Cannibalism in Chaco Canyon The charnel pit excavated in 1926 at Small House ruin by Frank H.Hкод для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 91:421-439 (1993) Cannibalism in Chaco Canyon: The Charnel Pit Excavated in 1926 at Small House Ruin by Frank H.H. Roberts, Jr. CHRISTY G . TURNER I1 Department of Anthropology, Arizona State Uniuersity, Tempe, Arizona 85287-2402 KEY WORDS Anasazi, Bioarcheology, Taphonomy, Violence ABSTRACT A charnel pit that contained the disarticulated and intentionally damaged remains of eight incomplete adult and subadult Anasazi skeletons was found and excavated in 1926 by F.H.H. Roberts, Jr., at an AD 900 ruin he named Small House, located in Chaco Canyon, northwestern New Mexico. Damage includes extensive perimortem cranial and postcranial bone breakage, cut marks, anvil-hammerstone abrasions, burning, many missing vertebrae, and fragment end-polishing. Together, these six types of perimortem damage are believed to be the taphonomic signature of prehistoric Anasazi cannibalism. The possible cause of the Small House episode is discussed within the framework of two explanatory models-random social pathology and institutionalized social control by violent means. o 1993 Wiley-Liss, Inc. on Polacca Wash, northeastern Arizona, by the late Museum of Northern Arizona archeologist,A.P. Olson (1966).The Polacca Wash analysis showed that some of the few whole and numerous fragmented bones had been burned, cut, and broken open like archeological animal bone food refuse. With the aid of carbon 14 dating of rib fragments, Hopi traditions, site location, episodic nature, population affinity assessment, comparisons with modern animal skeletons found on the ground surface in the region, and an absence of gravegoods and other signs of ceremonialism, it was proposed that the Polacca Wash human remains were the legendary massacred captives of the Hopi attack in AD 1700 on Awatovi, another Hopi village of several hundred inhabitants. The combined periCRITERIA FOR PROPOSING ANASAZI mortem damage and context of the Polacca CANNIBALISM Wash mass burial revealed three easily recThe first analysis of a Southwest Indian ognizable key taphonomic criteria (inten“burial” employing taphonomic, forensic, tional bone breakage, cutting, and burning) and other methods (quantitative, compara- for suggesting Southwest cannibalism. tive, chronometric, and ethnohistoric) was Since 1970, more than 40 Anasazi charnel conducted by Turner and Morris (1970). pits or scattered room deposits of human That study was based on the excavation of a bone with these and other taphonomic indiminimum (MNI) of 30 highly fragmented and disarticulated individuals deposited at Received July 9,1992; accepted March 10,1993 an isolated locality south of the Hopi villages This article describes and discusses in a regional context an unreported multiple human “burial” excavated in 1926 by Frank H.H. Roberts, Jr. The burial or charnel pit is another in a growing number of prehistoric southwestern U.S. episodic human bone deposits with massive perimortem damage that suggest cannibalism as determined by taphonomic and forensic analysis (Table 1). Before proceeding, some background on Southwest human taphonomy studies will help place this report in broader perspective since most previous studies on Anasazi cannibalism and violence have been published in regional or archeological journals, and may not be readily available to many readers of AJPA (Except, see White, 1992). 0 1993 WILEY-LISS, INC. 422 C.G. TURNER TABLE 1 . Published taphonomic reports on Anasazi episodic burial sites and bone deposits evidencine Derirnortern violence and cannibalism Site and reference Bone condition Breaks Cuts Damage percentage Abrasion Burn ’ Polish MNI Gnaw Pieces no.’ ~~ New Mexico Chaco Small House (1) Fence Lake (2) Burnt Mesa (3) Sambrito Village (4) Arizona Polacca Wash (5) Monument Valley (6) Ash Creek (4) Teec Nos Pos (7) Tragedy House (8J3 Lerow Wash (4) Canyon Butte (9) Colorado Mancos Canyon Mancos Canyon (111 Grinnell (12) Marshview Hamlet (13) Yellow Jacket (14) Utah Rattlesnake (15) 42SA12209 (16) Summary averages Good Good Good Good 58.5 99.9 98.0 99.8 7.9 0.2 2.8 4.6 6.6 Yes Yes Yes 3.9 0.5 35.4 16.7 10.5 Yes Yes ? 0.0 ? 1.5 0.0 11 (5) 152 1088 3389 474 Yes ? ? ? ? Yes 41.1 Yes ? 5.2 0.0 0.0 3.0 0.4 30 7 5 2 4 (33) 4 437 644 212 56 62 3443 248 33 29 7 6 14 1598 2106 380 528 2043 20 5015 691 1233 Good ? Poor Poor Good Good Good 80.3 80.1 97.6 5.4 11.3 99.7 87.5 Yes 0.9 3.3 10.7 1.6 2.6 4.0 Yes ? ? 0.0 1.6 Yes 2.4 Yes 12.6 25.9 1.8 1.6 1.7 4.4 Good Good Good Good Good 98.4 ? ? 18.5 ? Yes Yes Yes 21.5 10.8 30.7 8.9 - 97.9 99.9 Yes Yes 11.7 1.0 2.6 3.7 6.0 ? ? ? 0.0 0.5 7.0 Yes Good Poor Good 92.3 Yes 80.4 3.7 Yes 4.1 Yes Yes 5.8 3.0 Yes 12.0 ? ? 0.2 No 1.5 5 19.2 8 5 4 11.4 ‘References: (1) This report; (2) Grant, 1989; (3) Flinn et al., 1976; (4) Turner, 1983; Fay and Klein, 1988; (5)Turner and Morris, 1970; (6)Nass and Bellatoni, 1982; (7) Turner, 1989; ( 8 ) Turner and Turner, 1 9 9 0 (9) Turner and Turner, 1992a; (10) Nickens, 1975; (11) White, 1992; (12) Leubben and Nickens, 1982; (13)Turner, 1988; (14) Malville, 1989; (15)Baker, 1990;116)White, 1988,1991. Map locations of these sites can be found in references 9 and 11. ‘Bone number refers to whole bones and fragmentary pieces. 3Values based on only two individuals because of postexcavation bone loss. ‘Nickens’ [ref. 10) counts for MNI and total pieces were not used in the summary averages, but his breakage percent was included for average breakage. ’Not comparable. White (1992) considered a bone to be whole if a t least half of the bone was intact, nevertheless, breakage was extensive. cations of violence and cannibalism have been reported (Tables 1, 21, or are under study by several workers (listings that include published reports, work in progress and other cases can be found in Turner and Turner, 1992a,in preparation; Turner et al., in press; White, 1992). In 1983, a fourth taphonomic criterion was identified and defined for proposing cannibalism, namely anvil abrasions (Turner, 1983). The term perimortem was also coined at this time to define bone damage that had occurred at or around the time of death. A fifth criterion, many missing vertebrae, has recently been added to the proposed taphonomic signature of cannibalism because one vertebral fragment with anvil abrasions and cut marks was discovered in an episodic multiple burial from yet another site (Canyon Butte, Arizona). This discovery finally explained why so many vertebrae were missing or highly fragmented in all Anasazi episodic multiple burials with ap- parent cannibalism (Turner and Turner, 1992a; White, 1992). Seemingly, in all these sites, the vertebrae had been processed for oil, as had the long bones been cracked open to extract marrow. Burning, breakage, and anvil abrasions are illustrated in Figures 1-8. No cut or abraided vertebrae were found in the Chaco Small House assemblage, but examples can be seen in Turner and Turner (1992a) and White (1992) as can stone tool cut marks that are usually V-shaped in cross section with internal striations. A sixth damage type was identified by White (1992) based on his detailed reexamination of the seemingly cannibalized human remains excavated in Mancos Canyon, Colorado, first studied by Nickens 11975). This damage of the ends of long bone fragments White calls “pot polishing.” White showed by experiment that fragment tip polishing can result from stirring bone fragments in a pottery cooking vessel, and that identical pol- 423 CANNIBALISM IN CHACO CANYON TABLE 2. Demography of published Anasazi episodic burials Site Chaco Small House Fence Lake Burnt Mesa Sambrito Village Polacca Wash Monument Valley Ash Creek Teec Nos Pos Tragedy House Leroux Wash Canyon Butte Mancos Canyon Grinnell Marshview Hamlet Yellow Jacket Rattlesnake 42SA12209 Totals Percent Male 2 4 1 2 1 2 Adult' Female Sex? 2 2 2 1 (under reinvestigation) 2 15 1 4 Subadult Child 4 1 1 1 4 3 1 Aee? 1 3 8 1 1 1 1 (under reinvestigation) 1 2 1 3 2 1 22 14.1 1 1 5 1 1 1 15 1 1 1 4 3 3 4 1 15 9.6 47 30.1 26 16.7 1 1 9 2 2 7 7 2 43 27.6 3 1.9 Total 8 5 11 (512 30 7 5 2 4 (33) 4 29 7 6 14 20 4 156 'Analysts differed in their amount of experience and definitions of age classes. The values in this table should be viewed relatively. 'Values in parentheses not used for total number. ishing occurs on Anasazi animal bone food refuse. He also describes in great detail where the Mancos Canyon body parts were burned, and in so doing develops a compelling case for roasting as well as boiling. In addition, he compares the Mancos human remains with Anasazi animal food refuse, showing substantial similarities in the occurrence, quantity, and location of the damage criteria. Other studies on butchering and processing damage in Anasazi human and faunal remains include Dice (1993), Turner and Dytrych (in preparation), and Turner and Turner (1990). All six minimal perimortem damage criteria for proposing Anasazi cannibalism are abundantly and well illustrated in the reports cited in Table 1,especially the book-length study by White (1992). THE CHACO SMALL HOUSE CHARNEL PIT Briefly, the Chaco Small House site was named and excavated in 1926 by F.H.H. Roberts, Jr., but he never reported on this asDect of his Southwest fieldwork. which in&ded many more important archeological sites. ~h~ ten-or-so room small H~~~~is located in Chaco Canyon, northwestern New Mexico, and seemingly dates around AD Fig. 1. Adult and subadult cranial fragments with extensive perimortem breakage, especially facial. Burning occurred on right mastoid an d frontal areas in upper left adult, both being areas with minimal protective soft tissue (CGTneg. 5-20-92:l). 424 C.G. TURNER Fig. 2. Subadult temporal bone with burned left mastoid region (CGT neg. 5-20-92:14). 900. Beneath the floor of one room Roberts’ Navajo workmen discovered a pit filled with broken and burned bones (see Appendix A). Roberts’ field notes have measurements and other minor details about the charnel pit, including the observation that it contained a jumbled, broken, and partly burned series of human bone fragments. Because he was a very professional archeologist, had there been any gravegoods, or signs of ritual or ceremonialism, or had he found more damaged human bones elsewhere in the small pueblo, he would have noted such important details. In a 1957 personal letter to F. McNitt (1966, p. 3381, Roberts speculated without explanation that the Small House damaged human remains might have resulted from ceremonial cannibalism or human sacrifice. METHODS AND MATERIALS Sorting of the mixed remains was done by skeletal element, side, and age, leaving a residue of small unidentifiable long bone fragments. The series also contained one deer-sized nonhuman proximal femur fragment, with perimortem breakage. The human bones were either immature, highly fragmented, or gracile, so sex identification was limited t o only two questionable female elements (Table 3). The series had been carefully cleaned, so the surface of every piece of bone could be easily examined with a 20x hand lens and a 75 watt reflector flood lamp. Two femurs had been largely reassembled (“conjoined” to use White’s [1991, 19921 term), with the very strong dark dense glue frequently encountered in older Smithsonian specimens, sometime between accessioning and my examination. The bone condition was generally ivory-like in hardness, creamy white in color, showed very little root damage, and had no obvious sun bleaching or ground surface weathering. A few fragments were porous, crumbly, and fragile, but showed no sign of thermal alteration. None of the perimortem damage shows signs of healing o r infection. There CANNIBALISM IN CHACO CANYON 425 Fig. 3. Hammerstone impact fracturing with conchoidal depressed bone chips of anterior surface, proximal right subadult femur fragment (CGT neg. 5-20-92:18). purple staining of the interior surface of one frontal bone, presumably due to postmortem soil or microbial conditions. There are many skeletal elements missing from these eight individuals-most of the cranial elements; all of the occipitals, parietals, clavicles, and hand bones. Assuming lnventory eight individuals, only three of the expected Table 3 provides the skeletal element in- 192 vertebrae are represented. Because the ventory totaling 152 whole bones or frag- charnel pit assemblage contained some very ments. On the basis of the number of mandi- small bone fragments less than 2 cm in bles, eight individuals (MNI) can be length, it is likely that the butchering, identified, three more than Roberts (Appen- breaking, and burning of the bodies took dix A) and the museum catalog card indi- place nearby, meaning that the larger and cated. All the mandible fragments are num- missing vertebrae were destroyed or not bered 334,055, so our differences are not due buried, which will be discussed later. This overall representation, mainly skull parts, to postaccession mixing. Aside from periodontal disease and dental some shoulder and pelvic pieces, and many caries, including decayed deciduous molar long bone fragments, is a taphonomic charocclusal surfaces, the only pathology noted acteristic of most other Anasazi multiple was moderately severe bilaterally symmet- burials where cannibalism has been hypothrical cribra orbitalia in the orbits of one of esized (see reports cited in Table 1). Because the subadult frontal bones. There is light of the extreme fragmentation of the eight were 56 loose teeth that could not be matched up with the empty maxillary and mandibular tooth sockets, suggesting some alveolar bone loss before, during, or after excavation. The dental morphology fits the Sinodont American Indian pattern. C.G. TURNER 426 Fig. 4. Adult female (?) and subadult femurs plus two other long bone fragments that had been reassembled before this study. Fragment in upper left had burned, but fits with two adjacent unburned fragments, thus breaking had occurred before burning (CGT neg. no. 5-20-925). people, age and sex identification was difficult to determine, sex especially. The MNI of eight individuals and their ages were based on the number of mandibles and dental development. Perimortem damage Cut marks Perimortem damage that involved direct human acts includes cut marks on one frontal, one temporal, one skull fragment, three mandibles, two humeri, two femurs, and two unidentifiable long bone fragments. Thus, 7.9% (12/152) of the whole bone and fragment inventory has cut marks. The number of identifiable cut marks is comparable with the range and mean of cut marks in the assemblages listed in Table 1. Breakage Slightly more than half (58.5%;89/152) of the Small House inventory evidences peri- mortem breakage (Figs. la), which had the effect of exposing the brain and long bone marrow cavities, about 20% less than most of the other such sites. Perimortem breakage is readily differentiated from dry bone breakage by spiral fracturing, smooth unstepped fracture surfaces, and embedded bone chips. Abrasions The presence of perimortem anvil or hammerstone abrasions (Fig. 6) on 6.6%(10/152) of the Small House bones and fragments, helps visualize how the bones were broken, as well as demonstrates that the overlying skin, fat, and muscle had been removed prior to the shaft and head fragmentation. Had soft tissue been in place, the accidental abrasions of the impacted bone slipping off an anvil stone, or rotating slightly when hit with a hammerstone, could not have occurred. Abrasion frequency is similar to that reported in other episodic burials (Table 1). CANNIBALISM IN CHACO CANYON 427 Fig. 5. Representative long bone fragments showing perimortem breakage similar to animal bone food refuse. At least one piece had burned (upper center). The other dark fragments seem to have been discolored by soil conditions (CGT neg. no. 5-20-92:6). Burning As with most other Anasazi charnel pits or floor deposits where cannibalism has been proposed, the Small House assemblage has some burned bone (Fig. 21, here 3.9% (6/152) of the inventory, at the lower end of the range in Table 1. As elsewhere, some of the burning was unintentional-a few long bone fragments falling into a fire (Fig. 5). A few of the burnt or scorched pieces can be fitted to unburned fragments (Fig. 4). However, here as well as in other such skeletal series, there is some evidence of possible head and other body part roasting (Figs. 1, 2 ) . This possibility usually presents itself when there is scorching or charcoaling on external anatomical locations where little soft tissue overlies a given bone or region, for example, the upper vault, elbows, and lower mandibular borders. The Chaco Small House series has three such external burned spots from possible roasting-two on the mastoid processes of an adult and subadult, and one on an adult frontal bone. There are at least two “accidentally”burned long bone fragments. The small amount of burning or scorching is not indicative of a multiple cremation or accidental fire victims. Missing vertebrae There are only three vertebrae, each undamaged, out of an expected 192, or 1.6%. Given that there are many missing elements in the Small House assemblage, this low amount would not be meaningful were it not for the fact that missing and damaged vertebrae characterize all of the Anasazi assemblages suggesting cannibalism. Polishing Pot-polishing is present on 10.5%(16/152) of the assemblage, mostly on the end tips and spurs of long bone fragments. Polishing is difficult to identify because it usually occurs on a very small surface area, usually less than 0.25 mm2. During my first exami- C.G. TURNER 428 Fig. 6. Anvil abrasions on long bone breakage plane (CGT neg. no. 5-20-92:8) nation of the Small House series I missed finding any polishing because I was using good but diffused ceiling lighting. My second examination employed a strong reflector flood lamp mounted close to my field of 20x hand lens inspection, and 16 end-polished fragments were recognized. There was no polishing on the sides of these fragments. Natural damage Because there are no sure signs of sun bleaching or ground surface weathering cracks, and a complete absence of scavenger damage, that is, carnivore tooth puncture marks, gnawed ends of long bones, and the distinctive gnawing incisions of porcupines, wood rats, and other rodent teeth, it is likely that the remains were not left exposed on the ground surface for any extended period of time before deposition. This chronistic information suggests that the trauma, butchering, and subsequent burial were part of an episodic event that took place over a relatively short period of time, perhaps a few days or weeks at most. The absence of wood rat gnawing is noteworthy in light of Pepper’s (1920) repeated remarks about the several formal and considerate (mindful of the deceased) burials he excavated in Chaco Canyon that had been severely disturbed by “rats.” Violence Violence can be suggested on taphonomic grounds for the Small House series. One of the subadults (15-18 years old) has the maxillary anterior teeth smashed, with root sockets “blown” out, and fragments of root tips remaining in a few sockets (Figs. 7, 8). Another subadult had received a heavy blow to the head causing the first molar to be severely fractured. While this facial and dental damage could have resulted from breaking open the cranial vault to gain access to the brain, as White (1992) has suggested, the absence of anvil or hammerstone abrasions in the anterior tooth regions in all individuals is more suggestive of violence or mutilation. Elsewhere, for example, at Largo-Gallina site Bg 20 in northwestern CANNIBALISM IN CHACO CANYON 429 Fig. 7. Blown out maxillary anterior tooth sockets with retained fractured root tips. There are no anvil or hammerstone abrasions present here or in Figure 8 (CGT neg. 5-20-92:20). New Mexico, perimortem facial damage with blown out anterior teeth occurs in noncannibalized remains of mainly men who had met violent deaths as evidenced by embedded projectile points, head wounds, and their bodies heaped on the floors of abandoned or burned rooms (Turner et al., in press). But, as in all other seemingly cannibalized Anasazi series, the Small House remains have no projectile points in any bone or bone fragment, nor are there any traces of penetration wounds from pointed bone or stone weapons, facts that correlate well with the extensive body processing. Most head and body wounds cannot be identified in these cannibalized series because the heads and bodies are so extensively broken up. Whole or broken projectile or spear points would have been recoverable following defleshing and bone breaking. DISCUSSION The major perimortem damage of the Small House series resulted from human ac- tivity, not animal, geologic, or weathering agencies. The kinds and amount of damage correspond well with other Anasazi sites where cannibalism has been proposed (Table 1). The characteristics of the perimortem damage are butchering; bone breakage to expose brain, marrow, and oil cavities; and consequences attributable to roasting and boiling. The subfloor human bone deposit found by Roberts at Small House is yet another example of traumatized, multiple, and seemingly cannibalized individuals in the Anasazi culture area. All six of the minimal taphonomic criteria are present for hypothesizing cannibalism. As best as can be determined from Roberts’ excavation notes (Appendix A), it would appear that the burial was also episodic in duration. He related that the remains were deposited beneath the floor of room 4 and seemingly near a wall. The vertical extent of the bone mass may have been about 2 feet. Thus, the remains did not accumulate over months or 430 C.G. TURNER Fig. 8. Blown out alveolar border of adult mandibular right canine. The first molar was diseased, not perimortem damaged (CGT neg. 5-20-92:31). years of room filling, nor were they scattered horizontally. Roberts’neatly measured 1926 sketch map of the irregularly-shaped pueblo shows room 4 located in the center of about ten rooms. It is unknown whether the charnel pit was intrusive, dug before, or related to the construction date of room 4. It is possible that not all of the bone was recoverable or saved by Roberts’ Navajo laborers. Roberts noted that many of the bones showed signs of fire, yet less than 5% could be identified positively as having been burned. The relatively large size of most of the bone fragments suggests that the bone deposit was not screened, which may explain our difference. And, what is to be made of his puzzling remark that all of the skulls showed ‘‘great malformation”? There are only two nearly complete frontal bones, and neither of these are deformed in any mechanical or pathological way. There are no occipital bones, which among Pueblo period Anasazi are often deformed from cradling practices, so it is impossible to understand what it was that led him to record this unusual observation. Perhaps Roberts meant breakage instead of malformation. Southwest mortuary practices Taken as a whole, the Small House remains are certainly not a normal considerate and formal burial. Considerate burials of multiple individuals are rare in the Southwest (Stanislawski, 1963). The remains do not show any indications of ceremonialism, i.e., Roberts did not note red ochre, other mineral ministration, or any gravegoods, and the fragmentary and incomplete remains were dumped in a pit beneath the floor of an isolated common house instead of in a public, religious, or ceremonial structure. The assemblage is not a secondary burial, which are also very rare in the Southwest (Carle, 1941; Hagberg, 19391,because a few small burned fragments can be fitted to unburned pieces. The small amount CANNIBALISM IN CHACO CANYON of burned bone is not consistent with cremation, which was rarely done in the Anasazi area, although it was a common mortuary practice among the desert Hohokam. The individuals were not trapped in a burning room, to be later broken up and buried in a pit, because the amount of burning is limited and its location has no relation to the body areas that would be exposed in a conflagration (Turner et al., 1991). Small House itself was not burned. Scavengers were not responsible for the damage, and the generally good condition of the bone speaks for an episodic or short duration causality where flesh had been removed. The removal of the soft tissue and apparent boiling of the fragmented bone limited later bacterial activity that contributes to postmortem bone decay in many considerate inhumations. The most parsimonious explanation for the overall Small House bone condition, by elimination, is cannibalism. This conclusion is reviewed in more detail elsewhere for other Anasazi charnel pits and floor deposits (Turner and Turner, 199213). Anasazi episodic multiple burials are functionally different from all other Southwest mortuary acts and behavior. Most of the latter are typically considerate formal burials, usually singular, and often with gravegoods. Such burials occur in patterned body orientations and predictable cemetery or interment locations like village plazas (Salad0 culture), abandoned storage pits (Basketmakers), trash mounds (Anasazi), cremation areas (Hohokam), beneath room floors (children of all culture areas), and combinations of these and other mortuary considerations (McGregor, 1965). The episodic burials differ also from instances of Southwest trophy burials, isolated heads or other body parts (Wilcox and Haas, 19891, rock art representations of trophy heads (Cole, 19891, probable human sacrifice (at Casas Grandes, Mexico, Di Peso, 1974; Ravesloot, 1988), people trapped or killed in burned rooms (Turner et al., 1991), and all other known Southwest ethnographic and archeological ceremonial activity involving the dead. Unlike areas outside of the Southwest, for example, the Iroquois region (Ambler and Logan, 19881, human bone was rarely used 431 by the Anasazi for artifact production or religious ceremonies. Moreover, there is no taphonomic evidence suggesting that the episodic burials are residues of bone tool production or ceremonial preparations. Anasazi episodic multiple burials or bone deposits are functionally distinguishable from all other burial situations on taphonomic and contextual grounds-absence of ceremonial or any other gravegoods; variable burial locations in kivas, rooms, cemeteries, outside rooms, and away from habitations altogether; and the site locations where the episodic burials have been found so far are usually small, defenseless, and isolated. To date, the episodic multiple burials have two depositional qualities-unburied bone masses left on floors or mixed with room or kiva floor fill, and collections of bones and fragments buried in pits by unknown persons. Only in the latter condition, are they in any sense burials. Two proximate models for Anasazi cannibalism The question of what caused the Anasazi perimortem damage seems to be answered as far as circumstantial evidence permits. However, there are larger and more theoretical problems that taphonomic analyses have not fully resolved, namely why was there apparent cannibalism, and why was there so much of it? In 1983 I hypothesized social pathology (taken from social psychology theory) as a proximate explanation for these episodic human bone deposits with perimortem damage suggesting cannibalism and violence. Extant explanations at that time included emergency, ritual, revenge, gustatory, endo-, and exocannibalism. It was also proposed that the apparent violence and cannibalism were chaotic precursor events related to the late prehistoric abandonment of much of the Southwest (Turner and Turner, 1990, 1992a), although no sites suggesting cannibalism have so far been reported for the desert Hohokam or mountain Mogollon culture areas. This general absence among the Hohokam and Mogollon may simply be due to the relatively little archeological activity in these areas compared with the great amount that has been conducted in the 432 C.G. TURNER TABLE 3. Inventory of Chaco Small House skeletal elements and anthropogenic perimortem damage Skeletal element and side Skull Frontal Frontal Temporal, L Temporal, L Temporal, R Maxilla, L+R Maxilla, L Maxilla, L Skull fragment Mandible, L+R Mandible, L Mandible, R Mandible, L+R Mandible, L Mandible, L+R Mandible, R Mandible, L Vertebrae Cervical 1 Cervical 2 Lumbar (no. ?) Rib No. ? Pelvis Ilium, L Ischium, R Scapula Right Right Humerus Right distal Left whole Left whole Right distal Left proximal Proximal Proximal Ulna Left Left & right Radius Distal Midshaft Fragments Femur Right (reassembled) (1) Right (reassembled) (1) Patella Left Tibia Right, proximal Right, distal Proximal Distal Midshaft Fibula Fragments Fragments Long bone fragments Type uncertain Foot All types All tmes Loose iieth Number 2 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Age and sex SA A SA A SA 15-18 12-15 15-18 ? 3-4 12 12 12-15 12-15 A A ? Breakage 2 1 0 0 0 1 1 1 0 1 1 0 Pieces with perimortem damage Abrasions ~ Burning _ _ Cutting _1 0 0 1 0 0 0 0 1 0 0 0 I 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 1 1 0 1 0 0 1 1 0 0 1 0 Polish _ 0 1 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 SA A SA 0 0 0 0 1 SA 1 0 0 0 1 1 SA SA 0 0 0 0 0 0 0 0 0 SA A 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 1 1 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 SA A (F?) A A A A SA 0 0 0 0 2 2 A SA 0 0 0 0 0 0 0 0 1 1 A SA SA 1 0 0 6 0 0 0 0 0 0 0 0 1 A (F?) SA 8 5 2 5 0 0 0 1 0 ? ? 1 A 0 0 0 0 0 1 1 4 A 1 1 4 1 2 0 0 A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 4 0 0 0 0 0 0 0 0 2 7 1 14 0 0 0 0 0 0 0 0 0 ? 0 0 1 1 1 1 1 1 1 1 1 1 6 8 2 SA SA A 2 4 SA A 1 37 A 37 18 A 22 SA 0 0 56 A & SA ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (Continued 1 433 CANNIBALISM IN CHACO CANYON TABLE 3. Inventory of Chaco Small House skeletal elements and anthropogenic perimortem damage (Continued) Skeletal element and side Totals (whole bones and fragments, excluding teeth) Number 152l Age and sex Pieces with perimortem damage Breakage 2 A' 89 Cutting 12 Abrasions 10 Burning Polish 6 16 3.9 10.5 4 SA 1c 1 age ? 8 Percentapes 58.5 7.9 6.6 The reassembled femurs and two other long bones in Figure 4 were counted a s fragments. 2The number of individuals is based on mandibles, and their ages estimated by dental development. A, adult (>18years); SA, subadult (>12 years); C, child (<12years). plateau Anasazi area during the last 100 years (Cordell, 1984). On the other hand, the differences may have only Anasazi historical significance that will be discussed in a moment. Thus, today, as Tables 1 and 2 show, the literature on Southwest episodic charnel pits or floor deposits of perimortemdamaged human remains indicates an average of 11.4 individuals per episode, a range of 2-33 persons per event, and a minimum total of 194 individuals of all ages and both sexes that seem to have been cannibalized. A regional pattern of Anasazi archeological sites with perimortem-damaged skeletal remains is now well established (Table 1;for map locations see Turner and Turner, 1992a; White, 1992).This pattern of damage resulted from the processing of Anasazi victims that parallels the way large game animals were butchered and prepared for consumption (Dice, 1993; Turner and Turner, 1990; White, 1992). However, the sequence of processing may have varied. White (1992) feels that burning (roasting) may have preceded some butchering a t Mancos Canyon, whereas in some series the sequence looks to have been butchering, breaking, and then burning (Turner, 1983,1992). Flinn et al. (1976) reviewed the cannibalism literature in their study of the Burnt Mesa series, and opted for emergency cannibalism, as did Nickens (1975) for Mancos Canyon. Turner (1983) and Turner and Turner (1990) could not find any direct or indirect basis for suggesting starvation in most sites, so the new proximate social pathology model was proposed, in part, to avoid the alternative possibility, namely in- stitutionalized violence and cannibalism, and to avoid specific and ultimate explanations such as witchcraft (Merbs, 1989) or postmortem mutilation (Bullock, 1991) that could not be demonstrated by the taphonomic or contextual evidence. The possibility of institutionalization now has to be explored. Too many of these patterned episodic Anasazi multiple burials have turned up for them to be only random socially pathological events. It is also very doubtful if all of these multiple burials simply represent emergency or starvation cannibalism because of the large numbers of individuals involved in sites like Leroux Wash (33 MNI), Mancos Canyon (29 MNI), and Rattlesnake (20 MNI). Moreover, if emergency cannibalism was the sole causality, one or two such sites like the infamous Donner and Packer examples would be expected for the mountain Mogollon area where winter snowfalls are often deep and long-lasting. What is most striking is that several of the Anasazi sites with apparent cannibalism seem to date at about the same time period as the rapid rise of the complex and far-flung Chaco nucleus and nodal system of Great Houses and roads, although where dating is relatively secure a few series are earlier (42SA12209) or later (Polacca Wash). As far as I know, there is no other region in the world where there is both a wealth of archeological information for the rapid development or introduction of a new sociopolitical system associated with as many reported mass burials evidencing violence and possible cannibalism as there is in the Ana- 434 C . G . TURNER sazi area. With respect to the magnitude of the episodic burials, four more sites have been discovered and excavated within the last 4 years, three in northwest New Mexico and one in southwest Colorado. J.A. Turner and I have just completed examining other unstudied episodic multiple burials excavated before 1980, bringing the total to about 40 Anasazi sites with multiple human burials or bone deposits exhibiting varying degrees of probable cannibalism and violence (Turner and Turner, in preparation; Turner et al., in press). At least 473 individuals were involved, many evidencing cannibalism. Patently, this number of sites and total MNI is not just an archeological issue. Instead, it is one that requires the cooperation of archeologists, skeletal biologists, and bioarcheologists in order to bring diverse expertise to a complex problem hinting at systems failures and growing chaos in the post-AD 900 Anasazi culture and population. The alternative model, institutionalized violence and cannibalism, can be entertained if it is assumed that the Anasazi had evolved, borrowed, or had imposed upon them from Mesoamerica, complex social organization with ranked individuals and specialized organizations, including quasi-military war societies. Such seems to be the case for the Chaco Great House phenomena, evidenced by the monumental architecture and exotic material culture found in the large Chaco Canyon communities, the apparent social, religious, and political center of a network of some 75 prehistoric “towns”in New Mexico, Colorado, and Arizona linked together with more than 650 km of approximately 10 m wide straight roads (Cordell, 1984; Judge, 1979; Wilcox, submitted). In Chaco Canyon, there was planned and directed monumental architecture involving hundreds of rooms and great kivas. There were high status burials in Pueblo Bonito, the largest of Chaco Canyon’s great houses, along with rare or exotic items in various of Chaco’s buildings such as imported Mexican macaws and copper bells, unique Mexicanlike pottery incense burners and human effigy vessels, large amounts of locally mined and processed turquoise, and other elements expected of a society with substantial centralized authority (Cordell, 1979, 1984; Frisbie, 1978; Lister, 1978; Pepper, 1909, 1920; Reyman, 1978). Given the limited soil, water, and timber in Chaco Canyon, the relatively large number of inhabitants, as estimated by room numbers and other considerations, must have required nutritional and diverse material support from outside the canyon. All Chaco prehistory experts agree on this. Estimates vary, but Hayes et al. (1981) allow that well over 5,000 people occupied Chaco Canyon in the 12th century. These experts also agree that the Chacoan towns were well organized, and their inhabitants had farflung influences and contacts that in some way or other even reached into northern Mexico. Building on a huge body of Anasazi archeological literature, Wilcox (1992, submitted) has proposed that the Great House phenomena, within and outside of Chaco Canyon, represented a regional sociopolitical system or state involving trade, tribute, and military activity among some 24,000 people. In order to maintain the necessary “voluntary” support, Wilcox believes that the Chacoans would have needed an enforcing mechanism, which could well have included along with religious and trade mechanisms, a quasi-military group(s). This is easy to envision in light of the large powerful warrior societies that once existed in most eastern and western Puebloan towns (Beaglehole and Beaglehole, 1935; Dozier, 1954; Eggan, 1950; Ellis, 1951; Fewkes, 1902; Hawley, 1937; Parsons, 1939; Woodbury, 1958). Under orders from their religious and war chiefs, Anasazi warriors might have committed example-setting acts of extraordinary violence against tribute- or politically resisting individuals or entire communities, as has happened throughout the world and up to the present day. Historic Pueblo Indians were skilled and courageous warriors, and insofar as ethnographic analogy permits, so must have been their ancestors a few centuries earlier. There is no doubt that there was warfare and raiding among the Anasazi (Haas, 1990; Mackey and Green, 1979; Upham and Reed, 1989; Wilcox and Haas, 1989; Woodbury, 1958). Accounts of historic Puebloan vio- CANNIBALISM IN CHACO CANYON lence and possible cannibalism begin with the very first foreign contact. Esteban, a North African who was the first foreigner in the Southwest, was on his arrival at the Zuni villages ordered killed by Zuni leaders and “cut into a great many pieces, which were distributed among all the chiefs” (Bandelier, 1892, p. 14). In 1680, Puebloan leaders and their followers in many towns secretly planned and successfully brought off large-scale simultaneous attacks against Spanish colonists throughout New Mexico and Arizona, killing more than 400, and driving away for several years all who survived the revolt (Sando, 1979). In AD 1700, an attack planned by Hopi chiefs was carried out with the intent of destroying another Hopi town, Awatovi, and all its inhabitants. Awatovi may have had at the time 800 or more people (Fewkes, 1912). Historic accounts, Hopi legends, and taphonomic and archeological findings indicate that most of the Awatovians were killed, and some were probably cannibalized (Turner and Morris, 1970). The reasons for these Pueblo warrior actions against Esteban, the Spanish, and the Awatovians vary, but the point is that historic Puebloan communities had large numbers of capable warriors who executed the commands of their leaders and quasimilitary war chiefs. Hence, there is good reason to expect the same of Anasazi leaders and warriors a few centuries earlier. It is apparent that the Anasazi episodic burials are patterned, distinctive, and have taphonomic correspondence with the events surrounding the historic destruction of Awatovi and its townspeople. But, can the patterning and prehistoric distinctiveness be linked to institutionalized prehistoric behavior? Perhaps. First, it is reasonable to assume that all individuals in a given episodic burial were alive just prior to their being dispatched and butchered. The number of individuals represented in an episodic burial or bone deposit had to have been assaulted, constrained, or entrapped by an equal or larger number of adult antagonists or assailants. This tactical necessity, whether intra- or internecine, suggests the same degree of institutionalized organization as found in all Puebloan hunting and war parties (Parsons, 1939). 435 Second, the age, sex, context, and natural diagenesis of the Anasazi mass burials suggest rapid destructive action against any members of small isolated settlements, not the eating of helpless children and the elderly by the strongest individuals during protracted and severe winter storms. These mass burials are not Donner parties (Grayson, 1990). Third, the quasi-military enforcement aspect of the institutionalized violence model would benefit from finding an episodic, multiple, and cannibalized burial in Chaco Canyon itself, especially at the beginning of the Great House development, which this report provides. What is more, Small House is not the only ruin in Chaco Canyon where violence and cannibalism may have occurred. Even before Roberts excavated Small House, Pepper (1920) had concluded that cannibalism had been practiced in Chaco Canyon. His view was based on the finding of broken, burned, and scattered human bone on certain room floors in Pueblo Bonito and Penasco Blanco. Not all of Pepper’s, or other earlier and later archeologist’s, Chaco Canyon skeletal material has been located and restudied, but so far I have tracked down an adult male (Smithsonian 327139) from Pueblo Bonito with anvil abrasions on his frontal bone, blown out anterior teeth, and unhealed perimortem vault and nasal bone fractures. At the American Museum of Natural History there is a Pueblo Bonito adult male (W3672) with unhealed cut marks on the left temporal and perimortem breakage of the right temporal (see also Akins, 1986). Both of these men seem to have been treated in violent fashions, but they were not cannibalized. The Anasazi cannibalism hypothesis has not gone without challenge (Bullock, 19911, but has been rebutted (Turner and Turner, 1992b). The proximate social pathology hypothesis was proposed to avoid the logical alternative, namely institutionalized violence and cannibalism. Baker (1990) accepted cannibalism had occurred a t Rattlesnake ruin, and proposed that it could have been a form of social control. Baker may in fact be close to the truth if these episodic Anasazi multiple burials can somehow be scientifically linked to control mechanisms, 436 C.G. TURNER such as the destruction of Awatovi, and as implied by the rapid development of the Anasazi culture and population growth beginning around AD 900. If this could be done with some form of direct physical or indirect distributional evidence, then the charnel pits and bone deposits might be viewed as acts of coercive terrorism and example-setting. The enforcement argument is testable, in the sense that it 1) predicts episodic burials with the cannibalism signature should be earliest in the region where the rise of large communities first occurred, and 2) the burials should be spatially associated with these communities. There is one major problem with a generalized social control model, namely, the Hohokam also had a strongly developed regional system, but no episodic burials have ever been reported. However, this may be the fault of skeletal biologists who have studied Hohokam remains, because cremations, the most common Hohokam burial type, have never been systematically examined from a taphonomic perspective, and because uncremated Hohokam skeletons are often badly preserved due to desert soil and climatic conditions. CONCLUDING REMARKS In sum, the mixed remains of eight extensively fragmented individuals ranging in age from childhood to young adults were buried beneath a floor in a small pueblo in Chaco Canyon sometime after they had died, been dismembered, and their bones variously broken, cut, burned, and boiled. The lack of scavenger tooth marks or gnawing incisions, and the absence of surface bleaching and cracking, indicate that the remains were not left exposed to animals or weathering, instead they were deposited by unknown persons shortly after the event that led to the victims’ extensive skeletal damage. All six minimal taphonomic criteria deemed necessary before cannibalism can be proposed are present at Small House-breakage, cutting, anvil or hammerstone abrasions, burning, many missing vertebrae, and fragment end-polishing. The taphonomic bone damage signature of the Small House charnel pit matches that found in several other Anasazi mass burials where cannibalism and violence have been proposed and social pathology offered as the proximate explanation for bone damage. These burials could represent, in addition to random chaotic social pathology, an extreme form of deliberate social control, genocide, or tribute enforcement. The story of the rapid development of the large post-AD 900 Anasazi settlements, especially the Chacoan Great House phenomena, with its monumental architecture, revolutionary road system, numerous linked outlyer communities, elite burials, long-distance trade in the Four Corners area and Mexico, and the all important regional tribute, may have had its domineering and disruptive beginning at small defenseless farming settlements like Small House. It is unclear if Anasazi cultural efflorescence was due strictly to local evolution, or stimulated by Mesoamerican influences or migrants (Cordell, 1984). But, it is becoming ever more evident that even in northern Mexico, far from the Mesoamerican heartland, human sacrifice and cannibalism were institutionalized practices associated with ritual, warfare, and social and territorial control (Kelley, 1978). Did those practices extend even further northward into the Anasazi area? ACKNOWLEDGMENTS Permission to study, photograph, and report on the Small House skeletal series was granted by D.J. Ortner, Chairman, Department of Anthropology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. Assistance was provided by physical anthropology collections manager D.R. Hunt, and National Anthropological Archives librarians. Chaco Canyon National Historical Park archeologist, D. Ford, permitted our site inspection, and generously provided copies of Small House site records, maps, and ceramic analyses. Travel funds were granted by my Department’s Research and Development Committee, chaired by J.F. Martin. D.R. Wilcox, Curator, Museum of Northern Arizona, Flagstaff, generously lectured to my Spring, 1992, human taphonomy class on his ideas and evidence for the political evolution of the Chacoan multisettlement polity system. I also thank CANNIBALISM IN CHACO CANYON the two anonymous reviewers, M. Cartmill, and J.A. Turner €or their valuable suggestions on this report. LITERATURE CITED Akins N J (1986) A Biocultural Approach to Human Burials from Chaco Canyon, New Mexico. Reports of the Chaco Center No. 9. Santa Fe: Branch of Cultural Reserach, National Park Service. Ambler T, and Logan M (1988) The fluorescence and demise of Iroquoian cannibalism: Human sacrifice and Malinkowski’s hypothesis. Man Northeast 35rl26. Baker SA (1990) Rattlesnake Ruin (42Sa 18434):A Case of Violent Death and Perimortem Mutilation in the Anasazi Culture of San J u a n County, Utah. M.A. thesis, Department of Anthropology, Brigham Young University, Provo, Utah. Bandelier AF (1892) An outline of the documentary history of the Zuni tribe. J . Am. Ethnol. Archaeol. 3:l115. Beaglehole E, and Beaglehole P (1935) Hopi of Second Mesa. American Anthropology Association Memoir 44. Bullock PY (1991) A reappraisal of Anasazi cannibalism. Kiva 57:5-16. Carle P (19411 Burial Customs of the Indians of the Southwest. M.A. thesis, Texas Technological College, Lubbock. Cole SJ (1989) Iconography and symbolism in Basketmaker rock art. In JS Day, PD Friedman, and MJ Tate (eds.): Rock Art of the Western Canyons. Colorado Archaeology Society Memoir No. 3. Cordell LS (1979) Prehistory: Eastern Anasazi. In A Ortiz (ed.): Handbook of North American Indians. Volume 9. Southwest. Washington, D.C.: Smithsonian Institute, pp. 131-151. Cordell LS (1984) Prehistory of the Southwest. San Diego: Academic Press. Dice MH (1993) A Disarticulated Human Bone Assemblege from Leroux Wash., Arizona. M.A. thesis, Department of Anthropology, Arizona State University, Tempe. Di Peso CC (1974) Casas Grandes: A Fallen Trading Center of the Gran Chichimeca, Vols. 1-3. Dragoon, AZ:Amerind Foundation Pub. 8. Dozier E P (1954) The Hopi-Tewa ofArizona. Univ. Calif. Pubs. Am. Archaeol. Ethnol. 44:259-376. Eggan F (1950) Social Organization of the Western Pueblos. Chicago: University of Chicago Press. Ellis FM (1951) Patterns of aggression and the war cult in Southwestern Pueblos. SW J Anthropol7:177-201. Fay PM, and Klein PY (1988) A reexamination of the Leroux Wash skeletal material for evidence of human modification. Report on file, Museum of Northern Arizona, Department of Anthropology, Flagstaff. Fewkes JW (1902)Minor Hopi festivals. Am. Anthropol. 4~482-511. Fewkes JW (1912) Awatobi. In Handbook of North American Indians, Part I, Bulletin 30. Bur. American Ethnology. Washington, D.C.: Smithsonian Institute, pp. 119-120. 437 Flinn L, Turner CG 11, and Brew A (1976) Additional evidence for cannibalism in the Southwest: The case of LA 4528. Am. Antiquity 41:308-318. Frisbie TR (1978) High status burials in the Greater Southwest: An interpretative synthesis. In CL Riley and BC Hedrick (eds.): Across the Chichimec Sea: Papers in Honor of J . Charles Kelley. Carbondale and Edwardsville: Southern Illinois University Press, pp. 202-227. Grant SS (1989) Secondary burial or cannibalism? An example from New Mexico. Am. J. Phys. Anthropol. 78~230-231. Grayson DK (1990) Donner party deaths: A demographic assessment. J . Anthropol. Res. 46:223-242. Haas J (1990) Warfare and the evolution of tribal polities in the prehistoric Southwest. In J Haas (ed.): The Anthropology of War. New York: Cambridge University Press, pp. 171-189. Hagberg EB (1939) Southwestern Indian Burial Practices. M.A. thesis, Department of Anthropology, University of Arizona, Tucson. Hawley FM (1937) Pueblo social organization as a lead to Pueblo History. Am. Anthropol. 39504-522. Hayes AC, Brugge D, and Judge WJ (1981) Archeologial Surveys of Chaco Canyon, New Mexico. Washington, D.C.: National Park Service Publications in Archeology, 18A. Hurst WB,and Turner CG I1 (1990) Rediscovering the “great discovery”: Wetherill’s first Cave 7 and its implications regarding Basketmaker violence. Paper presented at Anasazi Basketmaker Symposium, Wetherill-Grand Gulch Research Project, Blanding, Utah. Judd NM (1927) Archeological investigations in Chaco Canyon, New Mexico. Explorations and field-work of the Smithsonian Institution in 1926. Smithsonian Misc. Coll. 78:15%168. Judge WJ (1979) The development of a complex cultural ecosystem in the Chaco Basin, New Mexico. In RM Linn (ed.): Proceedings of the First Conference on Scientific Research in the National Parks. Washington, D.C.: U.S. Government Printing Office, pp. 901-905. Kelley EA (1978) The Temple of the Skulls at Alta Vista, Chalchihuites. In CL Riley and BC Hedrick (eds.): Across the Chichimec Sea: Papers in Honor of J . Charles Kelley. Carbondale and Edwardsville: Southern Illinois University Press, pp. 102-126. Leubben RA, and Nickens PR (1982) A mass interment in a n early Pueblo 111kiva in southwestern Colorado. J . Intermountain Archeol. 1~66-79. Lister RH (1978) Mesoamerican influences at Chaco Canyon, New Mexico. In CL Riley and BC Hedrick (eds.): Across the Chichimec Sea: Papers in Honor of J. Charles Kelley. Carbondale and Edwardsville: Southern Illinois University Press, pp. 233-241. Mackey J , and Green RC (1979) Largo-Gallina towers: An explanation. Am. Antiquity 44:144-154. Malville N J (1989)Two fragmented human bone assemblages from Yellow Jacket, southwestern Colorado. Kiva 55:3-22. McGregor J C (1965) Southwestern Archaeology. Urbana: University of Illinois Press. McNitt F (1966) Richard Wetherill Anasazi. Revised 438 C.G. TURNER edition. Albuquerque: University of New Mexico Press. Merbs CF (1989)Trauma. In M Iscan and KAR Kennedy (eds.): Reconstruction of Life from the Skeleton. New York: Alan R. Liss, pp. 162-189. Nass GG, and Bellatoni NF (1982) A prehistoric multiple burial from Monument Valley evidencing trauma and possible cannibalism. Kiva 47:257-271. Nickens PR (1975) Prehistoric cannibalism in the Mancos Canyon, southwestern Colorado. Kiva 40:283293. Olson AP (1966)A mass secondary burial from northern Arizona. Am. Antiquity 31:822-826. Parsons EC (1939) Pueblo Indian Religion. Chicago: University of Chicago Press. Pepper GH (1909) The exploration of a burial-room in Pueblo Bonito, New Mexico. In: Putnam Anniversary Volume Anthropological Essays. New York: G.E. Stechert, pp. 196-252. Pepper GH (1920) Pueblo Bonito. Anthropology Papers, American Museum of Natural History, Vol. 27 (whole). Ravesloot J C (1988) Mortuary Practices and Social Differentiation at Casas Grandes, Chihuahua, Mexico. Anthropology Papers University of Arizona, No. 40. Reyman J E (1978)Pochteca burials a t Anasazi sites? In CL Riley and BC Hedrick (eds.):Across the Chichimec Sea: Papers in Honor of J. Charles Kelley. Carbondale and Edwardsville: Southern Illinois University Press, pp. 242-259. Sando JS (1979) The Pueblo revolt. In A Ortiz (ed): Handbook of North American Indians. Volume 9. Southwest. Washington, D.C.: Smithsonian Institution, pp. 194-197. Stanislawski MB (1963) Extended burials in the prehistoric Southwest. Am. Antiquity28:30%319. Turner CG I1 (1983) Taphonomic reconstructions of human violence and cannibalism based on mass burials in the American Southwest. In GM LeMoine and AS MacEachern (eds.): Carnivores, Human Scavengers and Predators: A Question of Bone Technology. Calgary, Canada: Archaeology Association University of Calgary, pp. 219-240. Turner CG I1 (1988) Another prehistoric Southwest mass human burial suggesting violence and cannibalism: Marshview Hamlet, Colorado. In GT Gross, and AE Kane (eds.): Dolores Archaeological Program: Aceramic and Late Occupations at Dolores. Denver: Bureau of Reclamation, pp. 81433. Turner CG I1 (1989)Teec Nos Pos: More possible cannibalism in northeastern Arizona. JGva 54:147-152. Turner CG I1 (1992)Anasazi cannibalism. Review of TD White, Prehistoric Cannibalism at Mancos BMTUMR2346. Rev. Archaeol. 23:7-13. Turner CG 11, and Dytrych MM (in preparation) Damage residua in Anasazi body processing. Turner CG 11, and Morris NT (1970) A massacre at Hopi. Am. Antiquity 35:320-331. Turner CG 11, Regan MH, and Irish J D (1991) Physical anthropology and human taphonomy of U:8:78, U:8:221, and U:8:225, Lake Roosevelt, central Arizona. Report to Statistical Research, Inc., Tucson. Turner CG 11, and Turner JA (1990) Perimortem damage to human skeletal remains from Wupatki National Monument, northern Arizona. Kiva 55:187212. Turner CG 11, and Turner JA (1992a) The first claim for cannibalism in the Southwest: Walter Hough‘s 1901 discovery at Canyon Butte Ruin 3, northeastern Arizona. Am. Antiquity 57:661-682. Turner CG 11, and Turner J A (1992b) On Peter Y. Bullock’s “reappraisal” of Anasazi cannibalism. Kiva 58: 18s201. Turner CG I1 and Turner JA (in preparation) Cannibalism in the American Southwest. 13th International Congress of Anthropology and Ethnology Sciences, 1993, Mexico City. Turner CG 11, Turner JA, and Green RC (in press) Taphonomic analysis of Anasazi skeletal remains from Largo-Gallina sites in northwestern New Mexico. J . Anthropol. Res. Upham S, and Reed PF (1989) Inferring the structure of Anasazi warfare. In DC Tkaczuk and BC Vivian (eds.): Cultures in Conflict: Current Archaeological Perspectives. Proceedings of the 20th Chacmool Conference, Department of Archaeology, University of Calgary, Canada, pp. 153-162. White TD (1988) Appendix C. Cottonwood Wash, southeastern Utah: The human osteology of feature 3, FS#27, Site 42SA12209. In J Fetterman, L Honeycutt, and K Kuckelman (eds.): Salvage Excavations of 42SA12209. Yellow Jacket, Colorado: Woods Canyon Archaeological Consultants, pp. 1-7. White TD (1991) Human Osteology. New York: Academic Press. White TD (1992) Prehistoric Cannibalism at Mancos 5MTUMR-2346. Princeton, NJ: Princeton University Press. Wilcox DR (1992) Current directions of macro-regional analysis in the American Southwest. Paper presented at 57th Annual Meeting of the Society for American Archaeology, Pittsburgh, April 8-12. Wilcox DR (submitted) The evolution of the Chacoan polity. In G Matlock, P Duke and M Malville (eds.) Chimney Rock Pueblo. Wilcox DR, and Haas J (1989) The reality of competition and conflict in the prehistoric Southwest. Paper presented at Advanced Seminar, Santa Fe. Woodbury RB (1958) A reconsideration of Pueblo warfare in the southwestern United States. Proc. Int. Congress Am. 33:124-133. APPENDIX A In reanalyzing and searching literature relevant to a Utah mass burial excavated in 1893 by Richard Wetherill (Turner, 1983; Hurst and Turner, 1990), a remark was noted in one appendix in McNitt’s (1966, p. 338) biography of Wetherill that was the stimulus for the present paper. McNitt quoted from a letter (January 1957?) he had received from archeologist Frank H.H. Rob- CANNIBALISM IN CHACO CANYON erts, Jr., that commented on an unpublished find Roberts had made a t a small ruin 9 miles east of Pueblo Bonito in Chaco Canyon, northwestern New Mexico. Roberts related that he had found beneath one of the room floors of the “Small House” remains of five burned and broken-up individuals, ages 10-18 years old: “The bones were in a jumbled mass and many of them showed the effects of fire. Most of the long bones had been split and the marrow removed while the skulls were broken into several pieces and were cupped. Whether they represented the remains of ceremonial cannibalism or of human sacrifice is a question.” Because Roberts was a member of a 1926 National Geographic Society-sponsored Chaco Canyon expedition, led by Smithsonian archeologist N.M. Judd (19271, it seemed possible that if the Small House remains had been saved, they might be curated at the Smithsonian Institution’s National Museum of Natural History. With the help of D.R. Hunt, the skeletal series was located, having been accessioned as 95112 (catalog number 3340551, on February 7, 1927, for five individuals. Although there were no records of the Small House remains ever having been studied, the series had been handled in the recent past because the small storage tray containing the post-cranial bones had a few teeth in a modern plastic bag. On May 20 and September 16, 1992, I studied lot 334055, and in the Smithsonian’s National AnthropologicalArchives reviewed Roberts’ 1926 field notebook (Roberts papers, Box l ) . Roberts made only the following comments about the charnel pit and its contents: Small House. Dwelling of the Early Period . . . Located 250 feet south of the Small Pueblo described in the preceding pages. Is at the foot of the talus tonguing out of the cliffs. In contrast to the Small Pueblo which was largely foundations and which show a pre-conceived plan, 439 this Small House just grew and as a result is very irregular in outline. There’s no regular divisions of rooms. etc. There seems to have been a number of surfaces of occupation. . . Lp. 251. Room 4 [various measurements] . . . Mass of skeletons 2.5 [presumably feet1 below foundation of west wall. Bottom of mass 4.5 below [p. 331. upright slabs in bottom of Room 4 . . . Many of the long bones were cracked & many of the bones showed sign of fire. Skulls were all in pieces and one of them had the pieces cupped, not smashed down. All of the skulls showed great malformation [p. 441. On page 9 of Roberts’ notebook is his sketch map of the site locality dated July 1926, showing Small Pueblo, Small House, a “dump,” and the location of these features relative to two adjacent arroyos that empty into the main Chaco wash. There is no sketch of the multiple burial, and no photographs could be located. There are no known artifacts associated with the multiple burial. Today, this small ruin is inventoried as 29852385 and LA42385 according to National Park Service site survey records compiled in July 1983 by archeologists P.J. McKenna and J. Miles. They found that the styles of pottery sherds a t Small House range mostly from AD 950-1225. Chaco Canyon Park Service archeologist D. Ford suggested that Small House was probably occupied at the earlier end of the range. On June 9,1992, with permission and help from Ford, J.A. Turner and I visited Small House to get a sense of the setting and environment. The 7 by 15 m roughly rectangular structure is situated in a grass, juniper, sage, and saltbush plant community thinly dispersed on deep loose sandy-silty alluvium at an elevation of ca. 1900 m. One explanation for the location of Small House is the presence of a seep in the 30 m high coalveined sandstone canyon wall 200 m to the west. The strongest impression gained was one of relative isolation and defenselessness.