1Survival and Longevity, Max Planck Institute for Demographic Research, 2Anthropological Database (ADBOU), University of Southern Denmark
Thursday All day, Park Concourse
This paper is about modelling the rate at which events occurred from an archeological context. We employ the method for seven Mediaeval Scandinavian cemeteries to estimate yearly burial rates: the yearly number of burials in a community. The estimated rate is a proxy for the changes in population size of the community analysed.
The archaeological indicator used is the way the lower arms were placed at the burial (arm position). The dating of the different positions is known. Observed frequencies of the various positions are used to estimate a temporal probability density function by Maximum Likelihood Estimation. The expected number of burials per year was calculated twice: once with the parameter values from the fitted model and once with an assumed constant burial speed.
The populations analysed comprise different types of communities. The rural graveyards belonged to two small farming villages, and a monastery with two socially different groups of burials. The four urban cemeteries were used by a parish in a large town, an economically stressed suburban population, and two possible leprosaria. All communities experienced fluctuations in the burial rate, and show peaks at different points in time.
The estimated beta distribution fits the observed values better than a flat distribution. Arm position is a good indicator, but cannot define an exact time point at which a burial happened. Short-term events, like a single epidemic, will not be captured by changes in the burial rate; still the measure is sufficient to display long term trends in population fluctuation.