London: Holy Trinity Church, Marylebone, designs for construction and interiors, 1820-34 (120)
Despite the Church Commissioners Act being passed in 1818, it was another two years before the three architects attached to the Clerk of Works Office: John Nash, Robert Smirke and John Soane were formerly approached by the Commissioners. In 1820, John Soane was asked by the Commissioners to design a church for the relatively populous and affluent area of Marylebone in London. This would be Soane’s first church commission, later to be followed by St Peter’s Walworth, and finally, St John’s Bethnal Green, all in London.
Indeed, by the time Soane’s Commissioner church was being devised, the area already had St Marylebone Parish Church (1813-1817) by Thomas Hardwick to the west, and to the east of Euston Square were the Old Parish Church of St Pancras which is claimed to go back to the Norman period; and the recently built New Parish Church of St Pancras (1819-1822) by the father and son team of William and Howard William Inwood. Nevertheless, there seems to have still been demand for another place of worship in the area of Marylebone.
The process of constructing Holy Trinity was a long, drawn-out affair, and took eight years to complete. Soane drew up initial designs in 1820, but found that this was at the start of a dispute between the Church Commissioners and the Marylebone Vestryman as to the site of the proposed church. At first, the Vestrymen favoured a location opposite Great Portland Street. Yet, the two sides could not agree on this location, and subsequently a further eight other sites were examined over a two-year period. Finally, the Vestrymen forced the issue on the Commissioners by purchasing land for a church which was — in the end — the very first site opposite Great Portland Street. The Commissioners did not respond to this purchase for almost a year; but faced with a fait accompli, they reluctantly agreed to this location. The Commissioners stated that the church should be designed to hold 2,000 parishioners and cost no more than £20,000.
Soane’s estimate for the first church he was planning was indeed £20,000. This was, for some unspecified reason rejected, and a second set of drawings from March-April 1821 was estimated by Soane at £19,760. None of this mattered, as the site had still not been finalised. Finally on 13 March 1822 when the issue of location was settled, Soane produced more elaborate and grander designs than before. When submitted, the cost according to the Building Committee Minutes was £28,600. Not surprisingly, given the strict total expense for any church set at £20,000, this scheme was also rejected. This may also have been around the time Soane had an architectural model made of his conception for Holy Trinity Marylebone (SM MR79), which bears some similarities to the 1822 design, especially when compared with some of the drawings produced in that year (SM 54/1/30-31; SM 54/3/2).
Throughout 1823, very little progress was made. In the collection of drawings in the Soane Museum, only two date to that year (SM 54/3/15-16). Soane was working on other projects, including his second Commissioner Church, St Peter’s Walworth, and the Privy Council and Board of Trade Offices in Whitehall, which took up a lot of time, but still this does not explain the relative inactivity concerning Holy Trinity.
By early 1824, Soane had not produced anything new for the Commissioners or Vestrymen to consider. As a result, the Vestrymen pressured the Commissioners to replace Soane the following year. Under fire, Soane produced more designs in early 1824, with a new estimate of £23,800. The Commissioners and the Vestrymen were not quite sure how Soane had made a reduction in cost of £4,200, and pressure still mounted for Soane to be removed. In March, the Vestrymen insisted that the Commissioners ask (or pressure) Soane to create designs for a neo-Gothic design for Holy Trinity. Despite Soane’s protests to the Commissioners, claiming they had ‘misunderstood’ his new plans, he put the office to work on the new designs. It could only really be at this time that Soane was also pressured to produce designs for an Anglo-Norman (Romanesque) version in addition. With regards to the latter, the documentary evidence is scarce, but a total of seven drawings for this scheme are in the Victoria and Albert Museum (Du Prey, 1985, 72-74 cat. Nos’ 222-224, 231-234, 237). The neo-Gothic designs (and the Romanesque) are essentially Soane’s previous classical design with an overlay of features from the Gothic and Romanesque styles. The historical context is Soane’s problems with the New Law Courts which commenced in March 1824. Soane had constructed a Palladian-style façade around the Law Courts, but was forced to change it to neo-Gothic by a Parliamentary Commission. Therefore, Soane was already under intense pressure, and it seems the Vestrymen saw an opportunity to increase the burden on Soane even further.
Fortunately for Soane, the alternative schemes did not come to fruition; but he had to work hard to convince the Commissioners that his original plans were best. This meant that some more work on the designs was carried out in autumn 1824. In November, Soane sent in revised estimates. The most expensive option was still £27,200, a second reduced option of £23,340, and the last which he marked as ‘B’ being £19,860. This last estimate was predictably approved. Nonetheless, the Commissioners and Vestrymen required further alterations to the interior of the Church, which delayed the process into 1825. More problems followed as the adjustments Soane had to make would cost £1,430, and included a ‘considerably enlarged tower’. This was met with approval by the Commissioners, but they would not provide extra capital for its execution. Of further embarrassment for Soane, in June 1825, the tenders turned out to be £1,115 higher than the new lower estimate. Soane explained the overshoot as being caused by a general increase in the cost of materials. A second set of tenders turned out to be even worse, some £2,600 over-budget, which Soane blamed on inflation, which it is true, as over the course of the mid-1820s’ it had risen considerably (1823: 6.8% 1824: 8.4% 1825: 17.4%; before falling into negative inflation the following year). Fortunately for Soane, he had received two private tenders, one from the stone-mason William Chadwick who had just finished working with Soane at St Peter’s Church, Walworth, and Daniel Sharp. In August 1825, Sharp’s tender of £19,524 was accepted (see SM 54/4/21; SM 54/4/26-8). A caveat was attached: if Sharp was asked to build the balustrade and the enlarged tower, it would cost no more than £1,113.
November 1825 saw the eventual decision to build. Nonetheless, there was one last problem, which, for once, all parties agreed on unanimously. The location of the church had been decided, and to give it an east-west axis. Unfortunately, surveys recorded the ground to be of very poor condition, and the foundations would take an extra £2,200 to rectify. Consequently, a decision was taken to re-orientate the Church to a north-south axis. As a result, the ground plans show the east-west-axis and then a pencil emendation to the north-south axis (SM 54/5/2). A nudge of the principal front to the south-west was also factored into the plans to at least give the Church a slight east-west orientation (SM 54/5/4). As it turned out, the ground was not much better on this axis either, and it took a further £700 to correct, but they decided to persevere.
Further, perhaps, unforeseen expense was required, as Soane’s amended designs had actually seen a reduction in seats of 290, which would cost an additional £278 to correct. Soane also calculated that the balustrade and enlarged tower, which were both approved in 1826, would cost £1,280. At this point, Soane put out a call for private donations to pay for these additions, guaranteeing that he would do the work at cost price and that it would not exceed the stipulated amount. The subscription was headed by the 4th Duke of Portland, William Bentinck (1768-1854). The work was carried out, and in the end cost the original price of £1,113 to execute.
The Church was consecrated on 31 May 1828, almost eight years after it was commissioned. The total cost was £24,596. This meant Soane’s 5% commission amounted to £1,229, a sum Soane would not receive until 1834. Holy Trinity, Marylebone was the second most expensive Commissioner Church after the neo-Gothic St Luke’s in Chelsea which was built from 1820-24 by James Savage at a cost of £28,109. To put the cost of these two Commissioner churches into perspective, the aforementioned St Pancras New Church cost £76,679, and the Parish Church at Marylebone was allegedly £80,000.
As it was, Soane also fulfilled the requirements that the Church Commission had set out for Holy Trinity in 1820. The church had 2,003 seats, a total of 1,269 pews, of which 734 (36.7%) were Free Seats.
Work continued after the consecration date, as the coffin-lowering mechanism seems to have been installed soon afterwards (SM 54/4/29), and a drawing of the front porch dates to 1834 (SM 54/3/1). Additionally, although not among the drawings for Holy Trinity, as it is for St Peter’s Church, Walworth (SM 64/1 /1, 8-10), the organ was made by Bishop and Son and installed in 1828.
Nevertheless, despite the ongoing disputes throughout the process from 1820-25, Soane must have thought very highly of his design. Soane had his office, or more likely, hired Joseph Michael Gandy to produce highly washed versions of the church in a variety of settings and configurations, all of which can be identified as Holy Trinity by the revised tower (SM 15/4/ 1-8; SM 60/194-5).
Two further designs by Gandy may be reflective of earlier 1824 prototype drawings by Charles James Richardson (SM Vol. 60/191; SM Vol. 60 /193). The first, is a highly washed drawing (SM 15/4/8) which may have been reserved for a Royal Academy Lecture. This shows the church designs lined up in a row, with the first set of three in the foreground being the Classical, Gothic, and Anglo-Norman designs for Holy Trinity Church, Marylebone. The Classical version of the Church is closest to the viewer, and so becomes more emphasised than the other styles.
The other is a framed watercolour (SM P259), which was exhibited at the Royal Academy Summer Exhibition of 1825 under the title ‘A groupe of churches, to illustrate different styles of architecture’ (RA No. 902). The centrepiece is Soane’s neo-Gothic, Anglo-Norman (Romanesque) versions of Holy Trinity, which is shown from the front and one long side, and the favoured neo-Classical design for Holy Trinity, Marylebone which is shown as a front elevation. Both the neo-Gothic and Romanesque versions are framing the neo-Classical Church on each side, as if acknowledging the inherent superiority of the latter. Each of the three types also has an accompanying rear perspective placed around the hills in the background. Of course, at this date Holy Trinity was still aspirational as construction had not yet begun, and the finished building differed in many respects to the watercolour, for example: the Doric in antis arrangement on the west front on both the drawing and watercolour eventually turned into Ionic columns supporting a porch, and a stone covering was included, but without the rusticated masonry envisaged in the drawing and framed watercolour.
The initial reaction to Holy Trinity Church, Marylebone was mixed. Christopher Davy’s review in the Mirror Newspaper (July 14, 1827 pp. 33-34, ten months before consecration), admired the tower and noted that the lower portion of the tower used columns based on those at the Temple of Vesta at Tivoli, near Rome. This was Soane’s favourite Roman temple, and casts of the capitals from the building appear in his collection (SM 75; SM 859; SM 958; SM 1027). He used the Vesta capital on many buildings, as for example the ‘Tivoli Corner’ for the Bank of England. Davy regretted the departure of a pediment from church architecture, but felt Soane had done as well as anyone to compensate by using a balustrade, especially as it meant the full tower was totally exposed. James Elmes in Metropolitan Improvements,1828, praised the Church in the opening sentence as ‘a very handsome and well built church’. He admired the pediment-free portico, even though he was not generally in favour of them, and claimed that Soane had executed the second best tower of the London churches after that of St Leonard’s in Shoreditch by George Dance the Elder (1740). Elmes still felt that the tower should have been on a separate base, and that the columns from the second portion of the tower, which he correctly identified as being based on the Tower of the Winds in Athens, were too inconspicuous compared to those on the first tier. Nevertheless, Elmes was not so enthusiastic about the Greek fret decoration, a point taken up by E. J. Carlos (see below), but Elmes also praised the ‘cinerary urns, or pyramidal sarcophagi’, by which Carr has correctly suggested that he actually meant the Soane caps and pinecone finials.
In 1826, Edward John Carlos, the architectural critic for the Gentleman’s Magazine, had reviewed St Peter’s Church, Walworth, for which he reserved some praise, especially for the interior. In 1829 he reviewed Holy Trinity, Marylebone. However, Carlos was far less favourable to Holy Trinity. He disliked the exterior because unlike St Peter’s which was brick with stone elements, at Holy Trinity the stone elements were interrupted at the corners by some exposed brick: ‘…a novel, it is true, but at the same time a tasteless style of execution’. He goes on to give a caustic assessment of the Greek fret decoration as ‘an ornament once very fashionable with the designers of fenders and tea-boards’ (Carlos, 1829, p. 297). Carlos would not excuse the lack of a pediment either — even though he barely mentioned this in his earlier St Peter’s review — as he felt it exposed the tower which already looked awkward and bottom-heavy, which is one aspect in common with Elmes’s assessment. Carlos did not write at great length about the interior as he saw it as extremely similar but not superior to St Peter’s. The similarity in basic conception of the two churches was something Carlos disliked the most:
'Though not absolute copies, there is that sameness of design which we have already censured as a fault in the works of inferior architects, and which we should not have expected in any building proceeding from the pencil of Mr Soane.' (Carlos, 1829, p. 297)
In 1833 Thomas Smith’s publication also gave a detailed description, and noticed the exposed brick corners, but unlike Carlos, Smith was generally positive, and perhaps more understanding of the limitations imposed on Soane by the Commissioners. This was especially true in terms of the tower, where Smith noted, ‘the parsimony, or rather meanness of public bodies, in the presentation of public works’ (Smith, 1833, p. 119) and how thanks to the public subscription the enlarged tower was a vast improvement over the original plan, especially the first tier:
'and it is no vague or unmerited compliment to the architect, to say that a more beautiful piece of eccesiatical architecture is not to be seen in the whole range of modern churches' (Smith 1833, p. 119)
This was in contrast to the second tier which was in Smith's opinion, underwhelming when compared with the first. However, Smith blamed the lack of funding on the project, rather than Soane as an architect.
It seems Carlos, unlike Smith, may not have been overly aware of all the problems Soane had faced with the project, and possibly assumed as more money was spent on this church than St Peter’s, it would be substantially different.
It is notable that these exposed brick corners — which do not appear in the corpus of drawings — were later covered up, perhaps by Soane in 1834. Thus, Holy Trinity is Soane’s only completely stone-covered church.
Smith also gives a more detailed description of the altarpiece, which does not appear in the corpus of drawings for Holy Trinity, but seems almost identical to the St Peter’s, Walworth version (SM 54/6/37), with the tablets showing the Ten Commandments, Lord’s Prayer and Creed. As with the review of St Peter’s by Carlos, Smith noted the sculpted dove with outstretched wings and irradiation, which does not appear in Soane’s designs for the St Peter’s altarpiece. As a result, Smith was more positive about Holy Trinity than Carlos.
In the twentieth century, Nicholas Pevsner was far more enthused about Holy Trinity, stating that Soane had built ‘the most elegant and expensive of the Neo-Classical churches of Marylebone’ and unlike Carlos, Pevsner thought the Grecian details on the Church had been used ‘with great elegance’, and Pevsner noted the Ionic columns on the side as ‘an unusual piece of care for the less important sides’ (Pevsner, 1991, pp. 40, 51, 600).
John Summerson was less impressed, but noted Soane had improved the relation between the horizontal and vertical when compared to St Peter’s, Walworth. Soane achieved this by projecting the front columns and adding a more substantial tower, although Summerson adds ‘but at Marylebone the Soane sensibility is lost and neither church is wholly successful’ (Summerson, 2003, p. 258).
This form of sympathetic appraisal can also be seen in the work of Carr who, like Summerson, believed Soane was hamstrung by a combination of the budget and the bickering of all the parties involved, and as a result it was not the church it possibly could have been. Du Prey commented that Soane had, after all the trial and tribulations of the design process, extracted ‘a handsome, if not very original church’ (Du Prey, 1985, p. 75).
Therefore, it is fair to say that since its consecration, the reception of Holy Trinity Church, Marylebone, has been mixed. It must be mentioned, however, that many of the critics seem to have been aware of the limitations imposed upon Soane.
At the present time, Holy Trinity is much changed: a new round apse at the back was constructed in 1876 by Somers Clarke; Soane’s latticed glass was replaced with stained-glass, as well as new pews and pulpits being installed in the Victorian period. By 1952, falling congregations meant that the parishes of Holy Trinity and Marylebone were reunited. Subsequently, Holy Trinity was converted to the base for the Society for the Promotion of Christian Knowledge (SPCK), had been a depository for the publishers Penguin, and in the early years of this millennium, the Church was deconsecrated and converted into an events space. Nevertheless, the exterior is still largely Soane’s, including the tower.
E. J. Carlos, ‘New Churches No. XXI: St. Peter’s Church, Walworth’, 1829, in TheGentleman’s Magazine, September, vol. 99, series 22/1, January-June, pp. 297-9; J. Elmes, Metropolitan Improvements Or London in the Nineteenth Century, 1831, pp. 81-4; T. Smith, A Topographical and Historical Account of the Parish of St. Mary-le-bone, comprising a copious description of its public buildings, antiquities, schools, charitable endowments, sources of public amusement, &c with biographical notices of eminent persons, 1833, pp. 118-122; A. Bolton, The Works of Sir John Soane F.R.S., F.S.A., R.A. (1753-1837), 1923; G. Carr, The Commissioners’ Churches of London, 1818-1837: A Study of Religious Art, Architecture and Patronage in Britain from the Formation of the Commission to the Accession of Victoria, 3 Vols, 1976 (PhD thesis, History of Art: University of Michigan); D. Stroud, Sir John Soane: Architect, 1984; P. Du Prey, Sir John Soane: Catalogues of Architectural Drawings in the Victoria and Albert Museum, 1985; B. Cherry and N. Pevsner, The Buildings of England: London 3: North West, 1991, (3rd ed.); D. Watkin, Sir John Soane: Enlightenment Thought and the Royal Academy Lectures, 1996; M. John, Sir John Soane’s Three London Churches, 2003, (Diploma, Architecture Association); J. Summerson, Georgian London, 2003 (5th Edition); P. Dean, Sir John Soane and London, 2006; M. H. Porte, Six Hundred New Churches: The Church Building Commission 1818-1856, 2006; R. Rossi, 'Sir John Soane and the Sepulchral Chapel', 2019, in Mausolus: Journal of the Mausolea and Monuments Trust (Summer 2018), p. 12.