Caerwent Forum-Basilica, Caerwent, South Wales

Ceramic Building Materials Assessment

by S Garside-Neville

I) Factual summary

A small amount of ceramic building material (CBM) from about 100 contexts was examined for the purposes of this assessment. The material consisted of samples from the North Aisle and the East Road, plus any complete or near complete ceramic building materials readily available when the museum was visited. See Table 5 below for a listing of all contexts examined. The tables are drawn from an Access 2 database, which also had the phasing and group information added to it.

Brick

The brick has a thickness ranging between 34-74mm. Breadth measurements (where present, see below) ranging between 231-288mm. There are two tiles with complete measurements and these lengths are 399mm and 410mm. These tiles with complete measurements are a lydion shape. In comparison with Brodribb's (1987, 142) measurements for British tiles, the Caerwent tiles are a little larger than average. Lydions, like all tiles, could be used for a variety of purposes. However, this type of tile in particular could be used as wall bonding or lacing courses. The variation in breadths and thicknesses hints at other sizes of tiles being present, but too fragmentary to observe.

Some bricks had particular features. A few bricks had a combed surface. This would probably be for keying, but was it was not a common feature within this small sample. It is possible that these bricks were used for cavity walling along with spacer bobbins and hold fasts. Indeed, one example has a notch cut out, which could have been the space needed for a hold fast. A more common feature was signature marks, in either a two or three arc shape. Signatures are thought to be associated with a particular tile maker, or may symbolise the mark of a tile guild. There was one example of a deer hoof print.

Context	Length	Breadth	Thickness
0	399	255	60
0	0	288	48
0	0	260	49
0	0	272	45
0	0	252	45
215	0	264	34
217	0	266	43
217	0	265	37
1703	0	231	39
2502	0	260	66
2505	410	268	49
3345	0	271	46
3477	0	278	57

Table 1 - Bricks with at least two complete measurements

Flue tile

Only a small amount of flue tile was examined. It had combed keying, which has been associated in London with an early second century or later date. There are two types of vent - circular and square, and there is some variance in the thickness of the form.

Context	Breadth	Thickness	Comments
5104	0	13	COMBED
5157	0	25	COMBED; SOOTED; SQUARE VENT
5206	0	27	COMBED
5207	0	16	CIRCULAR VENT, 70MM FROM TOP EDGE
5207	72	12	CIRCULAR VENT

Table 2 - Flue tile

Roof tile - tegulae

The thicknesses of the tegulae are between 13-46mm. Breadths are between 288-425mm, and lengths start at 470mm. External flange heights vary between 26-73mm. The two complete, or near complete, length measurements are longer than Brodribb's average (1987, 142), but the complete breadth measurements are narrower. Features include signatures of one to four arcs, lower cutaway types B & D (using the classification of Betts, 1985, 160), possible paint, and pawprints. There is also one example of a nailhole. This feature is thought in London to be a later feature found in tegula, where perhaps the roof pitches were steeper than previously and may have needed the support of a nail.

Context	Length	Breadth	Thickness	Flange Height	Comments
0	0	295	19	0	4 ARC SIGNATURE; ABRADED
0	0	297	22	0	3 ARC SIGNATURE; DELIBERATE UPPER CUTAWAY KNOCKED OUT OF LOWER CUTAWAY; REUSED
0	470	0	30	55	4 ARC SIGNATURE; CUTAWAY B
418	0	288	25	55	REUSED
418	0	288	26	51	UPPER CUTAWAY
418	0	305	23	54	UPPER CUTAWAY
604	*480	425	31	60	*MINIMUM LENGTH; 1 ARC SIGNATURE; REUSED; LOWER CUTAWAY KNOCKED OFF TO UPPER CUTAWAY
1194	0	333	24	40

Table 3 - Tegula with at least two complete measurements

Roof tile - imbrice

The imbrices have thicknesses ranging between 11-26mm. The breadth measurements are within the average range observed by Brodribb (1987,142), and the length measurements do not deviate far from the average for imbrice. Some of the imbrices from this site have a noticeable peak to the curve of the arch. There is one example which appears to have been moulded around the leg of the maker, rather than a wooden former.

Context	Length	Breadth	Thickness	Comments
2732	350	143	18	*MINIMUM LENGTH; FINGER STREAKED
2732	372	135	17	*MINIMUM LENGTH
2952	429	204	23
5157	380	0	17	REDUCED OUTER; LEG MOULDED

Table 4 - Imbrice with at least two complete measurements

Daub

There was one fragment of what was possibly daub from context 5189.

Fabrics

Three distinct fabrics were observed. These were identified with the aid of a x10 hand lens, and were gathered whilst quickly examining the Caerwent sample. The fabrics range from fine to coarse, and may or may not be from the same clay source.

Condition

Approximately 20-30% of the sample examined was unwashed, and this made observation of features very difficult, as well as being a health hazard due to excessive dust. This material should be washed before any further work is undertaken. The brick and tile was mostly fragmentary and some pieces show signs of reuse.

II) Research potential of the material

From the sample seen so far, it is uncertain whether the research aims and objectives detailed in the specialists notes can be achieved for the CBM. This is dependent on the retained material as a whole forming a reliable sample both chronologically and spatially. However, assuming the retained CBM is a reliable sample of that found on the excavations, it should be possible to characterise the ceramic building material from the forum-basilica and the nature of its use in this part of the Roman town.

The sample (forms and fabrics) could be used, in conjunction with the records of the discarded brick and tile, to elucidate construction and demolition activity at the forum-basilica, which would contribute to research objectives 1.11-12 and 1.14. The presence (or absence) of forms could provide indications of the building construction and architecture, addressing research objectives 1.21-2.

It is unlikely that the representivity of the CBM can be determined without comparable assemblages from elsewhere in Caerwent. On the other hand, it should be possible to set the forms and fabrics within a regional context, not least in relation to the material from Caerleon nearby. This would provide evidence for the manufacture of, and trade in, brick and tile, as defined in objectives 1.41-3.

In addition, the importance of analysing in detail the first CBM assemblage from Caerwent cannot be overstated. Characterising the forms and fabrics will greatly increase the usefulness of the brick and tile as a site dating tool, and will enable comparisons with other CBM assemblages recovered from Caerwent and further afield to be made in the future.

III) Methodological approach

(not included in web version)

Bibliography

Betts I M, 1985. A Scientific investigation of the brick and tile industry of York to the mid-eighteenth century. Bradford University (unpublished PhD thesis)

Brodribb G, 1987. Roman brick and tile. Alan Sutton

Table 5 - Context Listing

Context	Form	Length	Breadth	Thickness
0	RBRICK	0	252	45
0	RBRICK	0	260	49
0	RBRICK	0	272	45
0	RBRICK	0	288	48
0	RBRICK	399	255	60
0	TEGULA	0	295	19
0	TEGULA	0	297	22
0	TEGULA	470	0	30
215	RBRICK	0	264	34
217	RBRICK	0	265	37
217	RBRICK	0	266	43
418	TEGULA	0	288	25
418	TEGULA	0	288	26
418	TEGULA	0	305	23
604	TEGULA	480	425