Redefining the archaeological object

This is rather a short introductory paper, the purpose of which is to explain both why and how we are attempting to redefine the archaeological ‘object’ at Çatalhöyük. Overall, the aim behind this was to forefront interpretation as a guide to method, and facilitate the ‘making-sense’ of archaeological data in a more holistic way. It is an attempt at trying to challenge the practice of analysing artefacts as bounded categories of data by, instead, emphasising their inter-connectivity with context as a way to guide artefact analysis and recording methods. I don’t think there is anything particularly radical or innovative in this, and much of this can be seen as a response to the fragmentation that occurs in a large project where information overload often results in decreased information exchange. Although the database – which Tim Ritchey’s paper outlined in length this morning – provides a methodological tool to facilitate information flow, in practice it was recognised that a starting point is needed to begin the process of contextual thinking and the exploration of data relativity. This in addition to the fragmentation (fault-lines) that were occurring. The papers which follow this one can be seen as efforts to think in this way during – and I think that is a point worth emphasising – the process of data collection and assimilation. In other words, these results are not conclusions, but assessments of patterning at the contextual level using data that is currently available. The aim is that will direct further analysis, and lead to offshoots where other types of relationships, events or phenomena are explored.

Perhaps even before I begin, I should outline the background to why we actively attempted to do this. In part this extended from Carolyn’s work on site, where she had become increasingly interested in the way objects were defined, particularly in the context of identifying the mechanism and processes of the practical creation of archaeological knowledge. It was apparent that the product of specialists – particularly ‘object’ specialists – each analysing their own types of data during the project seemed to be creating problems of fragmentation in part caused from the pressures to keep up and ‘process’ the data as it was being generated. Largely in response to her observations, a series of events was initiated where we began to question the fragmentation of data, the reasons this came about, and mechanisms for the reintegration we were striving for. In turn this lead to the problematisation of the data that we were actually dealing with on a day to day basis.

To do this, I will first to define what I mean by the ‘material’ object and the circumstances in which it may inhibit the process of interpretation of contextual data. Following this, I will try to show that if the boundaries of the object are not fixed at the material level but redefined at the level of context, event or phenomena, a more integrated and holistic interpretation may result.

So, to problematise the ‘material’ object or thing. It seems to me that it is relatively straightforward to accept the notion that our traditional objects of analysis – and I am focusing here primarily on artefact analysis – are materially constructed, and are viewed implicitly as bounded, discrete entities. Examples are easy to find: ‘flint tools’, ‘ceramic vessels’, ‘bone’, ‘seeds’, and so on. Here we can draw a parallel between this type of archaeological ‘object’ and what Heidegger referred to as the ‘thing-in-itself’. This he defined as "that which can be touched, reached, or seen – i.e. what is present at hand", which, in effect, describes the principal behind the manner in which archaeological objects are separated for analysis, most obviously seen with the separation of things into bone, stone, plant, and ceramic categories, and so on. The primacy of the material object is why we tend to have ‘specialists’ on site such as lithic analysts, ceramic analysts, archaeobotanists and so on, and generally speaking not on-site specialists in things like ‘decoration’, ‘fire’, ‘food’ or ‘exchange’ that cross cut the traditional categories of artefacts. Given this, primacy is often bestowed on the ‘artefact’ as an bounded entity, and its cross-cutting relationships, although recognised as important, in practice are often treated as secondary at the point of analysis and entry into what was essentially the ‘primary’ record – the database.

It seems that the object is therefore defined primarily on the basis of its material composition. But the object as materially defined category, as well as being arbitrary, is also rather limiting, as it can hinders the process of integration and holistic synthesis – or more simply the ‘making-sense’ of contextual relationships that cross-cut artefact categories. This is not to deny, however, the value of the work of the archaeo-zoologist or ceramic specialist in their separate analysis and interpretation of their particular categories of objects. But, as I said, this in itself is not particularly holistic, as bones are disassociated and analysed separately from the pots, which are analysed differently from the lithics, and so on, the end result being that the categories of objects are interpreted in entirely disparate ways: the bone specialists may discuss animals, butchery, domestication, etc., lithic analysts talk about raw materials, core reduction strategies and tool functions, and so on. These ‘goals’ govern the methods of analysis of the different artefacts, the way things are recorded and, ultimately, the ensuing interpretation. This all stems from a decision to separate things, i.e. artefacts, on the basis of their material composition.

As one alternative, we can conceptually, rather than practically, attempt not to separate things in this way, and attempt to retain a sense of object relativity. A useful illustration of this may be to contrast the manner of examination of things like micro-stratigraphy, and macroscopic things like pits or dumps. In the case of the former, although stratigraphic layers are composed of several disparate types of things – grains of sand, plant fragments, trampled plaster, and so on – the ‘layer’ only really only exists in its contextual composite – yes, there are discrete particles of matter, but everything blurs together to create something interpreted as a ‘layer’, where boundaries between things are vague and dependent on the level of magnification. A similar situation can be proposed for macro-archaeological deposits such as a pits or dumps, etc.: these also exists in a contextual composite of events of cuts, fills and artefacts. The essence of the problem is that these all blur into each other and form a context that is, in a sense, really greater than the sum of its constituent parts. Although, ideally, everyone recognises the need to maintain this contextual integrity, in practise, as I have said, it is often difficult to recreate the original once excavated and parcelled out to specialists for a variety of reasons, some of which no doubt are specific to the particular circumstances and pressures under which we work at Çatalhöyük: what in practise seems to occur is that the objects are by default given primacy, and undergo disparate types of analyses. Indeed, the artefacts become the object – they become bounded entities on the basis of their (arbitrary) material being – despite the blurred boundaries that existed between them within their original context. In a sense the context has been lost through the attrition of its objects to the specialists.

A potential solution, therefore, may be to destabilise the object as the focus of analysis, and redraw the boundaries of the object from material to phenomenal. For clearly, to draw from Heidegger again, "within what boundaries we determine the meanings of the term ‘things’ always remains arbitrary. With respect to this, the scope and direction of our questions will change". So, if we try to redefine the boundary of ‘thing’ to encompass artefact and context, event, or phenomenon, in doing so we can, at least conceptually, work towards maintaining this blurring in the way that bits of sand, plant remains and plaster blur together to form a layer – a contextual coherency is, at least in theory, maintained at a macro-scale and should influence the manner of data collection and analysis.

However, given the practical difficulties of numerous specialists working on site doing context by context analysis – perhaps an ideal but difficult as of yet to implement – it seemed to be more realistic simply to start by looking at non-object or context specific phenomena as a route towards contextual synthesis. At a practical level – i.e. in the field – this is, in part, facilitated by large-scale relational databases where contexts can be recreated and relationships explored electronically (although this has brought its own problems, previously discussed). For instance, if a normative concept such as ‘rubbish’ is taken as an non-object or context specific phenomenon that is worth exploring, it is possible to examine the relationships between things in light of the idea of ‘rubbish’. This prompts us to begin asking questions such as, ‘what is rubbish and does it exist’ and if so, ‘how is it constituted’, ‘what are its associations’, and so on. In doing so, we begin not to define contexts by the objects they contain, nor objects by context they’re in, but, rather, each becomes contingent upon the other. The result is, in theory, interpretation that stretches beyond any particular type of artefact or material object, but encompasses the contextual whole and can drive method.

At a practical, or methodological level, this is a useful exercise insofar as it would be as equally valuable to be able to say that ‘the definition and deposition of waste differs between areas A and B’, or ‘the definition and use of decoration changes between phase Y and Z’, as it is to say ‘the obsidian cores between site A and B are the same’, or ‘the ceramic lugs between phase X and Y change’. Both types of information can be used not only as pieces of descriptive information, but also as guides to further analysis. If, for instance, as a lithic analyst I discovered that what I interpreted as production waste ended up in alternative contexts and had different associations depending on the particular phase or area of the site, then this would alter my methods of analysis and my interpretation of how rubbish was constituted. Although a fairly obvious statement, it is perhaps less so if it is observed that in some instances what I would interpret as ‘rubbish’ appears to ends up as an useful element of the mortar that binds mud-bricks. Normally, I may perhaps overlook mud-brick mortar as a useful context to collect and record lithic material and so would miss an important piece of my own data, but also miss being able to make a contribution to interpretations of what mortar is and where it comes from and how it should be excavated. Or, in a further example, contextual relationships between some types of obsidian and elaborate rooms has prompted me to start considering obsidian as an ascetic phenomena, part of the range of visual art and ideological expression at Çatalhöyük. This began the process of me experimenting with alternative ways of analysing the knapped-stone objects, such as symmetry of form and regularity of flaking. In both these ways, obsidian as an independent category to be interpreted in isolation of other objects and contexts is being challenged. The object is decentralised by the object boundary being redrawn to encompass the phenomena, the effect being that alternative ways of translating object to the inherently textual primary record.

So, both to conclude and introduce, as a group of analysts more often concerned with particular categories of objects, we are attempting to move towards a process that emphasises inter-relationships and contingencies at an early stage in the data collection process. Redrawing boundaries and attempting to destabilise the primacy of the artefact as object is a step towards this. The following papers strive towards this goal by taking various phenomenon as the object of analysis. They should be, however, considered works in progress, not attempts at conclusions but as entry points into continued data collection and interpretation.