The archaeological record provides evidence of humans permanently altering the appearance of their bodies for at least the last 10,000 years. Tattoos have been used for the last 5,000 years ( and humans have been permanently modifying the shape of their children’s heads for much longer than that (Brown 2010). While we will never be certain why this done in the remote past, the ethnographic and historical record for the last few hundred years suggests head shapes were modified for esthetic reasons (fashion), to distinguish one group of people from another, ritual, or even the belief that a particular head shape had health benefits for the child (Dingwall 1930). If head shapes were modified for esthetic reasons, then the child would be extremely unfortunate if the fashion changed during their lifetime. Most recently clinicians have modified the shape of children’s heads that have grown abnormally, usually as a result of a particular sleeping position or asymmetrical growth (positional plagiocephally). Corrective appliances may then return an infants head to a more normal shape within 6-12 months.

After birth a child’s head is soft and malleable. The frontal bone, that forms a single solid bone in adults, is divided in two by the metopic suture and the rear of the bone contains a soft fontanelle. Similarly the occipital bone, rather than being the single bone present in adults, is divided into four, with posterolateral fontanelle’s. The sutures between the bones that comprise the neurocranium are open and moderate external pressure can modify the shape of the cranial vault. However, unless the growth of the vault is frequently, or permanently, restricted in some way, internal pressure from the expanding brain, meninges and cereobrospinal fluid will result in the return of a normal, genetically determined, vault shape. After 6-12 months the fontanelle’s have usually fused, vault bones have started to thicken and harden, and the cranial sutures form a tighter bond between different sections of the vault. It is therefore much more difficult to modify the shape of a child’s head when they are older than 12 months, than it is for infants.

infant skull
Human infant skull and mandible showing the anterior fontanelle at the rear of the frontal bone and the partially closed metopic suture (copyright Peter Brown 2009).

In the recorded historical period infant’s head shapes were permanently modified using a variety of fixed appliances (tight bandages, cradle boards and caps), as well repetitive manual pressure by the mother’s hands. All of these procedures were usually carried out by the mother and thought to be in the infant’s best interests. However, depending upon the method used to alter the shape of the infant’s vault, the diligence and expertise of the child’s mother, and social pressures to produce a uniform head shape, there may be considerable variation in shape, both within communities, and between siblings. Some forms of deformation, for instance cradle boards (Cybulski, 1989; Manriquez et al., 2006; Lekovic et al., 2007), were also responsible for more extreme alterations in shape than head binding, caps or repetitive manual pressure (Miklouho-Maclay, 1882; Dingwall, 1931; Blackwood and Danby, 1955; Kiszelly, 1989; Benus et al., 1999), but importantly all methods alter the normal size and shape relationships found in the human cranial vault. An excellent geographic and historical review can be found in Dingwall (1930), with a recent discussion of the effects on growth and cranial anatomy in Brown (2010).

As pressure is usually applied to both the front and back of an infants vault, most methods of deformation usually result in an increase in cranial height with more curvature in the parietal bones (when viewed from the side), a flattened frontal bone and a flattened occipital bone. Other changes include distinctive changes in the normal pattern of cranial vault bone thickness, and symmetrical depressions on the frontal bone, where bone growth has not been able to follow its usual pattern (Brown 2010).


Recent historical example: The Arawe from New Britain

Historically the Arawe intentionally altered the shape of their infant’s heads through a series of tight bandages applied shortly after birth, and kept in place for the first four to six months of life (Blackwood and Danby, 1955; Brown 2010). Pressure was applied to the front, back and sides of the neurocranium, and this tended to produce a relatively tall and somewhat conically shaped head. The frontal bone was flattened, increased in length and usually contained both distinctive depressions and a raised area of thickened bone at the rear of the frontal. The curvature of the parietal bones was greatly increased, and the occipital bone flattened and elongated. Lateral radiographs (x-rays) demonstrate these changes in bone curvature and a distinctive pattern of cranial vault thickness, where bone growth has been altered.

Arawe cranial deformation process
Arawe skull
Arawe man Arawe xray


Earlier history: did Neandertal’s use artificial cranial deformation?

There has been some debate over the antiquity of this practice, particularly if it was employed by Neandertals. In 1982 Erik Trinkaus argued that the shape of the frontal bones, and possibly also the parietal bones, in the Shanidar 1 and 5 Neandertals were a product of intentional cranial modification (the Neandertal deposits in Shanidar cave are dated to between about 55,000 and 45,000 BP, Bar-Yosef 1989). The Shanidar Neandertals have flatter frontal bones than the European Neandertal average, and the parietal bones of Shanidar 5 suggested that it might have had a relatively high vault, with great parietal curvature. However, in 1999, Alan Thorne, Colin Groves and colleagues noted that the original reconstruction of the fragmentary Shanidar 5 was faulty, particularly for the parietal bones (Erik Trinkaus was also a co-author so he was admitting to an earlier misinterpretation). There was no longer any convincing evidence for artificial cranial deformation in Neandertals. They simply don’t have the characteristic indicators of cranial deformation, and Neandertal head shape is remarkably uniform over a large geographic area and for 10’s thousands of years. It is therefore surprising, that as recently as 2012, claims that Neandertals used to modify the shape of their children’s heads still appear in the medical literature (Obladen 2012).

Shanidar cave Shanidar 1 Shanidar 5
Shanidar Cave, Iraq Shanidar 1 Neandertal Shanidar 5 Neandertal


Upper Cave at Zhoukoudian, China

The Upper Cave skeletons were excavated in 1933 and 1934, with the archaeological assemblage discussed by Pei (1935, 1939) and the human skeletal materials briefly described by Weidenreich (1939) and in more detail by Wu (1960, 1961). Radiocarbon dates now extend from 10,175 ± 360 BP (ZK-136-0-4) for the upper part of the cave to 33,200 ±2000 BP (OXA-190) for the basal layers. Unfortunately, as I have discussed previously (Brown 1992), the published accounts of the excavation contain insufficient information to be certain of the stratigraphic relationship between the human remains and the dated animal bones. Wu and Wang (1985) argue that the older dates from the Upper Cave are well below the areas of human occupation, which they place at around 10,000 BP, while (Chen et al. 1989; Hedges et al. 1992; Hedges et al. 1988) suggest 29-24 ka BP for the cultural layers.

Parts of three human skeletons were excavated from the Upper Cave and one of these, UC102 referred to as the “Melanesian woman” by Weidenreich, as been argued to be artificially deformed (Brothwell 1975). However, the UC102 skeleton was lost at the same time as the Zhoukoudian H. erectus (Peking Man) fossils in 1941 and the original specimen is now only known from a plaster replica. Both the replica and original description by Weidenreich (1939) indicates that the UC102 cranium was broken and somewhat distorted postmortem. This complicates any anatomical assessment of the cranium, but as Donald Brothwell correctly indicated, the shape and proportions of the major cranial vault bones are consistent with cranial deformation. The cranium of Upper Cave 102 (UC102) is nearly complete but has severe post mortem damage. This has left the skull with a number of long and quite broad cracks running transversely across the parietal and occipital regions and the cranium as a whole is somewhat twisted to the right. The skull can be compared to the other well-preserved “female” cranium found at the site (Upper Cave 103), which does not share UC102's unusual morphology. Morphologically, the unusual features of UC102 include the lengthening and flattening of the frontal bone, with a marked flattening in the posterior two-thirds of the bone, and great cranial height. There is also a series of depressions on either side of the mid-line on the frontal bone, however, there is no evidence of a prebregmatic eminence. While the age of UC102 will always remain uncertain (10-29 ka BP) the evidence indicates that it was deformed during infancy and may be the earliest recorded evidence of this type of behavior.

Upper Cave Zhoukoudian Upper Cave 102 skull
Upper Cave, Zhoukoudian, China
Upper Cave 102


Terminal Pleistocene Australia: Kow Swamp, Coobool Creek and Nacurrie

Although there are ethnographic accounts of the use of artificial cranial deformation by Australian Aborigines (Robinson 1841 in Kenyon 1928; Brierly 1848 -1850; Macgillivray 1852), the reliability of these observations are uncertain as no examples of artificially deformed crania were found in the extensive museum collections (>5000 crania) that were present until the mid-1980’s. Ethnographic descriptions of head shaping using manual hand pressure in the Cape York region of northern Australia, may be the most reliable as this method, as well as other form of cranial deformation, have been recorded in the Torres Strait Islands and PNG (Macgillivray 1852).

Excavations at Nacurrie, Coobool Creek and Kow Swamp in the Murray River Valley, south-eastern Australia, located human burials that were subsequently dated to 9-13 ka BP (see relevant entries in the Australian index). The most securely dated of these burials is Nacurrie 1, which has an AMS collagen date of 11440 ± 160 years BP (NZA 1069), obtained from postcranial bone fragments (Brown 2010). In 1975 Donald Brothwell argued that the shape of the cranial vault in Kow Swamp 5 was the result of intentional cranial deformation, rather than normal growth processes. With the larger and better preserved Coobool Creek sample, Peter Brown (1981, 1989), confirmed and extended Brothwell’s conclusions, with similar results obtained by Anton and Weinstein (1999), Clark et al. (2007) and Durband (2008).

Kow Swamp 5 Coobool Creek 65 Nacurrie 1
Kow Swamp 5 (copyright Peter Brown 1981)
Coobool Creek 65 (copyright Peter Brown 1982) Nacurrie 1 (copyright Peter Brown 1983)

The terminal Pleistocene human skeletons from south-eastern Australia can be distinguished from those of people living in the same region during the late Holocene, by significantly greater cranial, dental and skeletal dimensions (Brown 1989, 2010). Within the skeletal material from Kow Swamp and Coobool Creek there is also considerable variation in cranial morphology. Some of this variation is consistent with the modification of an infant’s head shape, while others seem not to have had the shape of their head altered. Examples of crania that are clearly deformed include Kow Swamp 5, Coobool Creek 65 and Nacurrie 1 (Brown 1981; Clark et al. 2007; Brown 2010). In these crania, as well as others from the same location, the frontal bones are elongated and flattened, cranial vault height is increased, the curvature of the parietal bones is relatively high and the occipital bone is also elongated and flattened. The frontal bones have symmetrical depressions bordering the midline, there is marked thickening of the bone at the rear of the frontal with a prebregmatic eminence, and the bone in the middle of the frontal is relatively thin (constricted diploe). The normal relationship between cranial dimensions, for instance the length and curvature of the frontal bone, is also not present in these crania (Brown 2010). Comparison of the way these Australian crania were deformed, with known ethnographic examples like the Arawe from New Britain, suggests that the more subtle deformation in Australia (parietal bones not as curved) indicates that an infants heads were molded through repetitive manual pressure from their mother’s hands, and not by the use of fixed bandages or other appliances.

Interpretation of the cultural significance of cranial deformation in the terminal Pleistocene of the central Murray River region is more problematic. The available evidence indicates that this behaviour was restricted both in time and geography. Given the geographic proximity of Kow Swamp, Coobool Creek and Nacurrie, and the similarities in skeletal morphology, skeletal robusticity and body mass (Brown, 1989, 2010), it seems likely that they were a closely related group of people. Both male and female crania show evidence of deformation, and there is considerable variation, from clearly deformed to those that are difficult to classify (Brown, 1981; Anton and Weinstein, 1999; Durband, 2008). As neonates are usually in their mothers care, it seems most probable that head-shaping was carried out by the mother during the first six months postpartum. The reasons why head shapes were modified will remain a matter for speculation, as will the details of why it was discontinued.


Adebonojo, F.O., 1991. Infant head shaping. JAMA 265, 1179.

Anton, S.C., Weinstein, K.J., 1999. Artificial cranial deformation and fossil Australians revisited. J. Hum. Evol. 36, 195-209.

Benus, R., Masnicova, S., Lietava, J., 1999. Intentional cranial vault deformation in a Slavonic population from the medieval cemetery in Devin (Slovakia). International Journal of Osteoarchaeology 9, 267-270.

Blackwood, B., Danby, P.M., 1955. A study of artificial cranial deformation in New Britain. Royal Anthropological Institute Journal 85, 173-191.

Brierly, D.W., 1848-1850. Journals and manuscripts. Mitchell Library, Sydney.

Brothwell, D., 1975. Possible evidence of a cultural practise affecting head growth in some late Pleistocene East Asian and Australasian populations. J. Arch. Sci. 2, 75-77.

Brown, P., 1981. Artificial cranial deformation: a component in the variation in Pleistocene Australian Aboriginal crania. Archaeol. Oceania 16, 156-167.

Brown, P. 1989. Coobool Creek. Terra Australis 13. ANU, Canberra.

Brown, P. 2010. Nacurrie 1: mark of ancient Java or a in terminal Pleistocene Australia? J. Hum. Evol.

Clark, J.D., Dobson, S.D., Anton, S.C., Hawks, J., Hunley, K.L., Wolpoff, M.H., 2007. Identifying Artificially Deformed Crania. International Journal of Osteoarchaeology 17, 596–607.

Cybulski, J.S., 1989. Skeletal variability in British Columbia coastal populations: a descriptive and comparative assessment of cranial morphology. National Museum of Canada, Ottawa

Dingwall, E.J., 1931. Artificial cranial deformation. John Bale, Sons and Danielson, London.

Dobson, J., 1994. Baby beautiful: a handbook of baby head shaping. Heirs Press, Carson City, NV.

Durband, A., 2008. Artificial cranial deformation in Kow Swamp 1 and 5: A response to Curnoe (2007). Homo 59, 261-269.

Kenyon, A.S., 1928. The Aboriginal Protectorate of Port Phillip. Report of an expedition to the Aboriginal tribes of the western interior by the Chief Protector, George Augustus Robinson. Victorian Historical Magazine 12, 134-171.

Kiszelly, I., 1989. The origins of artificial cranial formation in Eurasia. BAR International Series (Supplementary) 50, London.

Lekovic, G.P., Baker, B., Lekovic, J.M., Preul, M.C., 2007. New world cranial deformation practices: Historical implications for pathophysiology of cognitive impairment in deformational plagiocephaly. Neurosurgery 60, 1137-1146.

Macgillivray, J., 1852. Narrative of the Voyage of HMS Rattlesnake, 2 volumes, London.

Manriquez, G., Gonzales-Bergas, F.E., Salinas, J.C., Espoueys, O., 2006. Intentional cranial deformation in archaeological populations of Arica (Chile): Preliminary geometric morphometrics analysis using craniofacial radiographs. Chungara-Revista De Antropologia Chilena 38, 13-34.

Miklouho-Maclay, N., 1882. Cranial deformation of new-born children at the island of Mabiak. Proceedings of the Linnean Society of New South Wales 6, 627-629.

Thorne, A., Groves, C., Check, M., Trinkaus, E., 1999. A new reconstruction of the Shanidar 5 cranium. Paleorient 25, 143-146.

Trinkaus, E., 1982. Artificial cranial deformation of the Shanidar 1 and 5 Neandertals. Curr. Anthrop. 23, 198-199.

Weidenreich, F., 1939. On the earliest representatives of modern mankind recovered on the soil of East Asia. Bulletin of the Natural History Society of Peking 13, 161-174.