Photographic analysis of a Shabe Yoruba burial

When you’re one of the only bioarchaeology grad students in a department with few other osteologists, almost anything involving human remains will eventually make its way across your desk. After getting back to the museum in early September, fellow graduate student Andrew Gurstelle  told me that he had come across a burial this past summer, and asked me whether I would mind taking a look at it if I had the time. I was initially extremely excited because I thought this would involve hands-on work with human remains. However, as he explained the context of his site, and his relationship with the local community, I gradually understood why there weren’t any physical human remains to be examined at all.

Map of Settlement History Surrounding Oke ShabeAndy is the Principle Investigator for the Savè Hills Archaeological Research Project (SHARP), which investigates the settlement history, demography and political economy of a 17th century Yoruba state. One of the goals of the project during the 2013 field season was to excavate stratified contexts like trash pits, in order to build a regional artifact chronology. Knowing the chronological relationship between different artifact types would allow archaeologists conducting exploratory survey to tell what time period a site was  likely from, just by examining surface scatters of material culture.  In summer 2013 SHARP focused on  Aukpon, a village abandoned during the 19th century after a series of wars in the region. Importantly, the site of Aukpon (also known as “Old Aensin”) is related to the contemporary village of Aensin, a settlement that has close ties to the abandoned settlement. Lineage heads and local elders supported SHARP’s excavations at Aukpon, and provided detailed oral histories of Aensin, Aukpon and the surrounding area.

Map of Archaeological Features at Aukpon

In 2013 the SHARP team decided to excavate two 2x2m units in the largest mound at Aukpon, which local informants suggested was likely the remains of a derelict midden. Initial excavations produced high densities of faunal remains, ceramic sherds, ash and charcoal, supporting the initial functional identification of the mound. However, at about 65cm below the mound surface, the team came across a well-preserved human pelvis. Further excavations revealed associated femora and lumbar vertebrae. After opening another 1x2m unit to reveal the extent of the human remains, the excavators uncovered a well-preserved and articulated skeleton that extended SSW-NNE across the midden (Figure 1).

Stratigraphic evidence suggested that a pit was dug into the midden for the purpose of interring this individual, after which it was covered over with midden refuse. No grave goods were found associated with the remains. After the burial, the midden continued to be used until the abandonment of Aukpon in the 19th century.

Figure 1. Midden burial from Old Ainsen

Figure 1. Midden burial from Aukpon

Importantly, SHARP had previously worked with the Directorate du Patrimoin Culturel of Bénin and the king of the Shabe to establish a ‘best practice’ protocol for handling human remains. This agreement stipulated that human remains would not be removed from the archaeological site. Accordingly, after careful excavation to expose the extent of the midden burial, and photographic documentation of the skeletal remains, the  individual from the Aukpon midden was reburied. However, in the course of the 2013 season excavators had accidentally removed fragmentary human tibiae and fibulae from a higher level of the midden. No bioarchaeologists were present on site, and these elements had been assumed to comprise faunal components of the midden fill. The fragmentary long bones were brought to the U.S. for analysis along with the rest of the midden fill. The human remains that were accidentally collected will be returned to Ainsen for reburial in 2014.

Andrew had taken a number of photographs of the midden burial, and wanted to know what I could tell him about the interred individual. Specifically, he was curious about the age and sex of the person that had been buried, because a few features of the grave were particularly unique for this region and time period. Midden burials are not common at Yoruba archaeological sites, making this a relatively notable example of historic mortuary practices. Additionally (as is visible in Figure 1), the skeleton was relatively complete except for the bones of the right hand. The residents of Aensin likened this amputation to the historically known practice of post-mortem removal of the head, hands and feet of powerful individuals, such as titled officials and lineage heads. This ritual removal of appendages or crania was performed  in order to disempower the spirits of such influential people, and prevent them from returning as malevolent forces. The community of Aensin preserves detailed written genealogies of lineage heads, so Andrew hoped that knowing the approximate time period of the midden burial and the sex of the interred individual might make it possible to identify the individual.

Because the Aukpon individual had been reburied,  the only way to estimate age or sex was through a photographic analysis of non-metric traits of the cranium and pelvis. Osteologically speaking it is important to handle bones when making estimations of sex, both to develop an appreciation of the robusticity of muscle attachments, and to be able to rotate them in multiple dimensions to assess their size and shape. However, archaeology often entails maximizing the information available from problematic data sets by any means necessary, so I told him I would be happy to look at the photographs.  I used guidelines from Standards for Data Collection from Human Skeletal Remains to score the non-metric attributes of the skull and pelvis.

Figure 2. Close-up of the pelvis of the Old Ainsen individual

Figure 2. Close-up of the pelvis of the Aukpon individual

When examining a close-up photograph of the pelvis, I noticed that the Greater Sciatic Notch (GSN) was preserved on the right side. At first glance it looked relatively wide, which is the more female form. Since I was not able to handle the os coxa itself I used Powerpoint to draw the shape of the GSN and then overlaid that shape on the figure from Buikstra and Ubelaker (Figure 2). This led to an estimation of 2/3 (Probable Female/Indeterminate). While someosteologists also use the relative size and shape of the sacrum as a means of estimating sex, I had never used this particular non-metric trait before, and decided not to try it for the first time using only a photograph.

Figure 3. Skull of the Old Ainsen individual, lateral view

Figure 3. Skull of the Aukpon individual, lateral view

The cranium also looked relatively female. The nuchal crest, mastoid processes and glabella were all visible in photographs, and all looked gracile, rather than robust. In Figure 3, you’ll notice that the back of the cranium is relatively smooth, the brow ridge/glabellar region does not protrude, and the mastoid processes are small. My examination produced a score of 2 (Probable Female) for all three traits.

 Figure 4. Mandible of the Old Ainsen individual, anterior view

 Figure 4. Mandible of the Aukpon individual, anterior view

This individual also lacked a pronounced chin. The mental eminence shown in Figure 4 is smooth and non-projecting. The left half of the mandible is partially obscured by sediment, but if the left side is symmetrical to the right side, the chin appears quite gracile. Additionally, the shape of the chin is rounded rather than square, (see Bass 1994), and the gonial angle is obtuse (Figure 5), both of which are more likely female traits than male traits.

Figure 5. Gonial angle of the Old Ainsen individual

Figure 5. Gonial angle of the Old Ainsen individual

All of the non-metric traits suggested that this individual was more likely female, which I found particularly convincing because there is a trend of masculinization with age for the crania: older individuals are more likely to appear male (see Buikstra and Ubelaker 1994:16). Since this individual was edentulous with the exception of a single tooth, the person was likely of an advanced age when they died. Because the cranium still appeared very feminine with regards to non-metric traits, despite the trend for masculinization with age in humans, I was fairly confident in my analysis. As confident as an osteologist can be while working from photographs, anyways.

Despite the congruence of the non-metric data, I decided to see if there were any metric traits that could also be used to support the morphological assessment. As I’d mentioned above, portions of the left tibial shaft had accidentally been brought back to the lab, so I began searching for methods for estimating sex using the tibial shaft. Iscan and Miller-Shaivitz have a 1984 AJPA paper in which they took tibial measurements from individuals of European and and African ancestry using the Terry Collection. They used a discriminant function to estimate sex that incorporated tibial length, circumference at the nutrient foramen and anteroposterior and transverse dimensions at the nutrient foramen. As is customary when taking metric measurements I measured each dimension three times and then took the average (this helps to control for intra-individual variability in measurements). Measurements of AP and transverse diameter were taken using digital calipers, and measurements of circumference were taking using a plastic covered-cloth tape. All measurements were taken at the base of the nutrient foramen, and the maximum dimension in each direction was used.


First three measurements (mm)


Anteroposterior diameter

33.64, 33.87, 33.53 32.40
Transverse breadth

24.48, 24.24, 23.96


Circumference 90.3, 90.1, 90.2


Because the proximal and distal ends of the tibia were missing for this individual, it was not possible to estimate the maximum length of the tibia, or use their discriminant function analysis. However, Iscan and Miller-Shaivitz reported the means and standard deviations for all individuals in each population/sex subsample. Using these, I found that:

  • Circumference was more than 1SD away from the average male measurement, but within 1SD of the average female measurement.
  • Transverse breadth was more than 1SD from the average male measurement, but within 1SD of the average female measurement.
  • AP diameter was within 1SD for both males and females.

So while I was not able to use the discriminant function analysis, the tibial metrics suggest that this individual would have been within the female range. While in no way a decisive confirmation of my initial estimation, this individual does seem to have tibiae that were gracile enough to belong to a female. The metric data do not refute the non-metric estimation of this individual as female.

Figure 6. Full extent of midden excavationFigure 6. Full extent of midden excavation

What’s particularly fascinating about this example of bioarchaeology in action is that it illuminates a component of Yoruba history that other wise would have remained dormant. As Andrew said,

“The case is fascinating, as it points to a different context for the role of women in Shabe’s recent political history. Like many other parts of West Africa, Europeans actively encouraged the exclusion of women from political life as the older political institutions were transformed to serve colonial interests. Thus, the lack of titled women in the present may be a recent phenomenon and that prior to the 20th century women had larger political roles”.

 Which just goes to show – it never hurts to have a bioarchaeologist around!

Image Credits: All maps were drawn by Andrew Gurstelle, all photos of the midden burial are shared courtesy of Andrew Gurstelle. All figures detailing sex estimation techniques are taken from Buikstra and Ubelaker 1994.


Bass, William M.1994. Human Osteology: A Laboratory and Field Manual of the Human Skeleton. Fourth Edition. Special Publication No. 2. Missouri Archaeological Society. Columbia: Missouri.

Buikstra, Jane and Doulgas Ubelaker (eds) 1994. Standards for Data Collection from Human Skeletal Remains. Fayetteville: Arkansas Archaeological Survey Research Series No. 44.

ResearchBlogging.orgİşcan, M., & Miller-Shaivitz, P. (1984). Determination of sex from the tibia American Journal of Physical Anthropology, 64 (1), 53-57 DOI: 10.1002/ajpa.1330640104

This entry was posted in Archaeology, Human Remains, Osteology and tagged , , , , , , , . Bookmark the permalink.

2 Responses to Photographic analysis of a Shabe Yoruba burial

  1. fantastic post, cheers for writing this up!


  2. Pingback: Bone Broke Year in Review | Bone Broke

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.