Archaeology: a guide to excavation and fieldwork



This guide is intended as an introduction and reference guide for archaeological excavation and related fieldwork. It does not pretend to be comprehensive, but will attempt to provide coverage for the most obvious categories of and approaches to fieldwork. The article is written from the perspective of somebody who has worked predominantly in the United Kingdom and Turkey, and so largely reflects these field conditions and procedures. However, an attempt has been made to generalize beyond specific national frameworks for excavation, in order to provide the broadest coverage possible.

Contributions are sought from those with expertise and experience in any relevant area of fieldwork and excavation.

Types of excavation and research

Two broad categories of fieldwork are conducted within archaeology. These are developer-funded archaeology and research-oriented archaeology. There is often cross-over between these two forms, which relate mainly to forms of funding and the structure of organization, yet also impact on issues of timing, data-recovery, data-publicity and, equally, the nature of recording and analysis.

Developer-funded archaeology (DFA)

DFA represents the private arm of excavation, constituting the majority of ongoing fieldwork within the U.K. at the present time. Private archaeological 'units' are employed by necessity of construction and planning developers in the course of their goals, in line with U.K. legislation PPG-15 and PPG-16. Units may also be associated with County Archaeology Departments (CAD), whose interests may be more in line with research-oriented excavation.

Broadly, the task of DFA is to excavate a site systematically and rigorously within a limited time-frame. Each period represented by deposts at the site must be equally weighed, while efforts are aimed at a 'total recovery' of data. In other words, each area of activity and context is to be identified, excavated and recorded. This data is often published as 'grey literature', or densely statistical, non-inferrential data which represents, in effect, a tabulation, in print, of the physical deposits in the ground. Conservation is often secondary to development. With the pace of agrarian and industrial development in some countries it is often difficult to maintain the balance between the proper record/treatment of the past, and an appreciation for the needs of local communities and developers. India and Pakistan provide two notable examples; here, significant amounts of agrarian development are based on local initiatives, while the inaccessibility of many sites ensures that regular tours of inspection cannot be made. As such sites are likely to be ploughed and eroded within the possibility of intervention by antiquities experts.

Research-oriented archaeology (ROA)

What is archaeological data?

The purpose of excavation is to identify and remove superimposed stratigraphic layers one-by-one, in reverse order, while recording as much contextual and associative data as is possible to obtain while removing layers (see Barker 1993, 71). Potentially, every element of this 'matrix' contains valuable archaeological data, largely due to the contribution of 'association' to understanding buried sites. Broadly, this means that the associations between objects, features and conditions in the ground can reveal the processes which led to their deposition. This is known as 'behavioral archaeology', characteristic of 1980s and 90s excavation in the USA.

Pre-excavation and survey

This section will consider the varying forms of archaeological data as they apply to fieldwork and excavation.

Survey and identifying a 'site'

While excavation is typically associated with 'sites', the nature, morphology, and formation of these 'sites' varies depending on the area in question, and the processes of transformation that have acted to modify the matrix in which archaeological deposits are contained. Image 1 represents a generic model of sites within an n-space; in this situation, the main site, a settlement area, represents the densest concentration of deposits and materials within the space. No site has a clear boundary, however, and so a marginal or ephemeral area is represented by the secondary circle. This is, in turn, represented by a wider area in which uncontaminated deposits (i.e., those which have not been moved subsequent to deposition and burial) are not contained. The wide and uncertain distribution of these materials is a result of n-transforms and C-transforms, or natural and cultural (n and c) activities which have altered the deposits at a later stage. These transformations may occur as a result of ploughing, animal burrowing, and river-washing. Difficulties are encountered when we wish to label particular concentrations of material; do they represent a dumping ground that was used only once, or a long-term activity area used (for example) by flint knappers? Excavation can partly answer these questions, but many research designs and fieldwork programs do not have excavation specified within their remit; as such we must apply provisional labels to these areas.*

Image 1, types of site

  • one means of attributing use-length to an activity area is to measure the depth of all human-created deposits directly, and to subsequently identify individual lenses (short-term deposition events/processes) within the total stratification. Of course, numerous deposition events can occur over a short period of time, and vice-versa.

The theory and practice of stratification

Stratification is the term used to describe the superimposed layering and intercutting of geological and anthropogenic layers within a site. This concept follows the Law of Superposition, which holds that earlier deposition events will be layered lower than those of a later event. On-site stratification is not, however, this simple or generic; layers can become disrupted and inverted through both natural and cultural actions and processes, leading to often complex stratigraphic 'puzzles' within a site. This is more likely to occur at a long-term urban site. As such, systems of recording and interpretation aim to schematically interpret these layers, and their relationships, in both spatial and temporal terms. The most commonly employed system is the Harris Matrix, which is described below.

The Harris Matrix

The Harris Matrix is a system for the record and interpretation of stratigraphic relationships within a site. If each context is given a numerical code (a context number), from 1-5, then these numbers can be represented in a schematic manner in order to represent the most-likely sequence of deposition. Image 2 conveys an example of a Harris Matrix for 5 generic contexts. In practice these relationships are often provisional, and can only be firmly anchored through the systematic dating (by either relative or absolute methods) of each context in turn. The occurence of diagnostic material in each layer can also provide clues as to their temporal relationships, though the mixing and disturbance of deposits (and materials therein) can occur over time.

Diagram 2, generic example of a Harris Matrix

Diagram 3 represents a series of superimposed layers within an excavated area of a site (Area 100). Diagram 4 represents the manner in which these contexts would appear as a result of Single Context Recording.