Jostein Aksdal

This paper presents a multi-proxy approach to coastal Stone Age demography. It uses the district Hor-daland, western Norway as a case and applies the proxies SPD (summed probability distributions) of radiocarbon dates and stray find distributions. These are compared to pollen-based landscape reconstructions. Large numbers of Stone Age sites have been surveyed and excavated in western Norway during the last few decades, mainly because of modern development and cultural heritage management. This work has produced significant amounts of radiocarbon dates. The data has, until now, not been sufficiently organized and systematized for the purpose of doing research on long-term changes. The same is true for the many stray finds, which are stored at University Museum of Bergen. During the last decades, methodological development in palynology has made compilation of data and new vegetation reconstructions possible. For the first time, these dispersed datasets from the district Hordaland are brought together for comparative purposes, with a specific goal to study relative demographic changes. The hypothesis is that during the Stone Age, demographic change accompanied big cultural transformations in the transition from LM (late Mesolithic) to EN (early Neolithic) c. 5950 cal BP and between MN (middle Neolithic) and LN (late Neolithic) c. 4300 cal BP. This study partly supports the hypothesis, as the changes in the SPD and the stray finds during the transition to the late Neolithic clearly reflect marked population growth, related to the introduction of agriculture, at the same time as the pollen data reveal forest clearance. The LM-EN transition is less clearly connected to demographic change. Generally, up until the transition to the LN, the data indicate that there was gradual demographic growth with marked fluctuations within a forested landscape. Although the proxies sometimes co-vary for the different periods, they may also display conflicting patterns, and this strengthens the argument that a multi-proxy approach to demographic studies is to be recommended.