6533b835fe1ef96bd129f0ee

RESEARCH PRODUCT

The Bronze Age in Lorraine: a proposed model of the settlement

subject

description

Thirty years of assiduous preventive archaeology practice in Lorraine have built up a stock of data that can be used for numerous archaeological problems with a spatial focus.For the Bronze Age, as for the other chronological periods, the archaeological occupations discovered during diagnostics and excavations are strongly correlated with current developments (motorways, TGV, housing estates, quarries, etc.). These occupations must be discussed in order to estimate their spatial representativeness. Similarly, the landscape characterisation, in which the occupations highlighted are situated, is an important step in defining the types of settlement.As most of the data comes from the national 'Bronze inrap' survey, the comparison of the general dynamics (Bradley et al., 2016) with the Lorraine dynamics, using the sums of probability distribution (SPD) method, makes it possible to highlight some very specific phenomena (decline around 2200 BC, very strong rise around 1250 BC, etc.). The trend observed broadly follows the trend observed in north-western Europe.The characterisation of the data also requires a critical look. The chronological, attribute or spatial data must lead to a standard, as subsequent processing will affect the totality of Bronze Age occupations. The aim is to define types of occupation and their evolution in Lorraine and over time.Thus, in spite of the weaknesses that have been detected, some global treatments have been undertaken and make it possible to form an image of the characteristics of occupations during the Bronze Age. What types of landscape hosted what types of occupation; how far from the main rivers; which plateaus were colonised?Seriations (using a simple presence/absence matrix) make it possible to propose an initial hierarchy of habitats, while the Correspondence Factorial Analysis (CFA) / Hierarchical Clustering on Principal Components (HCPC) pairing attempts to characterise the types of occupation in greater detail (without necessarily being successful).If we were to summarise the observations made, we observe a long stagnation of occupations, mainly confined to the Meuse and Moselle valleys throughout the Early and Middle Bronze Age. A few occupations ventured beyond the valleys during this period. Then, around 1250 BC, there was a very strong increase in the number and density of settlements, which sometimes extended far beyond the main valleys. Throughout the Final Bronze Age, the curve, although irregular, remained at a very high level, well beyond its "vegetative" state at the beginning of the period.One might think, after examining the curve of probability densities, that occupations, after a long phase of sleep, suddenly wake up and show a dynamism never seen before. Perhaps this is a paradigm shift that leads to an impassable ceiling. A ceiling, or rather an asymptote, which then continues during the Hallstatt period.Finally, a few avenues are envisaged to try to explain this unprecedented explosion of habitats, such as the link that these occupations have with the plant economy, and in particular, the generalisation of millet in the cultivated species.These impressions, which are still uncertain and rather fragile if we look closely at the initial data, deserve renewed attention. It is therefore not with a conclusion that we should end this rapid overview, but with perspectives for future work. These prospects should cover very broad fields of study, of which a brief inventory can be made:- The representativeness of the archaeologically explored areas, and how to introduce the Bronze Age occupations into a scheme that takes into account the stripped surfaces, the degrees of erosion, or the nature of the developments (implying certain biases of observation...)- The definition of habitat types. It must be possible to "fill in" the building plans! To be satisfied with the drawing of the post holes seems, at the moment, too reductive. Geophysics on stripped ground has proved its worth. Phosphates make it possible to assume stabling areas, magnetism, forge work etc. In the future, these approaches will have to be systematised as a complement to 'conventional' excavation operations- Dating and the Bayesian approach. Being satisfied with the typo-chronology of the ceramics is too reductive. A systematic radiocarbon dating policy must be considered, itself placed in a Bayesian event model. The dating and phasing of the occupations would gain in quality and give as a result a curve, with a probability surface, and no longer a simple 'flat' interval [lower date - upper date] however precise it may be.