 2008), (5) geomorphology (Clary et al. 1984), and (6) vegetation history derived from a lake sediment record at Laguna La Yeguada, located within the Monagrillo pottery production zone (Piperno and Pearsall 1998; Piperno 2006). Like their Preceramic predecessors in central Panama, Monagrillo potters were farmers who by 3500 cal BC had cleared extensive tracts of forest on the Pacific watershed for planting early forms of New World staple crops, e.g., maize (Zea mays), manioc (Manihot esculenta), sweet potatoes (Ipomoea batatas), squash (Cucurbita spp.) and peppers (Capsicum spp.). They were not hunter-gatherers and ‘incipient’ agriculturalists bound to the marine littoral – as formerly believed (Willey 1971). However, the degree and duration of sedentism at individual sites, and the seasonality of their occupations, are difficult to gauge from the existing record. On the Pacific watershed of Panama under present-day climatic conditions, rain-fed farming is only possible between May and December. Domestic refuse at the largest Monagrillo site (Zapotal [Pr32]) covers ca. 1 ha. This site, located near prime agricultural soils, produced a large number of plant-processing tools. The remains of a small dwelling were also found here. It is likely that it was a village (Cooke and Ranere 1992; Perry et al. 2007). Ag13 and Cl1 – two rock-shelters located 18 and 25 km from the present-day shore of Parita Bay – were probably occupied for long periods at a time (i.e., during the farming season). Game was regularly brought back to each site. Shellfish, crabs, and fish from the marine littoral, also regularly consumed, would have been obtained by making trips to the coast or through exchange. Palm nuts were processed intensively at the Ag13. The primary species – the Neotropical oil palm, Elaeis oleifera – requires swampy habitats, and is most productive in rainy season months. The processing of palm fruits (Atallea alenii) and balsam seeds (Humirastrum diguense) is in evidence at the only Monagrillo site known on the Caribbean watershed: Calaveras (Lp8). This shelter's inhabitants consumed maize and may well have lived year round in this perennially humid habitat. Hansell's (1979) preliminary growth-ring study of marine shells found in middens at the Monagrillo (He5) type site suggests that this settlement, dated between 2400 and 1300 BC (2800–1400 cal BC) and located then on the active marine shore of Parita Bay, was occupied mostly in the dry season (non-farming) months by people who spent the rest of the year elsewhere. Fishing in in-shore marine habitats was an important activity here and at Pr32. The high number of small shoaling fish in the middens, e.g, threadherrings (Opithonema spp.), small jacks (Carangidae spp.), and sea catfish (Ariidae spp.), suggests the use of gill-nets and/or inter-tidal weirs. Inland-coast seasonal transhumance is still practised in this region in the dry season. He5 and Pr32 may have provided the inland shelters with dried and salted fish – another practice that persists in the region (Zohar and Cooke 1997). Smaller rock-shelters in the foothills and mountains where Monagrillo pottery is scarce are likely to have been occupied intermittently or irregularly as hunting-and-gathering camps or as rest stations on paths. These data allow us to infer that Monagrillo communities interacted regularly, exchanging or transporting foodstuffs and other produce from many different habitats in an environmentally heterogeneous interaction zone. It can be assumed that pottery was one of these products. Theprimarygoal of this study is touse a detailed examination of the production processes and circulation of theMonagrillo pottery in order to improve knowledge about residential and interactive behaviour and pottery use before the appearance of those better -known regional societies that archaeologists consider to be chiefdomswithwell-defined social classes and extensive trade connections (e.g., Linares, 1977; Helms, 1979; Drennan 1996; Isaza-Aizuprúa, 2007; Haller, 2008). 3. Materials and methods