In his paper on structured deposition the author exclusively focuses on the Neolithic, Bronze Age and Iron Age archaeology of Britain. He shows the wide range of discussions from the the key paper by Richards and Thomas in 1984 to the most recent papers of the ‘hyperinterpretive turn’.
As the author correctly mentions, metalwork hoards have usually not been discussed under the banner of structured deposition. This is especially true for archaeological discourse in Germany. The English term ‘deposit’ is different from the German term Depot. By definition a Depot consists of a minimum of two objects which were deliberately put on the ground or dug into the earth. This term covers either Bronze Age hoards or Roman coin hoards: everything what was called ‘treasure’ in the 19th century. The term Deponierungen was originally used as an umbrella term for hoards, singly deposited objects, river finds and so on, and is also different from the term ‘deposition’.
Nevertheless the discussion has several connecting points. Bronze Age hoards have been discussed for many years either as hidden treasures or as ritual deposits. The division was quite simple and was already in use in the 19th century in Denmark. Finds with complete objects from bogs were seen as votive offerings, finds with scrap material as founders' hoards.
A fresh view arose with the use of structural methods. The individual elements of a hoard were examined with reference to their relation to all other elements within the hoard, or to the entire system, including single finds from rivers and grave goods. From this, conclusions were drawn concerning the meaning and function of the individual elements as well as those of the entire system. Following the method of Claude Lévi-Strauss promised to find ‘necessary relations’ between elements which were not the results of anecdotal coincidence but reflected a system behind which insightful coherence should be visible. Lévi-Strauss spoke of syntagmatic chains and paradigmatic rows in his analysis of myths. This approach opened up the field for new perspectives on the phenomenon. It became clear that hoards were structured. They did not contain just ‘anything’ but a certain selection of things which was highly significant in space and time. Furthermore, certain hoard models existed which spread through Europe in several waves.
Discussing structured depositions, scrap metal hoards are the most interesting cases. The fragmentation of objects in many hoards is perhaps the crucial point within the history of the interpretation of hoard finds. Traditionally they were seen as founders' hoards. The used objects were fragmented to melt them down, were hidden in pots or other containers and remained in the earth for different reasons – war or some other crisis. Already in the 19th century scrap hoards were seen as remains of an international scrap metal trade (Sherratt 2012).
Through several studies of the early 1990s it became clear that also ‘scrap metal hoards’ have a certain structure. They contain a specific selection of weapons, tools and ornaments in certain proportions. It seems normal that they were deposed in a structured way, such as with ingots, weapons, tools and ornaments about each other in layers. The objects were fragmented according to certain rules. The weight of the fragments in the hoards depends on the region. In western Hungary, fragments are heavier than in Bohemia. In central Germany the fragmentation of objects was made according to standard weight, which has been attributed to a premonetary system (Sommerfeld 1994). In other cases fragments were seen as the result of ritual violence (Nebelsick 2000). Fragments were sometimes collected in socketed axes to join them together for uncertain reasons. There is, however, evidence that fragmentation should be perceived as having taken place within complex (ritual) activity (Hansen 1996–98). Moreover, it is a striking fact that fragmentation was not connected with Early Bronze Age hoards but started in the transition from the Early to the Middle Bronze Age. Fragmentation was part of a new conception of the hoard which was quite successful. It consisted of fragments of ingots, weapons, tools and ornaments and became widely distributed and was still in use in some regions in the Iron Age (Hansen 2005).
There is in my view no longer any reason to speak of scrap metal hoards (Brucherzdepots). The fragmentation of objects in hoards has proven to be a cultural feature that is an integral component of the meaning and purpose of deposition. A recent study has shown regularity in the reuse of bronze votive offerings in the sanctuary of Olympia. Larger parts of bronze tripods, the most prestigious of votive gifts in the Geometric Period, were melted down; other single pieces were saved from being reused and were preserved in wells or in other heaps of votive debris within the sanctuary. Keeping these fragments in possession of the deity was possibly understood as pars pro toto (Kyrieleis 2006, 97).
Pattern in archaeological finds can always be understood in several ways. The regularity of fragment weights can be understood as the standardization of the votive gift. One could argue that fragmentation for practical reasons like remelting would lead to standardized results as well. Repetition is part of ritual activity as well as of standardized craftmanship.
The structure itself is not the explanation of the deposition. A degree of contextual information is necessary to understand a single observation of structured deposition. Structural analysis is not a mathematical equation whichs leads to identical results. It has to be adapted to different conditions and different contexts.
