Ranscombe Re-Fitting

ArticlesIssue 109

1 Aug

By David May

In this third and final article, the author explains how the feature that must make Ranscombe an important site is its sheer quantity of re-fitting.

Lithic refitting has been used in archaeological research since the end of the 19th Century, and more recently it has become a standard research method (Schurmans 2007:7). The process of refitting pieces from a production sequence can illuminate the different stages of flint tool production or the various stages of the châine opératoire.

In recent years attempts at automating re-fitting (Evans et al). have been undertaken on Boxgrove artefacts using modern technology. However, Ranscombe – unlike at Boxgrove – has a larger assemblage available, making an automated approach difficult with the available technology. Being retired, the author undertook re-fitting studies of the Ranscombe assemblage by the traditional method.

The re-fitting started back in early 2012 with the Shorne Woods Country Park Archaeological volunteers assisting with the digging of test pits. An excavator from one test pit found two similar flints a few centimetres apart, at the same depth, commenting that they looked as if they may re-fit. Some days later the author, having processed the flints, recalled this comment, located the flints in questions and re-fitting was achieved.

Fig 1: The first re-fitting group

At this time, more than 2,500 flints had been recovered from the Ranscombe site. Out of curiosity, the author looked through the assemblage and quickly found two more re-fitting flints. It soon became an addiction and one that continues to this day. At the time of writing, more than 3,000 hours over six years have culminated in 151 groups of re-fits comprising of 404 items or 2.9% of the assemblage recovered. The lateral separation of pieces going into re-fitting groups was a maximum of 2.5 metres, and some 80% of items were from a depth of between 35cm and 60cm below the ground surface.

The methodology for finding re-fits is simple. Good lighting is essential but avoid a mixture of colour temperatures (i.e. fluorescent, incandescent or LED). Cool white LED lighting provides the optimum.

Commence laying out much of the debitage in a suitable large area. The flints are pre-sorted by arranging in groups based upon physical characteristics

(e.g. body colour, colour patches, texture or fossil inclusions).

Finally, a time consuming, physically relaxing but mentally stimulating process of comparison of apparently similar flints.

As possible re-fits were found, the pieces in question are checked for key features:

a good mechanical fit with no light visible between the pieces;
ensure flint body colouration or patina is similar;
any fossils or other marks identical in both pieces and appearing to align when the flints are combined.

Fig 2: Example of aligning fossils

After that, platforms are examined for similarity and proper alignment allowing for bulb scars.

If doubts exist, the items in question are put aside and re-examined at a later date.

Fig 3: Pair with aligning cortex

When a re-fitting group is found, it is allocated and marked with the group serial number prefixed by the letters CJ. A photograph is then taken of the re-fitting pieces. These are then joined by an adhesive of a 50% solution of ethyl-methacrylate copolymer (Paraloid B72) in acetone with added fumed silica.

At times it was found that support between re-fitting pieces was required, in which case hot melt glue was used. Such an example is the group of 20 shown in fig 14.

The selection of adhesive type was to allow the re-fit group to be dis-assembled should the need ever arise. Any identifications of the disassembled pieces could be made from the earlier photographs.

Table 1: Re-fitting groups

Sometimes, flakes would be found that would fit onto an existing group. Very occasionally, existing groups were compared with the re-fitting of existing groups, into a larger group. In this situation, the lower group number reference was retained and appropriate re-numbering as required. With all scenarios, further photography was required for future reference (table 1).

Re-fitting groups are not solely formed of flakes. Table 2 lists items of interest and items with secondary working found re-fitting groups:

In some instances, it was found that several groups were of similar appearance with a high probability of being from the same nodule but would not re-fit.

In another case shown (fig 5) groups of 8 (top), 7 (middle) and 6 (lower) were of almost identical material. These three groups are formed from poor flint material but considered possible debitage from axe manufacture.

From the evidence collected it is surprising, that knapping of poor flint material is a common occurrence, as can be seen with fig 6.

Re-fitting should not be thought of as applicable to large flints, or those over 5cm in size (fig 7).

Table 2: Items in groups with secondary working

Group size (no of flints)	2	3	4	5	6	7	8	9	11	20
Quantity of groups	112	20	8	0	3	2	2	1	1	1

Item type	Quantity
Primary flakes with edge retouch	2
Secondary flakes with edge retouch	4
Tertiary flakes with edge retouch	1
Hammerstone	1
Axe thinning and sharpening flakes	4
Cores	2
Core tablets	2
Axe/adze preform	1

Fig 4: Two similar groups of two

Fig 6: Poor quality material group of three

Fig 5: Three groups of similar flint Fig 7: Small re-fitting group

Fig 9: Illustrating the variety of flint at Ranscombe

Fig 8: Illustrating the variety of flint at Ranscombe

Fig 10: Group of 11

Fig 12: Group of 10

However, finding these smaller groups is less easy to re- fit, and so they often have higher group numbers.

It is understood that these items are debitage from a core tool such as an axe.

Fig 13 is unusual in that the flake indicated by the lower arrow (not attached to a group of three) is probably an axe thinning flake which if turned over would re-fit making this a group of 4 pieces. Arrows indicating re-fitting area.

And finally, the piece de résistance (fig 14).

Note the core at lower left and the flake in the middle with the number has secondary edge retouch.

With three exceptions the re- fitting groups were from an area of 3 metres by 12 metres. One exception was from an area of Mesolithic and Neolithic activity approximately 250 metres away from where a hammerstone with a refitting flake was found.

Fig 11: Group of 6 with broken and abandoned preform all found within a 1 metre by 1.5 metre area