Title On Meroitic Nubian Crania, Fordisc 2.0, and Human Biological History Author(s) S. O. Y. Keita Identifiers Current Anthropology, volume 48 (2007), pages 425427 DOI: 10.1086/507184 Availability This site: PS | HTML | PDF (44.2k)
Posts: 951 | From: where rules end and freedom begins | Registered: Jun 2004
| IP: Logged |
posted
Yes, I have access to it. I don't have an account at any file hosting site, though. Here's the article in full below (from HTML version):
CURRENT ANTHROPOLOGY Volume 48, Number 3, June 2007
DISCUSSION
On Meroitic Nubian Crania, Fordisc 2.0, and Human Biological History
S. O. Y. Keita National Human Genome Center at Howard University, Room 615, 2041 Georgia Ave. NW, Washington, D.C. 20060, U.S.A. (skeita@genomecenter.howard.edu). 5 IV 06
Williams, Belcher, and Armelagos (CA 46:34046) have reported and discussed the results of a study of Meroitic Period Lower Nubian crania in which, using Fordisc, a forensic identification program, they analyzed 42 crania individually as unknowns in what can be described as a morphometric experiment employing forced allocation (Campbell 1984) that is based on phenetic affinity (McNeil 1978; Cain and Harrison 1960). The results were unexpected in terms of biogeography; the crania were not primarily affiliated with the nearest neighbor (an Egyptian series) but grouped (although not always with strong statistical support) with series from all of the continents and Pacific Oceania and a range of U.S. social census groups. These results deserve further comment with regard to their possible "meanings." Williams et al. seem to restrict their explanation to invoking the published finding of a greater within-group than between-group craniometric variation in one study and suggesting a role for nongenetic variation caused by the "conditions of growth" during development. They do not fully reconcile their claim that human biological variation is primarily clinal with the results of their study, which indicates a more complex situation. Clearly the demonstration of similarity using multivariate analyses does not always mean identity, close recent common origin, or even origin in an adjacent region.
Williams et al. anticipate and effectively refute the criticism that a major flaw of their study is its use of a group not present in the database, noting that a forensic worker can never really know if an individual's origin population is actually represented in the forensic tool's database. Technically, however, this criticism is correct and is grounded in the concerns of both forensic science and the best-use parameters of canonical variate or discriminant function analysis when used for identification. The goal of forensics is the actual identification of the unknown with a recognized social group known to law enforcement and not the discovery of relative similarity (phenetic affinity). The aspect of discriminant function analysis relevant to this discussion is its secondary use to minimize the probability of misplacing an unknown object, in this case a cranium, in the nonorigin population (Albrecht 1980). Given the goal of identification (Campbell 1978), the use of discriminant function analysis and its relative canonical variates for more than two groups (Albrecht 1980) requires in principle that unknowns belong to one of the groups in the analysis from which the functions were derived.
Williams et al.'s results are not to be dismissed and can be seen as "new" data. Fordisc can be used as a tool of comparative morphometric anatomy. It can be treated as an exploratory instrument whose database contains a range of craniometric patterns, rendering particular origins incidental. From this perspective the findings of Williams et al. can be said to illustrate anatomical craniometric diversity that invites explanation.
There are several perspectives that may help explain why the Meroitic Nubians were not "uniform" in their grouping. These perspectives employ both "history" in the narrow sense and biology (assuming no or little role for differences in measurement technique, which would apply to every forensic worker and are an issue for the sources of the base data in Fordisc). Meroitic Nubia was a complex state-level society that had long-standing interactions with its neighbors. This polity chronologically overlapped with the Graeco-Roman period in Egypt. Contact with outsiders is well documented (see Snowden 1970; Harris 1977; Burstein 1997). Although it is possible that Europeans, Egyptians (and other Africans), and Middle Easterners actually lived and procreated in Nubia, this is not true for the Australian, Taiwanese, and Japanese "aboriginals" (the last two represented by the Atayal and Ainu series) with whom some Meroitic Nubian crania are affiliated in the study. Interestingly, Angel and Kelly (1986, 56) reported that a Wadi Kubbaniya Paleolithic skeleton from Nubia had an "Australoid" facial appearance that was "not dissimilar" to certain Upper Paleolithic remains from Europe, while simultaneously being similar to contemporary "US Blacks" in a range of traits.
What would account for this range of resemblancesinfraspecific convergence, parallelism, admixture, chance, or all of these? It is perhaps best to consider these findings as reflective primarily of an indigenous northeastern African biological evolutionary history and diversity. Hiernaux (1975) reports that the range of values in selected metric traits from populations in the northeastern quadrant of Africa collectively largely overlaps the range found in the world. Given that this region may be the place from which modern humans left Africa, its people may have retained an overall more generalized craniometric pattern whose individual variants for selected variables may resemble a range of centroid values for non-African population values.
Reply
FRANK L'ENGLE WILLIAMS, ROBERT L. BELCHER, and GEORGE J. ARMELAGOS Department of Anthropology, Georgia State University, P.O. Box 3998, Atlanta, GA 30303 (frankwilliams@gsu.edu) (Williams)/Department of Social Science, Atlanta Metropolitan College, 164 Metropolitan Parkway, Atlanta, GA 30310 (Beleher)/Department of Anthropology, Emory University, Atlanta, GA 30322 (Armelagos), U.S.A. 7 VI 06
We appreciate Keita's comment on our analysis of Nubian crania in Fordisc 2.0, a program that uses discriminant functions based on a sample of cranial traits from known "populations" to attribute biological affinity to unknowns. We tested the efficacy of Fordisc 2.0 using a sample of 42 Meroitic Nubian crania. Fordisc 2.0 failed to cluster the Nubians and did not consistently attribute them to their nearest neighbor (ancient Egyptians) or other African populations. The use of Nubians was heuristic, given the geographical and historical nature of the people represented by these remains, although we anticipate similar results from Fordisc 2.0 using other human populations. Our analysis was challenged by Freid and colleagues (2005), who claimed that we failed to apply the methodology correctly and that since Nubians were not part of the reference population we should not have expected successful classification. Naar, Hilgenberg, and Armelagos (2006) responded by using one of the reference populations (Egyptian) and found that Fordisc 2.0 was problematic in the assessment of that group. Of 111 Egyptians, only 55 (49.5%) were correctly identified using a typicality of greater than 0.1 and a posterior probability greater than 0.5 as recommended by Freid and coworkers. Sixteen (14.4%) were classified as Egyptian but did not meet the statistical test with respect to typicality and posterior probabilities. Fourteen were classified as other than Egyptian with significant probability and typicality statistics. Twenty-six (23.4%) were misidentified but without a significant result.
Keita now criticizes us for a paper that we did not write. Specifically, he critiques our discussion of noninherited factors, namely, the conditions of growth, which we cite as contributing to craniofacial variation. Nutrition, disease, and ecology are well documented to influence patterns of growth and development across human groups (Boas 1912; Whiting 1958; Baker 1969; Frisancho 1970; Eveleth and Tanner 1990; Armelagos and Van Gerven 2003). No one has yet shown that the conditions of growth do not affect head (and body) shape and size, although one of the architects of Fordisc 2.0, along with another researcher, argues that cranial form is largely invariant across human history (Sparks and Jantz 2002) despite overwhelming evidence to the contrary, including studies of immigrants and their descendents (Boas 1912; Shapiro 1939; Bogin 1988; Gravlee, Bernard, and Leonard 2003). These issues have been dealt with by others using Nubian populations that lived along the Nile for the past 10,000 years (Carlson 1976; Carlson and Van Gerven 1977; Van Gerven, Armelagos, and Rohr 1976; Carlson and Van Gerven 1979; Van Gerven 1982; Goodman et al. 1986; Calcagno 1986).
The clinical nature of human variation is not expressed in Fordisc 2.0, and Keita is correct that individuals are essentially forced into one of the populations, or racial groups, within the program. Keita is also correct that we do not resolve the issue of polymorphism within human groups. However, other studies have shown that within-group variation is far greater than between-group variation (Lewontin 1972; Relethford 1994; Marks 1995). Our results simply confirm the heterogeneity of Meroitic Nubia and the lack of fit between conceptual models in osteological research and actual patterns of human biological variation.
Finally, accepting Fordisc 2.0 as an exploratory tool, Keita interprets our results to suggest that the heterogeneity of Northeast Africa generally reflects Pleistocene dispersals "out of Africa," which presumably occurred approximately 200,000 to 100,000 years BP (Stringer and Andrews 1989). Keita cites the typological studies of Angel and Kelly (1986) and Hiernaux (1975) to support his view that the heterogeneity of our sample reflects a worldwide representation of human variation. Rather than attribute the pattern to a single prehistoric event, we prefer to acknowledge the complexities of human history in Northeast Africa and elsewhere, where migrations and gene flow repeatedly occurred out of Africa, into Africa, out of Eurasia, and into Eurasia from prehistoric to recent times.
What our results were intended to show is that Fordisc is a poor predictor of population affinity using either the Howells data set or the Forensic Date Base because the heterogeneity of a single population is not captured by the analysis of craniofacial traits that the program employs. We also maintain that the program is conceptually flawed because human variation is an evolutionary phenomenon that is roughly distributed along geographical gradients with a predominant signal of greater heterogeneity within rather than between groups. The populations and races within Fordisc 2.0 differ from real biological units such as species because human groups are not constrained by prezygotic or postzygotic boundaries. The complexities of human biological history, including those exhibited by ancient Nubians, will never be adequately resolved using present-day artificial groupings of individuals that, at best, only weakly capture the diversity present within the linguistic, national, or racial entities they supposedly represent. And, finally, the objective of Fordisc 2.0 is not only to identify unknowns; by using the program (which often yields incorrect classifications) practitioners may be inadvertently reinforcing contemporary epistemologies that grossly misinterpret human biological diversity.
References Cited
* Albrecht, Gene, 1980. Multivariate analysis and the study of form, with special reference to canonical variate analysis. American Zoologist 20:67993. First citation in article * Angel, Lawrence J., and Jennifer O. Kelly. 1986. Description and comparison of the skeleton. In The Wadi Kubbaniya skeleton: A late Paleolithic burial from southern Egypt, vol. 1, ed. F. Wendorf and R. Schild, 5377. Dallas: Southern Methodist University Press. First citation in article * Armelagos, George J., and Dennis Van Gerven. 2003. A century of skeletal biology and paleopathology: Contrasts, contradictions, and conflicts. American Anthropologist 105:5364. First citation in article | CrossRef * Baker, Paul. 1969. Human biological variation as an adaptive response to the environment. In Evolutionary anthropology, ed. H. Bleibtreu, 30521. Boston: Allyn and Bacon. First citation in article * Boas, Franz. 1912. Changes in the bodily form of descendants of immigrants. American Anthropologist 14:53062. First citation in article | CrossRef * Bogin, Barry. 1988. Rural-to-urban migration. In Biological aspects of human migration, ed. C. Mascie-Taylor and G. Lasker, 90129. Cambridge: Cambridge University Press. First citation in article * Burstein, Stanley, ed. 1997. Ancient African civilizations, Kush and Axum. Princeton: Markus Wiener. First citation in article * Cain, A. J., and G. A. Harrison. 1960. Phyletic weighting. Proceedings of the Zoological Society of London 135:131. First citation in article * Calcagno, James M. 1986. Dental reduction in Post-Pleistocene Nubia. American Journal of Physical Anthropology 70:34963. First citation in article | CrossRef * Campbell, N. A. 1978. Multivariate analysis in biological anthropology: Some further considerations. Journal of Human Evolution 7:197203. First citation in article | CrossRef * . 1984. Some aspects of allocation and discrimination. In Multivariate statistical methods in physical anthropology, ed. G. N. Van Vark and W. W. Howells, 17792. Boston: D. Reidel. First citation in article * Carlson, Donald S. 1976. Temporal variation in prehistoric Nubian crania. American Journal of Physical Anthropology 45:46784. First citation in article | CrossRef * Carlson, Donald S., and Dennis P. Van Gerven. 1977. Masticatory function and post-Pleistocene evolution in Nubia. American Journal of Physical Anthropology 46:495506. First citation in article | CrossRef * . 1979. Diffusion, biological determinism, and biocultural adaptation in the Nubian Corridor. American Anthropologist 81:56180. First citation in article | CrossRef * Eveleth, Phyllis B., and James M. Tanner. 1990. Worldwide variations in human growth. 2d edition. Cambridge: Cambridge University Press. First citation in article * Freid, Donna, M. K. Spradley, Richard L. Jantz, and Steve D. Ousley. 2005. The truth is out there: How NOT to use FORDISC. American Journal of Physical Anthropology 128 (S41):103. First citation in article * Frisancho, A. Roberto. 1970. Developmental responses to high-altitude hypoxia. American Journal of Physical Anthropology 32:41017. First citation in article * Goodman, Alan H., George J. Armelagos, and Dennis P. Van Gerven. 1986. Post-Mesolithic craniofacial and dental evolution in Sudanese Nubia. In Science in Egyptology, ed. R. David, 20112. Manchester: University of Manchester Press. First citation in article * Gravlee, Clarence, H. Russell Bernard, and William Leonard. 2003. Heredity, environment, and cranial form: A reanalysis of Boas' immigrant data. American Anthropologist 105:12538. First citation in article | CrossRef * Harris, Joseph E., ed. 1977. Africa and Africans as seen by classical writers. (William Leo Hansberry African History Notebook, vol. 2.) Washington, D.C.: Howard University Press. First citation in article * Hiernaux, Jean. 1975. The peoples of Africa. New York: Scribner. First citation in article * Lewontin, Richard. 1972. The apportionment of human diversity. In Evolutionary biology, vol. 6, ed. M. Hecht and W. Steere, 38198. New York: Plenum. First citation in article * McNeil, J. 1978. Purposeful phenetics. Systematic Zoology 28:46582. First citation in article | CrossRef * Marks, Jonathan. 1995. Human biodiversity. New York: Walter de Gruyter. First citation in article * Naar, Nichole A., Dana Hilgenberg, and George J. Armelagos. 2006. Fordisc 2.0 the ultimate test: What is the truth? American Journal of Physical Anthropology 129 (S42):136. First citation in article * Relethford, John. 1994. Craniometric variation among modern human populations. American Journal of Physical Anthropology 95:5362. First citation in article | CrossRef * Shapiro, Harry. 1939. Migration and environment. New York: Oxford University Press. First citation in article * Snowden, Frank. 1970. Blacks in antiquity: Ethiopians in the Greco-Roman experience. Cambridge: Harvard University Press. First citation in article * Sparks, Corey, and Richard Jantz. 2002. A reassessment of human cranial plasticity: Boas revisited. Proceedings of the National Academy of Sciences, U.S.A. 99:1463639. First citation in article * Stringer, Chris B., and Peter Andrews. 1989. Genetic and fossil evidence for the origins of modern humans. Science 239:126368. First citation in article | CrossRef * Van Gerven, Dennis P. 1982. The contribution of time and local geography to craniofacial variation in Nubia's Batn el Hajar. American Journal of Physical Anthropology 59:30716. First citation in article | CrossRef * Van Gerven, Dennis P., George J. Armelagos, and Arthur Rohr. 1976. Continuity and change in cranial morphology of three Nubian archaeological populations. Man 12:27077. First citation in article | CrossRef * Whiting, Marjorie Grant. 1958. A cross-cultural nutrition survey of 118 societies representing the major cultural and geographical areas of the world. Ph.D. diss., Harvard University. First citation in article * Williams, Frank L'Engle, Robert L. Belcher, and George J. Armelagos. 2005. Forensic misclassification of ancient Nubian crania: Implications for assumptions about human variation. Current Anthropology 46:34046. First citation in article | Full Text
Posts: 1024 | Registered: Jun 2006
| IP: Logged |
quote:What our results were intended to show is that Fordisc is a poor predictor of population affinity using either the Howells data set or the Forensic Date Base because the heterogeneity of a single population is not captured by the analysis of craniofacial traits that the program employs. We also maintain that the program is conceptually flawed because human variation is an evolutionary phenomenon that is roughly distributed along geographical gradients with a predominant signal of greater heterogeneity within rather than between groups
UBBFriend: Email this page to someone!
Printer-friendly view of this topic