Hostname: page-component-7b9c58cd5d-6tpvb Total loading time: 0 Render date: 2025-03-15T01:27:49.053Z Has data issue: false hasContentIssue false

WESTMINSTER HALL’S LOST STUART DOOR PASSAGE REDISCOVERED

Published online by Cambridge University Press:  17 March 2022

Elizabeth Hallam Smith
Affiliation:
Houses of Parliament, Westminster, London SW1A 0AA, UK. E-mail: hallamsmithe@parliament.uk
John Crook
Affiliation:
Hillside, Brede TN31 6EJ, UK. Email: john@john-crook.com
Rights & Permissions [Opens in a new window]

Abstract

In 2018 unpublished archaeological evidence was discovered recording a doorway and passageway concealed inside the Romanesque wall of Westminster Hall, near the south-east corner. Although commemorated by a bronze plaque in situ, their existence had largely been forgotten. Further investigations revealed an access panel in the 1951 cloakroom fittings in adjoining St Stephen’s cloister: this was located, and the space accessed, seemingly for the first time since c 1952. The many features of interest found within included the doorcase and soffits of a great doorway and iron pintles for the doors; Purbeck flagstones on the floor; complex masonry and plaster from several different eras; graffiti by masons from the nineteenth and twentieth centuries; a still-functional Osram lightbulb dating from the early 1950s; and wooden joists supporting the masonry of the ceiling. Isotope dating of the timbers produced a date of 1659, and works accounts showed that the doorway and passageway were created in 1660–1, to form a ceremonial route for the coronation of Charles ii. Further archaeological and historical investigations have enabled the authors to establish a full chronology for the changing fabric and uses of the doorway and passageway from the seventeenth to the twenty-first century, and to trace the masons who walled in the space in 1851. They have also established why the brass plaque in Westminster Hall marking the space erroneously ascribes it with Tudor origins: that ‘fake history’ was created by an over-enthusiastic late-nineteenth century Clerk of the House of Commons.

Type
Research paper
Copyright
© The Author(s), 2022. Published by Cambridge University Press on behalf of The Society of Antiquaries of London

INTRODUCTION

Although many aspects of the archaeology and architectural history of the Palace of Westminster have been comprehensively studied, it still has significant secrets to yield up. One such secret is the concealed door passage at the south-west corner of Westminster Hall, the unveiling of which captured worldwide media attention in February 2020. Sir Lindsay Hoyle, Speaker of the House of Commons, who championed the research findings, said: ‘To think that this walkway has been used by so many important people over the centuries is incredible. I am so proud of our staff for making this discovery and I really hope that this space is celebrated for what it is – a part of our parliamentary history.’Footnote 1

The now hidden doorway and passage were created in 1660–1 as a processional entrance for Charles ii at his coronation, and from 1661 to 1807 served as the main route from Westminster Hall to the House of Commons, then located in the former St Stephen’s Chapel. Walled in since 1851 and forgotten since the mid-twentieth century, the passage was rediscovered in 2018, and consists of a short – and at first sight somewhat unassuming – cavity within the east wall of Westminster Hall, measuring 2.55m deep by 3.45m wide and 3.5m in height. This obscure and forgotten space has offered up a wealth of historical, political, architectural and archaeological evidence about a little understood area of the palace and about the people who frequented it. Here we consider the findings and highlight their significance.

LOCATION AND REDISCOVERY

Westminster Hall and parts of St Stephen’s cloister, which mark the two sides of the door passage, are, together with the chapel of St Mary Undercroft, the only buildings from the old Palace of Westminster that, despite much alteration, still remain standing above ground. Although Charles Barry shifted the centre of gravity of his new Palace of Westminster to the east, building out into the River Thames, the medieval survivors were almost as close to his new House of Commons as to the old, and alongside their historical significance they have long been at the centre of political life (fig 1).Footnote 2

Fig 1. Hist Eng, PSA01/08/00003, overlay plan showing the old and new Palaces of Westminster, 1932. The survivors are Westminster Hall, St Stephen’s cloister and St Mary Undercroft, shaded in orange. North is on the left. Image: © Historic England Archives, reproduced with kind permission.

The door passage opened into Westminster Hall in the second bay from the south, near its south-east corner (fig 2). It was cut in 1660–1 through the massive Romanesque east wall of William Rufus’s great hall, completed in 1099, and also through a wedge-shaped thickening on its outside face that had evidently been added when the first stone cloister of St Stephen was built for the royal college of canons based in neighbouring St Stephen’s Chapel, between 1384 and 1396.Footnote 3

Fig 2. Location plan of the doorway (red circle). North is at the top. Drawing: John Crook.

The east side of the door passage is located in the double-height Tudor reconstruction of that earlier cloister. It lies behind panelling, in the second bay from the south in the lower west walk, near the cloister’s south-west corner (fig 3). Probably part funded by King Henry viii (r. 1509–47), this cloister was constructed c 1512–27 at vast expense and may have been intended as a processional way for the king and canons.Footnote 4 After the college was dissolved in 1548, St Stephen’s Chapel passed to the House of Commons as its debating chamber. The cloister was put to work as the service quarters of a great house, occupied by the Auditors of the Exchequer from 1570 to 1794,Footnote 5 although its south-western corner was purloined in 1660 to create the new door passage. From 1794 to 1835 the house was occupied by the Speaker of the House of Commons.

Fig 3. Concealed panel of 1950 providing access to the door passage. Photograph: John Crook.

The existence of a former entrance to the pre-1834 House of Commons in this area was known from a brass plaque on the wall of Westminster Hall, dated 1895 (fig 4). Although the plaque went unnoticed by most parliamentarians and visitors, the lost entrance appeared on historic plans of the old palace and on an important academic reconstruction of the layout of this part of the old palace.Footnote 6 What had been forgotten since the 1950s was that the door passage still lay within the walls of the Hall: it had not been infilled.

Fig 4. The plaque in Westminster Hall. Photograph: John Crook.

The existence of the passage was, however, newly revealed in 2018 through archival research, which led the team to locate and enter it through a hinged access door cut in 1949 through the oak panelling and brick wall behind it in the west range of St Stephen’s cloister (fig 3).Footnote 7 The ability to physically access and record the space, and the unearthing of a wealth of related visual and archival evidence, have now enabled its history and significance to be retold.

DESCRIPTION OF THE DOORWAY AND DOOR PASSAGE

Archival and archaeological evidence for the creation of the door passage

When rediscovered in 2018, the door passage was at first assumed to date from the first part of the seventeenth century.Footnote 8 However, the assignment through dendro-isotope dating of a felling date of Spring 1659 for the timbers of the ceiling joists prompted a major re-assessment.Footnote 9 It was noted that up to 1660 the doorway would have had no obvious purpose since the route from Westminster Hall to the House of Commons and the southern end of the palace was via a medieval doorway nearby. As depicted in a drawing of 1559,Footnote 10 this was in the south wall of Westminster Hall, close to the south-east corner.Footnote 11 It was known to have been blocked off in 1660–1 when the dais at the south end of Westminster Hall was remodelled and extended for Charles ii’s coronation: part of major works in the hall that included repaving the floor.Footnote 12 A new access route to the Commons was therefore needed, and the dimensions of the doorway created for that purpose, as stated in the accounts, matched with those of the door passage.

The new entrance was formed by cutting a tunnel through the east wall of Westminster Hall (fig 5a), a process described in the accounts as ‘beating down’. The key entries in the accounts are:

Masons Imployed in … working of stone for the doorcase at the upper end of Westm[inster] hall cont[aining] Cxxvj foote, cutting [a] way for the s[ai]d doorcase through the stone wall there viij fo[o]t thicke being xj fo[o]t wide and xiiij fo[o]t high and setting the said doorcase w[i]t[h] freestone Cont[aining] Cxxvj fo[o]t and yoting in iron hookes into the same and laying the sell w[i]t[h] vij fo[o]t of Kentish stepp, … letting in an iron cramp to fasten the Kings head over the new doorway, … in Joynting and laying xv foote of Kentish step in two doorwayes at the low[er] end of the hall, in cutting parte of the Collums to enlarge the passage and parte of the wall to give light in the passage going up staires to the parliam[en]t house.Footnote 13

Bricklayers in working up the new doorway in the Hall on both sides and above the lintle Cont[aining] about one rod viij foote of brickworke a bricke and halfe thicke, and working up the wall on one side w[hi]ch was beaten downe to enlarge the coming into the Courte houseFootnote 14 Cont[aining] lxxvij foote of brickworke, in working up the new doorway in the Hall on both sides and above the lintle Cont[aining] about one rod viij foote of brickworke a bricke and halfe thicke, and working up the wall on one side w[hi]ch was beaten downe to enlarge the coming into the Courte house Cont[aining] lxxvij foote of brickworke.Footnote 15

For a greate double stocke locke to the new doores by the house of Com[m]ons staires, 00: 05: 00; … For a greate stay to the doore by the house of Com[m]ons staires wt xjld’ at vd 00: 04: 07.Footnote 16

Fig 5. Development of the opening, showing state in 1661 (a), 1846 (b), 1851 (c), 2021 (d). Drawing: John Crook.

The accounts thus show that an aperture 8ft thick, 11ft wide and 14ft high (2.43 × 3.35 × 4.26m) was made through Westminster Hall’s wall and that its walls were lined with brickwork one and a half bricks thick (about 380mm (15ins)). Because of the fourteenth-century extra skin, the east wall of the hall thickens northwards, but on the centreline measures 2.55m (8ft 5ins). Excluding the brick lining, the opening as broken through would have been about 3.48m (11ft 5ins) in width at the widest, east end. The height from the stone-flagged floor to the ceiling now averages 3.5m (11ft 6ins): it is likely that a taller opening of 4.27m (14ft) high was initially created but subsequently lowered. That would account for the fill visible above the joists, which includes post-medieval brick. The ceiling slopes upwards by around 40mm from east to west.

Exterior of the doorway in Westminster Hall

The only known contemporary illustration of the doorway’s exterior is in Brittannia [sic] Illustrata of 1727 (fig 6), where it is described as the entrance to the House of Commons.Footnote 17 Although this part of the engraving lacks detail, the location and shape of the doorway conform with the archaeological evidence outlined below.

Fig 6. Left: the only identified contemporary image of the doorway, seen at the south end of Westminster Hall, as shown in the rectified detail (right). The central doorway at the south end of Westminster Hall, between the two law courts, is also shown. Image: reproduced from Brittannia Illustrata (BL, RB.23.c.669, 1727, pl 26).

The doorway was blocked off on this side in 1807. That section of Westminster Hall’s east wall was later refaced by Sydney Smirke, probably in 1837, using Huddleston stone from Yorkshire,Footnote 18 and further repairs have taken place subsequently. In the 1850s Charles Barry constructed across the lower part of the doorway a raised platform area leading to the entrance to the chapel of St Mary Undercroft. Although by that time inaccessible, the existence and location of the hidden doorway were still known, and from 1895 it was marked by the brass plaque in Westminster Hall along with bronze studs in the wall, 2.67m apart, to indicate its position (fig 4). In 2014, repairs to the facing of this section of wall revealed part of the doorway’s southern stone jamb along with nineteenth-century brickwork (fig 7).Footnote 19 At that time, however, it was not realised that a void lay behind it.

Fig 7. Left: location of the plaque and the studs; the two pale blocks of Huddleston stone are those that were replaced in 2014. Photograph: John Crook. Right: remains of the right-hand (south) jamb of the stone entrance arch to the passageway as recorded in September 2014 during repairs to the walls. Photograph: © Mark Collins, reproduced with kind permission.

Interior: ceiling

The ceiling of the door passage (fig 8) is formed of oak joists laid flat, running north to south. These are supported on the brick side walls. However, in order to support the medieval walls when the cavity was broken through, the wall must have been needled, and the needle beams would have run east to west. Some larger timbers appear to be visible through gaps in the joists.

Fig 8. The ceiling of the door passage (west at top). Photograph: John Crook.

Interior: the west wall

When constructed in 1660–1 the west side of the door passage encompassed the great doorway leading into and out of Westminster Hall. Four iron pintles – the ‘iron hookes’ in the accountsFootnote 20 – that originally supported the two doors are to be seen on the two sides. The doors closed flush against the face of the square opening, without any rebate. The pintles are set in the stone blocks of the internal corners of the passage. The stone appears to be Kentish Rag, visible where the render has fallen away from the lower block. Removal of small areas of nineteenth-century plaster on this wall revealed the outline of the inner (east) face of the 1661 doorway (fig 9). The door head is flat and is formed of twelve Kentish Rag tapered voussoirs, which continued well above ceiling level.

Fig 9. Elevation of rear wall of passageway. Drawing: John Crook.

This side of the passage was first blocked up in 1807, but a smaller opening with a timber doorcase was inserted in the brickwork in 1846 (figs 5b and 10). That aperture was in turn blocked in 1851 with bricks varying in colour from yellow to salmon pink. Remains of the cut-back red brickwork of 1807 are visible around the edge of the unplastered area. The glass panel on the left covers a pencil graffito of 1851, and there is another dated 1950 to the right.

Fig 10. Rear (west) wall showing the short-lived opening of 1846. It was blocked in 1851, as recorded by a graffito behind the glass panel to its left. Photograph: John Crook.

Interior: the east wall

Any evidence for the original treatment of the opening on the cloister side appears to have been concealed by mid-twentieth-century panelling and the new plaster above it. The cloister pilaster buttresses flanking the opening would originally have acted as jambs, but a lintel must have been set above the opening. From here, the route to the House of Commons continued via the former two-bay porch at the south-west corner of the cloister, from whose first bay a flight of stairs within the thickness of the east wall of Westminster Hall led up to the lobby of St Stephen’s Chapel.Footnote 21 This access arrangement had been formed when the cloister was first built in the late fourteenth century, and was retained when the chapel became the Commons chamber.

The former east end of the door passage is blocked by brickwork of 1851, visible from within the chamber thus formed, including a central pilaster buttress. To the south of that, a patch of yellow brickwork is evidence for the rediscovery of the passageway on 21 June 1949 (fig 11, right). Subsequently a small rectangular opening was formed to the north of the pilaster, to maintain access (fig 17).

Fig 11. Left: north wall. Brickwork of 1660–1 is apparent behind the thin plaster render. Right: south wall. In the adjacent east wall is yellow brick infill of the opening made on 21 June 1949. Photographs: John Crook.

Current state: the north and south walls

The north and south side walls are finished in limewashed hair plaster, 4–6mm thick. On the north side the shadow of the brickwork is discernible behind the render (fig 11, left) and where the plaster has fallen away red bricks are visible set in a very white, soft lime mortar.Footnote 22 On the south wall of the door passage (fig 11, right), the thin plaster render also appears to be primary.

Both walls show evidence of small cavities, later infilled. The one on the north wallFootnote 23 coincides with the end of the door swings when opened and may relate to the ‘greate staye’ of the 1661 accounts, weighing 11lb and evidently of iron, holding back the door.Footnote 24 Its infill of red bricks and hair plaster are, however, earlier than any of the subsequent phases. The position and dimensions of a rectangular patch on the south wallFootnote 25 do not match it and the render appears to be modern cement. Its east edge seems to correspond with the door swing of the south door leaf, so it may be a primary feature related to the actual door. Its depth is not known. Several layers of limewash are discernible on the south wall face; and there are also some scrawled calculations on the latest limewash, and spurious graffiti.Footnote 26

Interior: the floor and sills

The floor of the door passage appears to be of the primary phase of 1661: the plaster render on either side seals the floor slabs, which probably abut the face of the brick skin to north and south. The slabs are of a white limestone, seemingly Purbeck stone from the Portland Formation brought from the Dorset coast,Footnote 27 consistent with the 1660–1 accounts.Footnote 28 The stone for the sills at either end of the new entrance was specified as Kentish Rag (called ‘Kentish Step’ in the accounts).Footnote 29

Analysis of the floor slabs is hindered by spreads of modern mortar, but towards the west end of the door passage the pavers in the centre are worn. Those at the sides, where people would not normally have trodden, retain their tooling, done with a comb chisel and evidently finished in situ.Footnote 30 A part of the pavement appears to have been replaced when the passageway was reopened in 1846, as an area lacking paving stones corresponds in width to that opening.

The floor level inside the chamber is 3.23m aOD. This is consistent to within 30mm with the level of the Purbeck marble paving of 1660–1 in Westminster Hall, as rediscovered by Sydney Smirke in 1835 and at that time thought to be medieval.Footnote 31 The present cloister floor is at 3.40m aOD. Given the insertion of service passages beneath the cloister floors in the nineteenth century, modernised in the twentieth, this is likely to be higher than the original level.

HISTORY AND USES, 1661–1850

The doorway and coronations, 1661–89

Created in 1661–2 as part of the works for the coronation of Charles ii, the doorway served as the main route from Westminster Hall to the House of Commons and also to the south end of the palace. Its use as an access way for coronations is likely to have been short-lived, but its iconographic significance continued for far longer, for the accounts show that a sculpted head of Charles i carved by a William Beard (otherwise unknown) was set above the doorway and secured with an iron cramp.Footnote 32 It was set in a cartouche of Beer stone.Footnote 33

During Charles i’s reign (1625–49), busts of the king had often been placed in prominent places to make a political point, and the Interregnum régime had ordered many of these to be decapitated. The symbolism of the placing of the late king’s head so close to the very spot in Westminster Hall where he had stood trial would therefore not have been lost on contemporaries.Footnote 34 In his History of the Coronation of James II, Lancaster Herald Francis Sandford marks it on a plan as ‘the dore through which the proceeding first entered Westminster Hall’ in 1685. It was thus the way that the king first appeared in splendour in Westminster Hall on the morning of the coronation and walked up to the high table to preside over the coronation breakfast, to the acclamation of the peers.Footnote 35

This use of the doorway for the first ceremonial arrival of the monarch may not have survived for long, given Queen Anne’s infirmities and later remodellings of the southern end of the hall. Nevertheless, the head would still have been highly visible during both the coronation breakfasts and the banquets that followed the crownings. In June–July 1702, not long after Anne’s coronation in April, the cartouche of Charles i was replaced by a bronze cast of a bust of the same king, set upon a pedestal above the doorway.Footnote 36 A fan print depicting the coronation banquet of George ii in 1727 shows it in situ, hovering above the proceedings, emblematic of the longevity and continuity of the monarchy (fig 12).Footnote 37

Fig 12. Fan print of the coronation banquet of George ii, 1727. The bust of Charles i is on the far left. Image: BM, Drawing 1891,0713.376, © The Trustees of the British Museum, reproduced with kind permission.

Route to the House of Commons and to the south end of the Palace of Westminster, 1661–1807

Up to the Restoration the south-west corner of the lower cloister had formed part of the much-prized Auditor of the Exchequer’s house. The sequestration of this area in 1660 by the Office of Works to form the new route to the House of Commons coincided with the reinstatement of former MP Sir Robert Pye to the post of Auditor of the Receipt of the Exchequer. It was a post that he had previously held from 1620 to 1653, but by the time of his return he was evidently incapable and perhaps was unable to resist the purloining of some important circulation space from his official residence.Footnote 38

As the acknowledged entrance to the House of Commons, the doorway and this corner of the cloister would now have been the route taken by the Speaker’s procession through Westminster Hall and into the House, climbing the stairs within the wall of Westminster Hall to reach the Commons Lobby at principal floor level. According to John Hatsell, active as Clerk of the House of Commons from 1768 to 1797, by the 1670s ‘and till long after’ the Speaker ‘always passed through Westminster Hall on his way to the House of Commons, and saluted the courts of Common Pleas and King’s Bench, the judges rising from their seats, with their caps on, to receive and return the Speaker’s salute’.Footnote 39

These two courts, meeting within staging at the south end of the hall that could be dismantled to make way for grand ceremonial events, remained a significant presence here in the eighteenth century. By 1718 a large additional doorway had been pierced through the centre of the south wall of Westminster Hall, accessed between their two courthouses and leading up steps into the Court of Wards and thence into the Commons Lobby,Footnote 40 and the Speaker may well have processed that way thereafter. But although this new central doorway was now the main route from Westminster Hall to the southern end of the palace at the higher (principal) floor level, our doorway was still seen as ‘the enterance of the House of Commons’, as shown by the 1727 Brittannia Illustrata print (fig 6).Footnote 41 And as contemporary plans show, at ground floor level it remained the way to reach the southern end of the palace (eg fig 13).Footnote 42 Its worn pavers are evidence of the high level of the traffic it sustained.

Fig 13. Detail from a plan of 1760–6 showing the doorway and the route leading from it at ground level through St Mary Undercroft (also at that time used by the Burgess Court of the City of Westminster), and then on into Old Palace Yard and the Long Gallery. Plan: Soane Museum SM 37/1/24, © Sir John Soane’s Museum, London, reproduced with kind permission.

Westminster Hall witnessed a number of changes over the next few decades: in 1739, the advent of William Kent’s demountable Gothick staging for the two law courts; in 1748–50, a major repair to the roof; and in 1780, the clearing out of the shops and stalls to allow both the raising of the floor level by one foot (305mm) (1780–2) and the refacing of the walls.Footnote 43

As for the bronze bust of Charles i, after several decades of removal for state trials and replacement thereafter,Footnote 44 in 1791 it was finally relocated from what was by this time its ‘obscure corner of Westminster-hall … to a safer and more honourable place’, the passageway from Westminster Hall to Old Palace Yard. This was as part of the works to accommodate the lengthy and complex trial of Warren Hastings (1788–95).Footnote 45 The bust was still regarded as important enough to be featured and illustrated by Thomas Pennant in his popular guidebook to London, where he attributed its design to Bernini.Footnote 46 Contemporary depictions of it, including Pennant’s own, show that it was, rather, after Hubert Le Sueur (fig 14).Footnote 47

Fig 14. Peter Mazell’s engraving of the Le Sueur bust of Charles i. Drawing: Royal Collection Trust RCIN 601957, © Her Majesty Queen Elizabeth ii, reproduced with kind permission.

In 1794 the Auditor’s House passed to the Speaker of the House of Commons, thereby reuniting the whole of the west cloister walk under his control.Footnote 48 The stairs leading up to the Commons from the west cloister were soon blocked off,Footnote 49 although the way through our doorway to the southern end of the palace was maintained at ground level until 1807.Footnote 50 This route went past the Le Sueur bust of Charles i, which by 1800 had unfortunately disappeared. Its subsequent fate is unknown.Footnote 51

The doorway from 1807–46

Between 1796 and 1807 James Wyatt repaired and remodelled the Speaker’s House as a grand Gothic Revival palace. As this lavish project drew to its protracted end, in 1807 the doorway was blocked up on the Westminster Hall side.Footnote 52 From that date, red bricks 60/65mm thick in a light grey mortar along with a narrow horizontal lacing timberFootnote 53 are visible inside the passageway. A plan of 1834 shows that the protruding door jambs on the hall side were retained.Footnote 54 The west cloister side, adjacent to a service area, remained open.Footnote 55

In 1819–23 the roof of Westminster Hall, by now in a very decayed state, was repaired by Sir John Soane,Footnote 56 and then in 1834–7 Sir Robert Smirke and his brother and partner Sydney Smirke refaced the walls of the hall with Huddleston stone and lowered the floor again, paving it with York stone.Footnote 57 Smirke’s scaffolding and ladders, and the masons working on his project, played a vital part in saving Westminster Hall from the conflagration that engulfed the old palace on the night of 16 October 1834.Footnote 58

At the southern end of the hall, the Smirkes’s refacing work revealed the blocked side of the doorway once more,Footnote 59 and a short-lived access route to the cloister through it is suggested by a site plan made in c 1835 for the design competition for the new palace.Footnote 60 By 1837 it was closed up again on the hall side, the door jambs and arch now having been hacked back flush with the main wall and the refacing masonry continued over them.Footnote 61

The doorway reopened, 1846–51

Although partly damaged in the fire of 1834, the Speaker’s House, including the west cloister walk, was soon pressed into service as temporary committee rooms, offices, kitchens, dining rooms and storehouses, as all around it Barry’s vast new Palace of Westminster took shape.Footnote 62 The use made of the door passage, then still open, is not recorded; but an 1843 plan for MPs shows it blocked off on the hall side, with no access between the cloister and the hall.Footnote 63

By 1845 a bypass route via the cloister was needed to allow MPs ready access to their entrance in Old Palace Yard, to their temporary chamber and to their facilities at the north end of the site during Barry’s construction of St Stephen’s Entrance and St Stephen’s Hall.Footnote 64 Passing round the perimeter of Westminster Hall,Footnote 65 at first this route started in the medieval porch of St Stephen’s ChapelFootnote 66 next to the south-west corner of the cloister and entered Westminster Hall via a new doorway cut through the wall very near its south end. In 1846, however, evidently needing this area for his works, Barry relocated the route’s starting-point to the north, reopening the blocked door passage by piercing a new aperture about 1.1m wide in the 1807 stone faced brickwork on the hall side (figs 5c, 910, 15).Footnote 67

Fig 15. MPs’ pocket plan of 1846 showing the doorway unblocked. Plan: Parliamentary Archives HC/LB/1/114/21, reproduced with kind permission.

That short-lived arrangement marked the final time that the door passage was used as an access route. By 1850 the aperture in its west side was no longer needed. Barry had by now – controversially – topped off his works in this area by demolishing the central window and central section of the Romanesque south wall of Westminster Hall to link it directly by a grand sweeping staircase to his new St Stephen’s Entrance.Footnote 68

THE DOORWAY AND DOOR PASSAGE LOST AND FOUND, 1851–2018

The door passage enclosed on both sides, 1851

In 1850–1 the lower cloister was magnificently and painstakingly restored for Charles Barry by his longstanding lead contractor Thomas Grissell.Footnote 69 Embellished and adorned with Minton tiles and stained glass designed by Pugin,Footnote 70 it was then once more put to use, but this time as a grand circulation and meeting space for MPs, and – to Barry’s great regret – as the members’ cloakroom.Footnote 71

Both sides of the door passage were blocked off as part of these works (figs 5c, 910). On the Westminster Hall side, Sydney Smirke’s refacing could now be reinstated over the 1846 aperture. Within the passage, two areas of yellow to pink brickwork from this blocking, on the west and east sides, are contemporaneous. From the same batch,Footnote 72 the bricks were set in a very hard grey cement mortar. The similarity of the jointing indicates that both sides were walled in by the same workman, who may be identified from his pencil inscription on the west wall’s plaster skim (fig 16). This reads:

This room was enclosed

by Tom Porter who was very fond of Ould Ale

The parties who witnessed the articles

of the wall was

R Congdon Mason

J Williams

H Terrey

T Parker

P Duv[a]l

These Masons w[ere]

employed refacing these groines [the vault of the cloister]

August 11th 1851

Real Demorcrats

Fig 16. Graffito of 1851 written by Thomas Porter, bricklayer’s labourer. Photograph: © Adam Watrobski, reproduced with kind permission.

The inscription not only provides a precise date for the enclosure of the space but also names of several of Charles Barry’s workmen, who had evidently used the passage as a place to socialise and talk politics. All of them appear in the 1851 census. Thomas Porter, the ringleader and scribe, was a bricklayer’s labourer living in Water Street near the Strand;Footnote 73 Richard Condon, a stonemason, was domiciled in St MaryleboneFootnote 74 and the other four, also stonemasons, all inhabited Ponsonby Place.Footnote 75

This graffito unlocks a rare and immediate insight into the lives of these workmen. Tom Porter, bricklayer or bricklayer’s labourer, perhaps the product of one of London’s elementary schools, was, in common with around two-thirds of working men at the time, literate.Footnote 76 The group’s self-identification as ‘real demorcrats’ clearly links them with the kind of radical Chartist views that ran strongly within the building trades of London. In 1841–2 the stonemasons working at Westminster had waged a spectacular and highly damaging strike against Barry’s building contractors Grissell and Peto, bringing progress on the palace to a halt, and were militant in pursuing reduced working hours in the next few years.Footnote 77 Bricklayers, and their labourers, whilst notorious for drinking and brawling, were likewise increasingly unionised.Footnote 78

How Tom Porter made his final exit from the door passage is unclear, although some of the brickwork on the eastern side at low level, where not broken through in 1949, appears to have been laid from the outside. Perhaps a temporary hatch was formed there, which was subsequently bricked up once Porter had crawled out. The whole of this new wall was now concealed behind cloakroom furniture for MPs’ coats.

The passage hidden, 1851–1949

A major dynamite explosion, one of the final episodes in the Fenian bombing campaign, which in 1885 greatly damaged the area of Westminster Hall just outside,Footnote 79 may have weakened the structure of the door passage’s ceiling but left its exterior fabric unscathed. A decade later, in December 1895, the brass plaque – described at the time as a tablet – was ceremoniously placed in Westminster Hall by the order of the First Commissioner of Works and Public Buildings (fig 4).Footnote 80 Its text indicates the existence of the doorway, dates its creation to 1547, and states that ‘King Charles i passed through this archway, when on 4 January 1641–2, he attempted to arrest in the House of Commons the five members of Parliament. This access to the House of Commons fell into disuse after the year 1680’.

The plaque was one of two erected in Westminster Hall at this time to ‘bring to mind memorable incidents in our national history’,Footnote 81 an initiative arising from the antiquarian zeal of the Clerk of the House, Sir Reginald Douce Palgrave.Footnote 82 Well regarded as a proceduralist, Palgrave’s real passion was history, which he treated in a far more cavalier fashion than he did his official work.Footnote 83 The plaque’s wording, which he supplied, was based on what he confesses in a little-known publication of 1869 to have been the enthusiastic suppositions of a ‘showman’.Footnote 84 Such was the distinction of Palgrave’s office that this text has until very recently commanded evidential credence. Meanwhile, the existence of the passage within the wall of Westminster Hall had perhaps already faded from memory by his time, and by the mid-twentieth century it had certainly been forgotten.

The passage rediscovered and concealed again, 1949–51

In 1940 the southern and eastern ranges of St Stephen’s cloister were damaged beyond repair by a high explosive incendiary bomb, leaving the northern and western sides battered but still standing. A complex and painstaking project to reconstruct the lost parts and to repair the rest was overseen by Sir Giles Gilbert Scott from 1948 to 1951, in tandem with his rebuilding of the bombed Commons chamber. During the installation of air ducts in the west cloister, Scott’s workmen stumbled upon the door passage on 21 June 1949: a considerable surprise to all concerned.Footnote 85 The space was carefully recorded in June and July in photographs and plans.Footnote 86

Any idea of opening up this significant find had no chance of success in the face of pressures to maximise the space available for the MPs’ cloakroom, which was to be reinstated as soon as possible.Footnote 87 Instead, Scott produced a scheme for a concealed access door,Footnote 88 which was implemented during the final stages of the cloister’s restoration. On the cloister side, he arranged for an inconspicuous hinged panel to be cut through his elegant new cloakroom furniture (fig 3). This gave access to the present entrance: yellow bricks,Footnote 89 set in a grey Portland cement mortar, were used both to block the temporary entrance and to construct the present entrance, which was given a concrete lintel (fig 17).

Fig 17. The opening formed in 1950. Photograph: John Crook.

To mark the work, one of Scott’s stonemasons added his own graffito on the northern part of the western wall: ‘1950 Alex Leiper Mason Stonehaven’.Footnote 90 The space was lit using vulcanised rubber electrical wiring and a galvanised metal switch. When rediscovered in 2018 this system sprang immediately into life, activating an Osram lightbulbFootnote 91 manufactured before the death of King George vi in 1952.Footnote 92

The passage forgotten, 1951–2018

In 1951 the west cloister reverted to its earlier use as the MPs’ cloakroom.Footnote 93 The door passage was probably now accessed on an occasional and casual basis, and two nonsensical graffiti scrawls (above) are likely to have been added to its walls between about 1951 and 1967. The latter date was when the west cloister was converted into desk space for backbench Labour MPs.Footnote 94

The electrical supply to the door passage was never cut off, but from about the 1970s any knowledge of the passage or of the existence of the access panel seems once more to have been lost within the parliamentary community at large.Footnote 95 From 1993 the west cloister walk was occupied by Shadow Chancellor Gordon Brown and his team, and finally, from 1997 to 2017, the Parliamentary Labour Party, who most recently placed a sofa in front of the access panel.Footnote 96 Cleared in 2017 for conservation work to begin, the lower cloister is currently unoccupied.

CONCLUSIONS

‘Infinite is the variety of fame and name with which [the doorway] may be coupled’, enthusiastically declared Reginald Douce Palgrave in 1869.Footnote 97 Whilst that does very considerably overstate the case, the rediscovery of this small and rather unglamorous space opens up many insights about the significance, evolution and uses of this now quiet corner of the Palace of Westminster.

The creation of the doorway and door passage in 1660–1 was only one element in a major re-ordering of Westminster Hall for the coronation of Charles ii. The nature and significance of this refurbishment campaign is increasingly being understood.Footnote 98 To that knowledge may now be added the location of this grand new access way. Moreover, the placing of the head of Charles i in such a prominent place over the new doorway may be seen as a highly symbolic gesture that reaffirmed Stuart dynastic legitimacy at the Restoration and at coronations for more than a century beyond that.

The rediscovery of the doorway will also allow a major rethink about the layout of this part of the old palace during the later seventeenth century. This is an era for which relatively little good evidence otherwise survives.

The doorway also assumed an equally significant political role as the principal route to the House of Commons from 1661 to 1794. Its intense level of use is attested by the worn pavers at the centre of the door passage. This was a path trodden for more than a hundred years by leading politicians and statesmen, as well as an access route to the southern end of the palace for people from all walks of life who lived, worked and visited there.

After its first blocking on the hall side in 1807, the door passage made two brief returns to active service, one fleetingly in c 1835, and a second in 1845–50, once more as the way between the cloister and Westminster Hall. The utilisation of the surviving parts of the old palace during Barry’s construction of its huge successor is not yet well understood. His need to redeploy the door passage to create such a circuitous route round his building site provides a rare insight into the scale of the problems that he faced with circulation during some of the most troubled years of this highly complex project.

In its final incarnation as usable space, in 1850–1, the door passage became a mess room for Barry’s workmen who were restoring the cloister, before they walled it in on both sides. The graffiti left by them on the west wall substantiates this use and provides a rare insight into the lives of these skilled and radical craftsmen.

Palgrave’s mounting of the brass plaque on the hall side in 1895 had a minimal impact in keeping the memory of the passage alive. Its low-key treatment after its unexpected rediscovery in 1949 is, moreover, emblematic of the longstanding tensions between the heritage and business uses of the cloister – unsurprisingly, given its proximity to the House of Commons and its value as political space. Hence the doorway and door passage soon returned to near oblivion for many decades. Yet the strong interest and support shown by the Speaker and by many other MPs at the time of their recent rediscovery marks a new and positive chapter in their history and has ensured that this space is now recognised as a worthwhile part of parliament’s history.

ACKNOWLEDGEMENTS

The discoveries reported here originated with the St Stephen’s Chapel Project at the University of York, emerging directly from a study supported by York and the Houses of Parliament, and generously part funded by the Leverhulme Trust. We are particularly grateful to Dr Mark Collins FSA for his support and input. We would also like to thank Dr Elizabeth Biggs, Dr John Cooper FSA, Dr Paul Hunneyball, Terry Jardine, Dr Henrik Schoenefeldt and Lady Alexandra Wedgwood FSA, as well as the two anonymous peer reviewers, for their help and advice. The main series of works accounts for the Palace of Westminster (principally TNA E 351, AO 1 and WORK 5) have been transcribed by Simon Neal for the St Stephen’s Chapel Project, and we are grateful to him for giving us access to these.

ABBREVIATIONS AND BIBLIOGRAPHY

Abbreviations

aOD

above Ordnance Datum

BL

British Library, London

BM

British Museum, London

Bodleian

Bodleian Library, Oxford

Gentleman’s Mag

The Gentleman’s Magazine

Hist Eng

Historic England Archives/National Monuments Record, Swindon

MoL

Museum of London, London

PA

Parliamentary Archives, London

PoW

Palace of Westminster, London

RCT

Royal Collections Trust, London

SAL

Society of Antiquaries of London

SM

Soane Museum, London

TNA

The National Archives, Kew

Footnotes

1. See eg Proctor Reference Proctor2020; Shaverian Reference Shaverian2020; also Hallam Smith Reference Hallam Smith2020.

2. Hist Eng, PSA01/08/00003.

3. Colvin Reference Colvin1963, 526–7.

4. Biggs Reference Biggs2020, 172–9.

5. Hatfield House, Cecil Papers 24/61 (112) and 24/62 (113).

7. Hallam Smith Reference Hallam Smith2020, 17–18.

8. On the basis of the information on the brass plaque and because it does not appear on a carefully measured drawing of the cloister dated 1593: Hatfield House, Cecil Papers 24/61 (112) and 24/62 (113).

9. This ground-breaking work in which the oxygen isotopes are analysed to allow dating as accurate as conventional dendro-dating was carried out by Dr Daniel Miles, FSA, of the Oxford Dendrochronology Laboratory in collaboration with Professor Neil Loader of the University of Swansea: Miles and Loader Reference Miles and Loader2020.

10. BL, Egerton ms 3320, fol 12v (Elizabeth i’s coronation); also consistent with a 1620s drawing long attributed to Wenceslas Hollar: BM, Drawing 1848,0911.748.

11. Indicated by John Stow: Wheatley Reference Wheatley1987, 418; later recorded by Lewis Nockalls Cottingham, FSA, and Sydney Smirke, FSA: PA, HC/LB/1/119, fol 31a; Smirke Reference Smirke1836a, 410. See also Barber et al Reference Barber, Holder and Phillpotts2013, 112.

13. TNA, WORK 5/2, fol 347.

14. This refers to Westminster’s Burgess Court, which from 1583 to 1766 occupied the western part of St Mary Undercroft. See eg TNA, E 351/3217; Westminster City Archives WCBG/WCB/1/13, 23 Oct, 5 Nov 1766 and fig 20.

15. TNA, WORK 5/2, fol 349.

16. TNA, WORK 5/2, fol 228.

17. BL, RB.23.c.669, Brittannia Illustrata, pl 66. The image was also published with a different commentary; see eg PoW, Parliamentary Art Collection, WOA 2811.

18. The so-called ‘Lower Magnesian limestone’ of late Permian date.

19. Recording of these in 2014 shows that the south stud of 1895 is 664mm too far to the south; the north stud is approximately correct, being only 89mm south of the northern stone jamb.

20. TNA, WORK 5/2, fol 347: ‘yoting in iron hookes into the same [doorcase]’.

21. The stair is illustrated in Brayley and Britton Reference Brayley and Britton1836, pl 16; for a plan, see Topham Reference Topham1795, pls iiiii and commentary.

22. Typically 650mm thick, and 64mm thick on average with joints 8–10mm thick. Seven courses measure a total of 510mm high.

23. This measured 345mm wide by 250mm high; its sill is 1450mm above floor level.

24. TNA, WORK 5/2, fol 228.

25. This is 540mm wide by 720mm high, the sill 1330mm from the floor.

26. On the more recent patch of render is the pseudo-historical ‘RIP Guy Fawkes 1606’. In common with the pencil date ‘1645’ on the north wall, next to the blocked cavity, this was probably added after 1950; see Hallam Smith Reference Hallam Smith2020, 21.

27. This stone, like the well-known Portland limestone, was widely used in southern England in the later 17th century.

28. TNA, E 351/3274: ‘Joshua Marshall for his taske in new paveing Westm[inster] Hall with Purbecke Stone and the passage at the end of the s[ai]d hall going up to the Parliam[en]t House …’.

29. TNA, WORK 5/2, fol 347. 7ft were required, consistent with a sill measuring 450mm from front to rear, set 200mm deep and spanning the known internal width of the doorway (2.38m).

30. Some of this tooling continues across joints to the adjacent slabs.

31. Smirke Reference Smirke1836b; Barber et al Reference Barber, Holder and Phillpotts2013, 114–15, 128, 139. In 1835–7 Smirke lowered the floor level in Westminster Hall by about 410mm to its present level averaging 3.20m aOD.

32. TNA, E 351/3274; TNA, WORK 5/2, fol 347.

33. A hard variety of Lower Chalk from the Beer quarries at Beer, near Seaton, Devon.

35. ‘A groundplott of Westminster Hall’, in BL, 604.i.19, following 108; ‘A groundplott of part of the City of Westminster’, BL, 604.i.19, following 64; BL, 604.i.19, 60.

36. TNA, AO 1/2446/139; TNA, WORK 5/52, fols 153, 157.

37. BM, Drawing 1891,0713.376.

38. Williams Reference Williams1953, 7–15; Aylmer Reference Aylmer1974, 311–13.

39. [Hatsell] Reference Hatsell1818, 241 note; Gerhold Reference Gerhold1999, 48.

40. The creation of this central doorway has long been associated with the trial of William Howard, Lord Stafford, back in 1680: Smith Reference Smith1807, 63; Howell Reference Howell1816, 1,296; Williams Reference Williams1953, 3. However, it is not there in a plan of c 1704 produced by the workshop of Sir Christopher Wren: Bodleian, Gough Maps 23, fol 40b. 1718 marks its earliest clear appearance on a plan: see SM, SM 37/1/25 and see also National Portrait Gallery, NPG 798, c 1725. Accounts at TNA, WORK 5/11, fols 180–1 (1668–9) and TNA, WORK 5/32, fol 161d (1679) may therefore relate to works on the hall’s south-western doorway rather than its central doorway.

41. It may also be discerned in the celebrated image by H-F Gravelot, Westminster Hall in Termtime, made before 1739: MoL, A15610. This was later repurposed by Charles Mosley, with a satirical poem, as ‘The First Day of Term’: eg PoW, WOA 673.

42. See also eg TNA, WORK 11/24/2; TNA, WORK 29/3239 (1746); TNA, MPE 1/327 (1793).

43. Colvin Reference Colvin1976, 388–9; Barber et al Reference Barber, Holder and Phillpotts2013, 124.

44. George Vertue’s notebooks, quoted by Howarth Reference Howarth and MacGregor1989, 112 n 9.

45. Gentleman’s Mag 1790, 1,189; Gentleman’s Mag 1791, 904; Kentish Gazette, 21 Jan 1791. See also Crook and Port Reference Crook and Port1973, 500–1. Antiquaries John Carter, FSA (in 1791–2) and John Topham, FSA (in 1795) both noted its recent relocation: SAL, Red Portfolio, John Carter, 1791–2; Topham Reference Topham1795, pl ii.

46. Pennant Reference Pennant1791, frontispiece and 94; RCT, RCIN 601957 with commentary at <https://www.rct.uk/collection/601957/bust-of-charles-i> (accessed 13 Jan 2022); Howarth Reference Howarth and MacGregor1989, 112 n 9.

47. Howarth Reference Howarth and MacGregor1989, 85–91, 89, fig 40 and 112 n 9. The earliest example of this design, featuring lion mask pauldrons, dates from 1636 and is at the Bodleian Library.

48. Williams Reference Williams1953, 19. For a plan, see SM, SM 37/1/28.

49. Topham Reference Topham1795, pl ii and commentary.

50. The doorway was used during the trial of Lord Melville in 1806: TNA, WORK 29/3243; and see Crook and Port Reference Crook and Port1973, 501.

51. Gentleman’s Mag 1800, 216. There is a surviving Le Sueur bust of Charles i in the Royal Collection, which was acquired in about 1820: RCT, RCIN 33467, with commentary at <https://www.rct.uk/collection/33467/charles-i> (accessed 13 Jan 2022). Although apparently similar to the lost Westminster example, its provenance is not known.

52. Williams Reference Williams1953, 19; J T Smith, ‘Foundation plan of the ancient Palace of Westminster’, in Smith Reference Smith1807, facing 125. Smith describes it as an ‘old passage to Cotton Garden now blocked up’.

53. 65mm deep and 1120mm above floor level.

54. Port Reference Port2011, no. 20.

55. Plan by Adam Lee of the Palace of Westminster, c 1807: MoL, A15453.

56. Crook and Port Reference Crook and Port1973, 502–3, 647–9.

57. Smirke Reference Smirke1836b; Barber et al Reference Barber, Holder and Phillpotts2013, 114, 128, 139.

58. Shenton Reference Shenton2012, 136–53, esp 142, 150; PoW, WOA 1669.

59. Port Reference Port2011, no. 20; TNA, WORK 29/3288.

60. PA, HC/LB/1/119, fol 15c.

61. TNA, WORK 29/79.

62. Shenton Reference Shenton2016, 63–4, 78, 150, 164–5.

63. PA, HC/LB/1/114/19.

64. The Times, 15 Jun 1846, 6.

65. It skirted the hoardings that contained the Royal Fine Arts Commission’s competition space for artworks for the new palace: PA, HC/LB/1/114/20. The route is noted in The Times, 4 Jan 1845, 6; Boase Reference Boase1954, 326–7; Anon 1843, 1 and 18.

66. This fine 14th-century survival was recorded by a number of antiquaries and is illustrated in Brayley and Britton Reference Brayley and Britton1836, pl 19.

67. PA, HC/LB/1/114/21–22.

68. Crook and Port Reference Crook and Port1973, 587, n 1. The results were excoriated in the chamber: House of Commons Debates, 10 Jun 1850 vol 111, c 1,005 [Sir D Norreys], < https://api.parliament.uk/historic-hansard/commons/1850/jun/10/supply-new-houses-of-parliament#column_1005> (accessed 13 Jan 2022).

69. TNA, WORK 11/9/7; The Builder 1851, 9/455, 25 Oct 1851, 674–5, <https://archive.org/details/gri_33125006201822/page/674/mode/2up> (accessed 13 Jan 2022); Port Reference Port1976, 113–19, 156–7.

70. For example, watercolour by Edward N Holmes, 1871, showing the restored north cloister: PoW, WOA 4265.

71. PA, BAR/1/55; TNA, WORK 29/1149; TNA, WORK 29/1173; Belcher Reference Belcher2015, 466, 494–6; Belcher and Wedgwood Reference Belcher and Wedgwood2018, 185–6.

72. Both sets of bricks measured 225mm × 1095mm × 65mm.

73. Thomas Porter, bricklayer’s labourer, b. 1811, living 9a Water St, St Clement Danes, Strand in 1851: 1851 census, Findmypast, TNA, HO 107/1512, 22. He also wrote ‘Tom Porter Bricklayer’ in very large letters on the plaster right at the top of the adjoining wall.

74. Richard Condon, stone mason, b. 1811, living 42 Nightingale St, St Marylebone in 1851: Findmypast, TNA, HO 107/1491, 83.

75. James Williams, stone mason, b. 1825, living 17 Ponsonby Place, St John’s, Westminster in 1851: Findmypast, TNA, HO 107/1479, 27; Henry Terry, stone mason, b. 1823, living 11 Ponsonby Place, St John’s, Westminster in 1851: Findmypast, TNA, HO 107/1479, 29; Thomas Parker, stone mason, b. 1799, living 36 Ponsonby Place, St John’s, Westminster in 1851: Findmypast, TNA, HO 107/1479, 45; Peter Dewal, stone mason, b. 1829, living 9 Ponsonby Place, St John’s, Westminster in 1851: Findmypast, TNA, HO 107/1479, 29.

76. Lawson and Silver Reference Lawson and Silver1973, 276–80.

77. Goodway Reference Goodway1982, 179–81.

78. Ibid, 181–2.

79. The Times, 26 Jan 1885, 10.

80. The Times, 24 Dec 1895, 10; Liverpool Mercury, 27 Dec 1895, 2.

81. The Times, 24 Dec 1895, 10.

82. Anon [1907], 39, 43–4; McKay Reference McKay2013; and see PoW, WOA 204.

83. For example, Gardiner Reference Gardiner1893, 310–11, excoriates Palgrave’s re-dating of Charles i’s death warrant.

84. Palgrave Reference Palgrave1869, 186–7 (Note C), quoting as his source Forster Reference Forster1860, 183: Charles i entered Westminster Hall via ‘a door at the south-east angle’. Hallam Smith Reference Hallam Smith2020, 16–17.

85. TNA, WORK 14/3127, memo dated Jul 1949.

86. Hist Eng, AL 1121; Hist Eng, Chest 42; TNA, WORK 14/3127, 12 Aug 1949 and annex: Ministry of Works, ‘Hse of P Westminster, Plan of Bricked in Cloister Doorway’, 5 Jul 1949.

87. TNA, WORK 11/445, 25 Feb 1949; TNA, WORK 14/3127, Jul 1949 (sic), 17 Aug 1949; PA, HC/SA/SJ/9/50, 8 May, 7 Sep, 13 Sep 1949.

88. TNA, WORK 14/3127, 19 Aug 1949 and annex.

89. Dimensions 220mm × 95mm × 67mm.

90. Electoral registers show that Alexander B Leiper, formerly of 26 New Street, Stonehaven, near Aberdeen, was living at 33 Millbank in 1950 and 1951: ancestry.co.uk.

91. 180mm in length: Hallam Smith Reference Hallam Smith2020, 18, 20.

92. As shown by the ‘i’ monogram on its end: Lamptech, Osram Sep 1951, esp 5.

93. Hallam Smith Reference Hallam Smith2019.

94. Ibid.

95. Our thanks to Lady Alexandra Wedgwood and Terry Jardine for their advice on this matter.

96. Hallam Smith Reference Hallam Smith2019.

97. Palgrave, Reference Palgrave1869, 186–7.

References

Primary sources

ancestry.co.uk, electoral registers for the parish of Westminster, 1950–1 (accessed 19 November 2019)Google Scholar
BL, 604.i.19, Sandford, F 1687. The History of the Coronation of the most High, most Mighty and most excellent Monarch James II, T Newcomb, LondonGoogle Scholar
BL, Egerton ms 3320, fol 12v, Anon, ‘Drawing of the processions, ceremonies etc, attending the coronation of Queen Elizabeth I, 14–15 Jan 1558/9’Google Scholar
BL, RB.23.c.669, Anon 1727. Brittannia Illustrata or Views of All the King’s Palaces etc, Henry Overton and J Hoole, LondonGoogle Scholar
BM, Drawing 1848,0911.748, Anon, ‘Westminster Hall, the south end’, c 1620sGoogle Scholar
BM, Drawing 1891,0713.376, Anon, ‘Unmounted fan-leaf, with the coronation banquet of George II’, 1727–30Google Scholar
Bodleian, Gough Maps 23, fol 40b, Anon, ‘Plan of the south end of Westminster Hall’, c 1704Google Scholar
Findmypast, TNA, HO 107 (1851 census), 1479, 1491, 1512 (accessed 19 November 2019)Google Scholar
Hatfield House, Cecil Papers 24/61 (112) and 24/62 (113), John Symonds, ‘Two plans of the lodgings, late Mr Peter’s, within St Stephen’s in Westminster’, 1593Google Scholar
Hist Eng, AL 1121, ‘St Stephen’s Cloister, Palace of Westminster, vol 3, 1949’, album of photographsGoogle Scholar
Hist Eng, Palace of Westminster Collection, Chest 42, Ministry of Works, ‘St Stephen’s Cloisters, plan of built up doorway in the east wall of Westminster Hall’, 23 August 1949Google Scholar
Hist Eng, PSA01/08/00003, G A Checkley, ‘Westminster Palace (Houses of Parliament), plan shewing the relative approximate positions of the existing buildings and of the Palace … in the 17th and 18th centuries’, 1932Google Scholar
Lamptech, historic lamp catalogues, Osram-GEC, September 1951, UK, Osram, the Wonderful Lamp, <http://www.lamptech.co.uk/Catalogues.htm> (accessed 13 January 2022)+(accessed+13+January+2022)>Google Scholar
MoL, Prints and Drawings Collection A15453, Adam Lee, ‘Plan of His Majesty’s ancient Palace of Westminster and adjacent buildings’, 1807Google Scholar
MoL, Prints and Drawings Collection A15610, H-F Gravelot, Westminster Hall in Termtime, before 1739Google Scholar
National Portrait Gallery, NPG 798, Benjamin Ferrers, The Court of Chancery During the Reign of George I, oil painting, c 1725Google Scholar
PA, BAR/1/55, letter from Charles Barryto Thomas Greene MP, 10 September 1851Google Scholar
PA, HC/LB/1/114/19–22, plans of the temporary Houses of Parliament published by Vacher, 1835–47Google Scholar
PA, HC/LB/1/119, volume of drawings and engravings relating to the Palace of Westminster before the fire of 1834, collected by John Britton; fol 15c, plan of the sites of the Houses of Parliament, Westminster Hall etc, c 1836; fol 31a, L N Cottingham, ‘Plans, elevations, sections and details at large of Westminster Hall by Cottingham Architect’, 1822, pl 2Google Scholar
PA, HC/SA/SJ/9/50, rebuilding of the House of Commons, file, 1944–51Google Scholar
PoW, Parliamentary Art Collection, WOA 204, William Ewart Lockhart, Sir Reginald Palgrave, oil on canvas, 1887Google Scholar
PoW, Parliamentary Art Collection, WOA 673, Charles Mosley, The First Day of Term, print, 1797Google Scholar
PoW, Parliamentary Art Collection, WOA 1669, George B Campion, Westminster Hall on Fire, watercolour, 1834Google Scholar
PoW, Parliamentary Art Collection, WOA 2811, Thomas Bowles, A View of the Inside of Westminster Hall, print, 1723–4Google Scholar
PoW, Parliamentary Art Collection, WOA 4265, Edward N Holmes, The North Cloister, watercolour, 1871Google Scholar
RCT, RCIN 33467, Hubert Le Sueur, Charles I (1600–1649), bronze bust, mid-seventeenth centuryGoogle Scholar
RCT, RCIN 601957, Peter Mazell, bust of Charles i, engraving, c 1791–7Google Scholar
SAL, Red Portfolio, John Carter, ‘Plan of the present remains of St Stephen’s Chapel, Westminster’, 1791–2Google Scholar
SM, SM 37/1/24, Anon, ‘Survey of the Palace of Westminster’, plan, 1760–6Google Scholar
SM, SM 37/1/25, ‘Survey of the Palace of Westminster’, plan, 25 January 1793 copy from an original of 1718 by William BensonGoogle Scholar
SM, SM 37/1/28, John Thomas Groves, ‘Plan of the late Duke of Newcastle’s house in the Exchequer, since the Speaker’s House, [ground floor]’, 1794Google Scholar
TNA, AO 1/2446/139, Auditors of the Imprest and Commissioners of Audit, declared accounts,transcribed by Simon NealGoogle Scholar
TNA, E 351/3217, Exchequer, Pipe Office, declared accounts, 1582–3Google Scholar
TNA, E 351/3274, Exchequer, Pipe Office, declared accounts, 1660–1, transcribed by Simon NealGoogle Scholar
TNA, MPE 1/327, copy of an original by John Soane,‘Plan of Westminster Hall, Houses of Parliament and adjacent ground and buildings’, 1793Google Scholar
TNA, WORK 5/2, 5/11, 5/32, 5/52, Office of Works, accounts, transcribed by Simon NealGoogle Scholar
TNA, WORK 11/9/7, Houses of Parliament, architect’s progress reports, 1842–60Google Scholar
TNA, WORK 11/24/2, accommodation for state trials in Westminster Hall, file, 1746Google Scholar
TNA, WORK 11/445, House of Commons, reconstruction of cloisters, file, 1945–51Google Scholar
TNA, WORK 14/3127, St Stephen’s cloister, Houses of Parliament, Westminster,works, file, 1947–51Google Scholar
TNA, WORK 29/20, printed in Port2011Google Scholar
TNA, WORK 29/79, Robert Smirke, ‘Plan showing the general arrangement of the proposed buildings for the Houses of Parliament’, 1835Google Scholar
TNA, WORK 29/1149, [workshop of Charles Barry], ‘Details of coat horses in cloisters and plan showing position of them’, 9 December 1851Google Scholar
TNA, WORK 29/1173, [workshop of Charles Barry], ‘Plan and details of coat bosses in lower cloisters’, 26 November 1851Google Scholar
TNA, WORK 29/3239, ‘Plan of Westminster Hall as at the trial of the Rebel Lords’, 1746Google Scholar
TNA, WORK 29/3243, ‘Erections in Westminster Hall for the trial of Lord Melville’, plan, 1806Google Scholar
TNA, WORK 29/3288, C J Richardson,‘Plan of the Parliamentary and other public buildings adjacent to Westminster Hall’, August 1835Google Scholar
Westminster City Archives, WCBG/WCB/1/13, Westminster Court of Burgesses, ‘Minute Book of the Court of Burgesses’, 1762–93Google Scholar

Published primary sources

Belcher, M (ed) 2015. The Collected Letters of A N W Pugin, V, 1851–52, Oxford University Press, Oxford Google Scholar
Belcher, M and Wedgwood, A (eds) 2018. ‘Letters from Pugin to Charles Barry’, True Principles: J Pugin Soc, 5 (3), 134–86Google Scholar
Howell, T B (ed) 1816. A Complete Collection of State Trials and Proceedings for High Treason and Other Crimes and Misdemeanours, Vol 7, Longman, Hansard and others, LondonGoogle Scholar
Port, M H (ed) 2011. The Palace of Westminster Surveyed on the Eve of the Conflagration, 1834, London Topographical Society Publications 171, London: facsimile and commentary of TNA, WORK 29/20–27, Thomas Chawner and Henry Rhodes ‘Plans of the Houses of Parliament, Westminster Hall and other adjoining buildings’, 1834Google Scholar
Wheatley, H B (ed) 1987. John Stow, The Survey of London, Everyman Classics, London Google Scholar

Secondary sources

Anon 1843. Illustrated London News, 62 (3), 8 JulyGoogle Scholar
Anon [1907]. The Illustrated and Popular Guide to the Houses of Parliament and the Palace of Westminster, The Pall Mall Press, London Google Scholar
Aylmer, G E 1974. The King’s Servants: the Civil Service of Charles I, 1625–42, 2nd edn, Routledge and Kegan Paul, London Google Scholar
Barber, B, Holder, N and Phillpotts, C 2013. ‘Excavations at Cromwell Green and Westminster Hall, 2005–06’, Trans London and Middlesex Archaeol Soc, 64, 109–41Google Scholar
Biggs, E 2020. St Stephen’s College Westminster: a royal college and English kingship, 1348–1548, Boydell, Woodbridge Google Scholar
Boase, T S R 1954. ‘The decoration of the New Palace of Westminster, 1841–1863’, J Warburg and Courtauld Inst, 17, 319–58Google Scholar
Brayley, E W and Britton, J 1836. The History of the Ancient Palace and Late Houses of Parliament at Westminster, John Weale, London Google Scholar
Collins, M, Emery, P, Phillpotts, C, Samuel, M and Thomas, C 2012. ‘The king’s high table at the Palace of Westminster’, Antiq J, 92, 197243 Google Scholar
Colvin, H M (ed) 1963. The History of the King’s Works, Vols I–II: the Middle Ages, Ministry of Public Buildings and Works, London Google Scholar
Colvin, H M (ed) 1976. The History of the King’s Works, Vol V: 1660–1782, HMSO, London Google Scholar
Crook, J M and Port, M H 1973. The History of the King’s Works, Vol VI: 1782–1851, HMSO, London Google Scholar
Forster, J 1860. Arrest of the Five Members by Charles the First, John Murray, London Google Scholar
Gardiner, S R 1893. History of the Great Civil War, 1642–49, Vol 4: 1647–49, Longmans, London Google Scholar
Gentleman’s Mag 1790. P Q, 60 (2), 1,189Google Scholar
Gentleman’s Mag 1791. Charles Thorndon, 61 (2), 904Google Scholar
Gentleman’s Mag 1800. An Architect, 70 (1), 217–7Google Scholar
Gerhold, D 1999. Westminster Hall, Nine Hundred Years of History, James & James, London Google Scholar
Goodway, D 1982. London Chartists, 1838–1848, Cambridge University Press, Cambridge Google Scholar
Hallam Smith, E 2019. ‘Parliaments, politics and people seminar. The “Gothic slum”: MPs and St Stephen’s cloister, 1852–2017’, The History of Parliament, <https://thehistoryofparliament.wordpress.com/2019/09/26/parliaments-politics-and-people-seminar-the-gothic-slum-mps-and-st-stephens-cloisters-1852-2017/> (accessed 13 January 2022)+(accessed+13+January+2022)>Google Scholar
Hallam Smith, E 2020. ‘A passage for kings and commoners’, Brit Archaeol, 173 (Jul–Aug), 1621 Google Scholar
[Hatsell, J] 1818. Precedents of Proceedings in the House of Commons, Vol II, new edn, Luke Hansard and Sons, London Google Scholar
Howarth, D 1989. ‘Charles I, sculpture and sculptors’, in MacGregor, A (ed), The Late King’s Goods, 73113, Oxford University Press, Oxford Google Scholar
Hunneyball, P 2002. ‘Conjectural plan of the Palace of Westminster, c 1640’, in Kyle, C R and Peacey, J (eds), Parliament at Work: parliamentary committees, political power and public access in Early Modern England, xiixiii, Boydell Press, Woodbridge Google Scholar
Lawson, J and Silver, H 1973. A Social History of Education in England, Methuen and Co, London Google Scholar
McKay, W 2013. ‘Palgrave, Sir Reginald Francis Douce (1829–1904)’, Oxford Dictionary of National Biography online edn, <https://doi.org/10.1093/ref:odnb/35366> (accessed 13 January 2022)+(accessed+13+January+2022)>Google Scholar
Miles, D and Loader, N 2020, ‘Dating the Westminster passageway with isotopic dendrochronology’, Brit Archaeol, 173 (Jul–Aug), 22–3Google Scholar
Palgrave, R F D 1869. The House of Commons, Illustrations of its History and Practice: a course of three lectures delivered to the Reigate South Park Working Men’s Club … 1868, Macmillan & Co, London Google Scholar
Pennant, T 1791. Some Account of London, 3rd edn, John Archer, LondonGoogle Scholar
Port, M H (ed) 1976. The Houses of Parliament, Yale University Press, New Haven Google Scholar
Proctor, K 2020. ‘Westminster: secret passageway discovery reveals 169-year-old graffiti’, The Guardian, 26 February, <https://www.theguardian.com/politics/2020/feb/26/westminster-secret-passageway-discovery-reveals-169-year-old-graffiti> (accessed 13 January 2022)+(accessed+13+January+2022)>Google Scholar
Shaverian, A 2020. ‘Secret passage dating to 1660 is found inside UK parliament’, The New York Times, 27 February, <https://www.nytimes.com/2020/02/27/world/europe/uk-parliament-secret-door.html> (accessed 13 January 2022)+(accessed+13+January+2022)>Google Scholar
Shenton, C 2012. The Day Parliament Burned Down, Oxford University Press, Oxford Google Scholar
Shenton, C 2016. Mr Barry’s War: rebuilding the Houses of Parliament after the Great Fire of 1834, Oxford University Press, Oxford Google Scholar
Smirke, S 1836a. ‘Remarks on the architectural history of Westminster Hall’, Archaeologia, 26, 406–14Google Scholar
Smirke, S 1836b. ‘Second letter from Sydney Smirke, Esq FSA, to Sir Henry Ellis, FRS Secretary, on the architectural history of Westminster Hall’, Archaeologia, 26, 415–21Google Scholar
Smith, J T 1807. Antiquities of Westminster: the Old Palace, St Stephen’s Chapel, etc, J T Smith, LondonGoogle Scholar
Topham, J 1795 [–1811]. Some Account of the Collegiate Church of Saint Stephen, Westminster, SAL, LondonGoogle Scholar
Williams, O 1953. Topography of the Old House of Commons, Ministry of Works, London Google Scholar
Figure 0

Fig 1. Hist Eng, PSA01/08/00003, overlay plan showing the old and new Palaces of Westminster, 1932. The survivors are Westminster Hall, St Stephen’s cloister and St Mary Undercroft, shaded in orange. North is on the left. Image: © Historic England Archives, reproduced with kind permission.

Figure 1

Fig 2. Location plan of the doorway (red circle). North is at the top. Drawing: John Crook.

Figure 2

Fig 3. Concealed panel of 1950 providing access to the door passage. Photograph: John Crook.

Figure 3

Fig 4. The plaque in Westminster Hall. Photograph: John Crook.

Figure 4

Fig 5. Development of the opening, showing state in 1661 (a), 1846 (b), 1851 (c), 2021 (d). Drawing: John Crook.

Figure 5

Fig 6. Left: the only identified contemporary image of the doorway, seen at the south end of Westminster Hall, as shown in the rectified detail (right). The central doorway at the south end of Westminster Hall, between the two law courts, is also shown. Image: reproduced from Brittannia Illustrata (BL, RB.23.c.669, 1727, pl 26).

Figure 6

Fig 7. Left: location of the plaque and the studs; the two pale blocks of Huddleston stone are those that were replaced in 2014. Photograph: John Crook. Right: remains of the right-hand (south) jamb of the stone entrance arch to the passageway as recorded in September 2014 during repairs to the walls. Photograph: © Mark Collins, reproduced with kind permission.

Figure 7

Fig 8. The ceiling of the door passage (west at top). Photograph: John Crook.

Figure 8

Fig 9. Elevation of rear wall of passageway. Drawing: John Crook.

Figure 9

Fig 10. Rear (west) wall showing the short-lived opening of 1846. It was blocked in 1851, as recorded by a graffito behind the glass panel to its left. Photograph: John Crook.

Figure 10

Fig 11. Left: north wall. Brickwork of 1660–1 is apparent behind the thin plaster render. Right: south wall. In the adjacent east wall is yellow brick infill of the opening made on 21 June 1949. Photographs: John Crook.

Figure 11

Fig 12. Fan print of the coronation banquet of George ii, 1727. The bust of Charles i is on the far left. Image: BM, Drawing 1891,0713.376, © The Trustees of the British Museum, reproduced with kind permission.

Figure 12

Fig 13. Detail from a plan of 1760–6 showing the doorway and the route leading from it at ground level through St Mary Undercroft (also at that time used by the Burgess Court of the City of Westminster), and then on into Old Palace Yard and the Long Gallery. Plan: Soane Museum SM 37/1/24, © Sir John Soane’s Museum, London, reproduced with kind permission.

Figure 13

Fig 14. Peter Mazell’s engraving of the Le Sueur bust of Charles i. Drawing: Royal Collection Trust RCIN 601957, © Her Majesty Queen Elizabeth ii, reproduced with kind permission.

Figure 14

Fig 15. MPs’ pocket plan of 1846 showing the doorway unblocked. Plan: Parliamentary Archives HC/LB/1/114/21, reproduced with kind permission.

Figure 15

Fig 16. Graffito of 1851 written by Thomas Porter, bricklayer’s labourer. Photograph: © Adam Watrobski, reproduced with kind permission.

Figure 16

Fig 17. The opening formed in 1950. Photograph: John Crook.