It is a pleasure to be invited to contribute to a debate on the nature of ‘structured deposition’, even though I do not have the expertise to comment on much of the detailed discussion of British prehistoric material presented in this excellent paper. The way in which my commentary approaches structured deposition is by posing two questions relating to the field of deliberate object fragmentation – an important aspect of many sites where structured deposition has been claimed (such as Kilverstone, Etton or Windmill Hill) – (a) can the identification of deliberate object fragmentation contribute to the understanding of specific deposits, sites or landscapes? (b) To what extent does the agency of humans and objects relate to structured deposition and deliberate object fragmentation?
The past social practice of deliberate object fragmentation (Chapman 2000; Chapman and Gaydarska 2006) related as much to ‘odd deposits’, such as the very fragmentary Gumelniţa ‘shrine model’ placed in a house on the islet of Căscioarele (Dumitrescu 1965), as to broader material culture patterning, such as the coarse pottery placed as grave goods at the Hungarian Copper Age cemetery of Tiszapolgár-Basatanya (Bognár-Kutzian 1963; Chapman 2000, 51–53; cf. incomplete decorated beakers in single graves: Woodward 2002). Balkan–Greek intra-site refitting has so far focused not so much on ‘odd’ types of object as on ‘common but special’ objects, such as fired clay figurines, shell bracelets and so-called ‘rhyta’. Insufficient research has been completed on the fragmentation of ‘mundane’ domestic pottery assemblages but such a programme is now in progress, this time focused on Iberian prehistory (Gonzales, in prep.). The issue to which Garrow alludes (p. 90) of using ‘odd deposits’ to help interpret material culture patterning is therefore very relevant to Balkan objects, since past interpretations of objects such as anthropomorphic figurines have inevitably coloured the meaning (if any) of structured deposition involving these things. It should be noted that, while the vast majority of objects were deposited in settlement contexts, the phenomenon of fragmentation is also well attested in the mortuary domain.
Many of the same interpretational debates concern both intentional fragmentation and structured deposition, particularly the probability of other reasons for fragments of striking objects appearing separate from the rest of the object. However, many of the potential causes of fragment deposition – accidental breakage, the distribution of ‘fertility’, the disposal of an object whose ritual powers are declining or finished or the deliberate destruction of an object to render it powerless – cannot readily account for finding only part of an object rather than the potentially refitting fragments of an object deposited whole.
The ‘fragmentation premise’ posits two practices: objects were regularly deliberately fragmented and the resulting fragments were often reused in an extended use-life ‘after the break’ (Chapman and Gaydarska 2006). There are now several examples of sites in which the parts of broken objects have led their own separate ‘lives’ – distinguished through use-wear analysis – before being brought together in final deposition (Varna and Dimini shell bracelets, Dolnoslav figurines: Chapman and Gaydarska 2006; cf. Iberian examples of Neolithic pottery: Gonzales, in prep.). But what are the implications for structured deposition of these fragments? It is always possible that figurine fragments could have been discarded at the end of particular ceremonies, just as other complete figurines may have been broken. But it seems improbable that two different parts of a once-whole figurine, with different life-histories, came together by chance in a specific context – especially if that context also contained other special objects. Thus the reintegration of object parts with different life-histories is a good instance of structured deposition in which four stages of an object's biography were presenced: the birth of a complete object, the fragmentation of an object, the reuse of different parts in different contexts of use, and the reintegration of the object as a part of the act of its final deposition.
The question of the social meaning of deliberate fragmentation has been enmeshed in the discussion of enchainment – the creation of links between people and things (and fragments of things). Ever since it was recognized that enchainment is an umbrella term covering a wide range of processes of relation building, the term has been problematized and, to some extent, it remains so even now (Brittain and Harris 2010). It seems reasonable to suggest that enchained links were just as germane to fragmentation processes as to structured deposition, insofar as the people making the structured deposition were positing relations between places, persons and things. Current research on the diversification of the interpretation of enchainment is predicated on the importance of revitalizing and extending this term for both structured deposition and fragmentation studies.
Two excellent examples of structured deposition were identified at Kilverstone (Garrow, Beadsmoore and Knight 2005) and Etton (Beadsmoore, Garrow and Knight 2010), even though this interpretation was only partly based on the basis of the refitting results at both sites. The implications of the Kilverstone and Etton refitting experiments showed contrasting patterns of fragmentation and deposition. The ditch segments at Etton must have been open for a long time – perhaps up to 500 years – with repeated visits but not such intensive deposition as at Kilverstone. At the latter, the pits often contained large, unweathered and unburnt sherds, in marked contrast to Etton, where the pits contained weathered, burnt sherds suggestive of the pottery being stored before deposition. There was a very strong inference drawn at both sites for the key social practice of deposition of material culture, whether pottery or flint, when a group moved onto a site and when a group abandoned the site. In summary, the ceramic assemblages on each of these settlements contained many, if not a majority of, orphan sherds, which represented between 2% and 25% of the vessel surface. It seems unequivocal that the deliberate fragmentation of vessels, not to mention lithics, contributed to the initial interpretation of these two sites as places of repeated structured deposition. The contrasting results of the two site refitting studies seem likely to indicate enchainment occurring on two different timescales. The material deposited at Kilverstone created enchained links between the site and other sites visited immediately before and after the Kilverstone occupations, since each pit group was dug at the same time and filled in quickly, perhaps within the time frame of a single visit. Here, the fragments presenced the living rather than the ancestors, marking out annual and multi-annual territories. At Etton, by contrast, the evidence for curation and complex treatment of the often weathered and burnt sherds suggests a greater time depth for enchained relations between the site and those visiting the site. These practices evoked the ancestors as much as the living, providing the enclosure with echoes of times long past. It is thus evident that we can answer the first question in the affirmative.
I now turn to the second question – the extent to which the agency of humans and objects related to structured deposition and deliberate object fragmentation. It seems to me that one major question underlying the debate over structured deposition is the extent to which people in the past were making their own conscious decisions which empowered them in their social world or, alternatively, were following their forms of habitus with little personal decision making in a robotic form of Bourdieu's Homo economicus. There is a distinctive, if not clearly articulated, view amongst some prehistorians that people in the past were much more like the latter than like the former and that any form of analysis that privileges the former is somehow illegitimate. Doubts such as this threaten Garrow's research aim of how to justify enhanced meaningfulness in material culture patterning.
It is, of course, fundamental to fragmentation research that all alternatives to deliberate fragmentation and deposition of fragments are explored before a conclusion in favour of that interpretation is reached. But if this procedure is satisfied through controlled argumentation, there is a presumption that not only the persons involved but also, potentially, the objects concerned have exercised some agency in the creation of enchained relations. What is not at all clear, however, is why we should construct a ‘symbolic’ interpretation of processes of fragment enchainment. Rather, everyday processes linking persons to persons, or objects to persons, would be better appreciated on their own merits without an a priori symbolic charge. However, this interpretational limitation would not be valid in the case of pit deposition displaying a strong symbolic message. Thus, in the case of the Hamangia Pit No 1 at Medgidia–Cocoaşe, on the Romanian Black Sea coast (Haşotti 1985; Chapman 2000), the use of whole as well as fragmentary objects to make a contrast between the fills and contents in the southern and northern parts of the pit enhanced the symbolic nature of the structured deposition. The agency of the ‘assemblage’ of varied raw materials constituting what must have been the full range, or close to the full range, of materials used in everyday Hamangia activities brought to life the productive capacities of the social group through citation of the raw material sources and the persons responsible for bringing them (back) to the Hamangia settlement for use and final consumption. The performance of placing the items of this assemblage in Pit No 1 focused attention on the persons playing important roles in the symbolic re-creation of Hamangia lifeways. In this instance, structured deposition and deliberate fragmentation reinforced each other in a striking symbolic practice. But there were many other cases where fragment enchainment lacked symbolic significance other than that inherent in forming social links. Only through a wide-ranging contextual analysis can we seek to make a proper evaluation of the extent of symbolic charge in any specific case.
In the same way that structured deposition is a sign of the alterity of the past (p. 114), it could also be argued that fragment enchainment indicates the specificity of past social practices. Although the object enchainment well documented in Melanesia (Strathern 1988) was one of the key insights leading to fragmentation theory in archaeology (Chapman 2000; Fowler 2004), Melanesian enchainment does not occur through broken objects, even if there are such instances known in the malangan statuary of New Ireland (Küchler 1988). It would appear that, wherever the fragmentation premise can be supported by clear local or regional evidence, there are new and different processes that cannot be understood purely with recourse to Melanesian ethnography. This is a further sign that both structured deposition and object fragmentation help to shape our views of a past radically different from a recent ethnographic past.
