The Fabric Accounts of St Stephen’s Chapel

The Fabric Accounts of St Stephen’s Chapel, Westminster, 1292-1396. Edited by Tim Ayers and transcribed and translated by Maureen Jurkowski. Hardback and ebook (EPDF),1533 pages, seven black and white illustrations (Woodbridge, Boydell and Brewer, 2020), £150. ISBN  9781783274444 (hardback) and 9781787446151 (ebook).

Review article by David King

In this review article, David King outlines the significance of this important edition, published in 2020, and also provides a full and helpful explanation of what the accounts reveal about the craft of stained glass in the second half of the fourteenth century.   

 

The work under review concentrates on making available the contemporary documentation, mainly in the form of accounts, for the building and decoration of the chapel, including information about the college of priests founded on the site by Edward III in 1348. Over sixty documents are used for this, all in the National Archives at Kew, except for two in the British Library. All are in highly abbreviated Latin, with many Anglo-Norman or English lexical items, except for one in Anglo-Norman. A complete transcript and translation is given of all the documentation for the chapel, but parts which relate to other buildings are omitted where practical. There is an informative introduction by Tim Ayers, Professor of History of Art at the University of York, CVMA author and president of the international CVMA. The bulk of the book consists of a transcription of the Latin text on the left-hand page with the translation on the right, both meticulously carried out by Maureen Jurkowski. Also included are several appendices, tables and indexes. The latter need a careful examination to see how they function, but once this is done, they prove very useful for accessing the information contained in the book.

What makes the whole of particular interest to art-historians is the large amount of information these accounts contain concerning the chronology of the chapel’s construction, but also the names and activities of masons, painters, glaziers and other craftsmen, with abundant details of the equipment and materials they used to carry out their work. The importance of this source has long been recognised by scholars and some documents have previously been transcribed and discussed from 1795 onwards, but this is the first full publication. In addition, remnants of the destroyed chapel have survived in the form of painted mouldings and wall paintings, and illustrations were made of the architecture and its painted decoration, including just two of fragments of the stained glass, before and after the fire in 1834. The lower chapel survives in a much-restored form, as does the later adjacent cloister. The introduction includes a useful survey of the historiography.

The rest of this review will be devoted to what the accounts reveal about the craft of stained glass in the second half of the fourteenth century, this aspect of the decoration of the chapel having been carried out largely between 21 June 1351 and 5 March 1352, with a second less intense campaign from 27th July 1355 to 30th May 1356. This will include information on the glaziers involved, the materials and tools at their disposal and the working methods adopted. No information is given in the accounts on what the glass depicted, apart from a single reference to quarries, and the illustrated fragments give only a glimpse, suggesting that it included (almost inevitably) figures, scrolls with texts, borders, canopy work, a bird, foliage and the royal arms. The accounts do suggest that the east window of the chapel was glazed first, followed by an east-west campaign on the side windows.

An important aspect of the provision of stained glass for the chapel is that the glaziers were recruited by impressment and came from a wide area of England. The orders for impressment appeared in the Calendars of Patent Rolls and are given in the book. On 30 July 1349, John of Brampton was commissioned to buy glass for the king’s chapel in London, Shropshire and Staffordshire, and to take workmen. On 28 March 1350 John of Lincoln, himself a master-glazier, was ordered to take glaziers and other workmen in 27 counties and to buy glass in most of them. In the following year, on 20 March John of Brampton and John Gedding, glaziers, were commanded by the king to take workmen, glaziers and others for the chapel, without the counties being specified. It has been assumed that the reason such an extensive degree of impressment occurred was at least partly due to the loss of glaziers in the Black Death, but another reason would have been that as the chapel was a royal building, the king had the power to impress, and the previous long period of building activity may have given some urgency to the next stage of decoration.²

Most of the glazing references are in one document, no. 40 in the book.³ This covers building and decoration of the chapel from 21 June 1351 to 25 August 1354, but the entries for the glaziers stop on 5 March 1352. The accounts for each week customarily begin with the wages for the masons, followed by the painters, and then the glaziers. A further section covers the materials and equipment bought. Forty-two glaziers are mentioned by name, divided into groups, the first being of master glaziers designing the windows and painting the cartoons on the whitened glazing tables, there being between two and seven in this group, mostly led by Master John of Chester. They did also sometimes paint glass or engage in unspecified work, but were consistently paid 12d a day. The second group consisted of the glass painters, varying between six and fifteen in number, but usually around a dozen; they received 7d a day, and were sometimes divided into two groups. They were occasionally described as ‘working’ on the glass and may have carried out procedures other than painting. Next listed were glaziers who did not design, draw or paint, but were responsible for a variety of other work, described in the accounts in a number of ways. For example, ‘breaking and fitting together glass’ probably refers to the two steps in the process either side of painting and firing, that is, cutting the glass and leading it up. ‘Laying out the glass and making solid’ could mean taking the individual pieces of glass after firing and putting them back on the table ready for leading and then doing the leading and perhaps soldering, or it could refer to laying out the cut glass ready for painting and later doing the soldering. ‘Laying out and fitting together’ suggests arranging the pieces after firing and then leading them up. One work description is a little more specific: ‘laying out glass for quarries’, presumably to be leaded up after firing, if, as seems likely, they were patterned quarries. This group of glaziers were paid just 6d a week. A final smaller category of glazier was paid at a lesser rate, usually 4½d a day, working as a pair, sometimes alone, and for the last ten weeks not present. Where their work is specified, it consists of mulling glass paint with a pestle and mortar or on an iron plate. 

Some of the weekly accounts for glazing in this first campaign name all the glaziers concerned, but others often have such phrases as ‘William of Walton with his aforesaid twelve fellows’. It is probable that this type of entry refers to glaziers already named, but some may have been paid whose names do not appear. 

A few of the glaziers named in the accounts are thought to have been resident in London and sometimes employed in the normal way, rather than impressed. One of these was John Gedding. Already in 1335 he had been paid for purveying glass for the chapel. When the 1351-2 glazing campaign started, he was paid 7d a day on three occasions as a member of the group of glaziers who painted the glass and once for working on the glass. The other references to him at this period are to him being paid both for purveying glass on the king’s commission, but also for supplying materials including silver filings, ale for washing the tables, jet for making glass paint and grozing irons, all of which suggests that he was based locally and could readily obtain supplies of basic essentials for glazing. This is confirmed by the fact that he was the master of the glaziers’ company of London in 1373, alongside John Brampton, who had purveyed glass for the chapel together with him in 1349 and 1351.⁴ In an earlier Westminster campaign in January 1340, John of Walworth had been paid for one panel of glass bought for the window in the chamber of the chaplains of the king’s chapel, and in October of that year he received 7s for three windows of white glass bought for the temporary chapel of St Stephen by the Receipt, each having seven square foot of glass at 4d a foot. This suggests that these two purchases arrived ready-made, having been produced elsewhere, probably in London. 

The other main area of information provided in the accounts and of interest here concerns the materials and equipment used by the glaziers. This is best approached by considering the requirements of each stage of the process of making a stained glass window and comparing them with the accounts, but first, two entries in the summary accounts of receipts and expenses must be mentioned by way of a caveat. Summary accounts were made in addition to the weekly accounts and were submitted at the end of a term of service of a clerk of the king’s works. One item in that for 18 November 1346 to 1 August 1349 records payment for glass of diverse colours, silver filings, jet, gum arabic and other things bought for the painting of the glass; another for Michaelmas 1350 to Michaelmas 1351 notes expenditure on glass of diverse colours, silver filings, gum arabic, arnement and beer for washing the glaziers’ tables and on ‘other things bought and used on the glazing of the windows of the said chapel’. The individual items bought are discussed below, but the significant point is that ‘other things’ were bought, which may not have been mentioned in the weekly accounts in detail. It is evident that some items needed to carry out the tasks mentioned are not listed in the accounts, such as paint brushes and soldering irons; these may have been brought by the glaziers as their own personal equipment, but they could have been bought for them under the heading of ‘other things’. 

The first procedure to be carried out by the glaziers in the workshop was the making of designs for each panel of a window and their transfer onto the glazing tables. The tables are mentioned, but their purchase is not, unless a single purchase of a table for the king’s work in 1352-3 was for this purpose. They could hardly have been brought by those using them, as these craftsmen came from far and wide, although there is one reference to portable tables. Perhaps they were made on site by the carpenters already there from one of the many purchases of wood mentioned in the accounts and were moved around as required. However, they were the most important piece of equipment for the glazing process and it is surprising that there is no clear indication of how they were provided. There is no mention in the accounts of a glaziers’ hut on site which could have had permanently installed tables. Several references are given to ale or beer being bought to clean and whiten the tables; one account mentions ‘ale bought for the washing and the making solid of glass’, which is rather curious. It is doubtful that the ale could have been used to whiten the table – elsewhere chalk with water or egg or casein was used.⁵ What is indicated here is that the use of the ale to clean the tables removed earlier designs to reveal the original whitened surface. 

Once the designs had become cartoons drawn on the table, the glass had to be cut. There is one reference to ‘breaking and fitting’ glass by the group of glaziers who were paid 6d a day, but none to the hot iron which was run along the glass to crack it, although both firewood and coal were bought for the glaziers which could have been used to heat the iron. On the other hand, there are several references to the grozing irons used to nibble the pieces of glass into the correct shape, including their purchase (twenty-four being bought on one occasion) and repair. It has been suggested that soldering irons may have also been used for cutting the glass and the fact that these too are not mentioned in the accounts is some slight indication that this dual use may have occurred and that the glaziers may have brought their own.⁶

The next process was the painting of the glass, carried out by the second group of glaziers, and also possibly on some occasions by the master glazier designers. As indicated above, there is no mention of the various types of brush used for this, but several references to components of the medium itself.⁷ Glass paint was a mixture of a pigment, ground glass and a binder. Theophilus in the twelfth century described it as a mixture of burnt copper filings and green and blue glass and around 1400 Antonio da Pisa advised the use of yellow glass rosary beads with the copper filings. As a binder for diluting the paint, wine or urine was advised by Theophilus, but the Italian glazier used tempera mixed with fig sap.⁸ The St Stephen’s accounts are by their nature less clear, but mention jet (geet), arnement and gum-arabic as ingredients. The first two provided the pigment, the third was a binder, but the ground glass mixed with them to make the paint adhere when fired did not need to be bought, as off-cuts and fragments would have been used, and thus did not appear in the accounts. The two pigments were bought both together and separately. Jet is a mineraloid substance related to coal found mainly in the Middle Ages on beaches, particularly at Whitby in Yorkshire. It can be ground into a powder, and would presumably produce a black paint when mixed with ground glass and a binder.⁹ Arnement is an iron-based pigment, originally called atramentum, which means a black substance, and thus was used as an alternative to jet. However, medieval glass paint was of two kinds, one giving a black or dark grey appearance and the other a red-brown hue, both being used sometimes on the same panel, and one wonders if arnement and jet produced paint of these two different colours; one account has the two low-paid glaziers ‘mulling diverse colours for the painting of glass’. An additional applied colour which was considered in the accounts as a kind of paint was the lemon-coloured silver stain, made from the ground silver filings mentioned frequently in the accounts. 

When the glass had been painted the next, very skilled, operation was firing it in a kiln. The only entries which related to this process are those for the purchase of firewood (talshid, talwode). One entry lists 900 talwode used for annealing and breaking glass, that is, to heat the kiln used for firing and to heat the soldering irons. Mentions of diverse ironwork may have covered the trays on which the glass was placed in the kiln, but of the kiln itself there is no hint in the accounts. Some recent experimental archaeology carried out in France has shown how Antonio da Pisa’s instructions for constructing and using a kiln made of clay-covered wood with iron bars could successfully fire glass, although a door made of chestnut and clay was added, following another treatise, as Antoine failed to mention one.¹⁰ The wood was thin bendable green sticks and both these and the clay, if a similar kiln was used at Westminster, could have been provided without cost, together with the chalk and ash layer on which the glass rested in the tray, thus explaining their absence from the accounts. How permanent the kilns would have been is not recoverable. 

The glass-painters or the master-glazier designers would have overseen the firing, as judging when the kiln was hot enough to be closed and how long to leave the glass in it required much skill and experience. Once the glass was fired, however, it was the glaziers on 6d a day who rearranged the pieces on the cut-line on the table, so that they could lead and solder the panels. Theophilus recommends for this a table large enough for two panels, so that the fired glass could be laid out in its correct arrangement on one half, and the pieces transferred to the other half as they were leaded up.¹¹ Much lead was bought for both plumbers and glaziers on the site and recorded in the accounts. It then had to be melted and poured into a mould to make the cames used for glazing. Lead moulds were valuable items and they may have been available on site from earlier work, as they do not appear in the accounts. Listed in the store inventories included in the accounts are iron pans for melting lead; these may have been hung over a fire or brazier from the tripod which was bought for the work of the painters and glaziers, in order for the lead to melt. To get the lead from the pan to the mould a ladle was needed, but the only one mentioned in the accounts was bought for the work of the plumbers. However, as it was acquired specifically for pouring lead, the writer of the account may have assumed that it was intended for the plumbers’ use. Also in the store was a knife for cutting lead, the only knife mentioned in the documentation. This could have had a number of uses for both plumbers and glaziers, among which was that of pairing off the flange on the side of the came. 

To hold the leading in place until it was soldered closing nails were bought in quantity to be hammered into the table top against the cames. Eventually the table top would be so punctured by nail holes that it would be reversed and the other side of the wood used.

For the leaded panels to become ‘solid’, as the accounts have it, the joints had to be soldered. Solder was bought in large quantities for the windows, but also tin for the soldering. This means that the glaziers made their own solder by melting together lead and tin. A flux was needed to make the solder run easily onto the joints when the soldering iron was applied. Two kinds were bought and used: tallow, which was rubbed straight onto the joint, and less often, rosin, into which the soldering iron was dipped before soldering a joint. Much tallow and solder was bought and tin (in the form of filings in one account) to make it. In a review in the previous edition of Vidimus, it was noted that 6d worth of rosin was bought when the Norwich Dominican friary was converted for civic use c. 1540, involving much glazing activity, and it was suggested that this was also used as a flux for soldering.¹² In the St Stephen’s accounts for 1351 sixteen pounds of rosin were bought ‘for the painting of glass’. That this is a slightly misunderstood interpretation of its use for soldering is perhaps more probable than it being used in glass-paint. The final soldering task was to fasten ties, almost certainly made from split lead cames, in horizontal rows onto the leads at regular intervals.

The installation of the stained glass panels into the window was then possible. This required mortar, mentioned innumerable times in the accounts, to cement the outer leads to the stonework after they had been inserted into a rebate. Horizontal saddlebars were set at each end in the stonework and the ties mentioned above twisted round them to hold the panels firmly in place.

One aspect of the making of a stained glass window which came before all the above-mentioned steps, but for which there is very little contemporary evidence to indicate how it was done, is the taking of measurements and the making of templates to make sure that the panels made fitted the stonework when installed, especially in the curved openings in the tracery and heads of the lights. Canvas was bought on a number of occasions and for different purposes. One was for making ‘false moulds’ to send to the quarries to assist in cutting the stones to the correct size. This is a use somewhat similar to making a template of a window opening for the glaziers. Another use was to fill the windows to protect the interior before the glazing was installed. One could imagine how the canvas was cut roughly to size and then fastened in place by lathes (bought in quantity) sprung against the stonework. When the glass was ready to be made, the canvas could then be cut to the exact shape and dimensions of the opening and used as a template. This is obviously conjectural, but canvas is similar to the stiff brown paper used by glaziers for this purpose in modern times. 

The methodology outlined above rests on the evidence of the accounts wherever possible and on known medieval and modern practice where the accounts do not help. Some details are conjectural, but is clear that the accounts are an important source for the contemporary glazing practices, confirming what is known from other sources and adding new details and variants.

The publication of these key documents will open the way for further research on their contents in many areas, not least concerning the many named craftsmen, several of whom have toponyms giving some indication of where they came from. Three of the glaziers were from Norfolk: Simon of Lynn, Robert of Norwich and Adam of Norwich, and to illustrate what might be achieved by widening the scope of enquiry to that of the origins of the named craftsmen, a further contribution to Vidimus will discuss what is known of these three glaziers.

In conclusion, it must be stressed what an important resource the publication of these accounts is for the study of this important building in particular and of medieval craftsmen in England on a wider front.  The editors are to be congratulated on the careful completion of this immense task.

Appendix – List of glaziers named in the accounts

• John Alrewich
• John Athelard
• John Brampton
• William Bromley
• John Burton
• Edmund Bury
• John Carleton
• John of Chester
• John Cosyn
• Nicholas de Dadyngton
• Thomas de Dadyngton
• William Deeping
• Thomas Dunmow
• William Ens
• John Gedding
• John Halstead
• William Hamme
• William Hereford
• Andrew Horkesley
• William of Horkesley
• Thomas Jonge
• Godman of Lenton
• John of Lenton
• William of Lenton
• Hugh of Lichfield
• William Lichfield
• John of Lincoln
• John Lord
• Simon of Lynn
• William Nafferton
• Adam Norwich
• Robert Norwich
• William Papplewick
• John Parson
• John Selves
• Geoffrey Starky
• Robert of Thame
• John Walton¹³
• William Walton
• John of Walworth
• John atte Wode
• Robert Yerdele
• Thomas le Young

Many of these names have alternative spellings. These are listed in the book in the index of people and places.

References

1. St. Stephen’s Chapel, Westminster: Visual and Political Culture, 1292-1941, 2017, https://www.virtualststephens.org.uk, accessed 24 May 2021. Website and digital modelling was undertaken by the Centre for Christianity and Culture at the University of York.[↩]
2. A list of the glaziers named in the accounts is given in an appendix to this review.[↩]
3. National Archives, E 101/471/6.[↩]
4. John Brampton may have been the son of William Brampton, who provided several panels of glass for Ely cathedral in 1339-40, when his unnamed son, who was also a glazier, was given an ex-gratia payment of 8d. F. R. Chapman, Sacrist Rolls of Ely, II, Cambridge, 1907, p. 98.[↩]
5. S. Brown, ‘The Medieval Glazier at Work’, in E. C. Pastan and B. Kurmann-Schwarz (eds), Investigations in Medieval Stained Glass, Leiden, Boston, 2019, p. 14.[↩]
6. Brown, ‘Medieval Glazier’, 15.[↩]
7. There is a reference in document 40 to Simon of Lynn, one of the named glaziers, being paid 12d for 1½ lb of pig bristles for the brushes of the painters. Pig bristle was too stiff for most glass painting needs, but this may suggest that some glaziers had brought painting equipment with them.[↩]
8. Brown, ‘Medieval Glazier’, 16.[↩]
9. Personal communication from Dr. Simon Cotton.[↩]
10. See the review by the present writer of C. Lautier and D. Sandron, Antoine de Pise. L’art du vitrail vers 1400, Corpus Vitrearum France, Études VIII, Paris, 2008, in Vidimus 44, October 2010.[↩]
11. Brown, ‘Medieval Glazier’, 18-19.[↩]
12. D. King, Review of C. Rawcliffe (ed.), The Norwich Chamberlains’ Accounts 1539-40 to 1544-45, Norfolk Record Society, vol. lxxxiii, 2019, Vidimus 135, May, 2021.[↩]
13. John Walton may be an error for William Walton. William is named on 29 occasions and John on only 6, and only on lists where William is not named. Each time John is named at the head of the glaziers on 7d a day, it follows entries with either John of Chester and John Athelard, or just John of Chester, at the head of the glaziers on 12d a day, so it would have been easy for the scribe to have written in error another ‘John’ instead of ‘William’. By and large the accounts were very carefully produced, but there are a few mistakes and inconsistencies.[↩]