6533b838fe1ef96bd12a4d14

RESEARCH PRODUCT

Productions en série vers 1500 avant notre ère. Des règles de fabrication au Bronze moyen entre la Manche et les Alpes à la lumière d’une étude morphométrique.

subject

description

Some European Bronze Age objects were produced by what has been described as serial metalworking. One particularexample is the Middle Bronze Age palstave, massively produced and used in Western Europe during the mid-second millennium BC.These artefacts were often voluntarily buried together in hoards, meaning they were removed from the production network, thus avoidingany recycling. They are found intact, either as rough castings or ready for use. These homogeneous objects are grouped in sets of severalitems, or in tens, or even in hundreds. Such discoveries have immediately led to numerous questions as to the possible interpretation ofthis behaviour. It is clear that prehistoric craftsmen must have been seeking to reproduce the models they had designed, as faithfully aspossible. Macroscopic observations reveal a quest for the same general shape and ornamentation, suggesting great homogeneity in productionduring this period. Many examples have been found of palstaves that were produced from exactly the same mould.Considering the entire production as a whole, and regarding all palstaves as belonging to the same type, some disparities are neverthelessvisible, even to the naked eye, particularly with regard to shape. The question is therefore to discover what degree of consistencyexists among types usually identified only with the naked eye. It becomes necessary to measure the degree of determination required toduplicate so many objects over such a vast territory (up to several thousands of km²). The overarching question is to understand how theproduction of metallic objects was organised during the mid-second millennium BC, from the English Channel to the Alps.Macroscopic observations are no longer adequate to answer such questions. It has become necessary to concentrate on methodologicaltechniques commonly used in the life sciences. Mathematical analysis systems are indeed capable of discriminating between populationsaccording to shape. So far, two methods have been selected: 1) the Discrete Cosine Transform (DCT) and 2) orthogonal polynomials.These analysis techniques were used to translate naked eye observations into mathematical expressions. The two typological groupsconsidered in these studies are the Breton type and the Norman type, named after their areas of highest discovery density. Mathematicalanalysis confirmed the statistical validity of these two typological groups, but with an overlap in the morphometric space. This resultproves that each group was seeking to reproduce a specific model, and also confirms that these prehistoric populations were aware oftheir territorial affiliation: 1) the Breton peninsula, for the Breton type and 2) the Seine Valley, for the Norman type. These statisticalmethods also make it possible to quantify the variability present in each type. Shape thus acquires a geographic identity, giving rise to acultural identification, even if the objects differ somewhat from the original model.As the distance from the two high discovery density zones increases, some palstaves presenting a visual similarity to standard modelsare in fact mathematically identified as outliers, distant from the centre of the morphospace. This result raises the question of possiblelocal copies, or even imitations, in areas outside the two high discovery density zones, taken to be the two major manufacturing centres.To conclude, this example of prehistoric metal production illustrates that the rule in manufacturing seems to be the desire to get evercloser to a reference model, but that all copies do not have to be perfect.The degree of congruence does not seem to be the most important aspect, as long as the objects appear similar to the naked eye.