CORE AND PERIPHERY

A core is a concentration of massed demographic and economic power. A periphery is a zone whose lesser demographic and economic power is in some measure due to its relationship to a core. Cores usually dominate, in normal times. They exercise political and military power over peripheries and enjoy a net surplus in the flow of materials or labor vital to core interests. Mesoamerica had multiple cores, surrounded by large peripheries. Cores have received more archaeological attention; thus, the evolution of this civilization tends to be written as the evolution of its core zones (Blanton et al. 1993:222–225). But investigations in peripheral areas have shown that the aggregate demographic and political power outside traditional core zones has to be given more weight in our explanations. Study of areas outside the known cores can also demonstrate that what we previously thought of as peripheries did not always function in that way. I will not make this latter claim for Peñoles. Its role vis-à-vis the Valley of Oaxaca changed over time, as Laura Finsten (1996) has written, but it has never been central to surrounding regions as a core would be.

This is a study of one aspect of periphery–core relations, economic behavior, and, in only a minor part of material life, pottery. The article suggests that the pre-Columbian pottery in Peñoles was just that expected at the ends of solar or dendritic market systems. In other words, I show here that the pottery from Peñoles is distinctive in certain respects and that it looks the way it does because of this area's longstanding “backcountry” economic position.

I take it as given that, after the rise of urbanism, the basic Mesoamerican economy was one of market exchange. This is not to say that other forms of exchange, including tribute, did not exist, for the institution of tribute is well known. But market systems were omnipresent at the time of the Spanish Conquest and essential for daily existence. In Mesoamerica, the bulk of exchange for most goods took place through markets where buyers and sellers met and where prices were determined largely by supply and demand.

In market exchange, economic behavior is quite different in cores and peripheries. Table 1 is my compilation of core and periphery economic characteristics. These characteristics are drawn from empirical and theoretical studies of recent peasant markets and their production and exchange systems (Appleby 1985; Beals 1973, 1975; Cook and Diskin 1976; Johnson 1970; Smith 1976; Underwood 1970). Carol Smith (1976) provided the classic summary of the cross-cultural phenomenon of solar and dendritic market systems. The main feature of solar market systems is their dominance by a primary market center, which is surrounded by secondary markets. In the dendritic variant, goods and services pass along linear arrangements of markets that decrease in size toward the periphery. People at the farthest reaches of the market system have few buying and selling choices and tend to be dependent on only one nearby market. Peripheries are market-sparse in terms of market choices; cores are market-rich.

Generalized characteristics of craft economies in cores and peripheries

Peripheries tend to supply raw materials to the core, and cores in turn supply transformed goods. The economic and political power of cores, expressed in part through market dominance, tends to drive down the price of peripheral products in relation to core goods. In the periphery, imports from the core are relatively expensive, there is not much choice or competition, and transportation costs are higher.

The expected economic behavior in Table 1 applies especially to peripheries and cores of solar or dendritic market systems. These are generalized characteristics, verbal representations of the ends of a continuum, and they need to be specified for particular cases. The next section attempts to specify what would be expected for the case of pottery under the conditions of Mesoamerican market systems.

POTTERY IN CORE–PERIPHERY SYSTEMS

Several factors may be particular to how goods such as ceramics are made and distributed in core–periphery situations. Pottery is a fairly flexible good. Pots have relatively low cost. They have elastic renewal rates, being replaced infrequently to very frequently. Unit profits tend to be low (cf. Underwood 1970). Potential specialists may thus select other occupations before becoming potters. In these ways, pottery is similar to low-end textiles, but ceramics usually do not have the high-end value that cloth can have. Pots are purchased largely on the basis of visual appearance. Pots might be tapped or sounded by prospective consumers, but pottery is not tried out or tried on, like shoes. Traits of visible appearance are the most important criteria for discriminating quality, cost, function, and variety. Pottery consumers can forgo costly styles of pots in favor of simpler functionality, as consumers do with clothing. Producers and consumers of pottery thus can alter their behavior to adjust to different costs in core or periphery situations.

Basic raw materials are typically available to potters in most localities in Mesoamerica. However, preferred clays, tempers, and pigments are not uniformly distributed. Peñoles, the area of this case study, is an upland, mountainous zone of granitic and old metamorphic rock, steep slopes, very narrow canyons, and fast-moving creeks. Its clays are saprolites occurring in relatively small deposits, and the clay soils tend to be coarse rather than fine. The neighboring Valleys of Oaxaca and Nochixtlan have more extensive deposits of finer alluvial clays, as well as the upland primary clays. Upland potters in highland Oaxaca have access to local clays, but if they wish to make a very fine clay body they may have to work harder at procurement or paste preparation. Thus, there is some basis in geology as well as economics for the widespread coarse brownwares in upland situations.

Given this distribution of raw materials, the nature of pottery as an economic good, and the general economic characteristics of cores and peripheries, we can make some predictions about pottery in the core versus on the periphery of Mesoamerican market systems. First, there should be less production, less demand, and less consumption on the periphery of market systems. There should be fewer pots, fewer potters, and fewer pots and potters per capita. Production should be less intensive than in the core. Producers on the periphery are at a competitive disadvantage compared with their peers in the core in terms of access to multiple markets and numbers of consumers. Peripheral potters can take less advantage of economies of scale, and this should be reflected in their production process and their output.

Peripheries will have a narrower range of pottery goods, either as imports or as locally made products. Demand is not sufficient for peripheral potters to make a wide range of pots. The price of imported pots in the periphery is relatively high because of the greater transportation costs, the lack of economies of scale, and the monopoly dominance enjoyed by markets at the ends of dendritic chains. Transportation costs weigh against the long-distance movement of larger vessels. The range of imported offerings may be restricted to smaller pots. Local pottery is expected to be of lower quality and plainer design. Potters will invest less time and effort per pot, from procuring materials to forming, decorating, and firing.

Two additional dimensions of potting have received attention in the archaeological literature and are expected to vary with the core–periphery economic position of potters. These are variation in the degree of standardization and the degree of skill evident in the potter's work. As Cathy Lynne Costin (1991) has pointed out, the degree of standardization is influenced by quite a few factors, including economies of scale, intensity of production, the relative number of production units, competition, and efficiency in costing and marketing from both sellers and consumers. “Within the technological limits of functional necessity and the constraints created by social expectations, vessel standardization generally reflects the relative number of potters working within a particular production system” (Costin 2001:213). Potters will share more information if they work near one another. Larger-scale or more intensive potting enterprises encourage more routine in the production process and may discourage idiosyncratic behavior.

Closely related to standardization, but not quite the same thing, is skill. “[A]rtisans who spend more time at their work, and those who work without significant hiatus, are expected to be more competent at their tasks because of increased repetition and experience” (Costin 2001:217). Costin measured skill by the degree of consistency in vessel-wall thickness and the presence or absence of “firing cores.” It should be clear that skill in this sense is only that manifest in the product at hand, not a general valuation of a potter's or a production unit's whole work or real capacity, for there is no doubt that a skilled potter could turn out both shoddy and finely crafted goods, depending on the target market.

Finally, I expect that peripheries tend to copy core styles, especially in gross visual appearance. Core producers and markets are likely to set trends. As with low-end textiles, it is fairly easy for peripheral pottery specialists to copy the general appearance of pottery made by others, even though the materials and technology employed may not be identical. For example, peripheral Pueblo II potters in northeastern Arizona “were able to manipulate local materials to produce reasonable facsimiles of almost any Anasazi or Mogollon pottery type from the surrounding regions” (Hays-Gilpin and van Hartesveldt 1998:194). In low-demand situations it is not worth a potter's time to copy high-end products or to bother with exact replicas. Consumers in the periphery can make infrequent trips to a distant central market to acquire the occasional high-end piece, and these infrequent trips will undercut peripheral potters' ability to produce high-end products for local consumption.

Especially in archaeological cases there are caveats about this principle of peripheries copying core styles. Frequently it is not known where a style originated and what is the original and what is the copy. It is often assumed that the area of greatest abundance and elaboration is the originator. Cores receive the earliest archaeological attention and are the first described in the literature, but this of course does not necessarily mean that core ceramics are not copies.

Likewise, peripheries may reproduce and elaborate distinctive styles that are valued and preserved in core and periphery alike—for example, the hundreds of locally specific cheeses and wines in modern France (Planhol and Claval 1994). Peripheries may develop distinctive styles marking micro-ethnic groups, as with locally produced textiles in Middle America and the Balkans (Cordry and Cordry 1968; Wobst 1977). In solar and dendritic systems, especially where there are primate centers dominating extensive peripheries, exchange at the periphery tends to run vertically back and forth toward the center rather than horizontally among neighboring groups on the periphery. Relations among peripheral groups are often competitive, a situation that may not be discouraged by the primate center. The clothing distinctiveness of ethnic groups in Mexico and the Balkans is consistent with their intergroup competition and peripheral political and economic position. Dress, however, is a more mobile display medium than pottery. Textiles can translate family labor into ethnic messages better than ceramics. Peripheries may develop distinctive styles of their own in some media, but pottery, I think, is a less likely channel for sending ethnic messages in a market system. (Highly crafted polychrome pictorial pieces, distinctive to place and made by specialists attached to noble houses, are a different phenomenon.)

PEÑOLES: A PERIPHERY

“Peñoles” (Figure 1) is our term (based on a sixteenth-century name for a set of villages in the region's center) for an approximately 1,000 km² sierra region adjoining the western side of the Valley of Oaxaca. It is completely mountainous and is separated from the valley by an agriculturally forbidding and never much occupied mountain divide. The western side of this divide, the heart of Peñoles, is a land of narrow ridges and V-shape canyons. The northwest gradually opens out onto the foothills of the Nochixtlan Valley. To the west, the terrain becomes even more rugged and forbidding. Maize yields in Peñoles tend to be low compared with those of the lower-lying valleys (Garvin 1994). The traditional ways to make cash income have been emigration and trade in mountain products (flowers, tree fruits, agave products, firewood, charcoal, and lumber) (Garvin 1994). Today, travel by foot from the populated areas on the western side of the divide to the markets in the valley takes ten or twelve hours.

Central Oaxaca from the Nochixtlan Valley to the eastern Valley of Oaxaca, showing places discussed in the text.

In 1990, our project, directed by Laura Finsten, carried out a systematic survey of the Peñoles region, finding some 545 sites, many of which are quite extensive and have excellent preservation of architectural remains. The information from this project suggests that Peñoles has always been peripheral to the Valley of Oaxaca and, in part, to the Nochixtlan Valley. It was slow to be occupied by sedentary villagers. The first good evidence of sedentary villages comes from Late Formative times. Terminal Formative and Early Classic sites are infrequent and small (Finsten 1996). Richard Garvin (1994:Chapter 6) has proposed that maize farming at the high elevations of Peñoles was too unproductive for much settlement until higher-yielding maize varieties were developed in Classic and Postclassic times.

Population and numbers of settlements increased in the Late Classic and peaked in the Postclassic period. Population densities were lower than those of the core regions. Nevertheless, the Postclassic kingdom centered on Itzcuintepec Peñoles had a population of 5,000 to 15,000. In colonial and recent times, Peñoles has been at the political fringes of the Etla, Zaachila, and Nochixtlan districts. The market orientation is toward the valley plazas of Oaxaca City or Zaachila, except for the northwestern edge, which has more to do with Nochixtlan. Finsten has shown that in many ways, Peñoles had its own internal development, not like that of either the Nochixtlan Valley or the Valley of Oaxaca. Yet in the main, the region was dominated from the outside, economically and politically, although the kind of domination changed over time (Finsten 1996).

THE POTTERY SAMPLE

The Peñoles project collected ceramics by the usual procedures for regional surveys in highland Mesoamerica. The main purpose for collecting was to date occupations, but we also wanted to make preliminary interpretations about economy and ceramic spheres. Our survey provided information about the dating of most site occupations and about the distribution of common ceramic types and wares. More in-depth studies will require larger, systematic collections. In addition to the usual tension that we feel in the field between the goals of regional coverage versus studying sites in more depth, this survey gave us three new problems: (1) Potsherds were not as abundant as in the valleys; (2) leaf litter and erosion meant that visible sherds were spottily distributed; and (3) a common disposal pattern was to throw refuse downslope. Therefore, our sampling strategy was sometimes a matter of ferreting out scarce artifacts. Where artifacts were relatively abundant, we collected temporally diagnostic sherds that were roughly representative of types and wares. We also made collections where the material was a good representation of special activities. Where sherds were scarce, we picked up everything we could find. We searched for the refuse toss zones. The actual collections are only a small fraction of the artifacts inspected, but it is the detailed lab study of collections that forms the basis of this paper.

We took 334 surface collections, which provide good overall coverage of the region. But a significant limitation of the collections is their small size. The total is only 4,475 sherds, a mean of 13 sherds per collection. The small numbers and the lack of a systematic collecting design preclude quantitative treatment beyond presence and absence. However, the collection procedures are probably biased against the hypothesis of peripheral economic behavior. Our collections favor decorated, temporally diagnostic sherds, not the plain wares and odd examples expected in the peripheral-economy hypothesis.

The present study does not explore change in core–periphery economic relationships over time. Most Peñoles settlement dates to the Classic period, and especially to the Postclassic period. Distinguishing differences in ceramic economy within or between these broad periods is not possible at present. As discussed later, one of the features of the Peñoles assemblages is the relatively high frequency of temporally undiagnostic sherds.

RESUMEN

En las regiones apartadas es común que se presente una alferería reducida en formas, sencilla en decoración, más tosca que fina, y aparentamente baja en calidad artesanal. Este artículo ofrece una explicación basada no en el atraso cultural, sino en la acción racional de productores y consumidores en el ambiente económico de los márgenes o términos de sistemas de mercado (tianguis).

Estudios de los mercados campesinos de Mesomérica y otras regiones del mundo han demostrado definidas regularidades en la acción de pequeños productores y consumidores, según su posición respecto a las plazas mayores del sistema, o en las periferias, donde las plazas son pequeñas y poco frecuentes (Smith 1976). Los consumidores localizados en el centro de estos sistemas regionales, tienen acceso a múltiples plazas y aprovechan las ventajas que trae la competencia entre numerosos productores y vendedores. En cambio, las personas en los márgenes no tienen acceso a la misma variedad de mercados y están sujetas a los precios dictados por monopolios; también los pequeños productores en las zonas apartadas no pueden sostener una especialización intensiva a causa de la poca demanda y la competencia procedente del núcleo del sistema regional de mercado.

Las condiciones económicas en las periferias de los sistemas de mercado implican definidas consecuencias para la alferería. Se espera una cantidad menor de producción y consumo de vasijas de cerámica, un rango de productos restringuido, la sustitución de productos (por ejemplo, vasijas de madera o de calabaza en vez de vasijas de barro), una escala e intensividad de producción reducida, una menor capacidad en cuanto a la forma, cocción, y acabado de las vasijas, un menor inversión de tiempo, labor, y materiales, diseños sencillos, y la imitación de los estilos de los centros urbanos. Los estudios arqueológicos previos se han enfocado en la identificación arqueológica de la especialización, sobre todo en los centros mayores. El presente estudio incorpora las mismas medidas de la cerámica (e.g., Costin 2001) pero se dirige al lado opuesto del contínuo de la especialización.

Este es un estudio del caso de la cerámica precolumbina de los Peñoles, una región montañosa y marginal entre los Valles Centrales de Oaxaca y el Valle de Nochixtlan. Tanto en tiempos modernos como en la época pre-Hispanica, los Peñoles ha sido una zona de abastecimiento de productos serranos—frutas de árboles, flores, madera, leña, carbon y minerales. El potencial para la intensificación agrícola es más limitada que en los valles mayores pero se ha practicado un sistema de riego a zanja utilizando los numerosos ojos de agua.

El recorrido arqueológico sistemático de los Peñoles, dirigido por Laura Finsten en 1990, registró aproximadamente 545 sitios arqueológicos, entre los cuales predominan las ocupaciones del período cláscio y, en mayor cantidad y extensión, las ocupaciones del período postclásico. Se hicieron 334 colecciones de superficie; sin embargo la muestra es pequeña, de sólo 4,475 tiestos en total. El tamaño reducido de la muestra y la falta de control estadístico en el método de muestreo restringe el estátus lógico de las interpretaciones, que se ofrecen como hipóteses para esta investigación específica.

La cerámica de los Peñoles es consistente con las expectivas del modelo económico. Parece que la cantidad de las vasijas es menor que la de los valles adyacentes. El número de tipos de cerámica es menor. A simple vista las vasijas parecen semejantes a los tipos de las zonas nucleares, aunque los barros y los detalles tecnológicos son distintos y variables. Hay gran variabilidad en los barros y en la cocción de vasijas. En diseño y acabado, la cerámica de Peñoles demuestra la sencilléz y un menor grado de inversión de tiempo y materiales como se espera. La decoración es mínima. La falta de indicios de la producción de cerámica, en comparación con los Valles Centrales donde se han encontrado muchos sitios de alfarería, se interpreta como un régimen de producción no intensivo y de baja escala. Las colecciones cerámicas de Peñoles sugieren un grado menor en cuanto a la estandardización y la destreza artesanal que los valles circundantes.

Esto no quiere decir que no se encuentra vasijas semejantes de baja calidad en los centros mayores. Es cierto que productos baratos se vendían en todos los mercados. Pero la tésis de este estudio es que por lo general la cerámica de las zonas periféricas de los sistemas de mercado tiende a ser más sencilla debido a la especialización no intensiva.