Although the origin of scalping in the New World is unknown, it was a widespread practice among Native American groups during the historic period, and was assumed prior to WWII to have been present in pre-Columbian times as well (Catlin 1975; Friederici 1907; Reese 1940). This assumption was based primarily on the observations of early explorers who witnessed the practice firsthand, and on linguistic evidence for the early existence of specialized terms in certain Native American languages relating to the practice. These terms included specialized words for scalping, the scalp itself, the scalping victim, and so forth (Friederici 1907). Archaeological evidence for the scalping custom in prehistoric North America, however, did not enter the literature until the 1940's (Neumann 1940; Snow 1941,1942). This evidence was presented as either a characteristic lesion of the frontal and parietals indicating survival of a scalping event, or as a distinct pattern of cut marks in small, parallel clusters that encircled the calvarium and showed no evidence of remodelling. In the past decade, with the advent of more sophisticated dating techniques, osteological evidence of scalping in pre-Columbian North America has been rapidly increasing. This paper will compare many of these scalping cases in an attempt to gain insight into the nature of the scalping custom prior to European contact. It will also examine the burial contexts of some of these scalping victims in order to determine whether scalping victims were afforded the same burial treatment as individuals dying of other causes.Table 1: Prehistoric Scalping Cases Considered
A list of the sites considered in this study are presented in table 1. Several criteria were used in determining which sites would be studied. First, only sites which are considered to be pre-contact sites were selected. This limitation served to maintain a manageable data set, while at the same time, eliminated the possible effects of European influence on the custom following contact. Second, only sites within the United States were considered. The impetus for this limitiation was two-fold. The primary reason for such a limitation was to keep the sample size manageable and the secondary reason was to avoid such difficulties as access to publications and language difficulties for sites described outside the United States. Finally, the selection process avoided sites in the Southwest which may show evidence of scalping but are also suggestive of cannibalism, as it may be that the cannibalism phenomenon had a larger political significance related to the cultures of Central Mexico (Turner, personal communication), and therefore might represent a mix of cultural data that would be difficult to sort out.
The scalping victims considered in this study came from the Southeast, the Midwest, and the Southwest. The Midwest is by far the region with the largest sample, estimated at between 400 and 500 scalped individuals. However, most of these victims came from Crow Creek Canyon, the site of a large-scale massacre involving a minimum of 486 individuals (Gregg et al 1981). The excavators report that not only were almost every person scalped, but they also were mutilated and dismembered, and many of their hands and feet appear to have been removed and taken as trophies. This custom has been documented historically for certain Native American culture groups in the United States (Friederici 1907). Thus, it appears that something quite different from small scale raiding activities was occurring at Crow Creek Canyon. Furthermore, the skeletal remains were so mixed as to preclude the possibility of assigning postcranial elements with the proper skulls, so detailed information about individuals is not available. Crow Creek is included in the study only because it gives evidence of the intensity of conflict in the Northern Plains during the mid-fourteenth century, and because two of the crania found at the site demonstrated a lesion of the frontal and parietals consistent with survival of a scalping incident. Presumably those two earlier scalping events were unrelated to the massacre itself, and therefore can be included in the analysis of scalping survivors discussed below.
There appears to have been little sexual bias in choosing which individuals would be scalped. Sex could be determined for 33 of the scalping victims considered here, and of these, roughly 40% were female, while 60% were male. The samples are not large enough to break these percentages down by culture group. However, in the few cases where four or more victims came from the same site, invariably at least one was female and one was male. This situation proved true for both the Plains region and the Southwest. Regionally, the female/male ratio remains fairly constant with 6/14 (43%) of the victims being female in the Southwest, 5/14 (36%) being female in the Plains, and 2/5 (40%) being female in the Southeast. These results indicate that while the males were somewhat more likely to be scalped than females, females were nevertheless a frequent target.
It could be argued, then, that scalping in pre-Columbian America was more likely the result of raiding activity than of large-scale warfare taking place on a battlefield. The evidence for such an assertion is that women were not generally known to be warriors during historic times, and therefore, would be more likely to be scalped in or near their own villages (although see Ewers 1994). In fact, among certain tribes, there appears to have been great honor associated with the infiltration of an enemy village to make a kill. Catlin (1975) reports that chief Four Bears boasted of sneaking to a point within view of an enemy village, spying on the village for several days, then killing two women in full sight of the other members of the enemy village. Swanton (1946) goes a step further and states that among the Creek, the taking of a woman's or child's scalp was considered a sign of even greater valor than the taking of a scalp on the battlefield, because it indicated that the warrior had penetrated all the way into the enemy's village, a feat which required great skill. Thus, it seems clear that the scalps of women were also highly prized as trophies among some Native American groups. Perhaps the 20-25% difference in the sex ratio of the victims found simply reflects the difficulties in penetrating an enemy village. Another possibility is that women may have had value as slaves as well as value as providers of scalp trophies, and therefore, were sometimes taken captive rather than scalped. A final possibility is that many more men were scalped when they were away from the village and their bodies were either never found, or else they were found, but burial within the village was somehow seen as unnecessary or even dangerous. Whatever the case, it seems clear that Native American groups who practiced the scalping custom during the prehistoric period had no proscription against scalping women, even women who were pregnant (Brooks 1994).
Just as being a woman was no protection from scalping, being a child also appears not to have always been a deterrent to becoming a victim of this custom. The youngest prehistoric victim of scalping found in this study was a child between the ages of five and seven years, and another was a subadult between 13 and 15 years old (Hollimon & Owsley 1994; Allen et al 1985). The five to seven year old from the Fay Tolton site in South Dakota may have provided trophies in two different raids. The presence on the cranium of a characteristic scalping lesion with some bone remodelling indicates that the child survived an intial scalping event by at least two weeks before being killed in yet another raid. Obviously there was no scalp left on this child to take as a trophy, but both hands appear to have been removed by breaking the radii and ulnae toward their distal ends, and these hands were probably kept as trophies, along with the head of another individual from the site who was quite obviously decapitated (Hollimon & Owsley 1994).
The Fay Tolton child is not the only scalped individual considered in this study who showed evidence of surviving a scalping event. Several individuals appear to have survived this treatment for a period from several weeks to perhaps several years or more. Out of the 33 individuals considered, eight (24%) appear to have survived scalping, one being the five to seven year old from the Fay Tolton site described above. Two more cases of survival of a scalping incident come from the Crow Creek Canyon massacre. These numbers are significant and suggest that survival must have been a relatively frequent occurrence, as it was in the historic period (Bruesch 1974).
Among the Arikara, and presumably among some related groups as well, there appears to have been some cultural significance associated with survival of a scalping incident that did not pertain to other types of injuries (Owsley 1994). In the Arikara language, the verb quot;to scalpquot; was synonymous with the verb quot;to ruinquot; and a man who survived a scalping incident was thought to not be quite human any longer. Such a man was forced to lead a solitary life away from the village, and to stay hidden to keep from shocking and offending the living (Gilmore in Owsley 1994). This figure eventually became the focus of many myths and legends, but was usually associated with a man, rarely with a woman (Owsley 1994). If this belief were widespread in prehistoric times, one would expect to find a differential pattern of survival and interment evident in the sex of the individuals who survived a scalping event. Such a differential presence of female survivors over male survivors would suggest that women were accepted back into the community more readily than men who survived such treatment.
Only three of the ten survivors discussed here could be sexed accurately. The other seven were either too young, provided too few bones, or were not described in sufficient detail to allow determination of sex. Of the three who were sexed accurately, all were female. These three females were from Sea Island Mound, Georgia, from Moundville, Alabama, and from Wallace Mound, Nebraska and were all interred in the same mounds in which other members of their tribes were buried. It is interesting that there have so far been no cases of scalping survival reported for the Southwest. However, although it is admittedly dangerous to base conclusions on such a small sample size, there is at least a suggestion here that women who survived scalping were accepted back into their communities in the Plains and Southeast, while there is no definite evidence in these data of men surviving a scalping incident and continuing to live within their tribes.
The possibility of sex-based differential treatment of scalping survivors leads to the question of the whether scalping victims were treated differently in death. For a few of the scalping victims, information about the mode of interment was scanty and no conclusions could be drawn. For the remainder of the cases, the burial types can be divided into three general categories: those who were clearly the victims of some kind of raid and were probably buried or left as they died by the raiders themselves, those who were interred in the normal or fairly normal manner for members of their culture group, and those that were treated in an unusual manner, probably by members of their own culture group.
Three of the sites, all of which were located in the Plains region of the U.S., contained quot;burialsquot; that appear to have been the direct aftermath of some sort of raid. At the circa AD 1330 - 1430 Heerwald site in Oklahoma, the remains of an 18-22 year old pregnant female were found in a pit (Owsley 1994). Two Harrel points were found lodged within her body, one of them imbedded deeply in the anterior centrum of a lumbar vertebra, the other found between the ribs. Damage to the left scapula and ribs indicate that a third projectile point passed through her body at some time just prior to or just following her death. The woman's remains were found carelessly tossed into a pit, her face twisted on the neck and pointing downward, while her torso was twisted in the opposite direction and her hips were found facing upward. The body of a child was found laying over her back and side. It is certainly possible that this treatment was given by her own culture group, but the projectile points make it clear that she died in a very violent encounter, and the careless mode of interment suggests burial or disposal in this pit by the raiders themselves.
At the Fay Tolton site in central South Dakota, the 5-7 year old child with a scalping survival lesion already mentioned was found lying on the floor of an earth lodge, both hands apparently removed as trophies (Hollimon & Owsley 1994). Three other individuals lay haphazardly on the floor nearby, a female aged 14-16 years with no evidence of trauma, another female aged 10-12 years with a projectile point embedded in her anterior tibia, and a male aged 25-29 whose location within the earth lodge is unclear. The presence of charred areas on the skeletons suggests that the bodies or the structure was burned sometime before decomposition of the flesh had taken place. The evidence in this case clearly indicates that a raid had taken place, and these individuals were left as they had died, never recieving any kind of mortuary treatment from their own people.
The third case of this type is by far the most dramatic. At the Crow Creek Canyon site, at least 486 individuals were massacred, the bones of men, women, and children found strewn in an area roughly 7 meters square and 1 meter deep (Gregg et al 1981). These skeletons were severely dismembered, with most of the hands and feet apparently amputated and many of the skulls showing evidence of having been cut from the rest of the body. Nearly all of these individuals showed evidence of scalping, and as already mentioned, two of the crania showed evidence of an earlier scalping incident which they had managed to survive for a period of time. Once again, it is possible that this deposit was the work of people from the same village or tribe rather than that of the raiders, but a massacre of such size would probably leave few individuals to dispose of this large quantity of remains in such a manner. Therefore, the more likely explanation is that those responsible for these peoples' deaths were also responsible for their interment.
Aside from these raid situations, the majority of scalping victims appear to have experienced fairly typical mortuary treatment. Of the 31 skeletons whose burial provenience is documented, 22 (71%) appear to have recieved burials which were typical for the culture group to which they belonged. Seven of these, however, were from Nuvakwewtaqa (Chavez Pass) in the central Arizona, and due to looting by pothunters the exact nature of their burials is unknown, except that they were disinterrred from a known cemetery area (Allen et al 1985). The nine skeletons that remain come from only three sites and three burial contexts. One of the sites is in the Southwest, one is in the Plains, and one is in the Southeast.
The Mason site is in Franklin County Tennessee (Owsley & Berryman 1975). It dates to between AD 700 and 1000, and contains the flexed burial of a single male adult in a shallow grave. The grave was dug in the floor of a large basin-shaped pit, sometime after which the pit itself was covered with limestone slabs and a fire was built over these slabs. It is not clear whether the limestone slabs and the fire built upon them were directly associated with the interment of this individual or not, but the authors state that this mode of interment was fairly unique, perhaps indicating that the scalped male was somehow seen as unusual or different.
The unusual mortuary treatment of scalped individuals from the Plains comes from the prehistoric Sargent site ossuary in Custer County, Nebraska (O'Shea & Bridges 1989). These individuals may have been members of the Central Plains Tradition in which ossuary burial was common, however, the unusual patterning of the bones within the burials differed from the Central Plains Tradition ossuaries in the near complete absence of postcranial bones. Twelve individuals in all were interred in this ossuary. Ten crania were arranged in parallel rows of five crania each, all facing to the west and lined up in a north-south direction. The few postcranial bones associated with the crania were scattered nearby the skulls in a haphazard way and may have all belonged to a single individual. Four of the twelve individuals, three males and one female, appear to have been scalped. The disarticulation of the skeletons in the absence of defleshing cut marks on the bones that are present suggest that some period of decay had taken place before burial.
O'Shea and Bridges (1989), in searching for an explanation for this unusual burial practice, used ethnographic evidence to bring possible understanding to this small ossuary. As it was common in the Central Plains Tradition to create ossuaries in which the postcranial bones are deposited in a pit with the skulls being lined up in rows on top of this deposit, the authors conclude that the Sargent Site ossuary is simply a variation on this theme. Since these Central Plains groups were known to use alternative methods of burial when the tribe was away from their permanent villages, the authors postulate that this ossuary may have come as a result of an attempted burial of individuals killed in a raid while members of the village were away from their permanent home. If this explanation is correct, it would indicate that, while the mode of interment of these scalped individuals was not typical, the variation was due not to their unusual status as corpses that had been scalped, rather it would indicate unusual treatment due to the circumstances the tribe itself in not being near their permanent home when members of their group were killed.
The final case of unusual burial circumstances is that found at the Vosberg site of AZ U:13:26b (ASU) in the central Arizona mountains. At this circa AD 1050-1250 site, two males, aged 35-45 and 30-40 years, and two females, aged 25 - 35 years, were found carefully placed within one room of an eight room compound. They lay extended side by side, had two local ceramic vessels in association with them, which presumably represent grave goods, and the oldest male was found to have a quantity of burned seed on one of his shoulders, as though it were of some sort of ritual significance. This same individual also had a fetish bundle of rabbit femurs on his chest. This bundle was probably hung around the neck by some sort of string or cord in life. All four of these individuals exhibited cut marks on the cranium indicative of scalping (Bueschgen and Case n.d.). No attempt at burial was made other than the placement of seven large rocks over the bodies. There is also some evidence that the structure in which they were placed may have been ceremonial in nature. The room appears to have been separated slightly from the rest of the compound. There was very little pottery or other artifacts consistent with an occupation area, and this discovery led to the possibility that this somewhat separate structure may have served some unusual purpose. The walls were roof supporting rather than freestanding as was the case in most other Vosberg structures, and finally, a pit lined with stones was found just inside the door, which fits the description of a feature tentatively identified by Morris (1970) as a foot drum.
There was also a fine lens of charcoal associated with the skeletons which indicates that the roof of the structure was burned down upon the skeletons sometime after their placement on the floor. This type of burial within a structure is certainly atypical in the Vosberg locality, where normal burial was usually single or occasionally double inhumations in a cemetery area with the inclusion of two or three tradeware pots into each grave. An argument for the placement of the bodies and the burning of the structure being part of the raid that probably resulted in these people's deaths seems weak because of the careful placement of the bodies and the apparent inclusion of a few grave goods, including the charred seed. The best explanation of this unusual burial may be that these people were treated differentially, either because of the unusual nature of their deaths, or because of some aspect of their relative status within the community.
This review and comparison of scalping cases from several regions of the United States has resulted in several interesting observations about the nature of the scalping custom in prehistoric North America. The two strongest conclusions that resulted from this study are that there appears to have been little sexual bias among Native American groups in choosing who would be a scalping victim, and there also appears to have been a remarkably strong propensity for the scalped to survive the ordeal and to continue on with their lives within the community to which they belonged. Although not as strongly supported, it also appears that children were not immune to becoming scalping victims, although the single good case described here cannot be used by itself to fully support such an argument. Another interesting indication derived from this study includes the apparent paucity of male scalping victims who managed to survive the scalping ordeal, which may show that the historic practice among the Arikara of shunning males who survived scalping and treating them as if they were ghosts or beings with supernatural powers was already well developed and widespread in prehistoric times. Yet the relative infrequency of unusual burial treatment for scalped individuals who did not survive argues against a widespread custom of differential mortuary treatment for individuals who died as a result of the scalping custom.
As more cases of prehistoric scalping come to light within localized areas, comparisons of scalping cases within cultural horizons may allow greater insight into the nature of the scalping custom prior to its encouragement and commercialization by European and Colonial governments. When a larger data set is available, perhaps comparisons of known cultural practices among the descendents of these prehistoric groups during the historic period will cast some light on scalping customs that can be inferred from the archaeological and osteological records. These data might provide a rare opportunity to reconstruct certain beliefs that may have been associated with the custom of scalping in the pre-contact New World.
Site Name Location Date Reference
Moundville Alabama AD 1000 - 1500 Snow (1941)
Grasshopper Arizona AD 1200 - 1350 Allen et al (1985)
Nuvakwewtaqa Arizona AD 1200 - 1350 Allen et al (1985)
Vosberg Arizona AD 1050 - 1250 Bueschgen and Case (n.d.)
Wilson Arkansas AD 1400 - 1700 Owsley (1994)
BLM Land Colorado AD 1160-1300 France (1992)
Sea Island Mound Georgia AD 1000 Ortner & Putschar (1985)
Fulton County Illinois Middle Mississippi Neumann (1940)
Hanging Valley Iowa AD 190 - 310 Tiffany (1988)
Blasky Mound Minnesota AD 1 - 900 Hollimon & Owsley (1994)
Sargent Site Nebraska (Custer) prehistoric O'Shea & Bridges (1989)
Site 25DK9 Nebraska (Dakota) prehistoric Miller (1994)
Site 25DX4 Nebraska (Dixon) prehistoric Miller (1994)
Wallace Mound Nebraska (Sarpy) AD 1050 - 1400 Hollimon & Owsley (1994)
Bahm Site North Dakota 6000 BC - AD 500 Hollimon & Owsley (1994)
Heerwald Oklahoma AD 1330 - 1430 Owsley (1994)
Nagle Oklahomoa late prehistoric Owsley (1994)
Wickham #3 Oklahoma late prehistoric Owsley (1994)
Crow Creek South Dakota AD 1325 - 1390 Gregg et al (1981)
Fay Tolton South Dakota AD 950 - 1250 Hollimon & Owsley (1994)
Mason Site Tennessee AD 680-980 Owsley & Berryman (1975)
Clarksville Virginia AD 800 - 1630 Hoyme and Bass (1962)
Spencer Lake Mounds Wisconsin AD 490 - 580 Neiburger (1989)
References Cited
Allen, Wilma H., Charles F. Merbs, and Walter H. Birkby 1985 Evidence for Prehistoric Scalping at Nuvakwewtaqa (Chavez Pass) and Grasshopper Ruin, Arizona. In Health and Disease in the Prehistoric Southwest, edited by C. Merbs and R. Miller, pp. 23-41. Anthropological Research Papers No. 34. Arizona State University, Tempe.
Brooks, Robert L. 1994 Warfare on the Southern Plains. In Skeletal Biology in the Great Plains: Migration, Warfare, Health, and Subsistence, edited by Douglas W. Owsley and Richard L. Jantz, pp. 317-324. Smithsonian Institution Press, Washington D.C.
Bruesch, S.R. 1974 A Cure for Scalped Head. Morristown Journal of Pathology 1:23-26.
Bueschgen, Wolf and D. Troy Case n.d. Evidence of Prehistoric Scalping at Vosberg, Central Arizona. Unpublished paper.
Catlin, George 1975 Letters and Notes on the North American Indians. Clarkson N. Potter, New York. Originally Published in 1933 in A Book of Americans by Rosemary and Stephen Benet, Holt, Rhinehart, and Winston, New York.
Ewers, John C. 1994 Women's Roles in Plains Indian Warfare. In Skeletal Biology in the Great Plains: Migration, Warfare, Health, and Subsistence, edited by Douglas W. Owsley and Richard L. Jantz, pp. 3325-332. Smithsonian Institution Press, Washington D.C.
France, Diane L. 1992 Scalping in a Pre-Columbian Skeleton From Dolores County, Colorado (Abstract). American Journal of Physical Anthropology Supplement 14:76.
Friederici, Georg 1907 Scalping In America. Smithsonian Institution Annual Report 1906 Pp. 423-438 Smithsonian Institution Press, Washington D.C.
Gregg, John B., Larry J. Zimmerman, James P. Steele, Helen Ferwerda, and Pauline S. Gregg 1981 Ante-Mortem Osteopathology at Crow Creek. Plains Anthropologist 26(94):287-300.
Hollimon, Sandra E. and Douglas W. Owsley 1994 Osteology of the Fay Tolton Site: Implications for Warfare During the Initial Middle Missouri Variant. In Skeletal Biology in the Great Plains: Migration, Warfare, Health, and Subsistence, edited by Douglas W. Owsley and Richard L. Jantz, pp. 345-353. Smithsonian Institution Press, Washington D.C.
Hoyme, Lucile E. and William M. Bass 1962 Human Skeletal Remains from the Tollifero (Ha6) and Clarksville (Mc14) Sites, John H. Kerr Reservoir Basin, Virginia. In Inter-Agency Archeological Salvage Program: River Basin Surveys Papers No. 25, edited by Frank H. H. Roberts, Jr, pp. 329-400. Smithsonian Institution Bureau of American Ethnology Bulletin 182. U.S. Government Printing Office, Washington D.C.
Miller, Elizabeth 1994 Evidence for Prehistoric Scalping in Northeastern Nebraska. Plains Anthropologist 39(148):211-219.
Morris, Donald H. 1970 Walnut Creek Village: A Ninth-Century Hohokam-Anasazi Settlement in the Mountains of Central Arizona. American Antiquity 35(1):49-61.
Neiburger, E. J. 1989 A Prehistoric Scalping: 600 A.D. Central States Archaeological Journal 36(4):204- 208.
Neumann, Georg K. 1940 Evidence of the Antiquity of Scalping from Central Illinois. American Antiquity 5:287-289.
Ortner, Donald J. & Walter Putschar 1985 Identification of Pathological Conditions in Human Skeletal Remains. Smithsonian Institution Press, Washington D.C.
O'Shea, John M. and Patricia S. Bridges 1989 The Sargent Site Ossuary (25CU28), Custer County, Nebraska. Plains Anthropologist 34(123):7-21.
Owsley, Douglas W. 1994 Warfare in Coalescent Tradition Populations in the Northern Plains. In Skeletal Biology in the Great Plains: Migration, Warfare, Health, and Subsistence, edited by Douglas W. Owsley and Richard L. Jantz, pp. 3333-344. Smithsonian Institution Press, Washington D.C.
Owsley, Douglas W. and Hugh E. Berryman 1975 Ethnographic and Archaeological Evidence of Scalping in the Southeastern United States. Tennessee Archaeologist 31(1):41-58.
Reese, Hans H. 1940 The History of Scalping and its Clinical Aspects. The Year Book of Neurology, Psychiatry and Endocrinology pp. 3-19. Year Book Publishers, Chicago.
Snow, Charles E. 1941 Possible Evidence of Scalping at Moundville. Alabama Museum of Natural History Paper No. 15(2):55-57. 1942 Additional Evidence of Scalping. American Antiquity 7:398-400.