Background

I was drawn to work in West Mexico by something long recognized and little understood: the formation of prehispanic connections between the urbanized populations of Mesoamerica and the emerging complex societies in regions to the north, including West Mexico, Northwest Mexico and the US Southwest. The most profound and earliest of these connections was the introduction of maize agriculture which reached the US Southwest ca. 2000 BCE. Over many centuries, as maize economies became entrenched, other material markers of such connections appeared, for example: 1) Turquoise, chemically matched to a source in Cerrillos, New Mexico (US), found in an archaeological site in the Yucatan Peninsula (Mexico); 2) theobromine-caffeine residue of the cacao plant, which is native to the tropical lowlands of Mesoamerica, found in ceramic vessels in Chaco Canyon, New Mexico (US); 3) scarlet macaws (Psittacidae Ara macao), native to essentially the same tropical regions, found sacrificed and ritually buried, in Paquimé, Chihuahua (extreme northern Mexico, and sites in the US Southwest; 4) copper and bronze bells, probably made in West Mexico, found throughout the US Southwest. These and other markers began to appear in notable frequencies ca. 900 CE.

The emergence of new exchange connections was entwined with the rise of large-scale regional centers in West Mexico, and in Northwest Mexico and the US Southwest (the “NW/SW”). Some archaeologists consider these centers to be cities, while others do not. In any case, these large settlements began to appear ca. 600-900 CE, a millennium later than the earliest cities emerged in Mesoamerica proper, representing an expansion of Mesoamerican civilization to the north. The generally increasing appearance of distantly obtained objects in these sites is a kind of globalization. In today’s world, globalization is defined as “the increasing extension, interpenetration and interdependence of production systems, corporations, markets. . . and networks of flows across national borders” (Martin et al. 2018). The NW/SW is thus a place to investigate the fundamental questions posed by urbanization and globalization: why did centers form, why did they make such connections, how did life change in the affected places, and how did institutions, especially those of exchange, evolve over time?

My students, colleagues, and I have addressed these questions at two geographic scales, the local and the macroregional. We have achieved perspective on the local scale by conducting field work at a single regional center, La Quemada, to test predictions derived from competing ideas about the processes that drove the expansion. The competition pits environmental drivers against political-economic ones. The field work was authorized by the Mexican government under the title of the La Quemada-Malpaso Valley Archaeological Project.

Working at one site was productive, but it could not answer questions about the bigger picture of exchange. So, we began to build a macroregional perspective by combining data on long-distance exchange from many different regions. This work is primarily archival and bibliographic, mining data site-by-site for occurrences of long-distance exchange items. We tabulate the items in overarching typologies of form and style, chronology, and contexts of deposition. The data constitute a grid of time, space, and artifact characteristics against which to evaluate hypotheses about the form of network(s) and how it or they changed over time. This international collaboration is known to its participants as the Connections Project, short for Connections and Impacts of Northwest Mexican Cultures.

Methods and Results

Studies at La Quemada have used a range of methods from archaeology and related sciences to explore ancient socioenvironmental change at the local level.  La Quemada is a hilltop fortress and a “primate center,” many times larger than the other settlements in its orbit. Charles Trombold has shown that 175 km of prehispanic roads connect La Quemada to over 200 tightly clustered, subordinate villages in a 120 km2 area. The areas beyond this cluster were lightly occupied, suggesting that the road-related settlements constituted an ancient community.

Our team first mapped the ceremonial center, adding hundreds of features to an existing map that showed the major topographic contours of the hill. The mapping revealed some 60 terraces. Those in the monumental central precinct were masonry structures, including two colonnaded halls, 13 pyramids or altars, massive staircases and causeways, numerous sunken patio complexes, and a 75 m ball court. Other terraces are located on the flanks of the hill outside that protected zone. The configuration of the settlement, plus the clustering of other smaller, settlements around it, makes it reasonable to infer that defense was a consideration in creating built space.

The map provided a sampling framework from which we selected one major terrace and several non-contiguous areas for excavation. Excavation showed that Terrace 18, at least, was occupied and used for both residential and ceremonial purposes. The 40 x 80 m terrace included one temple, 41 other rooms, seven sunken patios, parts of three platforms, a ball court, a causeway, and a staircase. These findings might not seem revolutionary, but this excavation was the first to be conducted outside the ceremonial precinct. Archaeologists did not even know whether the terraces were agricultural or residential.

The excavations yielded new information about the spatial configurations of built space. Several groups of structures surrounded sunken patios; those groups together surrounded a larger patio, which contains a small ball court with a 10 m playing surface. The arrangement is one that we refer to in the United States as a Russian-doll arrangement (a better term is fractal).

We also excavated 10 trash middens associated with different terraces, and we dug offsite trenches in the alluvial floodplain to collect evidence of environmental change.

To see whether climate change played a role in the founding and abandonment of La Quemada, and therefore possibly in the Mesoamerican expansion, we used earth-science methods such as palynology, geomorphology, sedimentology, and paleomagnetism to analyze the samples from the offsite trenches. We looked for changes in environmental indicators belonging to the periods before, during, and after the major occupation of La Quemada. For example, paleomagnetism might reveal a increase in the iron content of sediments, indicating increased erosion. That pattern should be corroborated by botanical evidence of changing land cover. The sediments themselves should change in particle size and angularity. The main finding of these studies was that no evidence of environmental change could be detected in association with the main occupation of La Quemada. If the occupation were associated with improved rainfall, we should see upward inflections in the red lines where they cross the widest gray line (representing greatest human population). The only red lines that do inflect upward are arboreal pollen, which may have increased because of a decrease in grasses, and human population, which represents the main occupation itself. Major changes were seen instead in the Colonial period, when mining and agricultural activities led to deforestation in the region.

Investigating human-environmental relationships led to further questions, which we conducted several systematic studies to answer. One that particularly interested me was the development of a mathematical model to help understand the role of agave (century plant) as a backup resource when maize failed due to drought. Because agave can store water to feed itself for years, it might have been a key to survival in this relatively arid region. A simulation shows that agave cultivation could have eliminated the need for an astonishing 93% of famine-related migrations.

Researchers had suggested that long-distance exchange was a stimulus to the formation of La Quemada, specifically, Phil Weigand argued that the site was a station on a “turquoise trail” connecting Tula, the Toltec capital in central Mexico to Chaco Canyon, New Mexico, where the great quantities of turquoise have been found.  This suggestion brought me to La Quemada because, trained as a US Southwestern archaeologist, I was intrigued by the chance to find tangible evidence of such a connection. Just how common was turquoise as a raw material and as finished objects in the local setting?

We were unable to confirm the intermediary relationship with evidence from excavation. For one thing, La Quemada turns out to be earlier than Tula and Chaco Canyon – they both are occupied from the late 800s till mid 1100s CE, whereas La Quemada is earlier, 600-800/850 CE. La Quemada also turns out to have very little turquoise. We found fewer than 50 blue-green stones, only a third of which were turquoise according to geochemist Alyson Thibodeau. If worked into beads, this quantity of turquoise would not make a necklace. In fact, only a few pieces were worked, notably including a tessera characteristic of Mesoamerican mosaics. Thus, our excavation data did not confirm the idea of La Quemada as an outpost on a turquoise trail from the US Southwest to Central Mexico.

The small number of turquoise pieces contrasts with the high frequencies from Alta Vista; about 17,000 were documented by J. Charles Kelley and Phil Weigand. The La Quemada data thus show that even contemporary sites that were quite closely related culturally, did not necessarily have the same degree of participation in interregional exchange.

Items common at other sites were rare at La Quemada. Of over 400 pieces of obsidian submitted for neutron activation analysis, none were from the US Southwest. John Millhauser’s thesis notes a single prismatic blade of Central Mexican obsidian. Four West Mexican sources are represented in the analysis, revealing La Quemada’s limited participation in that network. Shell was also infrequent in the excavation samples, and notably, shell bracelets and conch-shell trumpets were not recovered, although they are found in other regional centers. No copper or bronze items have been found at La Quemada. Likewise, petrographic and chemical analyses by Nicola Strazicich and Christian Wells did not identify any non-local ceramic wares. Andrea Torvinen’s dissertation, now being published as articles, deepens our understanding La Quemada ceramics. As of now we may say that in a range of materials, La Quemada appears to have been relatively inactive in the exchange networks that existed among other sites.

Before leaving La Quemada, I must describe one of its hallmarks, the massive deposits of human bones, which were mostly not burials, but displays, bundles, caches or discards. My colleagues Peter Jiménez and others who have worked in the monumental core of the site have found over 600 skeletal individuals piled in prominent places such as on the floor of the Hall of Columns. Our excavations on Terrace 18 revealed about 50 individuals, mainly in the form of skulls and long bones. Suspended human skulls and long bones formed a passageway outside the entrance to the temple; three groups of skulls and long bones were hung in bundles inside the temple; other sets were placed on stairways. Small bones such as vertebrae and phalanges were deposited in the trash middens. Only two individuals were placed in chambers beneath the floors. Bioarchaeologists Debra Martin and Ventura Perez identified two distinct patterns of cut marks on the human bones, which we believe correspond to enemies on the one hand and revered ancestors on the other. This interpretation implies a level of social violence, or else ritualized processing of human bone in ways that look warlike.

Once again, why was La Quemada’s participation in exchange networks limited, while other sites had phenomenal quantities of distantly obtained items; for example, the tens of thousands of turquoise pieces in Aztec and Mixtec mosaic plaques, or the hundreds of macaws and copper bells at Paquimé in the Chihuahuan desert? To find out, we have initiated the macroregional exchange study.

The Connections Project has been attacking the question of long-distance exchange by gathering and analyzing data about exotic objects and symbols from hundreds of prehispanic settlements in Mesoamerica and the US Southwest. Our questions concern the evolution of macroregional exchange relations. It is well known that by Aztec times (1345-1519 CE), far-reaching systems of taxation and commerce were in place, conforming to a model of expansionary peripheral extraction. Is Aztec mercantilism an appropriate model for distant acquisition in earlier periods, or are other forms and mechanisms more apt to explain the early development of the macroregional exchange system? Do changes in northwest Mexican participation reveal a time when commercialized market exchange originated in the world-system of Mesoamerica?

Defining the spatial and temporal distribution of evidence for exchange should help archaeologists answer such general questions. For example, the Aztec model predicts a single set of relations coordinating flows of goods and information into a single marketplace. Alternatively, there could have been different sets of relations among centers of production and independent acquisition and production of different goods. Also, how homogeneous or uneven was participation in the globalizing economy? Were all northwest Mexican regional centers intermediaries in the distant interactions? Some of the first northwest Mexican regional centers that archaeologists chose to excavate, beginning in the 1950s, abounded in turquoise, shell, copper, and other exchange items. Yet, no one has looked at the exchange data in the aggregate to assess the extent of shared participation by West and Northwest Mexican centers.

This project, of which José Luis Punzo and I are co-directors, is still in its early stages but already is yielding exciting results. We no longer see exchange as simply between Mesoamerica and the US Southwest, but among specific regions and regional centers, and we can see it with a time dimension. Examples of our initial major findings are: 1) the networks did not involve all sites in the region; during any given time period, only certain sites were involved; 2) not all regional centers contain all kinds of objects; 3) in many instances there were very large spatial gaps in the occurrence of specific interaction markers; 4) not all West Mexican or northwest Mexican sites were intermediaries.

Here is a partial tabulation of a set of markers of interregional interaction that we have been tabulating. All sites listed are major ceremonial centers in NW Mexico. As noted, the expectation that groups in NW Mexico served as intermediaries in a long-distance exchange system is not borne out in most cases. Indeed, what seems striking is that many centers appear not to be participating in long-distance exchange, or if they are, only in one or two materials.

Discussion and Conclusion

Combining intensive site study with extensive interregional data collection permits a perspective on the problem of ancient globalization. La Quemada’s limited participation in interregional exchange suggests variable West Mexican and northwest Mexican participation, and additional data from other sites confirms this. Indeed, participation at the level manifested in sites such as Paquimé, Alta Vista, and Amapa is unusual. It may be that at any one time, only a select few sites were highly active in acquiring distant objects, and the rest were only occasional recipients of these valued goods.

Incorporating this new perspective, the answer to why La Quemada’s participation was limited is relatively simple. Most sites in West Mexico and northwest Mexico were not highly involved in long-distance exchange, and moreover, La Quemada was occupied and abandoned before the level of macroregional exchange reached its peak. The data from most early sites like La Quemada (600-800/900 CE) are consistent with a pattern of voyaging to obtain special goods rather than market-directed, commercialized behavior. Such sites may have one or a few items that mark distant acquisition. In later centuries, however, the overall level of exchange increased. For example, Paquimé with its many macaws, shell items, copper and bronze artifacts, etc., dates to 1200-1400 CE, the Aztec and Mixtec turquoise mosaics are from 1350-1519 CE.

Although we have not touched upon it, platform mounds, plazas, ball courts mark other institutional change in the times and regions of interest here.

In sum, our research strategy allows us to characterize the distant in the local. Such presence existed in ancient Mesoamerica proper from 1200 BCE, extending to West Mexico by the early centuries BCE, but not until after 900 CE into Northwest Mexico, and the US Southwest as new sets of regional centers arise. In its earliest phases, this extension is somewhat tentative, and it is unlikely to be explained by an Aztec-like model of mercantilism. Before about 1200 CE, the instances of distantly obtained objects are too few and scattered to support a contention of commercialization. Even after 1200 CE, it is unclear whether foreigners were directly intervening in the NW/SW to acquire resources, or the NW/SW peoples were traveling great distances to acquire what from their perspective must have been rare, powerful objects. A great many questions remain to be answered about this evolving network. For example, it is not clear whether any central Mexican powers ever intervened in this region and brought it into alignment with their systems of dominance. What is clear is that at all time levels, the degree of globalization was uneven, always changing and mediated by local custom and practice.

I conclude with a sketch representing the occurrence of selected markers of interaction in comparison to their source areas. “Separate distributions” means different distributions in time and space for several items, suggesting non-pervasiveness and the existence of several distinct networks. We note “non-reciprocity” between regions, a lack of evidence for direct exchange in Mesoamerica prior to the Late Postclassic (ca. 1350 CE), almost five centuries after Mesoamerican items become common in the north. Turquoise is the only material that comes from northern sources, and turquoise artifacts are vanishingly rare in major Mesoamerican centers predating 1350 CE. For now, we conclude that although interpenetration and interdependence began to affect northern centers ca. 900 CE, there is little evidence for mutuality until late in prehispanic times, and even then there were large spatial gaps in the occurrence of many items, implying variable integration of the areas in the NW/SW.

Biographic Sketch

Ben Nelson is Professor Emeritus of Anthropology, School of Human Evolution & Social Change, Arizona State University, where he taught from 1995 till 2019. He is interested in cycles of sociopolitical complexity and connectivity in the prehispanic societies northwestern Mexico and the U.S. Southwest. He directs the La Quemada-Malpaso Valley Archaeological project in Mexico, which has used techniques from the earth sciences to explore environmental change, chemistry and petrography to evaluate exchange relationships, bioarchaeology to evaluate social violence, and ethnoarchaeology to understand the ways in which regional ritual traditions integrate indigenous populations. He also is co-director of the Connections and Impacts of Prehispanic North and West Mexican Cultures project, which addresses the question of long-distance exchange from a macro-regional perspective, gathering and analyzing data about exotic objects and symbols from hundreds of prehispanic settlements in Mesoamerica and the US Southwest. He served as Associate Director of the School of Human Evolution and Social Change from 2004-09, and as President of the Archaeology Division of the American Anthropological Association from 2009-11. He is author or editor of three books; his articles have appeared in journals such as American Anthropologist, American Antiquity, Arqueología Mexicana, Human Ecology, Journal of Anthropological Archaeology, Proceedings of the National Academy of Sciences (US), and Quaternary Research.