At The Portrait Painter’s
David Mannings describes how the painters of the eighteenth century conducted their studios and sittings.
In a well-known letter to Lady Mary Wortley Montagu concerning her sitting for her portrait by Sir Godfrey Kneller, Alexander Pope wrote that Kneller ‘thinks it absolutely necessary to draw the Face first, which he says can never be set right on the figure if the Drapery and Posture be finished before. To give you as little trouble as possible, he proposes to draw your face with crayons, and finish it up, at your own house in a morning; from whence he will transfer it to the Canvas, so that you need not go to sit at his house.’
The unfinished portrait of Richard Boyle, 2nd Viscount Shannon, in the National Portrait Gallery, shows quite clearly how much attention Sir Godfrey paid to getting the face right before spending valuable time on the costume or background.
But this method led to some unfortunate misunderstandings, and it seems ironical that in the period which saw the greatest achievements of British portraiture there were many intelligent critics and artists who looked down on mere ‘face-painting’ as it was called, and considered it little better than hack work.
To be fair, the expression ‘face-painting’ was perfectly justified. It was quite common in Sir Joshua Reynolds’s studio, later in the century, for a head to be pretty well completed before the rest of the figure was sketched in. Thus we find Samuel Johnson, clerk, son of Reynolds’s sister Elizabeth, writing to his mother about his portrait by Reynolds’s pupil James Northcote, which was being painted in his uncle’s studio, and so, we may guess, very much under Reynolds’s eye.
‘This morning cousin John and I went to sit for our Pictures; I shall be drawn in my Oxford Dress if I can get one in Town.’
This referred to the first sitting, on February 8th, 1775. On February 22nd he wrote to his sister Frances saying:
‘I cannot get a gown, and fear I shall not be able to get one at all; however my face is pretty nearly finish’d, and I assure you, tho’ it is my own, I am prodigiously pleased with it;... what would you like that I should be drawn in if I cannot get a gown? I shall be glad to have your advice.’
After several more sittings he wrote again, ‘My face is pretty nearly finished’, though the costume was still not done.
This was not exceptional, either in terms of the number of sittings required or in the way the head would be worked on intensively before much definite thought was given to the rest of the composition.
And it is hardly surprising that this tendency to treat different parts of the same portrait as virtually independent of one another should have gone hand in hand with the actual delegation of what were considered the less important parts to assistants. This led eventually to the emergence of the fully-fledged drapery specialist, who not only painted nothing else, but served not one but several ‘face-painters’. At the same time it is interesting that the change from delegation within the studio to ‘putting out’ work provoked a suspicion that does not seem to have attended the age-old practice of using assistants as such.
The heyday of the drapery-painter was the later 1730s and ’40s, and the master of the craft was undoubtedly the Flemish immigrant Joseph Vanaken. André Rouquet, a Swiss visitor to England, recalled:
‘that extraordinary sight at Vanhaken’s, where he used to have canvases sent him from different parts of London, and by the stage coaches from the most remote towns in England, on which one or more masks were painted, and at the bottom of which the painter who sent them took care to add the description of the figures, whether large or small, which he was to give them. Nothing can be more ridiculous...’
There were even attempts to monopolise Van-aken’s services. Two painters actually came to some such arrangement with him; and this alarmed their fellow portraitists, for ‘the best of them knew not how to draw a hand, a coat, or ground’.
Yet in the late 1750s, ten years after Vanaken’s death, when Reynolds, due to pressure of work, found it necessary to delegate parts of his portraits, the public was sometimes a little surprised by the practice. Lord Bath wrote to Mrs Montagu that, while visiting Reynolds to arrange a sitting,
‘I have discovered a secret... that I fancy, he is sorry I should know. I find that none of these great Painters finish any of their Pictures themselves. The same Person, (but who he is, I know not) works for Ramsey [i.e. Allan Ramsay] Reynolds, and another, called Hudson, my Picture will not come from that Person till Thursday night, and on Fryday it will be totally finished, and ready to send home.’
Yet the putting-out system was common to most London trades in the eighteenth century, from watch-making to silk-weaving, and Lord Bath’s surprise to find it was also used by painters seems odd. (The portrait itself is now in the National Portrait Gallery).
The only important painters who did not use assistants to paint drapery and backgrounds to their portraits were Thomas Gainsborough, William Hogarth, Jonathan Richardson (though he did occasionally collaborate with a landscape ‘specialist’), and Joseph Highmore. According to one early writer, William Sandby, Gainsborough did use a drapery-painter, namely Peter Toms, who worked for Reynolds too.
But it is hard to find much evidence to support this, and Philip Thicknesse, his friend and earliest biographer, insisted that he painted every bit himself. Indeed, Gainsborough despised the whole practice of drapery-painting, writing to his friend William Jackson to say
‘There is a branch of painting next in profit to Portrait and quite in your power without any more drawing than I’ll answer for your having, which is Drapery and Landskip backgrounds. Perhaps you don’t know that whilst a Face painter is harrased to death the drapery painter sits and earns 5 or 6 hundred a year, and laughs all the while...’
Most portrait commissions were probably made directly, more often by personal approach, as when Lord Bath visited Reynolds to arrange a sitting, than by post; but it is in the nature of such an informal procedure that little in the way of evidence survives. A satisfied customer would simply recommend the painter concerned to his acquaintances, and a clientele would be built up from direct personal contacts.
It was only where artists were forced to devise their own ways of gaining custom that they attracted the attention of people like George Vertue, the antiquary and engraver whose forty volumes of notes are a mine of information on this period. Thus he records the cunning method used by one B. A. Lens, son of the miniaturist to George I and II. Lens would arrange for his friends to take a miniature to the theatre and show it to each other across the auditorium from one box to another:
‘(One) askes who did it, the other answers Mr Lens, pray where does he live, thus in a full house - I don’t know, but if any person here can tell, pray speak out, that we may know where to find such an Ingenious man’.
Travelling portraitists would place advertisements in local newspapers, for instance in the Ipswich Journal'. ‘Mr Ferguson, Limner in China ink, who offers to draw a portrait on vellum and supply it with frame and glass complete for 15/-’.
The commonest method of self-publicity was through engraving. Reynolds got James McAr-dell to engrave his portrait of Lady Charlotte Fitzwilliam in 1754, that is, almost as soon as he had settled in London; and he had the job done at his own expense. We also find cases where artists hoping for really major patronage would go ahead and produce a portrait, usually in the form of an engraving, without waiting for a commission at all.
This would sometimes, if the artist was lucky, lead to the hoped-for commission. Highmore used this method to bring himself to the attention of George II and Queen Caroline, though it is uncertain whether he was actually granted a sitting as a result. He was successful, however, when he published some engravings to accompany the revival of the Order of the Bath by George I in 1725. His portrait of John Sidney, sixth Earl of Leicester, was one direct result of this piece of self-advertisement.
The normal thing was, of course, an appointment to sit. In a few cases we hear of artists going to their customers’ homes. The poet and landscape gardener William Shenstone wrote to a friend on September 14th, 1761:
‘My picture shall be sent to you as soon as I can possibly get Alcock over, who has promised to come every week for these 3 months past. I believe he will come soon’.
And there was a painter called John Foldstone who painted people in their own homes. He usually began in the morning, lunched with them, and finished his picture before dark. We can safely infer that only the less successful portrait painters were forced into this routine, and very little is in fact known about either of these two.
The true mark of success was the busy and well-attended studio, with sitters arriving by carriage one after another. Rouquet’s account of the French immigrant J. B. Vanloo’s organisation gives us a vivid glimpse of the system in full swing in the late 1730s: ‘He sent for several painters to his house, who helped him to dispatch the great number of pictures which he had always in hand’.
These were said to number hundreds, and Vanloo ‘was obliged to take five sittings a day; the man who kept the list of these sittings was very handsomely rewarded for putting your name down before it came to your turn, which was often-times six weeks after you had first presented yourself to have your picture drawn.’
Reynolds received, at the height of his first great period of success in the late 1750s, six sitters a day, and Edmond Malone tells us Sir Joshua usually rose about eight o’clock, and was in his studio before ten. But this may reflect only the routine of the winter months, for the working day in Georgian London started very early - six o’clock was normal in most trades - and another source tells us that Reynolds usually began painting at six, resumed after breakfast until lunch, rested on the sofa till teatime, and carried on working as long as the twilight permitted. In fact it is most unlikely that he started quite so early. Who would have posed at six a.m.? Clearly the light was a crucial factor, and it is unlikely that he took sittings in the foggy winter afternoons.
Careful analysis of Reynolds’s own Sitter-Books suggests that in the normal course of events he took only three sittings each day; the first would be at nine, the second at twelve noon, and the third at two P.M. By spring he occasionally increased this to four or even five sittings a day, in some cases starting as early as seven A.M. Probably he did not work longer hours than his fellow-artists. J. T. Smith tells us that the very successful portrait-sculptor, Joseph Nollekens, rose at dawn in the summer, and worked till eight o’clock, when he breakfasted.
We must bear in mind that tailors worked from six in the morning till eight in the evening, bookbinders from six till nine, while shopkeeping hours were usually from seven in the morning till eight at night around the mid-century, and actually grew longer as the century wore on. It is difficult to know with which group artists should be compared; but it is important to remember that in most trades that involved the manufacture of goods for sale the hours were very long.
Northcote has left us a description of the studio into which Reynolds’s sitters would be shown. (The ‘painting room’ was separate, and was used for displaying pictures). It was octagonal, about twenty feet long and sixteen feet wide, and lit from a small square window high above.
The sitter’s chair (now in the Royal Academy) was ‘raised eighteen inches from the floor, and turned round on castors’, which suggests something like the arrangement we see in Rowlandson’s 1798 aquatint, At the Portrait Painter's. By the time Northcote came to live with him, Reynolds was well established and very prosperous, with, in addition to an elegant room in which to paint, a gallery to show his work, the result of his remodelling of the grand house he had bought in Leicester Fields (now Leicester Square) in 1761.
Competition was sharp, and it is not surprising to hear of artists going to some trouble to ingratiate themselves with sitters. Richardson, for example, is said to have worn a long fashionable curly wig when painting important people, and Vertue praises Kneller’s pleasant conversation and manners:
‘It is always observed that his performance is with so much facility that no seeming confinement to sitting (as is usual) that People go away with more sprightliness than they came’.
According to another contemporary source, Kneller made use of an arrangement of mirrors, so that sitters could watch the progress of their own likenesses, a device that Reynolds also adopted in a modified form. Nollekens, we are told, once dressed himself up in a ‘truly ridiculous manner’ when ‘an illustrious personage was sitting for his bust’.
Reynolds seems, on the whole, to have managed an easy relationship with most of his customers. In particular, he was famous for his success with children, and the best evidence for this must be the sheer quality of his child-portraits; no artist since Van Dyck had caught so well the shyness and intensity of children.
Posing does not come naturally to them, and Reynolds must have had a ‘way’ with him that put them at their ease. Perhaps one factor that helped was his lack of ‘polish’, an almost rustic quality noticed and recalled by at least one child who knew him. Cornelia Knight, later Lady Companion to the Princess Charlotte of Wales, spent her childhood in London where she knew Sir Joshua’s young niece, Offie. She remembered that ‘his pronunciation was tinctured with the Devonshire accent; his features were coarse, and his outward appearance slovenly...’ But this may well have intrigued and relaxed his youngest sitters.
Although the use of a camera obscura had been demonstrated at the Royal Society in 1670, where Evelyn saw it used ‘to take the profile of ones face as big as the life; which it did performe very accurately... ’ we do not hear of mechanical aids being much used by portrait painters in the eighteenth century. On the other hand, the use of controlled lighting to help establish the form of a head was common.
To take just one example, we have Gainsborough at Bath starting his portraits in a partially darkened room, so that he would not be led to pay too much attention to details of features before he had got the broad shape of the head right. Only then would he let in more and more light until the picture was complete.
The single high window in Reynolds’s studio was intended to provide similar conditions to Gainsborough’s darkened room, a way of controlling the direction and amount of light falling on the model. In fact, the idea was an old one, and can be traced back to Leonardo da Vinci, who advocated painting portraits at twilight, in a courtyard specially prepared with walls painted black, in order to achieve ‘grace in the light and shadows’ and a suitable air of mystery.
Once settled in a suitable pose in the studio, how long would a sitter expect to wait before being presented with a finished likeness? How many times would the sittings be resumed? And to what extent could the sitter prolong the commission by insisting on alterations or corrections to the picture?
Of course, no hard and fast answers can be given to these questions, but there is a certain amount of evidence to enable us to judge what was the normal time taken, what was considered quick, and what seemed excessively slow work. Certainly, few would expect a portrait to be finished in one sitting. Vertue tells of a ‘tryal of skill’ between a bumptious young Kneller and the older and more staid Sir Peter Lely, when the younger painter...
‘soon show’d how capable he was by readily drawing and Painting the Kings Picture at first sitting being like & seemingly finisht at once, to the pleasing Surprize of the King & the whole court & no less admiration of Sir Peter Lely who according to his manner had in the same time only laid in his picture in a broad manner, to finish at other times’.
And in 1731 he noted, not entirely approvingly, that Highmore ‘paints much, and the faces often at one sitting as much finisht, as he can-never touches any more...’ Although High-more achieved thereby a good likeness, it led to ‘a hard and smaller manner than the life’.
In a letter to a prospective client, Reynolds described his methods as requiring...
‘in general three sittings about an hour and a half each time but if the sitter chooses it the face could be begun and finished in one day... When the face is finished the rest is done without troubling the sitter.’
Most painters reckoned to complete their portraits in stages, and the first stage would consist in mapping in the broad shapes and establishing the balance of light and shadow, leaving details of features for future sessions. On June 20th, 1732, Thomas Hill wrote to John Conduitt about a portrait that Hogarth was painting:
‘Hogarth has but in a manner made a sketch of Lady Caroline. Nothing appears yet to any advantage. The next sitting will, I hope show something good. I think he has succeeded perfectly well in Miss Kitty’s face and air’.
In 1739 Vertue described Ramsay’s ‘new’ technique: ‘a way of painting faces the first setting - all red - or of lake or vermillion, the whole mask or visage over, one colour...’ This struck Vertue as a curious way of doing things, but it was certainly normal to devote the first sitting to some sort of underpainting. Such was the native tradition; only foreigners painted faces bit by bit. As Reynolds observed in his 14th Discourse, the Italian Pompeo Batoni...
‘in his portraits completely finished one feature before he proceeded to another. The consequence was, as might be expected; the countenance was never well expressed; and, as the painters say, the whole was not well put together’.
The recent cleaning of the French painter Arnulphy’s Captain Osborn at Greenwich, painted about 1744, has shown that his technique consisted in precisely this piecemeal method, taking one feature at a time and finishing it before going on to the next.
The Rev. William Mason, poet and amateur painter, made friends with Reynolds soon after his return from Italy, and arranged for Robert, fourth Earl of Holderness, to sit for his portrait. He attended every sitting, and described Reynolds actually setting to work:
‘On his light-coloured canvas he had already laid a ground of white, where he meant to place the head, and which was still wet. He had nothing upon his palette but flake-white, lake, and black; and without making any previous sketch or outline he began with much celerity to scumble these pigments together, till he had produced, in less than an hour, a likeness sufficiently intelligible, yet withal, as might be expected, cold and pallid to the last degree.
At the second sitting, he added, I believe, to the three other colours, a little Naples yellow, but I do not remember that he used any vermilion, neither then nor at the third trial.’
Writing to his brother Samuel in 1771, Northcote says that Reynolds advised him to use as few colours as possible. On the other hand, he notes how quickly Reynolds’s pictures cracked because of the way he used varnish with the colours, and goes on to say that Sir Joshua ‘always paints on the bare cloth unprepared, after the manner of the Venetians, whom he much admires (don’t show this part of the letter to anybody because Sir Joshua would not choose to have it known)...’ Which demonstrates how conscious Reynolds was of not always practising what he preached.
Northcote also recalled his master’s advice to restrict the number of colours as far as possible, on the principle that more colours meant muddier effects. And another of Reynolds’s Italian jottings stresses the importance of keeping the colours fresh and clean. Above all, colours should not be rubbed into one another once they are laid, but should be applied freshly over the top. In other words, mix on the palette, not on the canvas.
‘Sir Joshua’s canvas was close to me’ (Lady Burlington told Sir Francis Grant), ‘he took quite a quantity of exercise when he painted, for he continually walked backwards and forwards. His plan was to walk away several feet, then take a long look at me and the picture as we stood side by side, then rush up to the portrait and dash at it in a kind of fury. I sometimes thought he would make a mistake and paint on me instead of the picture.’
This procedure was copied almost exactly by the young Henry Raeburn, who was totally under Reynolds’s influence, and may have actually spent some time in his studio in the early summer of 1784.
And yet it is difficult to generalize about Reynolds’s procedure. There does not, for instance, seem to be any ‘average’ for the overall time taken; sittings, however, were almost always one hour, and in the 1767 Sitter Book we have what seems to be a complete record of the sittings for the small portrait of Dr Armstrong which now hangs in Adelaide Art Gallery.
There were no fewer than nine one-hour sittings, and this may have been typical, for John Wesley later praised George Romney at Reynolds’s expense saying ‘He struck off an exact likeness at once, and did more in one hour than Sir Joshua did in ten.’
Gainsborough, so far as we know, took less ‘exercise’ while painting. Instead of walking backwards and forwards, he set up his canvas not alongside, but at right angles to the sitter, and he used very long brushes so that he could stand still and touch the canvas at the same distance that he viewed his sitter. J. T. Smith says Gainsborough allowed him frequently to stand behind him and watch him at work; but his statement that Gainsborough used six-foot brushes must be wildly exaggerated.
Smith says that Gainsborough stood while painting, and it is certain that this was Reynolds’s method too. Indeed, Malone says Reynolds introduced the practice in contrast to the more usual way of working sitting down. Once again, the evidence hardly permits us to be dogmatic; but, if artists’ self-portraits can be relied on, it seems that most painters, since the Renaissance at least, have sat down to paint. Probably the biggest single factor was the size of canvas.
Miniature painters, one assumes, will always have worked sitting down, if only to keep a steady hand, and any painter engaged on an eight- or ten-foot canvas must have stood up to reach it. The interesting question is whether Reynolds was the first to work standing up, irrespective of the size of the picture he was painting, simply because this was how he felt himself to be most efficient.
So far we have a fragmentary but nevertheless quite vivid idea of what went on in an eighteenth-century portrait painter’s studio while the sitter was there. It is interesting to see what happened after he or she left. The canvas, with the beginnings at least of a head on it, would be passed to an assistant, whose job it was to fill in, under the master’s supervision, much of the surrounding costume and accessories.
This seems to be the stage at which Reynolds abandoned the double portrait of Rockingham and Burke in the Fitzwilliam Museum. Northcote, referring to another picture, describes his own role in this procedure in his letters:
I am now about the drapery of a half-length picture by Sir Joshua of an old gentleman, one Mr. Calthrop which is a very fine head. This is the first I have ever painted from the layman, and I am much afraid how I shall do it... I shall make part of a building appear behind the curtain and a landscape in the background.
This was in April 1772; in November of the same year he tells of painting the drapery of the Duke of Cumberland, who was shown in his full Garter robes, and these were placed on the layman to be copied. We know that Gainsborough used a layman too, for he mentions it in a letter to Lord Shelburne; but it seems fairly certain that he did not normally hand half-finished portraits over to assistants.
Finally, if the customer was not satisfied he could, and sometimes did, require alterations to be made. ‘Again with Mr Reynolds,’ wrote Lord Cathcart in October 1753, ‘and was disagreeably surprised with the figure. After some reasoning he came to the opinion that it would not do. I breakfasted with him and stood to him a good while. I thought it was much improved, and he was extremely satisfied with the alterations so we parted in great good humour.’
The picture in question, now on loan to Heaton Hall, is one of Reynolds’s most ambitious essays in the Venetian manner, and the Earl’s patience must have been very welcome to the painter. In a letter dated December 13th, 1759, Sir Ralph Milbanke wrote to Thomas Pelham, afterwards First Earl of Chichester, about a portrait of young Master Pelham (formerly in the Hirsch coll.):
I was with Reynolds this morning to see what progress with little Tom. He has been at work upon him & promises to complete him this week.
I have desired him to make an alteration in the hair, which he agrees with me in. He has made it too close and smooth, which is an affront to little Tom’s mop pate. He says he always intended some alteration in the hair since he saw little Tom at Sheere, so I shall attend to see him have justice done.
And it was not always the client who prolonged sittings in this way. An unpublished letter amongst the Grey family papers at Garrowby Hall shows that sitters were occasionally called upon to give up more of their time because the artist himself was dissatisfied:
Devonshire House June 4. 1759
Dear Sister,
Mr. Reynolds the Painter did not approve of the picture he had done so that I was forced to sit over again for it, it is now finished, so I hope to send it you very soon. I am going to Chatsworth to morrow but must be back again by the 22nd of this month when I shall sit again & then it will be finish’d so I hope to send it over soon...’
Your most Affec. Brother
Devonshire
Gainsborough was sometimes called upon to alter his portraits too, as we know from his letters. At the same time he would insist on a proper sitting to do this:
‘I shall be extremely willing to make any alterations your Lordship shall require,’ he wrote to the Earl of Dartmouth, ‘when Her ladyship comes to Bath for that purpose, as I cannot (without taking away the likeness) touch it unless from the Life’.
Sometimes drastic alterations at a late stage must have been necessary; and this may perhaps explain why X-rays of Reynolds’s Field Marshall George, 1st Marquess Townshend, in Toronto, show that the head was painted on a small piece of canvas added to the large canvas. This procedure was sometimes followed by those artists, like Ramsay (in the double portrait of the seventh Earl of Wemyss and his wife, at Gosford) who used a drapery-painter, and was in fact the regular method used by some travelling portraitists, like Hamlet Winstanley, as described by Vertue:
He ‘travelled about and drew pictures from the life in oyl colours, often on small pieces of cloth only the face, pasted them when sent to London on larger Cloths...’
The rest was filled in, usually, by the indispensable Vanaken. A signed and dated example, a portrait of Thomas Patten of Bank Hall, has recently been purchased by Warrington Art Gallery. In Reynolds’s case, however, it is more likely that we have an example of fairly major repairs needed at a stage when it would not have been worth starting the whole portrait again.
It is a commonplace to say that the history of art cannot be separated from social and cultural history. Early Georgian portraiture is a particularly rich subject that the student of the period ignores at his peril.
Yet professional art historians too often seem excessively concerned with the finished product and many people who enjoy looking at the splendid portraits that fill our country houses are left to wonder what kind of organisation went into their commission and manufacture, and how they were received by the public.
Of course there are still many gaps in our knowledge of the procedures followed, but I think it is true to say that these gaps are mainly in points of detail. The broad outlines seem clear enough, and throw additional light on a fascinating subject.