The importance of pre-depositional processes

All five of the respondents picked up, in one way or another, on the important role that processes prior to deposition have to play in ‘structuring’ the material culture actually deposited, and hence in our understandings and interpretations of structured deposition. I was very glad about this, as this single point was perhaps the most important element I wished to emphasize within my paper. As Chapman notes, citing the examples of Etton and Kilverstone, different settlement practices and rhythms of occupation resulted in quite different patterns of deposition at those two sites. Similarly, as Fontijn has discussed in detail within his own previous work (Fontijn Reference Fontijn2002) and close to the start of his response here (p. 121), what eventually comes to be deposited is only ‘the last part of a longer sequence of acts’. Those acts very much influence what is deposited how and where. Pre-depositional processes are vital to any understanding of deposition.

The notion of habitus and practice

Several of the respondents picked up on the issue of how ‘practice’ comes to be represented materially: the relationship between structure and agency caught up in the notion of ‘habitus’, and by association the knowledgeability and intentionality which lies behind any human action, and the material patterns which result from those actions. Fontijn, for example, states (p. 123) that ‘material culture patterning implies that any society has preconceived ideas of where and how to do particular things’. Similarly, Chapman implies that my conceptualization of the causes of material culture patterning tends towards a view which characterizes people as ‘following their forms of habitus with little personal decision making in a robotic form of Bourdieu's Homo economicus’ (p. 132). Thomas suggests (p. 126) that my paper could have focused more on habitual practice and the way in which material culture comes to be patterned as a result of (unconsidered) symbolic orders and conceptual schemes.

In relation to these discussions of practice, I agree with Thomas that I could perhaps have discussed the unconsidered material reproduction of symbolic schemes more, but broadly disagree with Fontijn and Chapman. In relation to Fontijn's point, I agree that material culture patterning certainly can come about as a consequence of society's symbolic beliefs and culturally specific norms, as Moore (Reference Moore and Hodder1982; Reference Moore1986), for example, so clearly showed. However, it does not have to come about as a result of these; as stated in the main paper, variability in the archaeological record (ancient or modern) can just happen (more on this below). In relation to Chapman's point, it is important to stress that an argument which makes a case for material culture patterning having been caused by the ‘mundane’ practices of everyday life certainly does not have to imply that people necessarily always behaved in a mundane, economic and/or boringly rational way (again, more on this below).

In his discussion of ‘practice, agency and intentionality’ Thomas mentions (p. 126) (alongside Bourdieu's notion of habitus) the distinction drawn by Giddens between discursive and practical consciousness – a distinction that I have always found helpful to work with. As stated above, I agree that I perhaps passed too quickly over non-discursive or ‘practical’ elements of practice (which nevertheless were influenced by symbolic orders and conceptual schemes) of the sort that could have led to material culture patterning. Again, Moore's work showed just this kind of thing – the Marakwet people's rubbish-disposal patterns were largely unconsidered (i.e. they did not consciously invoke or reference the symbolic order every time they disposed of something), but nonetheless were very clearly influenced by the ‘symbolic’ schemes of their society (dung, associated with the fertility of goats, could not be mixed with ash, associated with women, since goats and women represented two different and opposed types of fertility – Moore Reference Moore and Hodder1982, 78; see also figure 5). I too think that comparable beliefs probably did often lead to material culture patterning in the past as well.

The reason why I did not dwell on this aspect of depositional practice in more detail is that, ultimately, I took the fact that the types of patterning Moore describes would have come about in the past largely as a given, which did not really need debating since all sides would broadly agree. The main issue I wanted to focus on was what I see as an overemphasis within many discussions of structured deposition on patterns created as a result of discursive consciousness – material culture patterning created intentionally and explicitly, as a kind of ‘text’. As a counterbalance to this tendency, I wanted to stress the fact that material culture patterning and variability could come about purely through non-discursive practice, and that it could come to be patterned without even being influenced by any underlying symbolic scheme. Once we manage to stop seeing all (or, at least, much) material culture patterning as the result of people in the past consciously constructing highly symbolic material-culture texts, and more regularly consider the possibility that it was largely unintended and unintentional (but nevertheless still meaningful – both to them and to us – in terms of practice), time and space should be freed up for the discussions which Thomas rightly feels are not made in detail within my paper.

The role and validity of oppositions

A number of the respondents also picked up on my use of opposed concepts to frame the debate. Thomas (p. 125), for example, suggests that the main opposition used (odd deposits versus material culture patterning) easily slides into an opposition ‘between odd and everyday, meaningful and meaningless, ritualized and non-ritualized, and so on’; I can certainly see what he means. Similarly, Fontijn questions (p. 123) the helpfulness of employing terms like ‘ritual’ and ‘everyday’. Berggren also touches on similar issues at various points in her response.

In the following section, I would like to defend the use of at least some of these oppositions. The main pair of oppositions that I actually was intending to employ were ‘ritual versus everyday’ and ‘odd deposits versus material culture patterning’. As stated in the main paper, I do of course recognize that such oppositions might be viewed as problematic, especially if it is ever assumed that they had meaning in terms of people's perceptions in the past (see Brück Reference Brück1999b and Bradley Reference Bradley2005 for discussions of these issues). Equally, I understand that for some the use of any such opposition does not perhaps conform closely enough to postmodern conceptualizations of the fluidity of categories and meanings in the present. But, even now – having thought hard about these issues whilst writing the paper, and again in responding here – I do stand by my use of these oppositions. I feel that they are useful and help to frame a complex debate.

In relation to the ‘ritual versus everyday’ opposition, it is worth noting at the outset that ‘ritual’ is of course a term which always proves difficult to define. In discussing ritual within his response, Thomas (p. 127), for example, refers to Lewis's definition, whilst Berggren (p. 119) chooses to use Bell's more recent, but not entirely dissimilar, definition. The debate over ritual versus rationality was not something I particularly wanted to get into in the paper. As Bell put it (Reference Bell1992, 69), ‘a good deal of writing about ritual involves extensive exercises in cleaning up all the data and terms that are not included in the main definition . . . the nearly-but-not-quite-ritual behaviour’. I fully recognize the points made by Brück in her Reference Brück1999 paper on the subject, particularly in relation to the fact that people in the past would not necessarily have drawn a distinction between ritual and rationality in their own lives. However, turning back to Bell – whose fairly fluid and context-specific definition of ritual is very helpful, especially since it does not oppose ritual to rationality – I do still think that certain acts in the past would have been ‘ritualized’ and others (those which I have termed ‘everyday’ and which Bell would call ‘quotidian’) would not. Confusingly, these could even be physically and materially the same acts, performed in different contexts. The helpful ‘tip-of-the-iceberg’ metaphor which Thomas (p. 125) refers to in his response – where clearly odd deposits helped archaeologists to recognize a larger, more hidden body of meaningful depositional practice – is another way of viewing essentially the same thing. He is focusing more on how well we are able to identify ‘meaningful’ deposition in the present, rather than necessarily on what people felt about a deposit in the past. However, it is important to stress that the visible ‘ritualized’ tip of the deposition iceberg itself ends deep underwater at a non-ritualized (or ‘everyday’) base.

To illustrate the point I am trying to make in relation to the ‘odd deposits versus material culture patterning’ opposition (and again bearing in mind Bell's definition of ‘ritualized’ action), sometimes in prehistory people would have deposited things with accentuated ceremony – this would often have led to odd deposits but could equally have led to material culture patterning as well. At other times, however, people would have deposited material culture without any accentuated ceremony, but nonetheless influenced by cultural rules and conventions; in this case, we can gain insight into the latter by investigating the material culture patterning created as a result. Finally, it is important to stress that sometimes material culture entered the archaeological record without any such explicit rules affecting what was deposited when and where. It is this point, I think, which people find problematic in my argument, and so I will try to explain better what I mean. In the Kilverstone pits, for example, various different materials were deposited, but exactly what was deposited in each pit seems to have depended simply on what was available in the pre-pit context at the time that pit was filled. Cultural conventions as to what should be deposited where did not come into it. Any variability or patterning within the pits’ contents was created prior to the act of deposition, by the ebbs and flows of ‘everyday’ practice (which would of course have been very much affected by cultural rules and conventions). In saying this, I am not suggesting that the act of depositing material in a pit was not meaningful (or even ‘symbolic’ or ‘ritual’, if we choose to use those terms) – it almost certainly was. But I am saying that the spatial prevalences and contextual combinations of artefacts across the site were not meaningful – other than in relation to practices which themselves were almost entirely unrelated to the acts of deposition which characterize (the archaeologically visible element of) that site.

In making this point, here and in the main paper, my aim has always been to remind us to focus on and take seriously this end of the depositional spectrum (or indeed iceberg). It is important that we do not just see all patterning as an outcome of ritualized acts of deposition, or even of culturally significant conventions as to what should go where. As Thomas neatly puts it right at the end of his response (p. 127), it is important that we investigate how ‘habitual cultural practices mesh with the more random processes’ if we are ever really to understand how the structure of deposits relates to past practice.