YORYM 1976-81.7 – COPPERGATE, YORK

 

Ceramic Building Materials

 

S Garside-Neville

 

1.0 Introduction

There were 204 boxes of ceramic building materials (CBM) from the Coppergate excavations of 1976 to 1981. The CBM was recorded on a pro forma.  The fabrics were matched with the existing fabric samples held at the York Archaeological Trust.  The form and complete measurements (mostly thickness) were also recorded.  Where appropriate, comments were made, such as observations on condition, reuse and so on.  No weighing was carried out due to the random on-site sampling policy.  Discarding of recorded fragments took place when it was deemed that the fragment could not yield any further information.  In consequence, about 75% of the sample was discarded, leaving a total of 46 boxes of CBM.  These boxes are broadly divided into the following:

 

  1 box post-medieval material

13 boxes medieval material

  7 boxes Anglo-Scandinavian material

24 boxes Roman material

  1 box fabric queries

 

It should be noted that the boxes of Anglo-Scandinavian material and medieval material are loosely sorted, so that there is likely to be more pre-Conquest material in the medieval boxes.

 

There were several difficulties encountered with the material.  In particular, the presence of other categories of finds, which included animal bone, pottery (large amounts of amphora were mis-sorted as tile), crucible and loom-weights.  These required rebagging and labelling.  Approximately one third of the sample was packed in paper bags.  Due to the length of time in storage, these bags had decayed.  This caused,  in some cases, the mixing of contexts within a box.  In other cases, the bags had also stained the tile so that it was  difficult to read the context number.  The dust caused by these decayed bags was unpleasant.  Paper-bagged material needed to be rebagged and labelled, which also took extra time.

 

1.1 Summary

Number of boxes:  204

Number of small finds: 112

Number of fragments examined: c. 16,000

Number of boxes discarded: 158

Number of boxes retained: 46

 

2.0 The sample

2.1 Roman material

A large amount of the material examined was Roman.  However, 92% of that material was found in non-Roman contexts, and is therefore residual.

 

Forms present:

          Flue tile - keying: combed, finger-smoothed and scored

          Tegula and Imbrex - large and small sizes

          Brick  - of various sizes

          Antefix - two small fragments

Fabrics:

          These mostly matched the samples already held

Tile stamps:

          Legionary

          Possible civilian

 

2.1.1 Recommendations

There is only a small section of the sample that comes from Roman contexts.  Due to this, the potential for this sample to provide useful information about either the sample or the Roman CBM industry is limited.  However, there is a collection of tile stamps, including a probable civilian stamp that warrant further study.  There are also two fragments of antefix.  In addition, there is a large collection of tegula flanges, which could be recorded to provide comparisons for other sites.  After recording, these flanges could then be discarded.  However, this might take place as a separate project rather than for this site report.

 

2.2 Anglo-Scandinavian material

In this sample, the definition of 'Anglo-Scandinavian' material is the CBM that occurs in pre-medieval contexts.

 

Fabrics:

          Broad range, some of which continue into the medieval period.  Local and possible imported material

 

 

Forms present:

          Floor tile - The small finded floor tile from this sample numbered 22, however, after recording the bulk finds, a further 54 pieces were          found.  Some of the designs have not been previously published. 

          Curved and flanged - It was initial examination of the CBM      records that indicated that the curved and flanged material began to   appear in Anglo-Scandinavian contexts.  Curved and flanged is           similar to the Roman roofing system, but tends to be smaller in   size,   and is sometimes glazed.

          Specialised roofing material - There is a small category of      material the function of which is uncertain.

 

2.2.1 Discussion

The potential of this section of the sample is very high. Approximately 20% of the curved and flanged material comes from well-dated securely stratified Anglo-Scandinavian contexts.   It points to a previously unknown CBM industry of a very early date.  The Coppergate material could be the earliest known ceramic roof tile in Britain and Europe, and therefore will be of international significance (pers comm Alan Vince).  The contexts are securely dated and there is a large enough sample to make the evidence convincing.    Perhaps the CBM industry in York not only supplied floor tiles to the Church, but also produced roofing material to a wider market.  However, it is possible that some of the material was imported.

 

2.2.2 Recommendations

It is recommended that fabric analysis is carried out.  Some of the fabrics seem to be used into the medieval period (to manufacture peg tile).  One fabric (M9) is very refined, and atypical in appearance to any of the usual York fabrics.  It has a mica content which is not a characteristic of York clay, being more typical of the Beverley area (pers comm Alan Vince). 

 

Analysis of fabrics through thin sections should be carried out to confirm the location of the clay sources.  In the case of M9, it may be more appropriate to analyse the sanding to find out the whether it is actually from local sandstones.  This will aid the dating of various industries, some of which may be short-lived, or perhaps continue into the medieval period. Also, it was noticed that the fabrics of some of the loomweights were similar to that of M9, which may be worth confirming.

 

Comparison with pottery fabrics must take place, in an attempt to relate the two ceramic industries to one another.  The input of a pottery specialist will be needed for this to take place.

 

Typological study of the forms is recommended.  Variations in flange shape may be significant.  Some curved tile may prove to be ridge tile.  The floor tile designs need to be matched with published material.

 

2.3 Medieval material

Forms present:

          Floor tiles (small sample)

          Peg

          Nib

          Curved and flanged

          Ridge (including crested)

          Brick

          Louvers

Fabrics:

          It may be possible to pinpoint when certain common fabrics begin       to appear. 

          New fabrics appear after the Conquest

 

2.3.1 Discussion

The potential of this material is fairly high.  Due to the layers of well-stratified dumping, this material can be used to securely date the occurrence of forms and fabrics in York.  The material is associated with tenements, so that the tile may be able to suggest phases of re-building, demolition debris, dumping, status and so on. 

 

Very little peg occurs pre-1200 which suggests little contamination occurred.  Nib occurs in low numbers throughout the medieval period, suggesting use on a monor scale.  However,  one of the fabrics associated with the Nib tile is very distinctive, and may be imported from outside the area.  About 20% of curved and flanged tile appears pre-1150, so that a total of 40% if this material seems to be of an earlier date than previously thought.  There is some brick present in very early contexts. 

 

2.3.2 Recommendations

Input from a pottery specialist will be needed to relate the ceramic industries.  Further analysis of the presence of curved and flanged tile in pre-1150 contexts needs to be undertaken.    Characterisation of mesaurements of peg tile dimensions can take place.  The features and dating of the Crested ridge tiles can be studied further.

 

 

3.0 Analysis methodology

- Close examination of the CBM records relating the site and the forms and fabrics found

- Further research into the retained finds (eg. floor tile, curved and flanged, peg tile, and so on)

- Need for scientific analysis of  Anglo-Scandinavian and medieval material