Human Paleoecology during the Magdalenian in the Swabian Jura of Southwestern Germany

Abstract

The Swabian Jura is well-known as an important region for the study of Paleolithic archaeology. The Magdalenian, though, has received little attention in this region compared to other periods of the Paleolithic. Open questions, therefore, remain regarding regional subsistence strategies, settlement patterns, and environmental conditions during this time. There have been, for example, almost no modern, quantitative studies of environmental conditions during the Magdalenian that are specific to the Swabian Jura. In this dissertation, I use remains from Langmahdhalde, a rock shelter in the Lone Valley that is one of the first archaeological sites in the Swabian Jura with intact Magdalenian remains to have been discovered in decades. I use the faunal remains from the site to explore trends in human subsistence behavior, use of the rock shelter, local environmental conditions, and the resettlement of the Swabian Jura during the Late Glacial. To do so, I take four methodological approaches. First, I use traditional zooarchaeological analysis to understand human behavior. Second, I conduct stable isotope analyses on bone collagen of horse (Equus ferus) and reindeer (Rangifer tarandus) from the site. Third, I apply a model for reconstructing past environments, called the Bioclimatic Analysis (Hernández Fernández, 2001a, 2001b; Hernández Fernández and Peláez-Campomanes, 2005, 2003), to the microfaunal assemblage. Finally, I do a taphonomic analysis of the microfaunal assemblage. My results regarding human behavior during the Magdalenian in the Swabian Jura are consistent with current understandings of the Central European Magdalenian. The primary taxa at the site are hare (genus Lepus), small carnivores, reindeer, horse, and medium birds. There is evidence of butchery, marrow extraction, antler working, skinning, and needle making at the site. The results of the stable isotope analyses show that Late Glacial horses were more adaptable to local environments than reindeer and that their delta13Ccoll and delta15Ncoll values are better indicators of local environments than those of reindeer. Further, I document evidence of the loss of the preferred habitat of reindeer in the region. The Bioclimatic Analysis demonstrates that Late Glacial environments surrounding Langmahdhalde were generally open tundra but with more precipitation, warmer and shorter winters, and longer vegetative activity periods than modern tundra environments. The models also suggest that the landscape was mosaic in nature and likely even had stands of trees, indicating that the region was more heterogeneous than previously thought. This heterogeneity probably means that there was a higher diversity of plant and animal species on the landscape than in modern tundra regions. Finally, my taphonomic analysis of the microfauna indicates that in the majority of the horizons at Langmahdhalde, several predators, mostly species of owl, deposited the microfaunal remains. These predators include both opportunistic feeders and specialists who prefer specific prey. The microfaunal assemblage can, therefore, most accurately provide reconstructions of paleoenvironments when the presence or absence of taxa, not taxonomic abundance, is used in analyses. As the Bioclimatic Analysis uses presence/absence data, this suggests that the paleoenvironmental interpretations from the site are robust. Further, the hunting ranges of the predators responsible for the assemblage suggest that my paleoenvironmental interpretations apply to the Lone Valley and its surroundings. I end by arguing that the Swabian Jura (at least the Lone Valley) offered Magdalenian hunter-gatherers a greater diversity of resources than other regions, even those to the west, during this time. It is possible that this is one of the reasons that the resettlement of the Swabian Jura during the Late Glacial was successful.