A Group of Tin-Glazed Earthenware Kiln Waste
from the Brislington Pottery
The Recovery of the Kiln Material
During the course of his research into Bristol’s 18th- and 19th-century pottery industry the writer had made contact with the late Edward Duffett, a descendant of the Duffett family of potters. Coincidentally, Edward Duffett lived at 30 Wootton Road, St. Anne’s, some 150 metres east of St. Anne’s Chapel, and in 1975 he contacted the writer to report that during shallow excavations for a new pond in his back garden he had started finding large quantities of tin-glazed earthenware pottery (Fig. 8, Site 3). Although over the years he had often found sherds of tin-glazed pottery in his garden, he was aware that in this case there was an exceptional concentration of pottery which appeared to form a layer just below the topsoil. With Edward Duffett’s permission the writer excavated the rest of the site of the pond and recorded the finds recovered.
Before the construction of the modern housing estate, of which Wootton Road forms a part, the area was farmland and the find-spot would have been situated in the field called ‘Four Acres and Potters Leaze’ on the 1791 Gore-Langton estate map (Fig. 12), some 250 metres south of the River Avon where the land slopes up from the narrow flood plain of the river.
Although the full extent of the dumped material could not be ascertained, as it extended outside the area of the proposed pond, it covered at least six square metres and was up to 20 centimetres thick. The layer containing the kiln waste consisted mainly of tightly packed sherds of vessels and kiln furniture mixed with some ash, charcoal and red sand. There were no obvious tip-lines in the dump and it seemed likely that the material had all been deposited on one occasion rather than over a period of time; possibly to fill a depression in the surface of the field.
The importance of this find is that the kiln waste came from one well-stratified deposit; other kiln material believed to come from the Brislington pottery was retrieved from now unknown locations and contexts over a period of time without proper records being kept (see Chapter 3 above).
In view of the quantity of sherds in the waste dump it was necessary to be selective over the type of material retained, although the writer now regrets that course of action. All the glazed sherds and kiln furniture, but only the rims, bases, handles, larger body sherds and other diagnostic features of the biscuit ware, were kept. The biscuit ware retained for study weighs 17.35 kilogrammes, the glazed wares 1.08 kilogrammes and the kiln furniture a further 9.89 kilogrammes.
The much larger proportion of biscuit wares to glazed wares in the dump seems to be quite typical of tin-glazed earthenware waste groups from elsewhere. It seems likely that where vessels had been through the more expensive processes of decoration, glazing and second firing they were more likely to be sold as seconds rather than thrown away as waste.
The glazed wasters especially show signs of having been damaged during manufacture; vessels had collapsed in the kiln or become stuck together where their glazed surfaces had touched during firing, the glaze had badly bubbled or run, and there is evidence of over-firing.
The once-fired biscuit fabric is hard and various shades of pink. The twice-fired glazed vessels have a much softer, light yellow or buff-coloured fabric. Some sherds bear traces of what appears to be a white slip, presumably the remains of unfired glaze. The colour of the tin-glaze ranges from white to light pink while the majority of the decoration on the glazed vessels is executed in shades of blue. Only a few sherds with polychrome decoration were found and the colours are limited to purple, green and yellow.
The writer has illustrated the main types of vessels, the kiln furniture and some of the larger examples of decorated wares (Figs. 37-49), and has followed as closely as possible the general vessel typology established by Bloice (1971) in his report on the material from Norfolk House which comprised three groups having an overall date range of c.1680 to 1737. Comparisons are made between the Brislington material and the wasters from the Limekiln Lane pottery, which are dated c.1715 to 1725 (Jackson & Beckey 1991), and the wasters found in Temple Back which were almost certainly made in the Water Lane pottery between c.1730 to 1760 (Price forthcoming). An attempt has been made to put the Brislington waste in a wider context by comparing it with published material where the authors have tried to date and provenance the vessels on the basis of form or decorative style.
Plates are defined by Bloice as vessels with a height of less than one-fifth of their rim diameter in which the upper part of the side bends outwards in the form of a flange (Fig. 37, nos. 1-5). While the plates from Brislington have a definite flange they fail to conform to Norfolk House Type 1 and 2 plates in that they all have pronounced footrings and their sides slope upwards more steeply; infact they are more akin to the Norfolk House Type 3 dishes in their style of footring and appear to be a cross between plate and dish forms, a type of vessel commonly known as a ‘charger’.
They do not match any of the plate profiles given by Britton (1982, Appendix 1), Garner & Archer (1972, 81) or Ray (1968, 238-239) to accompany their illustrations of complete examples of tin-glazed earthenware vessels found in the Bristol Museum, Robert Hall Warren and other collections, although a similar form is shown by Britton (1987, Type D) in his catalogue of the collection in the Museum of London and by Archer (1997, Type A) in his catalogue of the Victoria and Albert Museum collection. Britton has dated plates of his Type D form to between 1675 and 1680 and tentatively ascribed them to the Southwark or Lambeth factories, while Archer dates his Type A plates between c.1620 and 1710, suggesting they were made in London or Brislington. These forms do not occur in the wasters from Limekiln Lane or Temple Back, possibly because both groups were produced after c.1715.
Dishes are vessels with a height between one-third and one-seventh of their rim diameter with no sharp changes in the direction of their side profile (Figs. 37-38, nos. 6-13). The dishes from Brislington are similar to the Norfolk House Type 3 examples and have horizontal, downturned or slightly everted rims. Due to the similarity in the forms of the plates and dishes from Brislington, particularly in the lower body and footring, it is possible that some of those the writer has typed as dishes may be the lower portions of plates. Fig. 37, no. 6 has a pierced footring in order that the vessel could be suspended by a string or wire for display. It has also been suggested that the footring was so-shaped that a wire or string could be put around it to support the vessel for display purposes (Garner & Archer 1972, 7-8). There were no dishes amongst the forms found at Temple Back and only one illustrated from Limekiln Lane.
Bowls are basically hemispherical vessels on footrings, having a height of a third or more of their rim diameter (Fig. 39, nos. 14-23). Although the dishes from Brislington are basically similar to Norfolk House Type 1 dishes they have a more prominently downturned rim, but share the same style of footrings – some being pronounced and of small diameter (nos. 20, 22 & 23) while others are thinner, shorter and of greater diameter (nos. 19 & 21). The rim forms are different to those from Limekiln Lane where they have flatter or even slightly everted rims and also differ from the majority of those from Temple Back which have straight rims.
Bowls with lobed handles are vessels similar in form to normal bowls but having a single, flat lobed handle, usually pierced, and possibly made in a mould, applied to the outside of the rim (Fig. 40, nos. 24-28). The handles from Norfolk House have small lobes with a single or triple pierced decoration usually in the shape of a heart and are on bowls with straight or slightly everted rims. The lobed handle bowls from Brislington have downturned or straight rims while the lobes are more pronounced than those found at Norfolk House and they have one or two pierced holes which are crudely executed. Ray (1968, plate 5, no.17) illustrates a similar lobed handle bowl to the Brislington examples and ascribes that to Brislington, dating it to c.1670 to 1690.
Bowls with a single lobed handle are commonly called bleeding-bowls: used by apothecaries to catch the blood drained from patients who were being ‘bled’ as a cure. However, Archer (1997, 280) terms these vessels ‘porringers’, which could have either one or two handles, and were used for serving semi-liquid foods eaten with a spoon called ‘spoonmeats’ or ‘pottages’.
No porringers or bleeding-bowls were found in the kiln waste from either Limekiln Lane or Temple Back and it is possible that these vessels were not produced after the early 18th century, although Britton (1982, 82) dates some bleeding-bowls, presumably on the style of their decoration, to between 1720 and 1740.
Cups are open vessels with a vertical handle, whose rim diameter is always greater than their base diameter and height (Fig. 41, nos. 29-42). The examples from Brislington, like those from Norfolk House, do not have footrings and are generally hemispherical in shape (nos. 32-37, 41-42), although a few have different shapes (nos. 29-31) and may not be cups as strictly defined by Bloice. Nos. 41 and 42 have speckled manganese-purple decoration. A number of cups similar in form to those from Brislington are illustrated in the literature and all are dated to the 17th century (Archer 1997, 244-249; Britton 1987, 121-125; Hume 1977, fig. VI, nos. 1-3). Britton calls these vessels ‘caudle cups’, caudle being a warm sweetened and spiced drink, although he does not explain his use of the word caudle, as the cups were presumably used for drinking all kinds of liquid refreshment. Cups with speckled manganese-purple decoration are also illustrated by Hume (1977, plate 14, no. 13) and Archer (undated, plate 50) and these are dated to the second half of the 17th century. It is perhaps significant that the groups from Limekiln Lane and Temple Back, dating to the 18th century, did not contain any cup sherds.
Mugs are basically tall, narrow vessels with vertical handles and simple rims whose rim and base diameters are approximately the same (Fig. 42, nos. 42-50). The examples from Brislington have kicked bases and the majority also have shallow parallel grooves close to base and rim, possibly resulting from being finished on a lathe. No mugs were amongst the groups of wasters from Limekiln Lane or Temple Back, while only one example is illustrated from Norfolk House. Only a few simple, straight-sided mugs are illustrated in the literature but those that are have been dated to the 17th century (Hume 1977, plates 22 & 23; Britton 1986, nos. 66 & 72; Archer 1997, C5) while those with more elaborate splayed bases seem to have been produced in the 18th century (Archer 1997, C17, C18 etc; Britton 1986, no. 152).
Mug handles were generally small and plain: the ribbed handle illustrated (Fig. 42, no. 52) appears to be too large and elaborate to have come from a mug, although mugs were the only vertical-sided vessels found in the Brislington group which would have had such a handle.
The narrow, vertical-sided (Fig. 42, no. 51) and wide, low-sided vessels (Fig. 43, nos. 53-54) have been classified by Bloice as storage vessels and those from Brislington are very similar in form to Norfolk House Type 1 and 2 storage vessels, which also occurred at Limekiln Lane.
The albarello-type containers are substantially straight-sided but have diagnostic constrictions in the sides immediately below the rim and above the base (Fig. 43, nos. 55-65). Both of the types found at Norfolk House – those with straight sides between the constrictions (Type 1) and those with convex sides between the constrictions (Type 2) – are represented in the Brislington group, occurring also in the Limekiln Lane and Temple Back groups. These types of vessel are generally regarded as having been used by apothecaries as containers for storing dry drugs, the constriction around the neck probably being used for tying down a parchment cover to keep the contents fresh.
A number of examples of this type are illustrated by Archer (1997, J1-J11), their period of use being throughout the 17th and early 18th centuries. They are usually decorated with simple blue geometric designs such as dots or crosses and a few small fragments of sherds decorated in this manner are amongst the Brislington group (not illustrated). The production of drug jars labelled with the name of the contents apparently began in the middle of the 17th century, the earliest dated specimen bearing the date 1652 (Drey 1978, 130).
Other containers used by apothecaries are classified by Bloice as the cauldron-type: basically a hemispherical body, with the rim diameter greater than that at the base, on a solid foot. While a number of these of varying sizes were found at Norfolk House, Limekiln Lane and Temple Back, there were no obvious examples amongst the Brislington assemblage.
Sherds of vessels with relatively narrow bases and flaring sides (Fig. 44, nos. 66-67) and one sherd of a narrow, inward sloping neck above a globular body (Fig. 44, no. 68) are almost certainly parts of jugs. Examples of jugs are illustrated by Britton (1982, plates 11.15-11.17; 1986, plates 6-14) and Archer (1997, plates 164/165 & C1), where they tend to have globular bodies below a parallel neck, the latter sometimes being quite squat, with a single handle joining the neck and body. The form of the neck from Brislington (Fig. 44, no. 68), with its slightly inturned rim and external parallel grooves, more closely resembles the necks on Rheinish stoneware jugs – particularly Bellamines – and it is possible that the potters were trying to imitate the imported products, the grooved neck perhaps intended to provide a purchase for a string used to tie down a parchment cover. This style of jug has a wide date range from the 16th to the 18th centuries.
The sherd of a globular vessel with thumb-impressed decoration around the neck and an applied handle (Fig. 44, no. 69) is probably part of a flower vase, although it does not exactly match the majority of illustrated examples (Archer 1997, C11-13; Archer & Morgan 1977, plate 12; Hume 1977, plate 17), the closest parallel being that illustrated by Britton (1986, no. 56); it is one of the less elaborate examples, having no spouts and with simple handles. Globular-shaped flower vases, often on pedestal bases and with a number of cylindrical nozzles around the upper part of the body, appear to date to the 17th century. Vases having the shape of an urn, again often set on a pedestal base, are invariably assigned to the 18th century.
The pedestal base (Fig. 44, no. 70) might be part of one of these flower vases, the base of a goblet (Archer undated, plate 32) or the bottom of a candlestick (Garner & Archer 1972, plate 29a; Britton 1986, no. 50).
Only one lid handle (Fig. 44, no. 71) was present in the Brislington group and it probably belongs to the lid of a posset pot, a similar one being illustrated by Archer & Morgan (1977, plate 15).
The strangely shaped, tightly rolled terminals on straight stems (Fig. 44, nos. 72-73) are from salts, the terminals projecting vertically or at a slight outward angle from a flat rim. A damaged example from Brislington (not illustrated) has one of the stems attached vertically to just such a rim. This type of salt was in the form of an open-topped dish on a pedestal base, the upward projecting terminals perhaps being intended to support a cover, although Archer & Morgan (1977, 39) suggest the terminals enabled it to be used as a stand for a dish on a crowded table at the end of a meal. Salts of this form are comparatively rare, Archer & Morgan considering that although this style of salt was already out of fashion it was still being produced in the last quarter of the 17th century. Certainly all the examples illustrated have been dated to no later than c.1675 (Archer & Morgan 1977, plate 18; Britton 1986, plate 44; Garner & Archer 1972, plate 27B; Archer undated, plate 31). One terminal of this kind is illustrated among the Norfolk House group although it appears not to have been recognised as part of a salt.
The tightly rolled terminal or knob (Fig. 44, no.74) appears to have been applied to a V-shaped piece of clay, perhaps serving a decorative function on a vessel.
The fragment of flat clay with a design crudely scratched into its surface before firing (Fig. 44, no. 75) may have been a rough design for a decorative motif intended to be used in the pottery or simply a graffito by one of the potters.
The kiln furniture found in the Brislington group consists of saggars, girders, tiles and trivets, in a friable light buff fabric.
The saggars are all of the Norfolk House Type 1 form: cylindrical vessels which have a flat base with a splayed basal angle, almost vertical sides, simple or crudely formed rims and with U-shaped openings cut into their sides from the rim (Fig. 45, nos. 76-82). The only difference between the Brislington saggars and the Norfolk House Type 1 saggars is that the latter had a single hole cut into the centre of their base, while there is no evidence for such a central hole at Brislington. Vessels, particularly those which were thin-bodied and could not support much weight or which had been glazed and were likely to adhere to each other if touching in the kiln, were placed inside the saggars for firing and the saggars would have been stacked in a pile on top of each other. A few of the Brislington saggar sherds have splashes of blue glaze on their internal surfaces: evidence that they contained glazed wares during firing.
Fragments of girders show that they consisted of thick rectangular slabs with projecting flanges on the two long sides giving a squat H-shaped cross section (Fig. 46, no. 83). These seem to be identical to the girders found at Norfolk House and were used on edge in the kiln to support shelves made out of flat kiln tiles (examples present at Brislington but not illustrated), thereby assisting the stacking of pots within the kiln. Unlike the Norfolk House and Limekiln Lane girders and tiles there is no evidence that the surfaces of those from Brislington had been sanded to prevent pots sticking to them. Generally one side of the Brislington kiln tiles, which were about 15 millimetres thick, had been splashed with white and blue glaze, as were some of the girders. It is likely that Fig. 46, no. 84 is the broken end of an H-shaped girder but as there were a number of these within the assemblage it is possible that they were a type of bar also used for providing support in the kiln.
Trivets are triangular slabs with one flat surface while the other has pinched-up points at the three angles (Fig. 46, nos. 85-86). Glazed vessels, especially plates and dishes, would have been set on them in the kiln to prevent the vessels sticking to the kiln shelves; the trivets being easily knocked off after firing. The two types of trivet present at Norfolk House – Type 1 with only slightly concave sides and Type 2 with very concave sides – are represented at Brislington where the complete examples vary in size from 46 to 100 millimetres between two points of the trivet.
The glazed sherds recovered were usually quite small and only the largest decorated examples have been illustrated (Figs. 47-49, nos. 87-110). The wares generally have tin-glaze applied to the internal and external surfaces of the vessels except in the case of plates and dishes where the underside, or hidden surface, is always covered with an impure tin-glaze or lead glaze which fired a dark yellow. Such an impure tin-glaze or lead glaze was noted at Norfolk House, where it was confined to Type 3 dishes, but was absent in the Limekiln Lane and Temple Back material. Archer (1997, 18) comments that most 17th-century chargers have plain lead-glazed backs, presumably for reasons of economy, the tin content in glaze being a more expensive ingredient than lead.
The colour most often used for decorating vessels was blue – in various shades depending on the thickness of its application. Yellow and light green were used quite frequently but purple was confined to the speckled manganese-purple cups or only appears as narrow brush strokes to pick out details on vessels decorated in other colours. There is no evidence that red was used on the Brislington material.
The decorative motifs take the form of stylised floral patterns or geometric designs and seem to be quite crudely and heavily executed. Figures 47 and 48, nos. 87-102, are executed in blue only, while Figures 48 and 49, nos. 103-110, incorporate light green, yellow and purple into the designs. The only motif which can be identified is the head of a unicorn decorating the base of a plate where the outline is in blue with the horn and mane picked out in yellow (Fig. 49, no. 105).
The surviving fragments of decoration are unfortunately too small for any positive statements to be made about motifs and patterns which might be confined to the Brislington pottery during this period and thus enable us to identify Brislington products amongst other material. The fragments of floral and geometric decoration which have survived could well apply to the London potteries of this period: they are to be found amongst the material from Norfolk House, the only other published kiln group of comparative date.
Perhaps it is not surprising that the forms of vessels and their styles of decoration are common to Brislington and London, bearing in mind the origin of the Brislington potters. It is generally the practice amongst ceramic historians to attribute particular tin-glazed earthenware vessel forms and styles of decoration to a particular pottery or, sometimes, even to a particular pot painter and the question arises as to whether such attributions can safely be made. Contemporary documents prove the mobility of potters and pot painters, not only between the various potteries operating in Brislington and Bristol, but later between other production centres such as London and Glasgow.
It is likely that their production and decorative techniques travelled with them. For example, the potter John Harris, who had been apprenticed to John and Hester Weaver in Bristol in 1733 was later recorded as a potter at Crown Street, St. Ann’s, Soho, London. Perhaps typical examples of the mobility of potters were John Longbottom and John Grace who had been employed at the Limekiln Lane potteries. It seems likely that they were the same potters who gave evidence of their employment at the Delftfield pottery in Glasgow in 1750; if so, then they had both also been employed at some time in a pottery in Lambeth.
The Date of the Kiln Waste
A broad date range can be obtained for the production of the kiln material by examining the types of vessels that are present in the group. Of particular interest are the plates, bowls with lobed handles, cups, mugs, flower vases and salts.
The plate forms found at Brislington are characteristic of a type known as ‘chargers’. They conform to Britton’s Type D plates, examples of which he has dated to beteen 1675 and 1680, and Archer’s Type A plates dating to between c.1620 to 1710. The latter is quite a broad date range but suggests that the major output of these wares was during the 17th century, especially as the glazed examples have impure tin- or lead-glazed backs apparently typical of the 17th century.
The bowls with lobed handles, commonly called ‘bleeding-bowls’ or ‘porringers’, occur in the three groups of waste from Norfolk House which have a date range of c.1680 to 1737, although the latest group could have been deposited as early as c.1725. It may be significant that they were not found in the Limekiln Lane or Temple Back groups which are post c.1715 in date. It seems likely that bowls with lobed handles were essentially a late 17th-century/early 18th-century product, which may not have been fashionable after the first quarter of the 18th century. A similar example to those from Brislington has been dated by Ray to c.1670 to 1690.
The majority of the cups from Brislington are of the distinctive hemispherical shape with the absence of a footring and cups of this form illustrated in the literature are invariably dated to the 17th century. However, they occur in all three groups from Norfolk House which shows they were still being made throughout the first quarter of the 18th century, although they do not occur at Limekiln Lane or Temple Back suggesting that in Bristol, at least, their production did not continue much after c.1715. The use of speckled purple decoration which occurs at Brislington has been discussed by Bloice (1971, 148) who quotes Ray as suggesting there is a gap in the use of that style of decoration between c.1700 and 1725. Bloice is of the opinion that the evidence from Norfolk House would suggest its continued use throughout that period and that it then ceased.
The mugs from Brislington are all of the simple, straight-sided form which are paralleled in the material from Norfolk House, although the mugs only occurred there in the earliest of the three kiln groups, dated c.1680 to 1710. No mugs were found in the Limekiln Lane or Temple Back groups and the 18th-century mugs illustrated in the literature all have more elaborate splayed bases which seems to be a diagnostic feature of the later forms.
The globular-shaped flower vase from Brislington is typical of those produced during the 17th century, although slightly less elaborate than some which have cylindrical nozzles, the 18th-century vases apparently generally assuming the shape of an urn.
The Brislington salts are important evidence for dating the group. All the examples illustrated in the literature have been dated no later than c.1675. One terminal from a salt is illustrated amongst the sherds from the earliest group of waste from Norfolk House which dates from c.1680 to 1710.
Trivets seem to have gone out of use after the early part of the 18th century; they only occur in the c.1680 to 1710 kiln group at Norfolk House and Dr. Graham Dawson, who has extensively studied the tin-glazed earthenware kiln groups from London, considers that trivets were only used until about 1720 (pers. comm.). Only one trivet fragment was found at Limekiln Lane and none were in the Temple Back group which tends to confirm the pre-1720 date for these items of kiln furniture.
The Brislington kiln group is not large enough to be taken as a complete representation of the products of the pottery at any one time and care must be exercised when suggesting a date of manufacture based on the absence of particular types of ware from the material. However, certain important types of vessels do not occur: wall tiles, cauldron-type containers and Norfolk House Type 2 saggars: the latter cylindrical with triangular cuts in the sides to take pegs for supporting the wares.
It is considered that tin-glazed earthenware wall tiles were not manufactured successfully in England until the last quarter of the 17th century. In 1677 Dr. William Johnston wrote that ‘he went to Lambeth and Vauxhall to inquire about the tiles, but found them at both places unfurnished, they telling me they were yet at a loss both in the mixture of their clay and way of burning them …’ (Archer 1997, 41). In the 1690s the Glass Sellers Company were still stating that English potters could not supply high-quality ‘White and Painted Tiles for Chimnies …’ which they could only obtain in Holland. However, wasters of domestic tiles were found at Norfolk House, even in the earliest group, so perhaps there were attempts to produce these in the London potteries in the latter part of the 17th century.
Until further work is carried out on other kiln waste groups it is not clear whether the absence of the cauldron-type containers, used for drugs and ointments, and the Norfolk House Type 2 saggars is significant, but it may imply that the Brislington group pre-dates the earliest group from Norfolk House.
The comparison of the Brislington kiln group with other kiln groups and examples of vessels held in public and private collections suggests that a broad date for its manufacture was during the second half of the 17th century and into the first decade of the 18th century. However, based on the factors outlined above, the writer considers that the group may be more closely dated to the last quarter of the 17th century, but much more research needs to be carried out on kiln groups from Brislington, Bristol and elsewhere before its date can be refined further with certainty.