Volume 33, Issue 2 p. 147-167
Article

Iron age skulls from Southern Africa re-assessed by multiple discriminant analysis

G. P. Rightmire

G. P. Rightmire

Department of Anthropology, State University of New York at Binghamton, Binghamton, New York 13901

Search for more papers by this author
First published: September 1970
Citations: 33

Abstract

Multiple discriminant analysis of Bushman, Hottentot and Bantu Negro skulls allows construction of a multidimensional space or statistical framework within which various problematical Iron Age skeletal remains from southern Africa can then be located, and the question of their affinities with the extant groups examined. Comparison of subfossil crania from Zambia, Rhodesia, Botswana and South Africa with the modern population samples in a 27-variable discriminant analysis reveals the following:

  • (1)

    Six skulls from the Iron Age site K2 (Bambandyanalo) and nine tentatively associated with the Leopard's Kopje Iron Age complex of southwestern Rhodesia and Botswana are assigned as Negro, and many are in fact excluded at a 95% probability from being Bushman or Hottentot. Previous descriptions of these, and especially the K2, remains as “proto-Hottentot,” Hottentot, “large Khoisan” or, most often, as “Bush-Boskopoid” thus seem unjustified, at least on metrical grounds, and probably true Negroes had crossed the Limpopo into what is now South Africa by 1000 A.D.

  • (2)

    Assignments for the Ingombe Ilede (Zambia) skulls and for individuals unearthed in Harare “pot burials” near Salisbury are inconsistent; some of these skulls are almost certainly Negro, whereas others, though not excluded statistically from all Bantu distributions, are better classified as Hottentot. The (more recent?) Kakamas specimens from burials along the Orange River appear to constitute a mixed sample rather than one representative of some homogeneous “true Hottentot” population.

  • (3)

    Nearly all of the subfossil remains tested fall within the multivariate ranges of variation calculated for the several modern African groups. In view of this, attempts to discern in such material a variety of “types” or “strains” (e.g., “Bush-Boskop,” “Kakamas,” “Gerontomorph,” “Europoid,” etc.) might better be discarded in favor of a classification based on Bushman, Hottentot and Negro categories only. Multiple discriminant analysis demonstrates the existence of no other separate groups among the crania examined.

The full text of this article hosted at iucr.org is unavailable due to technical difficulties.