Human Cremation in Mexico 3,000 Years Ago

Human cremation in Mexico 3,000 years ago

William N. Duncan*†, Andrew K. Balkansky‡, Kimberly Crawford‡, Heather A. Lapham‡, and Nathan J. Meissner‡

*Department of Anthropology, St. John Fisher College, Rochester, NY 14618; and ‡Department of Anthropology, Southern Illinois University, Carbondale, IL 62901

Edited by Joyce Marcus, University of Michigan, Ann Arbor, MI, and approved February 22, 2008 (received for review November 10, 2007) Mixtec nobles are depicted in codices and other proto-historic ological contexts and osteological analyses of the burials, and documentation taking part in funerary rites involving cremation. evaluate the evidence based on known aspects of the Mixtec The time depth for this practice was unknown, but excavations at mortuary program. These early examples of cremation, along the early village site of Tayata, in the southern state of Oaxaca, with other factors, could reflect the emergence of ranked Mexico, recovered undisturbed cremation burials in contexts dat- society in the ancient Mixteca Alta and indicate that proto- ing from the eleventh century B.C. These are the earliest examples historic methods of marking social status had precursors of a burial practice that in later times was reserved for Mixtec kings extending back 3,000 years from the present. and Aztec emperors. This article describes the burial contexts and human remains, linking Formative period archaeology with eth- Archaeological Context nohistorical descriptions of Mixtec mortuary practices. The use of Survey and excavation indicate that Tayata was among the cremation to mark elevated social status among the Mixtec was largest villages of the preurban Formative period in the established by 3,000 years ago, when hereditary differences in Mixteca Alta of Oaxaca, Mexico (18, 19). During the 2004 field rank were ﬁrst emerging across Mesoamerica. season, the area adjacent to one of several excavated houses yielded the cremated burials along with diagnostic pottery and archaeology ͉ Mixtec ͉ bioarchaeology ͉ mortuary practices ͉ Mesoamerica radiocarbon samples dating from the late Early Formative to Middle Formative transition. During these phases (20–22), he Mixtec were one of the original Mesoamerican civiliza- Tayata expanded significantly in size and initiated construc- Ttions (1, 2), but until recently little was known about the tion of nonresidential buildings and public spaces. The time- archaeological precursors of Mixtec cities and states (3–5). The frame for these developments at Tayata in calendar years was discovery of cremated human burials (Ϸ1000 B.C.) from exca- the late twelfth through tenth centuries B.C., when significant vations at Tayata, an early village in the Mixteca Alta of Oaxaca, changes in social complexity were occurring in many parts of Mexico, suggests that status-marking funerary rites existed Mesoamerica (23–25). The Tayata cremations were associated centuries before urban states arose and millennia before histor- with a house and middens containing evidence for deer, dog, ical sources reference the cremation of Mixtec kings and Aztec and fish consumption, pottery firing, marine shell ornament emperors. Our study documents one of the key building blocks production, and access to exotic valuables such as obsidian. in the initial growth of social complexity: the time when early Dog remains were also interred with the cremations and found village societies divided into ranks and a person’s status in life in the middens in ways suggestive of ritual feasting (Fig. 4). was reflected at death, reinforcing the economic differentiation This area in and around the residential structure associated that was fundamental to early village life (6, 7). However, to with burials 1 and 2 stands apart from other contexts at Tayata understand the beginnings of those complex societies that given its higher frequencies of both dog and fish remains—a were the building blocks of later Mesoamerican civilizations, it pattern that also holds when the fauna recovered from this helps to begin with anthropological information from historical residence is compared with contemporaneous assemblages in periods. the Valley of Oaxaca. This specific residential area at Tayata One way that the proto-historic Mixtec marked social status contains twice as many dog remains and more than ten times was through burning the body of the deceased in various ways the number of fish bones as houses 16 and 17 at San Jose´ (8–11; Fig. 1). We define reduction of a corpse through Mogote, a relatively elite house complex dating to the late burning as cremation, typically to ash, although there is some Early Formative (26). latitude in the use of this term regarding how complete the Burial 1 had fragments of a dog skull, maxilla, mandible, left incineration must be (12). Postclassic Mixtec lords are shown and right humeri, left tibia, metapodial, and a phalanx included Ϸ being cremated in the codices, so we know that after A.D. with the burial and in layers above the human remains. The 1000 this practice was regarded as appropriate for nobles and distal portion of the left tibia exhibited cutmarks. Virtually all royalty. Mortuary burning for cremation could occur with

surfaces of the dog remains were encrusted with the same ash ANTHROPOLOGY deceased lineage members who were deserving of veneration. concretions as the human remains and were not calcined or Likewise, the processing of ancestors’ remains to make carbonized (see supporting information (SI) Text). The ex- mummy bundles, violation of enemies’ mummy bundles, and ception was the left tibia fragment, which may account for why cannibalism of enemies all could involve burning the body of cutmarks were visible on it. The lack of redundancy among the deceased in different ways. Although well documented in skeletal elements and the consistency of age among the dog historical records (especially for Nahua peoples; 13), the time remains (all visible epiphyses were fused) indicate that only depth of differential burning and formal cremation among the one animal was represented. The associated dog and human ancient Mixtec was unknown (cf. 14, 15). In the Maya area, for example, it has long been documented that mortuary burning recorded in ethnohistorical sources (16) had roots stemming Author contributions: A.K.B. designed research; W.N.D., A.K.B., K.C., and N.J.M. performed back to the period between 300 B.C. and A.D. 300 (17). research; W.N.D., K.C., and H.A.L. analyzed data; and W.N.D., A.K.B., H.A.L., and N.J.M. The two burials from Tayata demonstrate that considerable wrote the paper. time depth for formal cremations is found in other areas of The authors declare no conﬂict of interest. Mesoamerica. In the best-documented example, the body was †To whom correspondence should be addressed. E-mail [email protected]. burned in situ, still articulated and undisturbed (Fig. 2). In the This article contains supporting information online at www.pnas.org/cgi/content/full/ second example, the remains were also burned in situ, but were 0710696105/DCSupplemental. disturbed after burial (Fig. 3). Here, we describe the archae- © 2008 by The National Academy of Sciences of the USA

www.pnas.org͞cgi͞doi͞10.1073͞pnas.0710696105 PNAS ͉ April 8, 2008 ͉ vol. 105 ͉ no. 14 ͉ 5315–5320 Downloaded by guest on September 30, 2021 A B

Fig. 1. Plates 81 (A) and 82 (B) from the Codex Nuttall showing Mixtec cremation (ﬁgure created by modifying 11).

remains suggest intentional burial as part of the same mortuary Valley of Oaxaca (26). Excavations of a midden north of the event. The cutmarks and skeletal elements represented from residence yielded a semicomplete skeleton of a single dog the dog might indicate feasting as part of the mortuary ritual, along with numerous isolated skeletal elements from at least but the alternative interpretation that the dog was a primary four other dogs. In addition to an unusually large number of offering included in the burial cannot be discounted or ruled dogs, the area in and around the residential structure associ- out. Burial 2 had only two dog canine teeth found in context. ated with burials 1 and 2 was also the only elite context to However, burial 2 was disturbed postmortem, which might contain fish remains in any frequency (just less than one fifth have destroyed other dog elements. of all identified taxa from the area). Analysis of the fish Food consumption patterns further mark this area of the site remains is ongoing. Several specimens appear to belong to the and its residents as being somewhat apart from others at the Perciforme order, which includes a variety of marine and site. The area in and around the residential structure associ- freshwater perch and perch-like species. Fish are extremely ated with burials 1 and 2 contained more than three times the rare (and generally absent) at other highland Mexican sites of number of animal remains compared with other elite areas at this period. Tayata. Domestic dog (Canis familiaris) and white-tailed deer We infer from these observations that the cremated indi- (Odocoileus virginianus) were the two most common mamma- viduals were members of a high-status household given their lian taxa represented, followed by cottontail rabbits (Sylvilagus food consumption patterns, their participation in craft pro- spp.), jackrabbits (Lepus spp.), pocket gophers (Orthogeomys duction and exchange, the presence of the dogs and other grave spp.), and opossums (Didelphidae family). This area contained goods with the cremations (further described below), the a surprisingly large quantity of dogs. Dog remains were five formal arrangement of the burials, and their occurrence in a times more prevalent here than in other areas of the site. long-term elite residential area. This same residential zone was Butchery scars in the form of cut and hack marks were eventually covered with a public building and small plaza in identified on eight dog bones in the faunal assemblage from which various features were dedicated with dog and human the 2004 excavations at Tayata and all eight of these specimens crania. Other forms of burial not involving cremation were came from the area around this residence. Similarities in found at the site, suggesting that formal cremation was re- butchery events (including scar type, length, and placement; served for only some members of the community. number and orientation of scars) between deer and dog bones The stratigraphic relationships of the house and cremation indicate dogs were also butchered for consumption, which is an burials place these features together in time, such that the two expected outcome based on previous studies of dog use in the burials might have constituted a single event. The individual in burial 1 (described in greater detail below) was in primary

Fig. 2. Burial 1, in situ. Note that the burial is complete, articulated, and placed face down. Fig. 3. Burial 2, in situ.

5316 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0710696105 Duncan et al. Downloaded by guest on September 30, 2021 Fig. 4. Map of the burials, middens, and household architecture at Tayata.

context and wore a shell-bead necklace; the burial matrix contained numerous fragmentary Early to Middle Formative figurines (e.g., 20, 27), burned dog parts, and debris from the production of marine shell ornaments. Burial 2 was only partially articulated with numerous missing skeletal elements and no formal offerings in the burial matrix (although several Fig. 5. A burned ﬁgurine of a seated individual from the midden associated loose shell beads were recovered). This burial appears to have with the cremations. been disturbed by plowing in recent years, but remaining in situ skeletal elements suggest that this individual might have been surface of the bones. Less than 10% of the external surface of seated in the grave—a common posture for authority figures the bones was covered with the ash (Fig. 6). Mild warping was in ancient Mesoamerica (well known from the Zapotec area in found on the left parietal. the Valley of Oaxaca; e.g., 27). Burned clay figurines from one The majority of the long bones and the clavicles had ash on of the household middens included the status-marking fea- their surfaces, were brown in color, or both. The left femur, tures of a seated individual with earspools, mask, and head- tibia, and fibula, left and right humeri, ulnae, and radii all dress (Fig. 5). A charcoal sample from the ashy matrix of burial exhibited deep, longitudinal cracking and splitting (as opposed 2 produced a carbon 14 [accelerator mass spectrometer to transverse splitting; Fig. 7). Some transverse cracking and (AMS)] date of 1000–890 B.C. (1 sigma calibration; NSF- stepped fractures were evident (as on the right fibula), but the Arizona AMS Laboratory, sample no. AA63366; calibrated by majority was longitudinal. The left fibula and ulna and right OxCal 3.9); another carbon 14 date from one of the adjacent humerus had curved longitudinal breaks. The carpals, tarsals, household middens was slightly earlier in time. metacarpals, metatarsals, and phalanges all had varying Osteological Analysis amounts of ash on them. The presence of warping on the skull, the shielding of the ash layer, and the longitudinal, irregular The remains were inventoried and assessed for age, sex, trauma,

curved fractures suggest that the burning occurred while the ANTHROPOLOGY pathology, and taphonomy as outlined in Buikstra and Ubelaker bone was green (see SI Text). Thus, in burial 1, the complete, (28). Detailed osteological data for burial 1 are presented in the articulated body was placed face down, burned in situ relatively SI Text. For Burial 1, the preservation of the os coxae precluded their use in sex assessment. The structures of the skull, although, indicate that the individual was most likely a female. Age assessment was based on a combination of third molar eruption, pubic symphyseal morphology, and cranial suture closure, all of which suggested a young adult between 18 and 25 years old at the time of death. The presence of ash, fracture patterns, and the color and shape of the bones in burial 1 all reflected extensive heat modification on virtually all of the remains and was consistent with an in situ cremation of a primary interment, relatively shortly after death. There were some unidentified calcined fragments in the assemblage, but the majority of the bones were not calcined or carbonized. On each of the bones in the Fig. 6. Internal (A) and external (B) views of the occipital bone from burial calvaria, a thick gray ash layer covered over 90% of the internal 1. Note that the internal side exhibits more ash. (Scale bars, 1 cm.)

Duncan et al. PNAS ͉ April 8, 2008 ͉ vol. 105 ͉ no. 14 ͉ 5317 Downloaded by guest on September 30, 2021 and (iii) processing to remove certain elements followed by burning the rest of the remains. The absence of cutmarks and the articulation of the feet and ankle bones are inconsistent with the last scenario. Also, the presence of perpendicular fractures, vis-a`-vis those in burial 1 (Fig. 7), suggests that the bones were broken postmortem, when dry. This could reflect the removal of certain skeletal elements in antiquity, but the presence of bone fragments in the plowzone above the primary interment suggests that the first scenario is the most parsimo- nious. The fracture patterns seen in burial 2 likely reflect postmortem damage. Both burials reflect primary cremations, but burial 2 was seated and then disturbed at a later point in time. Discussion Examples of burned human bone have been reported as early as 6,000 years ago from elsewhere in Mesoamerica (29), but most Archaic and Formative period examples appear to be the Fig. 7. Comparison of break patterns between long bones in burials 1 and products of cannibalism (30, 26). Unlike the Tayata burials, 2. (A) The right radius from burial 2. (B) The left ulna from burial 1. Note that these early examples reflect burning as part of complex the bone in A exhibits transverse splitting. Those in B exhibit longitudinal mortuary processing, rather than in situ cremation of a recently curved splitting. (Scale bars, 1 cm.) deceased corpse. Since the 1960s various authors (e.g., 31, 32), citing ethnohistorical sources, have suggested that Mesoameri- shortly after death, and buried. The remains were not dis- can cremation was practiced most widely in the Postclassic. Although previous excavations suggest that Middle Formative turbed from the primary interment. There is no evidence to cremations may have been present in Morelos, Mexico, Grove suggest that other behaviors known to have been practiced at (ref. 33, p. 64) notes that the evidence is ultimately ‘‘extremely the time and involving burning, such as cannibalism, produced tenuous.’’ More recent excavations have uncovered what ap- the deposit (see below). pears to be the earliest crematorium in Mesoamerica dating to Like burial 1, burial 2 is a primary deposit that reflects in situ the Classic period, although the remains from Loma Alta were burning, with extensive heat modification of the remains. The pulverized after being incinerated (34). Burials 1 and 2 from only articulated portions of the skeleton, though, were the ankle Tayata therefore appear to be the earliest in situ cremations so and foot bones. The majority of the remains consisted of radius far found of complete bodies in Mesoamerica. Finding two and ulna fragments, along with carpals, metacarpals, one patella, contemporaneous cremation burials from the same site sug- tarsals, metatarsals, and phalanges. Rib, vertebral, and flatbone gests that mortuary burning was well established and earlier in fragments were recovered, although they were sparse. No frag- date than once thought, at least among the Mixtec. ments from the skull, humerus, femur, tibia, or fibula were Discerning acts of veneration and violation remains a gen- recovered. There is no duplication of skeletal elements, and thus eral problem in mortuary studies (35). Because cremation can no evidence that there are multiple individuals represented in be used in multiple ways, it is worth discussing why these the grave. burials reflect the veneration given to persons of elevated The individual was between 15 and 25 years of age at the status rather than persons deemed unworthy—be they crimi- time of death. The distal radial and iliac crest epiphyses were nals, captured enemies, sacrificial victims, or others suitable partially fused, with a clear epiphyseal line with distinct depth for violation. In the Valley of Oaxaca, Marcus and Flannery list present in both cases. There were no remains present that fall 10 lines of evidence that they used to infer the emergence of outside of this age range. The sex of the individual was rank, arguing that no single datum alone would be sufficient unknown. The in situ images of burial 2 indicate that the pedal (ref. 7, p. 110). The evidence for rank society at San Jose´ phalanges exhibited evidence of calcination. White flecks of Mogote includes the deliberate cranial deformation of elite ash are also visible in matrix surrounding the bones. With the children; differential access to jade earspools and magnetite exception of the tarsals and metatarsals, all of the remains mirrors; differential access to deer meat, pearl oyster, and present had a coating of ash on Ͼ90% of their surfaces. Where Spondylus shell; a dichotomy between seated/kneeling (elite) there was no ash, the bones were a pale brown (Munsell burials and prone (lower-status) burials; figurines showing 10Yr74). The tarsals and metatarsals had Ͻ70% of their individuals in contrasting positions of authority and obeisance; surfaces covered in ash. Additionally, all of the long bones four-legged stools like those carried by the attendants of chiefs exhibited transverse/perpendicular splitting but not the irreg- in other New World rank societies. There is a similar range of ular longitudinal splitting found in burial 1 (Fig. 7A). information from Tayata suggesting that the cremated indi- Although reconstructing exactly what activities led to burial viduals were members of a high-status family, and that the site 2 is less straightforward than burial 1, there are reasons to itself was prominent within the Mixteca Alta. Tayata was the think burial 2 was a primary, seated burial that was disturbed likely source of the founding population for one of the region’s after interment. The foot and ankle bones were the only first urban states (18, 36). articulated portions of the skeleton, but the fact that the os Several further observations cause us to doubt that the coxae are visible immediately behind the foot and ankle bones Tayata cremations were intended as punishments or acts of (to the north in Fig. 3) indicates that the individual was likely vengeance. In burial 1, one could reasonably ask how we know seated. The articulation of the foot and ankle bones, along with that the hands and feet were not tied. It is a difficult propo- the color of the surrounding matrix, indicates that the remains sition to falsify, but the hands in burial 1 are in front of the were burned in situ. The three most likely scenarios that could individual. At the Feathered Serpent Pyramid at Teotihuacan, have produced such a deposit are (i) an unintentional post- 77% of the burials with known arm positions had their arms mortem disturbance of the primary grave, (ii) burning to crossed behind their backs, which was one of the reasons for deflesh the bones and then remove certain skeletal elements, concluding that they were human sacrifices (37). It is also

5318 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0710696105 Duncan et al. Downloaded by guest on September 30, 2021 worth asking why burial 1 reflects elevated status if a distinc- roots deep in prehistory, at a time when hereditary inequality tion between seated and prone burials constitutes evidence for was first emerging (41). This result, and the cultural continuity rank at San Jose´Mogote? Although seated positions reflect it suggests, has broader implications. First, the Tayata burials what is thought to be the highest honor in mortuary treatment demonstrate that Mixtec mortuary practices and emerging in the Valley of Oaxaca, there is evidence that individuals ideologies concerning social rank were far more elaborate at placed faced down in prone positions were still honored at that this time than most archaeologists assumed. We now know that site. The individual described in Fig. 5 by Marcus (38) is face Mixtec from Tayata marked social distinctions by cremating down, prone, and accompanied with multiple grave goods. A the complete bodies of the recently deceased, a practice not similar emergence of seated burials as a part of the transition known to have been shared by contemporary groups such as to institutionalized authority has been documented in the the Zapotec or Olmec. Second, having further evidence that a Maya area (39). tradition known from proto-historic times finds its origins over In burial 2 the ankle and foot bones were immediately in two millennia before bolsters Mesoamerican archaeologists’ front of the os coxae, indicating that the individual was not use of the direct historical approach which posits that specific sitting in a lotus position, as seen in the figurine in Fig. 5, but cultural practices were continuous through time (in general may have had been in a flexed position with the legs tucked terms, allowing that specific meaning or significance might close to the chest. Does the distinction between these two change; e.g., 42). The Tayata cremations thus make visible key positions indicate that the individual in burial 2 was not of an elements of the ancient Mixtec worldview concerning death elevated status? Again, the alternative scenario is difficult to and the ancestors thousands of years before Spanish contact. falsify completely, but in the Valley of Oaxaca, as elsewhere in Third, these cremations provide archaeologists with tangible Mesoamerica (40), as institutionalized authority was forming, alternative scenarios to cannibalism for understanding and seated or flexed positions reflected elevated status. Although interpreting burned human remains in the Formative period. iconography and figurines may more often show individuals in The Tayata cremations demonstrate that a detailed bioar- the lotus position, the conflation of the two positions likely chaeological examination, considered in the full archaeologi- reflects the bundling of higher status individuals as a part of cal context, can reveal the varied uses of burning in prehispanic the mortuary treatment. This is seen elsewhere in the Valley mortuary programs. of Oaxaca at San Jose´Mogote (38) and at Teotihuacan in the Feathered Serpent Pyramid in later times (37). Thus, although ACKNOWLEDGMENTS. We thank Joyce Marcus and the anonymous review- ers for their comments and assistance on this article, Justin Bedard, Ayala the two Tayata burials may not reflect equal status among the Amadio, and Chikaomi Takahashi for preparing the human and faunal deceased, there is reason to think that both reflect elevated remains for analysis, the municipal authorities and people of Santa Cruz status. and Santa Catarina Tayata for the opportunity to study their past, and the The discovery of human cremation Ϸ3,000 years ago in the ofﬁcials at the Instituto Nacional de Antropologı´ae Historia and the Centro Regional de Oaxaca for authorizing this work and providing institutional Mixteca Alta is important because it suggests that later support. This work supported by National Science Foundation Grant methods of marking social status in mortuary contexts have 0431390.

1. Palerm A, Wolf ER (1957) Ecological potential and cultural development in Me- 20. Drennan, RD (1976) Fa´brica San Jose´ and Middle Formative Society in the Valley soamerica. Pan Am Union Social Sci Monograph 3:1–37. of Oaxaca, Memoirs 8 (Museum of Anthropology, Univ of Michigan, Ann 2. Spores R (1969) Settlement, farming technology, and environment in the Nochixtlan Arbor, MI). Valley. Science 166:557–569. 21. Flannery KV, Marcus J (1994) Early Formative Pottery in the Valley of Oaxaca, 3. Balkansky AK (1998) Origin and collapse of complex societies in Oaxaca (Mexico): Mexico, Memoirs 27 (Museum of Anthropology, Univ of Michigan, Ann Arbor, MI). Evaluating the era from 1965 to the Present. J World Prehistory 12:451– 22. Spores R (1972) An Archaeological Settlement Survey of the Nochixtla´n Valley, Oaxaca, 493. Publications in Anthropology No. 1 (Vanderbilt Univ, Nashville, TN). 4. Blomster J (2004) Etlatongo: Social Complexity, Interaction and Village Life in the 23. Coe M, Diehl R (1980) In the Land of the Olmec (Univ of Texas Press, Austin, TX), Vol 1. Mixteca Alta, Oaxaca, Mexico (Wadsworth, Belmont, CA). 24. Flannery KV, Marcus J (2000) Formative Mexican chiefdoms and the myth of the 5. Flannery KV, Marcus J, eds (2003) The Cloud People: Divergent Evolution of the “Mother Culture.” J Anthropol Archaeol 19:1–37. Zapotec and Mixtec Civilizations (Percheron Press, Clinton Corners, NY). 25. Sharer RJ, Grove DC, eds (1989) Regional Perspectives on the Olmec (Cambridge Univ 6. Drennan RD, Peterson CE (2006) Patterned variation in prehistoric chiefdoms. Proc Natl Press, Cambridge, UK). Acad Sci USA 103:3960–3967. 26. Flannery KV, Marcus J (2005) Excavations at San Jose´ Mogote 1: The Household 7. Marcus J, Flannery KV (1996) Zapotec Civilization: How Urban Society Evolved in Archaeology, Memoirs 40 (Museum of Anthropology, Univ of Michigan, Ann Mexico’s Oaxaca Valley (Thames & Hudson, London). Arbor, MI). 8. Bellas M (1997) The body in the Mixtec codices: Birth, puriﬁcation, transformation, and 27. Marcus J (1998) Women’s Ritual in Formative Oaxaca: Figurine-Making, Divination, death. PhD dissertation (University of California, Riverside, CA). Death and the Ancestors, Memoirs 33 (Museum of Anthropology, Univ of Michigan, 9. Smith ME (1973) Picture Writing from Ancient Southern Mexico: Mixtec Place Signs Ann Arbor, MI). and Maps (Univ of Oklahoma Press, Norman, OK). 28. Buikstra J, Ubelaker D, eds (1994) Standards for Data Collection from Human Skeletal

10. Spores R (1984) The Mixtecs in Ancient and Colonial Times (Univ of Oklahoma Press, ANTHROPOLOGY Remains (Arkansas Archaeological Survey Research Series, Fayetteville, AR). Norman, OK). 29. MacNeish R (1962) Second Annual Report of the Tehuacań Archaeological- 11. Nuttall Z (1975) The Codex Nuttall (Dover, New York). Botanical Project (Robert S. Peabody Foundation for Archaeology, Phillips 12. Warren M, Schultz J (2002) Post-cremation taphonomy and artifact preservation. J Academy, Andover, MA). Forensic Sci 47:656–659. 30. Turner CG, Turner JA (1999) Man Corn: Cannibalism and Violence in the Prehistoric 13. Umberger E (1987) Events commemorated by date plaques at the Templo Mayor: American Southwest (Utah Univ Press, Salt Lake City, UT). Further thoughts on the solar metaphor. The Aztec Templo Mayor, ed Boone EH (Dumbarton Oaks Research Library and Collection, Washington DC), pp 411– 31. Ruz Lhuillier A (1968) Costumbres funerarias de los Antiguos Mayas. (Universidad 449. Nacional AutońomadeMe´ xico, Me´xico City, Me´xico). 14. Ritchie W (1969) The Archaeology of New York State (Natural History Press, Garden 32. Welsh W (1988) An Analysis of Classic Lowland Maya Burials, International Series 409 City, NY). (British Archaeological Reports, Oxford, UK). 15. Wright J (1966) The Ontario Iroquois Tradition, Anthropological Series 75 (National 33. Grove D (1970) The San Pablo pantheon mound: A Middle Preclassic site in Morelos, Museum of Canada Bulletin, 210, Ottawa). Mexico. Am Antiquity 35(1):62–73. 16. Tozzer A (1941) Landa’s Relacioń de las cosas de Yucatań (Peabody Museum of 34. Carot P, Susini A (1989) Una práctica funeraria insólita en el Occidente: La cremación American Archaeology and Ethnology, Cambridge, MA), paper no. 18. y pulverización de osamentas humanas. Trace 16:112–114. 17. Ricketson O, Ricketson EB (1937) Uaxactuń, Guatemala; Group E, 1926–1931 (Carnegie 35. Duncan W (2005) Understanding veneration and violation in the archaeological Institution of Washington, Washington, DC). record. Interacting with the Dead: Perspectives on Mortuary Archaeology for the New 18. Balkansky AK (1998) Urbanism and early state formation in the Huamelulpan Valley of Millennium, eds Rakita GFM, Buikstra JE, Beck LA, Williams SR (Univ of Florida Press, southern Mexico. Latin American Antiquity 9:37–67. Gainesville, FL) pp 207–227. 19. Balkansky AK, et al. (2000) Archaeological survey in the Mixteca Alta of Oaxaca, 36. Balkansky AK, Pe´rez Rodrı´guezV, Kowalewski SA (2004) Monte Negro and the Urban Mexico. J Field Archaeol 27:365–389. Revolution in Oaxaca, Mexico. Latin American Antiquity 15:33–60.

Duncan et al. PNAS ͉ April 8, 2008 ͉ vol. 105 ͉ no. 14 ͉ 5319 Downloaded by guest on September 30, 2021 37. Sugiyama S (1995) Mass Human Sacriﬁce and Symbolism of the Feathered Serpent 40. McAnany P (1998) Ancestors and the Classic Maya built environment. Function and Pyramid in Teotihuacan, Mexico. PhD dissertation (Department of Anthropology, Meaning in Classic Maya Architecture, ed Houston SD (Dumbarton Oaks Research Arizona State Univ, Tempe, AZ). Library and Collection, Washington, DC), pp 271–298. 38. Marcus J (1999) Men’s and women’s ritual in Formative Oaxaca. Social Patterns in 41. Marcus J, Flannery KV (2004) Proc Natl Acad Sci USA 101:18257–18261. Pre-Classic Mesoamerica, eds Grove DG, Joyce RA (Dumbarton Oaks Research Library 42. Spores R (2003) The origin and evolution of the Mixtec system of social statiﬁca- and Collection, Washington, DC), pp 67–96. tion. The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civiliza- 39. McAnany P, Storey R, Lockard AK (1999) Mortuary ritual and family politics at Forma- tions, eds Flannery KV, Marcus J (Percheron Press, Clinton Corners, NY), pp 227– tive and Early Classic K’axob, Belize. Ancient Mesoamerica 10:12–146. 238.

5320 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0710696105 Duncan et al. Downloaded by guest on September 30, 2021