Social and cultural consequences of a late Holocene eruption of Popocatépetl in central Mexico

ARTICLE IN PRESS

Quaternary International 151 (2006) 19–28

Social and cultural consequences of a late Holocene eruption of Popocate´petl in central Mexico Patricia Plunket, Gabriela Urun˜uela Department of Anthropology, Universidad de las Ame´ricas-Puebla, Cholula, Puebla, Mexico, c.p. 72820 Available online 5 May 2006

Abstract Volcanic disasters often have been invoked as prime movers in the culture history of ancient civilizations. They have been used to explain large-scale migrations, the destruction of cities, famine, and demographic collapse. In this paper we explore the geological, archeological, and sociological records in order to provide insights into the complex nature of human responses to a major volcanic event of Popocate´petl in central Mexico that took place 2000 years ago. We suggest that the population implosion experienced by two emerging highland cities in the first century AD—Teotihuacan in the Basin of Mexico and Cholula in the Puebla Valley—was due to both the immediate consequences of the volcanic event and the disaster-driven acceleration of social processes already underway when the catastrophe struck. We conclude that a better understanding of the relationship between human populations and volcanic hazards and disasters permits a more realistic assessment of the social and cultural significance of eruptive phenomena in the prehispanic period. r 2006 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction A perplexing problem in the archeology of ancient Mexico revolves around the emergence of the great urban center of Teotihuacan in the central highlands. It is clear that the city experienced a huge population increase during the first century AD at the same time that much of the surrounding countryside was abandoned, but archeological opinion is divided regarding the causes of these population movements. One argument holds that Teotihuacan had become a powerful state capable of coercing most of the inhabitants of the Basin of Mexico to abandon their homes and move into the city; the second proposal views these movements as a response to major eruptive activity of the Xitle volcano, which is located in the fertile southwestern sector of this area. In this paper, we explore both of these ideas and conclude that indeed volcanism was probably responsible for many of the demographic changes noted by archeologists, but we propose that it was a different volcano—Popocate´petl—that provoked the relocation of tens of thousands of people. We further suggest that this major VEI-6 eruption at the beginning of the Common Era Corresponding author. Tel.: +52 222 229 2048.

E-mail addresses: [email protected] (P. Plunket), [email protected] (G. Urun˜uela).

was an important catalyst in the emergence, not only of Teotihuacan, but also of Cholula, a major prehispanic city in the neighboring Valley of Puebla (Fig. 1). Volcanic disasters have long been invoked by archeologists as prime movers in the culture history of various parts of the world. Not only have they been blamed for the demise of early civilizations (Thera in the Aegean (Hoffman, 1999)) and the destruction of important cities (Pompeii in Italy (Scarth, 1999) and Cuicuilco in central Mexico (Siebe, 2000)), but also for mass migrations (Ilopango in El Salvador (Dull et al., 2001) and the White River Volcano in Alaska (Harris, 2000)), famine (Tambora in Indonesia (Sigurdsson, 2000)), and demographic collapse (Ilopango (Dull et al., 2001)). Unfortunately, in the absence of written texts that provide a precise sequence of events, it is difficult to identify and correlate cause and effect. Below we explore both the geological and archeological records in order to provide insights on the variable nature of human responses to a major volcanic event of the Popocate´petl volcano 2000 years ago. 2. Geological background and archeological contexts Although Popocate´petl has erupted over 30 times since the 14th century (Simkin and Siebert, 2000a), none of these

1040-6182/$ - see front matter r 2006 Elsevier Ltd and INQUA. All rights reserved. doi:10.1016/j.quaint.2006.01.012

ARTICLE IN PRESS P. Plunket, G. Urun˜uela / Quaternary International 151 (2006) 19–28

20

Basin of Mexico

Teotihuacan

L. Texcoco

Tlalancaleca Cuicuilco Xitle

Malinche Xochitécatl Valley of Iztaccíhuatl Puebla-Tlaxcala 0m Coapan

Chalco

.5m 1m

Amecameca

Tetimpa

Cholula

Popocatépetl Sites mentioned in text Contour interval 330m 0

Colotzingo

20 km

Fig. 1. Map of central Mexico showing sites discussed in the text and lateral extent of the airfall deposit produced the Plinian-style eruption of Popocate´petl 2000 years ago (adapted from Hirth, 1984, Fig. 6.1; Panfil, 1996, p. 18).

explosive events appear to have had major destructive consequences for local populations. Geologists and archeologists have documented two earlier eruptions, however, whose volcanic explosivity was of a very different order (e.g., Seele, 1973; Panfil, 1996; Siebe et al., 1996; Plunket and Urun˜uela, 1998; Panfil et al., 1999). Our discussion will focus on the earliest of these, a Plinian-style eruptive event that took place during the first century AD. It produced a column that rose between 20 and 30 km before depositing a minimum of 3.2 km3 of yellow andesitic pumice over more than 240 km2 in an arc extending at least 25 km east of the volcano’s crater (Fig. 1) (Panfil, 1996, p. 16; Panfil et al., 1999). Siebe (2000, p. 61) has classified this as a VEI-6 eruption. The volcanic explosivity index describes, on a scale of 0–8, the size of explosive eruptions by providing measures of the volume of material expelled and the height of the eruptive column; in general terms, anything above 5 is considered to be very large (Pyle, 2000). A VEI-6

classification signifies a catastrophe of major proportions, such as Krakatoa in 1883, which only occurs about once every century on a worldwide scale; these are highly explosive and 100% fatal (Simkin and Siebert, 2000b). Following the Plinian phase of this eruption, lavas flowed over 50 km2 of the eastern piedmont of the volcano leaving between 20 and 40 m of olivine-bearing andesitic rock, known today as the Pedregal de Nealtican, that dammed and diverted drainages, altering the surface hydrology of the western Puebla valley (Panfil, 1996; Panfil et al., 1999). The collapsing eruptive column buried at least one large dispersed farming village on the northeastern flank of the volcano, Tetimpa (Figs. 2 and 3), and thus provided archeoologists with an exceptional opportunity to study village life and modes of rapid abandonment during the Terminal Formative period (100 BC–AD 100) in central Mexico (Plunket and Urun˜uela, 1998, 2003; Urun˜uela and Plunket, 1998, 2003). The primary archeological contexts

ARTICLE IN PRESS P. Plunket, G. Urun˜uela / Quaternary International 151 (2006) 19–28

21

Fig. 4. Amphora and jars abandoned at a Tetimpa house. Fig. 2. View of Popocate´petl volcano from excavations at Tetimpa. The volcanic stratigraphy is visible underneath the modern cornfields; the Pedregal de Nealtican lava flow appears as a dark band between the cornfields and the volcano.

Fig. 3. Remains of a Tetimpa house dating to the first century AD.

afforded by Tetimpa’s domestic structures and activity areas (Fig. 4) include the charred remains of roofing materials and food debris that have permitted us to securely situate the Plinian eruption in the first century AD by radiocarbon dating (Table 1). The most reliable date from the village setting is an Accelerator Mass Spectrometer determination on a carbonized corn cob found sealed inside a jar left outside a kitchen. This yielded a reading of 2010740 BP (Beta-146572). Three of the radiocarbon dates listed in Table 1 are obviously too late since they are not consistent with the chronology of the ceramic assemblages. We suspect that organic fertilizers have percolated through the unconsolidated lapilli, contaminating some of the charred materials, particularly those that are in direct contact with the pumitic ash. Thus, the date on the corn cob, protected by the jar, appears to be the most reliable result since it was obtained in primary context from an annual plant that was never in contact with any contaminants. In general, the radiocarbon dates from the houses buried by the lapilli indicate that the eruption probably took place towards the middle of the first century AD.

3. The emergence of Teotihuacan: volcanic catastrophe or political coercion? Most often, archeologists have framed their considerations of the effects of volcanic disasters on ancient culture history in terms of the immediate impact of the catastrophe itself (however, see Sheets, 1979; Dull et al., 2001). For example, in the Basin of Mexico, the destruction of the important prehispanic center of Cuicuilco by extensive lava flows from Xitle volcano (Fig. 1) has been used consistently to explain the dramatic demographic changes of the period between 100 BC and AD 100 that were involved in the emergence of the Teotihuacan state. In essence, authors postulate that the inhabitants of Cuicuilco were forced to leave and seek refuge on the opposite side of the Basin at Teotihuacan, thus inflating the population and accelerating the processes leading to social complexity in that emergent city (e.g., Porter Weaver, 1993, p. 79; Blanton et al., 1997, p. 123; Cowgill, 2000, p. 259; Sarmiento, 2000, p. 360; however, see Sanders et al., 1979, p. 106). Recent geological studies, however, have demonstrated that the lava flows did not encase Cuicuilco until at least the middle of the third century and perhaps as late as the beginning of the fifth century AD (Co´rdova et al., 1994; Gonza´lez et al., 2000; Siebe, 2000), thus eliminating the Xitle eruption as a causal force in this rapid social and cultural rearrangement. Siebe (2000) has suggested that the Popocate´petl eruption of the first century AD that we have documented on the other side of the volcano at Tetimpa, and not the Xitle event, was the source of destruction that led to these demographic changes. Siebe (2000, p. 61) places this event between 250 and 50 BC although archeological data and radiocarbon dates from primary contexts at Tetimpa support a later placement as outlined here. While the eastern slopes of the volcano that form the western rim of the Valley of Puebla were buried under the Plinian fallout, the NW flank, which descends into the Basin of Mexico, was ‘‘devastated by pyroclastic flows,’’ and Siebe (2000,

ARTICLE IN PRESS P. Plunket, G. Urun˜uela / Quaternary International 151 (2006) 19–28

22

Table 1 Radiocarbon dates from excavated contexts under the yellow pumitic ash in the ancient village of Tetimpa, Puebla, on the northeastern flank of the Popocate´petl volcano Laboratory number Years BP Context Charred material associated with ceramics on patio floor

AD 650

AD 620–680

1560740 1610780 1800760 1800760

Charred material from a wattle-and-daub storage bin charred material from censer in front of household shrine Charred material from household hearth Charred material from household shrine

AD AD AD AD

1860770 1910760 1930780 1940760 2010740

Charred material from smashed censer Charred roofing material Charred material from household hearth Charred material from wattle-and-daub storage bin Corn cob inside a sealed olla

AD 135 AD 100 AD 80 AD 75 10 BC

AD 430–550 AD 385–545 AD 145–330 AD 135–265 AD 290–325 AD 75–240 AD 55–160 AD 5–147 AD 15–130 50 BC–AD 40

AD AD AD AD AD AD

2030760 2070780 2150780

Charred material from a roasting pit excavated into the patio floor 40 BC Charred material from household hearth 50 BC Charred material from household shrine 181 BC

146574a 135900a 099431 135901

Beta 135902 Beta 099433 I-18291 Beta 099430 Beta 146572 AMS Beta 146575 I-17815 I-17814 a

cal 2 Sigma

1390760

Beta 146573 Beta Beta Beta Beta

cal Intercept cal 1 Sigma

a

530 430 240 235

560–720 740–760 410–600 250–625 90–395 80–390

AD 5–340 20 BC–AD 245 91 BC–AD 315 45 BC–AD 225 100 BC–AD 70

100 BC–AD 40 190 BC–AD 90 182 BC–AD19 357 BC–AD 116 357–49 BC 390 BC–AD 20

These dates are considered too late based on a comparison with ceramic chronologies.

p. 61) argues that survivors on both sides of the volcano had to relocate. The prehispanic settlement pattern maps of the Basin of Mexico that were developed by archeologists 30 years ago (Sanders et al., 1979) show an impressive and consistent population build-up along the southern and eastern piedmont and lakeshore between 650 and 100 BC. By the end of this period, population densities in this area had peaked at over 40 inhabitants per km2 and the total population of the southern and eastern sections of the Basin is estimated at about 76,500 (Sanders et al., 1979: Tables 6.17 and 6.18). The settlement map that describes the patterns of the next archeological phase (100 BC–AD 100) shows dramatic changes in these areas; densities are listed as ‘‘inappreciable’’ with a total population that is too small to calculate (Sanders et al., 1979: Tables 6.17 and 6.18). At least 10 regional centers (large, nucleated communities with monumental architecture), over 30 large and small villages, and almost 70 hamlets disappear from the maps (Fig. 5). This was the first population decline registered in the history of the Basin of Mexico, and using the archeological estimates we can calculate that the Basin lost about 30% of its inhabitants at some point during this 200 year period (Sanders et al., 1979, p. 183). The settlement pattern maps also indicate that prior to 100 BC the northern part of the Basin was virtually devoid of villages and hamlets, although Teotihuacan was already developing into a small city with a population of perhaps 20,000 (Sanders et al., 1979, p. 102). The nucleation of population at Teotihuacan was greatly enhanced between 100 BC and AD 100—precisely, the time period in which the southern and eastern parts of the Basin underwent the impressive abandonment cited above. About 80–90% of the Basin’s population became concentrated at the city and for the first time numerous hamlets were established to the

north and west of the urban center. Sanders et al. (1979, p. 107) comment that, Somehow y [Teotihuacan] was able not only to incorporate the rest of the Basin’s population politically, but to relocate them physically as well. The motive for such extreme population nucleation is still unclear to us. Presumably political control was the principal factor involved. However, we know of no other situation in the historical or archaeological record in which so large a sedentary regional population was involved in such a drastic relocation y. The apparent substantial Phase Four population decline suggests that this extreme population nucleation was not without a certain stress, manifested in the loss of population. The population estimates indicate that something happened to about 70,000 people (Sanders, 1981: Table 6.2). Around 50,000 may have relocated to Teotihuacan, but 20,000 remain unaccounted for. Millon (1981, p. 217) has proposed that Teotihuacan’s incredibly rapid growth was part of the emergent state’s ‘‘comprehensive plan for depopulating the countryside and resettlement in the city’’, even though it would have severely stressed the Teotihuacan Valley economy (Sanders, 1981, p. 180). Modern studies of disaster (e.g., Nolan, 1979; Oliver-Smith and Hoffman, 1999) indicate that relocation, particularly permanent relocation, is no simple matter. They demonstrate that place plays a crucial role ‘‘in the construction of individual and community identities’’ and that the ‘‘removal of a community from its ‘ground’ y may be profoundly traumatic’’ (Oliver-Smith, 1996, p. 308). At face value, the intentional displacement of 50,000 people and the consequent abandonment of some of the best agricultural lands in the Basin of Mexico by an emergent expansionist archaic state, no matter how aggressive and

ARTICLE IN PRESS P. Plunket, G. Urun˜uela / Quaternary International 151 (2006) 19–28

Basin of MŽxico (300-100 BC)

23

Basin of MŽxico (100 BC-100 AD)

Teotihuacan Teotihuacan

L. Texcoco

L. Texcoco

Cuicuilco

Supra-regional center Regional center Large nucleated village Small nucleated village

Cuicuilco

Iztacc’huatl

Iztacc’huatl

PopocatŽpetl

PopocatŽpetl

Large dispersed village Small dispersed village Hamlet Tezoyuca hilltop center

Small ceremonial precinct Excavated site, not surveyed

Contour interval 330m 0

20 km

Fig. 5. Distribution of settlements in the Basin of Mexico in the Late Formative Period (300–100 BC) and the Terminal Formative Period (100 BC–AD 100) (adapted from Sanders, 1981: Figs. 6.13 and 6.15).

powerful its armies, no matter how persuasive its ideology, appears to be an awesome and puzzling undertaking. We address the problems of forced relocation below. Our dating of the VEI-6 eruption of Popocate´petl to the first century AD offers another scenario that provides a different perspective on the settlement pattern data. We would not suggest that the eruption impacted the Basin of Mexico in exactly the same way that it did for the western Valley of Puebla where over 100 km2 were permanently lost under volcanic deposits and relocation was unavoidable (Plunket and Urun˜uela, 2003), but both the immediate and long-term consequences for the settlements along the slopes

of the Sierra Nevada may have been significant. Not only were the southern and eastern sections of the Basin of Mexico prime agricultural areas, but their freshwater lakes and forested piedmonts were richly endowed with natural resources, making them valuable hunting and gathering territories. As pyroclastic materials fell on the forested slopes the ash would have suffocated wildlife. The hot pyroclastic flows, with temperatures between 250 and 600 1C may have ignited multiple, highly destructive forest fires that would have spread quickly through the Sierra Nevada during the dry season, and the melting glacier would have formed destructive lahars that rushed down the

ARTICLE IN PRESS 24

P. Plunket, G. Urun˜uela / Quaternary International 151 (2006) 19–28

steep canyons into the Amecameca river and the Chalco area (see Nakada, 2000, p. 948; Siebe, 2000; Capra et al., 2004). Some of the 20,000 people missing from Sander’s population estimates may have perished in these events. We have suggested elsewhere (Plunket and Urun˜uela, 2000) that the Plinian eruption took place during the dry season, from October through May. The prevailing winds generally blow west to east at this time of year, explaining at least partially why the eruptive column collapsed over the northeastern flank of the mountain (Delgado et al., 1995). The archeological evidence also supports this interpretation. The agricultural fields adjacent to Tetimpa’s houses were not planted and there is little or no corn in the domestic storage facilities, indicating that households had used most of their existing food supply. This, in addition to the loss of seeds for next season’s planting, must have contributed to short-term food shortages. The destruction of important hunting and gathering areas would have been devastating, and the smoke from the fires added to the ash could have caused significant shortterm climatic changes as well, including frost and drought, that may have resulted in famine and consequent population relocation (Nolan, 1979, p. 330; Sigurdsson, 2000, pp. 10–11). In their discussion of the natural environment of the Basin of Mexico, Sanders et al. (1979, pp. 81–89) emphasize that fluctuations in the onset of the monsoonal rains and winter frost have led to devastating crop failures in historic times, and they classify the Basin as a moderateto-high risk area for agriculture. Under these circumstances, a VEI-6 eruption could easily shift the climate towards cooler and drier conditions on the 2000-m plus plateau. Establishing whether the scenario we suggest here is close to reality will require the combined efforts of geologists and archeologists focus on the fate of the Terminal Formative sites in the southeastern Basin of Mexico, documenting the impact of this eruption on local settlements.

4. Natural disasters and the acceleration of social change Anthropologists have observed that disasters tend to accelerate changes that are already underway (Nolan, 1979; Oliver-Smith, 1996, p. 313; Hoffman, 1999; Driessen and McDonald, 2000). They motivate social actions, and social action motivates change (Hoffman, 1999, p. 311), providing contexts for new agendas, new power relations, and the emergence of new leaders as part of the recovery process (Oliver-Smith, 1996, p. 310). In the Basin of Mexico, archeologists have identified a number of trends for the period immediately prior to the eruption that provide important contextual clues for interpreting the impact of the events of the first century AD: (1) significant population growth along the piedmont of Iztaccı´ huatl in the Texcoco area, in what may have been a high-risk zone; (2) population centralization at Cuicuilco and Teotihuacan; (3) political fragmentation and conflict; and (4) possible

problems with food supplies at larger settlements (Sanders et al., 1979, pp. 101–103). As in other regions of ancient Mesoamerica, the Terminal Formative in the Basin of Mexico was a time of rapid population growth and consequent social change. In this context, a major volcanic eruption may have served as a catalyst to accelerate those changes not only by highlighting the inability of local authorities to appease and control the forces of nature, but even to ensure the necessary resources for basic subsistence. Thus, it may have provided a situation ripe for decision-making at all levels of society that led to both intended and unintended consequences. Studies of modern volcanic disasters demonstrate that decisions to remain or relocate are made in response to both real and perceived social and natural hazards in the extra-village environment and not just in response to the volcanic conditions themselves (e.g., Nolan, 1979, p. 328); the trend is for communities to seek greater interaction, not isolation, from the larger society (Nolan, 1979, p. 314). We now look at the aftermath of the Paricutı´ n eruption in the state of Michoaca´n, Mexico, during the 1940s in order to provide a contextual parallel to the post-eruptive world faced by the inhabitants of central Mexico at the beginning of the first millennium. We realize that the size and style of these two volcanic events were considerably different, but we think the variable responses recorded during the modern episode illustrate the social and cultural complexity of relocation processes. 5. The towns of Paricutı´ n In her study of the five towns situated around the nascent Paricutı´ n volcano in 1943, Nolan (1979) found a general tendency for villagers to reject local community isolation in favor of greater interaction with national Mexican society, even at the risk of losing some of their Tarascan traditions. However, her analysis of the social impact of the eruption also describes considerable variability in the strategies employed by each community to deal with the catastrophe and the degree to which each allowed outside forces of change to determine their future. She demonstrates that in the end each response was strongly conditioned by local history and internal social stratification before the onset of the volcanic period (1943–1952), in spite of the towns’ shared experience as disaster victims. Below we summarize Nolan’s (1979) most important observations regarding the social and cultural consequences that Paricutı´ n had on the five towns closest to it. The most successful town, San Juan, was also the largest community. It served as the political center for the region and its church was home to a powerful religious image— the Lord of the Miracles—that served as a focus of religious devotion for both the meztizo majority and the Indian minority. When the lava finally began to cover the cemetery of San Juan, the entire town left in a procession led by the Bishop of Zamora who carried the statue of the Lord of the Miracles, in effect following a divine command

ARTICLE IN PRESS P. Plunket, G. Urun˜uela / Quaternary International 151 (2006) 19–28

to reestablish their community in a new location. Twenty years later, San Juan was once again the municipal seat, a vibrant town that offered many opportunities for its growing population. A new church had been built for the Lord of the Miracles and pilgrims arrived from far away to worship the image whose prestige had been enhanced by the community’s successful survival of the catastrophe. San Juan was the only town where the traditional system of mayordomı´a (a stewardship charged with caring for a church and its saints) survived. At the same time, however, new traditions had appeared that focused on the pioneer experience. The lack of solidarity that characterized the other communities was reflected in the more heterogeneous and less unified decisions that stemmed from the volcanic disaster. Some were absorbed by the city of Uruapan as barrios or quarters, while the younger generations of others migrated to more distant urban areas, particularly when they had relatives already living there. However, strong community solidarity did not necessarily result in a successful post-volcanic adaptation. The town of Angahuan, for example, chose to remain in its original location even though its lands had suffered serious environmental damage. In making this choice, community members isolated themselves both socially and economically from the outside world, and although they engaged in guiding tourists over the devastated landscape, this activity eventually led to the commercialization and consequent devaluation of their Indian identity. Life in Angahuan became marginal at best and offered little opportunity to its young people. In light of our discussion of the archeological record relating to the first century eruption of Popocate´petl, it is relevant to consider here that the federal government of Mexico tried to evacuate all of the towns around Paricutı´ n, on occasion through the direct intervention of General La´zaro Ca´rdenas, a popular ex-president from the state of Michoaca´n. The success of these evacuation attempts seems to have depended largely upon the degree to which each community was allowed to make its own decisions about when to leave and where to go, but in some cases, like the town of Paricutı´ n itself, the army had to intervene and force the inhabitants into trucks. The individual histories of the towns affected by Paricutı´ n demonstrate not only the variable nature of their strategies, but also that a modern state with a strong centralized government and military structure lacked the capacity to resolve the problems of relocation for five communities whose total population was less than 5000 people. Although archeologists (Sanders et al., 1979; Millon, 1981) have suggested that the emergent Teotihuacan state, with a population of between 20,000 and 40,000 undertook the forced resettlement of perhaps as many as 50,000 people for reasons of political control, this seems unlikely when we consider the Mexican government’s experience with the towns of Paricutı´ n during the 1940s.

25

6. The towns of Popocate´petl The ways in which the emerging Teotihuacan state may have reacted to the eruption and the recovery process need to be balanced by the decisions made by individual towns and villages. What were the advantages of moving to Teotihuacan? The potential for the development of hydraulic agriculture by swamp reclamation and irrigation from springs in the Teotihuacan Valley may be part of the answer, particularly if famine resulting from short-term climatic shifts, the loss of stored food and seed for planting, and the privations caused by pre-existing internal social inequality were a major consideration, even though this would have required the maximal utilization of this resource (Sanders, 1981, p. 180). Teotihuacan authorities, on the other hand, may have welcomed the population influx within their sphere of influence if conflict were an important aspect of life in the Basin of Mexico, but they may also have found it difficult to effectively exclude the migrants if environmental and social conditions were extremely difficult. The sudden abundance of labor, viewed against a dramatic and horrifying visual backdrop associated with traumatic dislocation, may help explain the surge in the building of civic-ceremonial architecture at Teotihuacan during the first and second centuries AD, which resulted in some of the largest structures in the New World. Much of this construction took place in the century after the VEI-6 eruption of Popocate´petl (Sugiyama, 2004, Table 4.2). Although the timing may be coincidental, we believe that the massive building projects at Teotihuacan represent not only the creation of a new kind of settlement, but also the development of a new accommodation between humans and their gods. In the western Valley of Puebla we see a similar process taking place and we suspect a similar causality. Unfortunately, the Puebla-Tlaxcala region does not have as detailed settlement data as does the Basin of Mexico, but the region’s population just prior to the eruption was probably over 100,000 (Dumond, 1972, p. 119). The large, complex regional centers, close to the Sierra Nevada— Colotzingo (Urun˜uela, 1989), Coapan, Xochite´catl (Serra and Palavicini, 1996; Spranz, 1996), and Tlalancaleca (Garcı´ a Cook, 1981)—were abandoned around the first century AD. The Classic period (AD 100–650) population became concentrated at Cholula and a few other settlements but the total number of inhabitants appears to have diminished (Dumond, 1972, p. 115), perhaps by as much as 30% (Garcı´ a Cook, 1981, p. 263). As we have mentioned, the process of relocation for many people residing along the eastern flank of Popocate´petl would have been less open to debate than in the neighboring Basin of Mexico since a significant expanse of the productive piedmont was covered by volcanic deposits that prevented reclamation (Plunket and Urun˜uela, 2000). The forest fires and the consequent decimation of hunting and gathering territories in addition to possible drought and increased risk of frost in a decidedly non-egalitarian social system, however,

ARTICLE IN PRESS 26

P. Plunket, G. Urun˜uela / Quaternary International 151 (2006) 19–28

Fig. 6. Radiocarbon dates from the initial stages of monumental architecture in Cholula, Puebla.

would have created the same socially charged environment in the Valley of Puebla-Tlaxcala as in the Basin of Mexico. Twelve radiocarbon dates associated with early monumental architecture in Cholula suggest that the main components of the city’s public structures were designed and built at the beginning of the second century AD (Plunket and Urun˜uela, 2005, Tables 1 and 2) (Fig. 6). Eleven of these determinations were obtained on samples recovered from the earliest structures of the Great Pyramid of Cholula, while the twelfth came from a hearth associated with a large prehispanic platform underneath the sixteenth century Franciscan monastery of San Gabriel. All of the dates support a construction sequence that initiates in the latter part of the first or beginning of the second century AD. Both Cholula (Fig. 7) and Teotihuacan (Fig. 8) appear to have initiated the construction of their monumental architecture during the latter part of the first century or the first part of the second century AD (Millon et al., 1965; Marquina, 1970). Although they developed in different ways, we think that the population implosion that each experienced 2000 years ago was due to both the immediate and long-term effects of the VEI-6 explosion of the Popocate´petl. These two cities received huge numbers of refugees and later migrants as the communities situated on the slopes of the Sierra Nevada relocated to what they perceived to be less vulnerable environments; some may have intended to return to their lands while others knew they could not. However, all of them must have contributed to the impressive acceleration of social processes already underway before the volcanic disaster as each of these cities tried to cope with one of the largest human catastrophes of the prehispanic period. Hoffman (1999, p. 306) has observed that some disasters ‘‘create legend’’, while others have noted that religious responses to catastrophe can be highly adaptive (Nolan, 1979, p. 331) as people deal with conflicts surrounding ‘‘the allocation of blame and the allocation of resources’’

Fig. 7. The Great Pyramid of Cholula surmounted by the church of Our Lady of the Remedies.

Fig. 8. The civic-ceremonial center of Teotihuacan; the Pyramid of the Sun appears on the left side of the Street of the Dead.

(Driessen and McDonald, 2000, p. 92). We find it suggestive that the two great cities of the central highlands both initiated ideologically rooted monumental building programs using the labor of those who decided to relocate in urban environments in the aftermath of the eruption, and we suspect that the motivation behind these programs had much to do with the emergence of a new consideration of the relation between humans and the sacred, new ways of tempering or avoiding divine justice, new political agendas, and new power structures. The ancient Aztec ‘‘Legend of the Suns’’ describes how the third epoch of creation was destroyed by a volcanic eruption when it rained fire, the rocks boiled, and houses burned, and with its destruction came the creation of a new world (Bierhorst, 1992). Perhaps as Hoffman says, the Plinian event of the first century AD was indeed the stuff of legends. Acknowledgements We would like to thank the Consejo Nacional de Ciencia y Tecnologı´ a, the Sistema Regional Ignacio Zaragoza, the

ARTICLE IN PRESS P. Plunket, G. Urun˜uela / Quaternary International 151 (2006) 19–28

Mesoamerican Research Foundation, the Foundation for the Advancement of Mesoamerican Studies, and the Instituto de Investigacio´n y Posgrado of the Universidad de las Ame´ricas-Puebla for their generous support of our research at Tetimpa. The Consejo de Arqueologı´ a of the Instituto Nacional de Antropologı´ a e Historia granted the official permit for this project and has provided assistance and advice throughout our research that are much appreciated. We are particularly grateful to the townspeople of San Nicolas de los Ranchos and San Buenaventura Nealtican for allowing us to work among them.

References Bierhorst, J., (trans.), 1992. History and Mythology of the Aztecs: The Codex Chimalpopoca. University of Oklahoma Press, Norman. Blanton, R., Kowalewski, S., Feinman, G., Finsten, L., 1997. Ancient Mesoamerica: A Comparison of Change in Three Regions, second ed. Cambridge University Press, Cambridge. Capra, L., Poblete, M., Alvarado, R., 2004. The 1997 and 2001 lahars of Popocate´petl volcano (Central Mexico): textural and sedimentological constraints on their origin and hazards. Journal of Volcanology and Geothermal Research 131 (3–4), 351–369. Co´rdova, C., Martı´ n del Pozzo, A., Lo´pez Camacho, J., 1994. Paleolandforms and volcanic impact on the environment of prehistoric Cuicuilco, southern Mexico City. Journal of Archaeological Science 21, 585–596. Cowgill, G., 2000. The central Mexican highlands from the rise of Teotihuacan to the decline of Tula. In: Adams, R.E.W., MacLeod, M. (Eds.), The Cambridge History of the Native Peoples of the Americas, vol. II: Mesoamerica, Part I. Cambridge University Press, Cambridge, pp. 250–317. Delgado, H., Carrasco, G., Cervantes, P., Corte´s, R., Molinero, R., 1995. Patrones de viento en la regio´n del Volca´n Popocate´petl y Ciudad de Me´xico. In: Volca´n Popocate´petl, estudios realizados durante la crisis de 1994–1995. Centro Nacional de Prevencio´n de Desastres-Universidad Nacional Auto´noma de Me´xico, Mexico City, pp. 295–324. Driessen, J., McDonald, C., 2000. The eruption of the Santorini volcano and its effects on Minoan Crete. In: McGuire, W., Griffiths, D., Hancock, P., Stewart, I. (Eds.), The Archaeology of Geological Catastrophes. Geological Society Special Publications 171, London, pp. 81–93. Dull, R., Southon, J., Sheets, P., 2001. Volcanism, ecology, and culture: a reassessment of the Volca´n Ilopango TBJ Eruption in the Southern Maya Realm. Latin American Antiquity 12 (1), 25–44. Dumond, D., 1972. Demographic aspects of the classic period in PueblaTlaxcala. Southwestern Journal of Anthropology 28 (2), 101–130. Garcı´ a Cook, A., 1981. The historical importance of Tlaxcala in the cultural development of the Central Highlands. In: Sabloff, J. (Ed.), Archaeology, Supplement I, pp. 244–276; Bricker, V. (Gen. Ed.), Handbook of Middle American Indians. University of Texas Press, Austin. Gonza´lez, S., Pastrana, A., Siebe, C., Duller, G., 2000. Timing of the prehistoric eruption of Xitle volcano and the abandonment of Cuicuilco pyramid, southern Valley of Mexico. In: McGuire, W., Griffiths, D., Hancock, P., Steward, I. (Eds.), The Archaeology of Geological Catastrophes. Geological Society of London Special Publication 171, London, pp. 205–224. Harris, S., 2000. Archaeology and volcanism. In: Sigurdsson, H. (Ed.), Encyclopedia of Volcanoes. Academic Press, New York, pp. 1301–1314. Hirth, K., 1984. Trade and society in Late formative Morelos. In: Hirth, K. (Ed.), Trade and Exchange in Early Mesoamerica. University of New Mexico Press, Albuquerque, pp. 125–146.

27

Hoffman, S., 1999. After Atlas shrugs: cultural change or persistence after a disaster. In: Oliver-Smith, A., Hoffman, S. (Eds.), The Angry Earth: Disaster in Anthropological Perspective. Routlege, London, pp. 302–325. Marquina, I. (Ed.), 1970. Proyecto Cholula. Serie Investigaciones 19. Instituto Nacional de Antropologı´ a e Historia, Mexico City. Millon, R., 1981. Teotihuacan: city, state, and civilization. In: Sabloff, J. (Ed.), Archaeology, Supplement I, pp. 198–243; Bricker, V. (Gen. Ed.), Handbook of Middle American Indians. University of Texas Press, Austin. Millon, R., Drewitt, B., Bennyhoff, J., 1965. The Pyramid of the Sun at Teotihuaca´n: 1959 Investigations. Transactions of the American Philosophical Society 55 (Part 6), Philadelphia. Nakada, S., 2000. Hazards from pyroclastic flows and surges. In: Sigurdsson, H. (Ed.), Encyclopedia of Volcanoes. Academic Press, New York, pp. 945–955. Nolan, M.L., 1979. Impact of Paricutı´ n on five communities. In: Sheets, P. (Ed.), Volcanic Activity and Human Ecology. Academic Press, New York, pp. 293–338. Oliver-Smith, A., 1996. Anthropological research on hazards and disasters. Annual Review of Anthropology 25, 303–328. Oliver-Smith, A., Hoffman, S., 1999. The Angry Earth: Disaster in Anthropological Perspective. Routlege, London. Panfil, M., 1996. The Late Holocene Volcanic Stratigraphy of the Tetimpa Area, Northeast Flank of Popocate´petl Volcano, Central Mexico. M.S. Thesis, Department of Geosciences, Pennsylvania State University, College Park. Panfil, M., Gardner, T., Hirth, K., 1999. Late Holocene stratigraphy of the Tetimpa archaeological sites, northeast flank of Popocatepetl volcano, central Mexico. Geological Society of America Bulletin 111 (2), 204–218. Plunket, P., Urun˜uela, G., 1998. Preclassic household patterns preserved under volcanic ash at Tetimpa, Puebla. Latin American Antiquity 9 (4), 287–309. Plunket, P., Urun˜uela, G., 2000. The archaeology of a Plinian eruption of the Popocate´petl volcano. In: McGuire, W., Griffiths, D., Hancock, P., Stewart, I. (Eds.), The Archaeology of Geological Catastrophes. Geological Society Special Publications 171, London, pp. 195–204. Plunket, P., Urun˜uela, G., 2003. From episodic to permanent abandonment: responses to volcanic hazards at Tetimpa, Puebla, Mexico. In: Inomata, T., Webb, R. (Eds.), The Archaeology of Settlement Abandonment in Middle America. University of Utah Press, Salt Lake City, pp. 13–27. Plunket, P., Urun˜uela, G., 2005. Dating Cholula. Report submitted to the Foundation for the Advancement of Mesoamerican Studies, Inc., Crystal River, Florida. http://www.famsi.org/reports/02042/index. html Porter Weaver, M., 1993. The Aztecs, Maya, and their Predecessors: Archaeology of Mesoamerica, third ed. Academic Press, New York. Pyle, D., 2000. Sizes of volcanic eruptions. In: Sigurdsson, H. (Ed.), Encyclopedia of Volcanoes. Academic Press, San Diego, pp. 263–269. Sanders, W., 1981. Ecological adaptation in the Basin of Mexico: 23,000 B.C. to the present. In: Sabloff, J. (Ed.), Archaeology, Supplement I, pp. 147–197; Bricker, V. (Gen. Ed.), Handbook of Middle American Indians. University of Texas Press, Austin. Sanders, W., Parsons, J., Santley, R., 1979. The Basin of Mexico: Ecological Processes in the Evolution of a Civilization. Academic Press, New York. Sarmiento, G., 2000. La creacio´n de los primeros centros ceremoniales. In: second edManzanilla, L., Lo´pez Luja´n, L. (Eds.), Historia Antigua de Me´xico, vol. I: El Me´xico Antiguo, sus Areas Culturales, los Orı´ genes y el Horizonte Precla´sico. Instituto Nacional de Antropologı´ a e Historia, Instituto de Investigaciones Antropolo´gicas, Editorial Miguel Angel Porru´a, Mexico City, pp. 335–362. Scarth, A., 1999. Vulcan’s Fury: Man Against the Volcano. Yale University Press, New Haven. Seele, E., 1973. Restos de milpas y poblaciones prehispa´nicas cerca de San Buenaventura Nealtican, Puebla. Comunicaciones 7, 77–86.

ARTICLE IN PRESS 28

P. Plunket, G. Urun˜uela / Quaternary International 151 (2006) 19–28

Serra, M.C., Palavicini, B., 1996. Xochite´catl, Tlaxcala, en el Periodo Formativo (800 a.C.-100 d.C.). Arqueologı´ a 16, 149–157. Sheets, P. (Ed.), 1979. Volcanic Activity and Human Ecology. Academic Press, New York. Siebe, C., 2000. Age and archaeological implications of Xitle volcano, southwestern Basin of Mexico-City. Journal of Volcanology and Geothermal Research 104, 45–64. Siebe, C., Abrams, M., Macı´ as, J., Obenholzner, J., 1996. Repeated volcanic disasters in Prehispanic time at Popocate´petl, central Mexico: past key to the future? Geology 24 (5), 399–402. Sigurdsson, H., 2000. Introduction. In: Sigurdsson, H. (Ed.), Encyclopedia of Volcanoes. Academic Press, San Diego, pp. 1–13. Simkin, T., Siebert, L., 2000a. Appendix 2: catalog of historically active volcanoes on Earth. In: Sigurdsson, H. (Ed.), Encyclopedia of Volcanoes. Academic Press, San Diego, pp. 1365–1383.

Simkin, T., Siebert, L., 2000b. Earth’s volcanoes and eruptions: an overview. In: Sigurdsson, H. (Ed.), Encyclopedia of Volcanoes. Academic Press, San Diego, pp. 249–261. Spranz, B., 1996. Totimehuacan und Xochitecal: zwei Pla¨tze in Puebla und Tlaxcala, Mexiko, in pra¨klassischer Zeit. Tribus 45, 138–150. Sugiyama, S., 2004. Governance and polity at Classic Teotihuacan. In: Hendon, J.A., Joyce, R.A. (Eds.), Mesoamerican Archaeology. Blackwell Publishing, Oxford, pp. 97–123. Urun˜uela, G., 1989. Investigaciones arqueolo´gicas en Colotzingo. Puebla. Notas Mesoamericanas 11, 110–119. Urun˜uela, G., Plunket, P., 1998. Areas de actividad en unidades dome´sticas del Formativo Terminal en Tetimpa, Puebla. Arqueologı´ a 20, 3–19. Urun˜uela, G., Plunket, P., 2003. Testimonios de diversos tipos de abandono en Tetimpa, Puebla. Trace 43, 84–96.