| Abstract: |
Demographic processes in pre-industrial societies are difficult to determine, because of the lack of written records and the complications of reconstructing population structure from archaeological information. Paleodemographers use skeletal material to reconstruct the living populations, but these cemeteries, which consist only of those individuals who have died, are not necessarily a representative sample of the people who were alive. The main methodological issues in paleodemography are hidden heterogeneity, selective mortality, and demographic nonstationarity. The first chapter of this dissertation presents a theoretical four state model that has the potential to address the analytical problems of paleodemography. The model uses health markers from skeletons (lesions) as a fundamental element in understanding paleodemography. The second chapter introduces a reduced, and more readily interpretable, three state version of the model that emphasizes the relationship between lesions and selective mortality This model is tested successfully using simulated paleodemographic data. Sensitivity tests are performed on the sample sizes needed to implement the model, and robust results can be recovered with initial populations between 100 and 200 individuals. The third goal of this dissertation is to use this three state model to test the effect of selective mortality on lesions from skeletal material from Tirup, a medieval Danish cemetery. Chapter 3 introduces Tirup cemetery and explores ideas of why this population seems to differ from any modern population. Four hundred sixty-eight skeletons were studied, and the paleodemographic profile indicates that about half of the people buried at Tirup cemetery were children. The average adult age at death is very low, similar to many other paleodemographic sample. Chapter 4 uses the three state model to test the effect of several lesions on the risk of death. One lesion, incisor enamel hypoplasia, appeared to reduce the risk of death. Several lesions increased the risk of death: first molar enamel hypoplasia, cribra orbitalia, short femur, and femur infection. Having a canine enamel hypoplasia, humerus infection, maxillary first molar caries, metopic suture, or rhomboid fossa did not significantly influence the hazard. The final chapter presents future directions for this research.
|
