Biology Reference
In-Depth Information
someone who was in their twenties at death will be smaller than for someone who was in
their sixties at death. This is why we refer to our analysis of age-at-death as an estimation,
rather than as a determination.
Age estimation plays a large role in biological anthropology. In bioarchaeology and paleo-
demography, age estimates for populations of skeletons can give us insight into the health of
the population (see Smith [Chapter 7] and Konigsberg and Frankenberg [Chapter 11], this
volume). They are indicative of growth processes as well as disease processes. Such inferences
regarding health were previously made on the assumption of a simple relationship between
the ages-at-death and the health of the living population.
Wood and colleagues' (1992)
seminal
paper The Osteological Paradox: Problems of Inferring Prehistoric Health from Skeletal Samples ques-
tions (among other things) the practice of using skeletal information to make inferences about
the health of a population, given that the skeletons represent the individuals who died, not the
individuals who were healthy and lived (or at least those able to fight off disease). Discussions
of the osteological paradox
2
and resulting research have been a driving force of the major
improvements in statistical techniques for age-at-death estimation over the past several
decades (e.g.,
Boldsen et al., 2002
). These techniques, namely transition analysis and various
multifactorial methods (statistical combination of several age estimates), are discussed here.
In biological anthropology, age-at-death estimation is a critical part of the biological
profile. At least a cursory estimation of age is required upon the initial examination of human
remains to determine if the individual was an adult or a juvenile because methods for adults
(assessing degenerative changes) differ from those used for juveniles (assessing develop-
mental changes), as mentioned above. Several different areas of the skeleton demonstrating
age-related changes have been utilized to create age-at-death estimation methods, although
some have become much more ingrained in practice than others. Refining and improving
techniques, as well as developing new and better methods, is still an area ripe for research.
Currently, most biological anthropologists use “phase” methods like
Suchey
e
Brooks (1990)
withmeans and standarddeviations although the consensus seems to be that Bayesian analysis
is a more appropriate way to estimate age-at-death. Bayesian analysis uses prior information
(for example, the agedistributionof apopulation) toprovidemore robust age-at-death estimates
for particular individuals or populations. This kind of analysis is compatible with phase
methods; it ismerely a different way of approaching the calculation of the age-at-death estimate.
In this chapter I will introduce subadult age estimation, beginning with a review of bone
growth, followed by a discussion of the methods used for estimating age from juveniles.
Adult age estimation methods will be treated in depth, including consideration of multifac-
torial methods (see
Table 3.1
), and the chapter will conclude with a case study illustrating the
utility of Bayesian and transition analysis to the multifactorial age problem.
SUBADULT AGE-AT-DEATH ESTIMATION
Humans are characterized by a life history that includes altricial young and a distinctly
prolonged postnatal ontogeny (
Zeveloff and Boyce, 1982
). This period of development, while
2
See also DiGangi and Moore (Chapter 1) and Smith (Chapter 7), this volume.
