Some Refitting Studies At Boxgrove, West Sussex, England

In my last post, I discussed the topic of refitting ancient stone tools and how such studies can provide us with important information about past human activities (see previous post). In this article, I would like to take the subject one step further and to show how refitting has been applied to the middle Pleistocene site of Boxgrove, and how it has helped improve our understanding of early human organization in the production of stone tools. Although a number of refitting studies have been carried out at Boxgrove, I will focus on a project that I previously completed relating to a specific part of the site, called Unit 4C, Area Q/2D (thanks to the cooperation and support of Mark Roberts, Norah M. Moloney, Michael J. Shott, Chris Bergman, Francis Wenban-Smith, Nick Ashton, Darren Lankstead and Matthew Pope).

A short introduction to Boxgrove

The Acheulean site of Boxgrove is located in a stone quarry in West Sussex, England. It has been dated to around 500,000 years ago, and has been under ongoing investigation by the Institute of Archaeology Field Unit (University College London) since the early 1980s (under the direction and of Mark Roberts and Matthew Pope). The main archaeological or soil horizon (within what is referred to as Unit 4 or the Slindon Silt Member) contains large numbers of flint artifacts, and is associated with being occupied by Acheulean flint knappers. The lithics for my refitting study came from this unit in Area quarry 2 (See diagram below).

The lithics studied from Area Q2/D comprised almost entirely of more than 700 flakes. Over a hundred refits were obtained by systematically conjoining the flakes, much like a 3-D jigsaw puzzle.

The depositional integrity of the area

To determine whether the artifact concentrations related to early human activities, as opposed to simply disturbances via natural agents (such as burrowing animals, plant roots or water flow), the distribution of refitted flakes were recorded in detail. Lithic finds were counted and plotted on graphs, and then distances were measured between refitted flake pieces. Most of the refits were found to be in close proximity to one another, and when results (such as the distribution of artifact size and the total number of refits obtained) were compared to other site studies, it was established that the artifacts had suffered relatively little movement and could generally be associated with past human behavior.

Organization within the site

In order to learn more about the early cognitive abilities of the flint knappers in this area, flake-reduction sequences and the nature of manufacturing technology were observed. Flakes were first set apart according to whether they had been produced by soft (i.e., both soft organic and soft stone materials) or hard (i.e., hard stone) hammer percussion. Interestingly, it was determined that over half of the flakes had been produced by soft hammer percussion, and that most of the debitage came mainly from stage 2 of the reduction sequence (trimming and shaping). Such findings can have many implications for hominid behavior. The use of both soft and hard hammer percussion modes would imply that they undoubtedly possessed divergent levels of skill. Also, if bones or antlers were utilized, these would not have been as commonly available as stone for convenient use, and it may therefore imply that these early humans carried out some form of mental planning. It must be noted, however, that there are some limitations to distinguishing between different types of percussor, and soft cortical flint nodules may well have produced the same characteristics as antler or bone. Nevertheless, the fact that many antler and bone soft hammers have been found frequently at Boxgrove would suggest that such an assumption would not be implausible.

Spatial distribution of lithics across the landscape

The characteristics of assemblage composition were compared with another adjacent unit area (Q/2A) to find out if there were any connections between the two. Such an examination can provide clues about behavioral patterns such as planning and mobility in transport distances, and this is particularly important for the Palaeolithic period. Studies of the debitage from Area Q/2A came from stage 3 of production (final modification of the tool), whereas Area Q/2D has been used particularly for stage 2 of the reduction sequence (trimming and shaping).  It would seem that stage 1 (acquisition and roughing out of the nodules) had been performed elsewhere. This implies that flint knappers exploited certain areas for particular stages of production, and then the unfinished tools were transported to other areas for the next stage of the reduction process. This again tells us that a certain level of forethought had been involved in the planning of the tool from the conception to completion.

Summary

Despite its limitations (when it comes to understanding site formation processes, recognizing discrete stages of the reduction process and distinguishing between hard and soft hammer percussion modes), refitting analysis of lithic artifacts from Area Q2/D has demonstrated that it can provide many useful insights relating to the cognitive abilities of the earliest colonizers of Northern Europe around half a million years ago (as previously attested by work in other areas at Boxgrove).

Source

Rees, Diane A. 2000. The Refitting of Lithics from Unit 4C, Area Q2/D Excavations at Boxgrove, West Sussex, England. Lithic Technology, vol. 25. no 2  

Archaeology Glossary

ACHEULEAN (or ACHEULIAN): A Palaeolithic industry of biface or handaxe manufacture (after St Acheul, France), found across Africa and much of West Asia and Europe.

ACQUISITION: Procurement of the raw material

DÉBITAGE or DEBRIS: Waste material created during stone tool production.

FLAKE: A piece of stone material removed from a core or another flake by striking.

KNAPPER: Someone who works stone to create a tool by applying force to its surface, either by percussion or by pressure.

LITHICS: Artifacts made of stone.

NODULE or CORE: A rounded to irregular relatively hard mass of mineral, used as the basic lump source material for a stone tool.

PALAEOLITHIC (or PALEOLITHIC): Meaning the ‘Old Stone Age’ (between approximately 2.5 million and 20,000 years ago). It pertains to the prehistoric period from the time crude tool manufacturing activities began until the appearance of fully modern human hunting and gathering societies. This era demonstrates an evolution in human brain-size, together with a refinement in stone tool technology.  

PLEISTOCENE: An interval of geological time that is part of the Quaternary period. The epoch contains evidence of humans and their development, and ended around 10,000 years ago.

REDUCTION PROCESS/SEQUENCE or “CHAÎNE OPÉRATOIRE”: The cumulative and irreversible removal of fragments from a core during the creation of a stone tool.

REFITTING: Involves at least partial reconstruction of pieces of stone so that they almost resemble their original form.

SOIL HORIZON: A distinctive layer in the land area that differs from both the overlying or underlying layers (e.g., by color and texture), that has been caused by the differing environmental conditions occurring over a period of time.

Further Information/Links

Boxgrove Dispatches Blog: Current news from the Middle Pleistocene Research Project at Boxgrove, UK. 

Roberts, M. B. and Parfitt, S. A. 1999. Boxgrove: A Middle Pleistocene hominid site at Eartham Quarry, Boxgrove, West Sussex  

Putting the Pieces Together: Refitting ancient stone tools

A SHORT HISTORY OF REFITTING

“Refitting” (sometimes also known as “conjoining”) can be traced back to F. C. J. Spurrel’s work at the Palaeolithic site of Crayford, England, in 1880. However, the technique’s potential has now become more widely known, thanks mostly to the relatively recent work at several later Stone Age sites, such as Pincevent and Meer in Europe, Kapthuring and East Turkana in Africa, and the Negev Desert in Israel.

WHAT IS REFITTING?

The creation of chipped stone tools involves a reduction sequence, which is characterized by the cumulative removal of raw material. When flaking stone, each stage involves the reduction of the original mass and the creation of tool fragments or manufacturing debris. This process of stone reduction is always irreversible, as each removal is permanent. Although it may be possible to make some adjustments along the way, unlike other materials such as metal or clay, it is not possible to start again with the original mass. The knapper, therefore, needs to plan his approach carefully, so that the best results are obtained during the reduction process. The basic stages of the reduction sequence include: procurement of the raw material and preliminary roughing out, trimming and shaping, and then final modification of the tool. Another (European) term for this sequence is called chaîne opératoire, although the name also includes the stage that occurs after completion of the finished tool (i.e., the maintenance/modification of the tool through use and final discard of the artifact). Lithic materials preserve well on archaeological sites due to their inorganic nature, enabling evidence from every stage of the reduction process.

Refitting includes at least partial reconstruction of the pieces of stone so that they almost resemble their original form. Flakes that have been struck from a core can be reassembled, by rejoining or fitting pieces together, much like a 3-D jigsaw puzzle. Although the process can be very tedious and time-consuming, it has the potential to tell us many things about past activities.

WHAT CAN IT TELL US?

Traditionally, refitting has been used to help reconstruct artifacts for illustrative or display purposes, or in order to obtain a better appreciation of the artifact’s original appearance. Refitting studies are now, however, being used to glean a greater understanding of archaeological record. By employing refitting studies, archaeologists are able to make inferences about:

* The depositional integrity of the area

It is possible to establish that the artifact concentrations were not merely the result of natural disturbances, but surviving evidence relating to human activity. For example, tight spatial clustering of refitted material would imply that there was little disturbance of the area.

* Organization within the site

Flake-reduction sequences and the nature of manufacturing technology within different parts of a site unit can be distinguished. For instance, large flakes with completely cortical dorsal surfaces would indicate that the very beginning of the reduction process was carried out in that area. Insights about early cognitive abilities may also be obtained from the use of manufacturing technology, such as hard-hammer (stone) or soft-hammer percussion (entailing the use of antler, bone or wood).

* Spatial distribution of lithics across the landscape

Archaeologists can compare the different characteristics in assemblage composition between two different unit areas in order to find out whether there are any associations. This may provide clues relating to behavioral patterns, such as forethought and mobility in transport distances, which is particularly important for the Palaeolithic period. For instance, two separate areas may have been used for particular stages of tool production.

WHAT DOES THE REFITTING PROCESS ENTAIL?

Although refitting has been used on a variety of materials (pottery, bone, shell, glass, metal and wood) and can involve various methodologies, for the purpose of this post I will discuss the approach I have used for stone tools.

First and foremost, it is important to ensure that all pieces to be refit are labeled accordingly, so that they can be properly returned to the correct storage container at the end of the project. It is advantageous to have a sufficient work area, so that the whole collection can be laid out and viewed at the same time, and for the whole duration of the refitting process. If this is not possible, trays may be used to compare batches based on a certain criterion. In order to protect the artifacts from breakage, a protective foam sheet may be laid on top of the surface area.

Once taken out of their bags, the lithics, can be sorted into groups, based on attributes such as cortex variations, color, inclusions, technological stages, or into whole pieces and broken pieces. For example, broken pieces can be split into proximal ends (those with bottoms missing), mesial parts (those broken at both ends) and distal ends (those broken at the top). The size of the break may also be taken into consideration for the order in which they are sorted. Some minimum size limit of flakes may be established for large collections or for those where there are project time constraints. It is advisable to correlate the pieces with a contour map and site grid, in order to view the lithic densities and to understand the significance of where the pieces come from. For example, it may be beneficial to separate flakes into specific groups based on the densities in their distribution.

The lithics are refitted in a systematic way. Comparison of the pieces can be quite a laborious process, as some trial and error is involved at the beginning. However, with time it is possible to recognize various characteristics (such as color, size, texture or shape) from pieces that may have come from the same core. However, it is also important to note that in some cases, features can differ dramatically between pieces within the same block (e.g., if some of the flakes had been thermally altered through heating). Once the refitting for one batch has been completed, the next group may be done in the same manner, after which time the refitter may also compare the two groups for further possible refits. All conjoinable pieces should be recorded as found and glued together using a reversible adhesive (i.e., one that may be easily removed).

Conjoining is a very lengthy process, and it is difficult to find every potential refit. It is therefore necessary to decide how much refitting is sufficient for the goals of the analysis. Once all refitting has been accomplished, it is important to look at the lithic markings to see where they came from.  Index cards can be used in order to record refit numbers and sketches may be made to document the position of lithics (i.e., whether joins are above or below other pieces). It can also be helpful to input the data onto a spreadsheet (each flake would be recorded as a separate entry and include such details as size, distance and reduction stage). Locations of the refits can then be plotted on a scaled graph in order to infer meaning from the patterns constructed.

SUMMARY

Although we don’t yet know the full potential of refitting, and many scholars are aware of the fact that it does have some limitations (for example, the recognition of discrete stages in the reduction process is not as straightforward as one would assume, and site formation processes are not entirely understood), refitting can nevertheless provide a fundamentally useful tool when viewed in the context of other pertinent information (such as microwear analysis). With careful and methodological application, refitting studies can provide us with crucial information relating to many aspects of archaeology.

BIBLIOGRAPHY

* Ciezla, E. 1990. On Refitting of Stone Artefacts. In The Big Puzzle: International Symposium on Refitting Stone Artefacts, edited by E. Cziesla, S. Eickhoff, N. Arts and D. Winter, pp. 9-44. Studies in Modern Archaeology, Volume 1, Holos Verlag, Bonn.

* Collins, M. B. 1975. Lithic Technology as a Means of Processual Inference. In Lithic Technology: Making and Using Stone Tools, edited by E. Swanson, p. 15-34. Mouton, The Hague.

* Gowlett, J. A. J. 1992. Ascent to Civilization: the Archaeology of Early Humans. McGraw-Hill, New York.

* Rees, Diane A. 2000. The Refitting of Lithics from Unit 4C, Area Q2/D Excavations at Boxgrove, West Sussex, England. Lithic Technology, vol. 25. no 2

* Shott, M. J. 1996. Stage Versus Continuum in the Debris Assemblage from Production of a Flutted Biface. Lithic Technology 21:6-22

ARCHAEOLOGY GLOSSARY

CORE: The basic lump source material for a stone tool.

Cortical surface or Cortex: A tough, dull-white powdery and porous coating found on the outer surface of newly exposed flint nodules. This may also include smooth weathered surfaces.

Debris or débitage: Waste material created during stone tool production.

Flake: A piece of stone material removed from a core or another flake by striking.

Knapper: Someone who works stone to create a tool by applying force to its surface, either by percussion or by pressure.

Lithics: Artifacts made of stone.

Microwear or usewear analysis: the study of the working edge of an artifact for signs of use. Patterns of wear, damage or residues may be observed via a high-powered microscope.

Palaeolithic or Paleolithic: Meaning the ‘Old Stone Age’ (between approximately 2.5 million and 20,000 years ago). It pertains to the prehistoric period from the time crude tool manufacturing activities began until the appearance of fully modern human hunting and gathering societies. This era demonstrates an evolution in human brain-size, together with a refinement in stone tool technology. 
REDUCTION PROCESS/SEQUENCE or “chaîne opératoire”: The cumulative and irreversible removal of fragments from a core during the creation of a stone tool.

FURTHER INFORMATION
Lithic Technology
The Lithic Studies Society