+ Expand

Artist or Maker: Burne-Jones, Edward Coley (1833-1898)

+ Expand

Provenance: William Brockbank, of Brockhurst, Didsbury, Manchester; (+) Christie's, London, 27 February 1897, lot 33 (180 gns to Agnew's).
with Agnew's, from whom bought by Douglas W. Freshfield, of Wych Cross Place, Forest Row, Sussex, by 1898.
His sale; (+) Christie's, London, 2 November 1934, lot 6 (40 gns to Barbizon House).
with Barbizon House, from whom bought by Cecil French the same month, according to a label on the reverse.
Presented by his Executors to The Watts Gallery, Compton, in 1954.

+ Expand

Exhibited: London, Dudley Gallery, General Exhibition of Water Colour Drawings, 1872, no. 196.
London, New Gallery, Works of Sir Edward Burne-Jones, Bart., 1898-9, no. 16, lent by Douglas Freshfield.

+ Expand

Literature: Burne-Jones's autograph work-record (Fitzwilliam Museum, Cambridge), under 1871.
Athenaeum, no. 2310, 3 February 1872, p. 150.
Illustrated London News, vol. LX, no. 1691, 3 February 1872, p. 118.
Times, 13 February 1872, p. 4.
Academy, vol. III, 1872, p. 67.
Art Journal, 1872, p. 74.
Malcolm Bell, Sir Edward Burne-Jones: A Record and Review, 4th ed., London, 1898, pp. 42, 130.
Fortunée De Lisle, Burne-Jones, London, 1904, pp. 125, 183.
City of Birmingham Art Gallery; Catalogue of the Permanent Collection of Paintings..., Birmingham, 1930, p. 32.
Martin Harrison and Bill Waters, Burne-Jones, London, 1973, p. 191.

+ Expand

Notes: The Property of the Watts Gallery, Compton

Sold to ensure the long-term care of the Watts Gallery Collection of works by and about G F and Mary Watts

A puzzle

Anyone conversant with Burne-Jones's career is aware of the Troy Triptych, the ambitious scheme to tell the story of the fall of Troy in triptych form that he embarked on in 1870. Familiar, too, is the fact that the Triptych as an entity was never completed, although it had immense repercussions for his future development in terms of individual designs within the ensemble that he re-cast, often on a greatly enlarged scale, as independent easel pictures.

With one possible exception to which we shall return, The Triumph of Love represents Burne-Jones's earliest attempt to realise the four allegorical figures - Fortune, Fame, Oblivion and Love - that were designed for the Triptych's predella. But what is their exact status? Were they, as some evidence seems to suggest, painted as part of the Triptych itself, to be inserted into its elaborate Renaissance-style frame? Or are they in effect highly finished watercolour studies for the four allegories, to which Burne-Jones would subsequently have given more definitive expression?

The problem may never be fully resolved, but in this and the following section the evidence is considered and a possible solution proposed. We shall then turn to the more general question of the Triptych's sources of inspiration, before noting other versions of the predella compositions and discussing the interesting exhibition history and provenance of The Triumph of Love.

The first mention of the Triptych occurs in Burne-Jones's autograph work-record, now in the Fitzwilliam Museum, Cambridge. Among the entries for 1870 he simply states that he has 'designed the triptych of Troy'. Our four figures are the subject of the next reference, the first entry for 1871: 'I painted in water colour on canvas 4 figures of Fortune, Fame, Oblivion and Love for the Troy picture'. This would seem to mean that the figures were intended to form part of the Triptych itself, and were the first section to be so completed. But a doubt occurs as soon as we come to the very next entry for this year: 'A drawing in pencil of Venus with Graces and a background of lovers, also for Troy'. This is a reference to the highly-finished pencil study for another of the predella compositions, Venus Concordia, now in the Whitworth Art Gallery, Manchester (fig. 2). Since this drawing is undoubtedly a study for a yet-to-be painted version, it raises the question of whether the phrase 'for the Troy picture' which is used of our panels means what we originally assumed. Could they too be studies rather than the final works?

References to the paintings in the early Burne-Jones literature do little to clarify the issue. When Malcolm Bell, the artist's nephew by marriage, published the first monograph in 1892, basing it closely on the work-record, to which he clearly had access, he described the paintings as 'four allegorical figures for the (Triptych's) predella... sketched in water-colour upon canvas'. When they appeared in Burne-Jones's memorial exhibition at the New Gallery in the winter of 1898-9, they were identified as having been 'designed as a predella for "Troy"'. These statements, while they could be interpreted as meaning that the figures were painted as part of the Triptych's predella, are hardly categorical assertions to this effect, and by the time Fortunée De Lisle produced her account in 1904 (a more useful publication, incidentally, than its inclusion in Methuen's series of 'Little Books on Art' might suggest), a distinct note of vagueness has crept in. She merely tells us that 'four fine panels of "Fortune", "Fame", "Oblivion", and "Love"' were among the works 'designed' for the Triptych, making no distinction between them and paintings that were only based on 'Troy' compositions. Nor did the panels fare better when they passed through the saleroom. At the Brockbank sale in 1897, they were simply 'four drawings in one frame', and no description of any kind was given when they appeared at the Freshfield sale in 1934, at a time when the artist's reputation was in almost total eclipse.

As for the closer scrutiny to which Burne-Jones has been subjected since he returned to favour in the 1960s, this has never addressed the question of the panels' true identity. In Harrison and Waters' pioneering monographs of 1973 it is claimed that they are 'studies from (sic) the predellas (sic) of the Troy Polyptych', and their medium (not for the first time, as we shall see) was described as oil. No other publication, so far as we know, has considered the problem, nor have the panels been included in any of the Burne-Jones exhibitions that have taken place in recent times.

The Birmingham 'Story of Troy'

Any attempt to tackle the issue must take account of the unfinished canvas known as The Story of Troy in the Birmingham Art Gallery (fig. 4). Dating from the early 1870s, this was painted largely by assistants from Burne-Jones's designs, but has documented touches by the master himself. It is one of the artist's oddest productions, quite unlike anything else from his hand. A picture within a picture, it not only shows the Triptych as it would have looked if carried out in three dimensions but how it might have appeared in situ. It was presented to Birmingham in 1922 by the artist's children, Philip Burne-Jones and Margaret Mackail, and first published in the 1930 catalogue of the collection's paintings. The entry included notes by Philip and T.M. Rooke, Burne-Jones's most faithful and long-serving assistant. Rooke had joined the studio in 1869, a year before the Triptych was designed.

The picture provides the only comprehensive image we have of the Triptych's iconography. The three main panels are no more than outlines, put in by Rooke from Burne-Jones's drawings, but the predella is comparatively highly finished.

The story begins in the predella panel beneath the central subject, showing the marriage feast of Peleus, King of Thessaly, and the sea-nymph Thetis, destined to be the parents of Achilles (The Feast of Peleus; for a clearer image, see fig. 7). Zeus presides over the assembled deities, and Eris, the goddess of discord, stands hunched at the far right. Uninvited to the nuptials, she has thrown down an apple inscribed provocatively 'For the fairest', thereby sowing dissent when it is claimed by Venus, Juno and Minerva, who are seen standing behind the table to the left. The ensuing beauty contest, judged by the shepherd Paris on Mount Ida, is represented in the large panel above (The Judgement of Paris)

In the panel to the left of this subject, Paris, protected by Venus, to whom he has awarded the prize, carries off Helen to Troy (The Rape of Helen), while on the right Helen is re-captured in the burning city, which has been sacked by the Greeks led by her vengeful husband, Menelaus (Helen captive in Burning Troy).

The predella panels beneath these lateral panels show Venus presiding over scenes of peace and carnage, thus symbolising the passions, love and hate, unleashed by the dramatic events depicted above (Venus Concordia and Venus Discordia; see figs 2-3). These panels and the central Feast of Peleus are then divided and flanked by the four smaller allegories represented by our Triumph of Love. Reading from left to right, they show Fortune turning her wheel, to which a king, a slave and a poet are bound, followed by three winged figures: Fame, blowing a trumpet, overthrowing Fortune; Oblivion, wielding a scythe, conquering Fame; and finally Love, with bow and arrow, subduing Oblivion. Each vanquished personification reappears in the subsequent panels, thus making the whole sequence symbolise the ultimate triumph of love in human affairs.

But the picture tells us far more than how Burne-Jones intended to represent his theme. The eight painted compositions are set into a massive frame in the neo-Renaissance style. Each is first given an individual frame of metal relief, 'studied', according to Rooke, 'from plaques in the South Kensington Museum', and these in turn are incorporated into an elaborate structure of green and white marble with sculptural embellishments. The three main panels are divided by columns of green marble with Corinthian capitals, and flanked by deep pilasters of white marble adorned with bronze medallions representing four Trojan heroines: Oenone, Cassandra, Polyxena and Iphigenia. The entablature above, also of white marble and robust in form, includes a frieze with what Rooke calls 'babies struggling, involved in draperies', and the putti motif re-appears in the six free-standing bronze figures of boys seen at the bases of the columns and pilasters. They are supported by a wide green marble ledge, another of which juts out beneath the predella.

Even this by no means exhausts the information that Burne-Jones provides about his concept. The whole ensemble rests on four corbels projecting from a stone wall. Red draperies, embroidered with emblems and mottoes, hang to either side. Fruit and foliage are strewn across the marble ledges, while leafy branches and more drapery are festooned above the entablature. Had the triptych ever graced some rich man's home, we can imagine these elements being the weekly concern of his Aesthetically-minded wife or the responsibility of a parlour-maid considered to have good taste.

The great question raised by this extraordinary production is whether it represents the Triptych itself in an unfinished state or is an elaborate and unparallelled attempt to show what it would have looked like if carried out in three dimensions. 'It remains unclear', Stephen Wildman has written, 'whether the artist intended to complete the unfinished oil now at Birmingham, or whether it was meant as a vast design for an architectural ensemble that would incorporate paintings and sculpture'.

The issue is important in the present context because if the canvas is the actual Triptych, then it would appear to have a direct bearing on the status of our panels. Unfortunately the 1930 catalogue entry is of limited use in helping as reach a conclusion. Philip Burne-Jones's note on the picture merely states that it 'represents all that exists of the artist's elaborate scheme for a large painting illustrating... the story of Troy', a comment could mean almost anything. Nor are modern critics by any means of one view. Harrison and Waters believed the painting to be the Triptych itself, whereas for Wildman the surviving evidence 'suggest(s) that Burne-Jones did indeed have a three-dimensional execution in mind.'

We believe that Wildman is right and that the Birmingham painting is in the nature of an exceptionally ambitious and comprehensive sketch. Our main reason for believing this is the canvas's astonishing illusionism. This might be understandable in a sketch; indeed it is foreshadowed in a closely related but earlier drawing, described by Rooke as 'the original plan', that is also in the Birmingham collection (fig. 5). But everything we know about Burne-Jones's artistic priorities - his preference for placing his figures in a clearly-defined foreground plane, his documented unwillingness to make use of distance as a pictorial resource, his strong tendency to base his paintings on existing decorative designs for tiles, stained glass, and so on - all this makes it highly unlikely that for a potentially finished work he would have opted for such a riot of trompe-l'oeil effects as we find in the Birmingham canvas. It would have been contrary to all his deepest artistic convictions; even in a sketch it is unsual and uncharacteristic.

We do not, however, believe that our panels were painted for the Triptych to which the Birmingham painting looks forward. Rather they seem to have been destined for an earlier version that had been superceded in the artist's mind when that painting was commenced. The Story of Troy being in oils, it seems to follow that the Triptych it anticipates would have been in this medium too. But our panels are in watercolour, albeit worked with bodycolour on canvas so that they closely resemble oil. They are also exceptionally highly finished, sometimes being touched with gold, and even incorporate the fictive frames, inspired by 'plaques in the South Kensington Museum', to which Rooke refers. These frames moreover not only bear the names of the personifications in question, carefully inscribed in Roman capitals, but are embellished with rosettes worked up in relief, probably in gesso, and glazed.

Nothing, in other words, could be less like a conventional study, or more what the paintings would have looked like if they were done 'for the Troy picture' in the most literal sense of the phrase. But how do we reconcile this with our assumption that the Triptych anticipated by the Birmingham canvas was to be in oil? The possibility that some of its subjects would be in oil and others in watercolour that looked like oil can surely be dismissed.

A solution to the problem may be offered by the famous Angels of Creation (Fogg Art Museum, Cambridge, Mass.), not as they are now, each panel in a separate frame, but as they appeared in the original, all-embracing frame that is recorded in an old photograph (fig. 6). Another composite work, consisting in this case of six panels, the painting is exactly contemporary with the Troy Triptych, having been conceived in 1870-71 and completed in 1876. The frame, which of course in this case was undoubtedly three-dimensional, was specially designed by Burne-Jones and is closely comparable to that of the Triptych, being classical in style, with a heavy entablature, a 'predella' element, and supporting corbels. Last but not least, the figure subjects, which in scale are almost as large as the main panels of the Triptych as represented in the Birmingham painting, are in watercolour on canvas, the medium of our four allegories.

Here, then, is a hint of how Burne-Jones might have envisaged the Triptych in 1870-71, with the eight designs carried out in watercolour on canvas and set into a three-dimensional frame. Our panels would have formed part of it, complete with their painted inner frames or borders executed partly in low relief.

But even if this was the original plan, it must very soon have changed, for within a few months Rooke was copying the panels into the Birmingham canvas. In his account of this picture he states that, in addition to outlining the three main compositions, he worked extensively on the predella, laying in the two 'Venus' designs and the four allegorical figures. The former were copied from the two highly-finished pencil drawings that Burne-Jones had recently completed (figs 2-3), Rooke being allowed to add colour 'under direction' from his employer. The smaller allegories were reproduced from our versions in 'a green under-painting' that made them look like bronze reliefs, and 'afterwards brought into tone and colour by the master's hand'.

It is true that Rooke was eight-eight when his account was published, and that his memory was sometimes at fault. He describes our panels as being 'in the Ionides Collection at the Victoria and Albert Museum', which was certainly never the case. But there is no reason to doubt his essential accuracy. He dates the copying of the 'Venus' designs to 'the winter of 1871-2', and this was probably when the replication of our panels took place as well.

Indeed, there is virtual confirmation of this in the fact that in February 1872 our panels were framed together and exhibited at the Dudley Gallery (see below), presumably no longer being required for the copying process. Theoretically they could have been separated again, but it seems more likely that by this date Burne-Jones had abandoned any idea of inserting them into the Triptych and was regarding them as an independent work of art.

Meanwhile work on the so-called Story of Troy continued. In 1872 Burne-Jones designed the central subject for the predella, The Feast of Peleus, and the following year, according to Rooke, this too was copied into the large canvas. The copyist on this occasion, however, was not Rooke himself but the young American Frank Lathrop, a nephew of Nathaniel Hawthorne, who was another of Burne-Jones's assistants.

Why had Burne-Jones changed his mind, decided to re-think the project and embark on a sketch for it so ambitious that in time it would be mistaken for the Triptych itself? Such a radical revision may have been prompted by the development of the predella compositions in 1871-2. It is noticeable that in the early sketch (fig. 5), which probably dates from 1870, the year the Triptych was conceived, the three main panels are filled in but the predella is a complete blank. The designs for this came later, and in the Birmingham painting the emphasis is reversed, they being the focus of attention while the three subjects above are represented only by the incomplete outlines laid in by Rooke.

This may not be an altogether satisfactory explanation, but it should never be forgotten that the early 1870s were a period of unparallelled creative activity for Burne-Jones. 'I have sixty pictures, oil and water, in my studio', he told his friend Charles Eliot Norton in 1871, 'and every day I would gladly begin a new one.' His third visit to Italy that autumn inspired an even greater productivity; the work-record for 1872 has an astonishing thirty-four entries, many denoting a whole series of designs and often anticipating important paintings and decorative projects; the full development of which stretched far ahead. The Troy Triptych was only part of this creative abundance, and as the spate of ideas flowed unchecked, its evolution could have taken any number of turns. We shall probably never know the full story.

There are, however, one or two further hints that the scenario sketched here is substantially correct. Rooke records Burne-Jones's wife, Georgiana, observing that she hoped he would not 'break his heart' over the project when she saw it all 'schemed out' in the Birmingham canvas. The phrase 'schemed out' strongly suggests that the work looks forward to something greater, i.e. a three-dimensional realisation of what it represents. Nor does it seem likely that she would have worried about him 'breaking his heart' over the canvas itself. Only the fulfilment of his dreams in all their three-dimensional glory would seem a big enough undertaking to have aroused her misgivings.

It may well be, moreover, that the Triptych in its revised form was not only to be in oil rather than watercolour but larger than originally envisaged. When Rooke copied our paintings into the Birmingham canvas, he kept to the 'same size', but when Lathrop copied The Feast of Peleus in 1873, he did so on a reduced scale from the oil painting on panel, now at Birmingham, that was 'somewhat larger' (fig. 7). Hitherto it has always been assumed that this painting was an independent easel version, but its medium, scale and meticulous finish suggest that it may originally have been intended as a predella panel for the revised and enlarged Triptych. Once again it is revealing to see how Burne-Jones describes the picture in his work-record. He lists it as being 'for Troy', just as he had listed our allegories as being 'for the Troy picture'.

Burne-Jones's work-record often needs decoding. For many years the complex development of the Briar Rose series proved elusive, but by reading between the lines of the record it was eventually explained. Similarly with the Troy Triptych. The problem is that Burne-Jones uses the phrase 'for Troy' in several senses. When he is talking about the two 'Venus' drawings (figs. 2-3), he means that they are preparatory studies for the project. When he refers to our allegories, he seems to imply that they were painted as part of the Triptych as it was originally conceived. And when The Feast of Peleus (fig. 7) is the work in question, he probably has the revised and enlarged version in mind. If this is the case, then the allegories and Peleus have something important in common. Our paintings are the only part of the original concept to be completed, and Peleus is similarly unique in respect of the updated version.

There are admittedly features of Rooke's catalogue note that puzzle. He observes that the bronze putti were added to the canvas 'later' at the suggestion of Charles Fairfax Murray, yet another assistant, 'to diminish the peril... that the whole would be cut up for the sake of the separate subjects', a contingency to which Murray's knowledge of early Italian art made him keenly alert. This suggests that the figures were only devised in order to keep the painting intact, whereas in fact they are already present in the early sketch (fig. 5) and would surely have been realised as free-standing bronzes had the three-dimensional Triptych been completed. Possibly the putti were planned from the start, at first deemed extraneous to the canvas's purpose, but eventually incorporated to prevent it being dismembered. But then they no more seem to be afterthoughts than the curtains, foliage, and other features of the elaborate mise en scène.

Rooke further states that, to the same end, 'festoons and chaplets of
jewels were... to be hung from the capitals at (the) top, across the
picture in the Crivelli manner'. As there is no indication of these
'festoons' either on the canvas or in the sketch, they are even more
problematic. But again it may have been a matter of adaption. Or was the octogenarian Rooke confusing them with the swags of drapery and foliage seen above the Triptych in both preparatory drawing and canvas? His account, as we have seen, is not infallible.

Burne-Jones continued to mull over the Triptych for another couple of
years. In 1874, according to the work-record, he 'designed many
figures for Troy', and he was still 'at Troy' in 1875. But by now it was clear that the Triptych as an entity would never be achieved, and The Story of Troy was taken to a studio, known as the Iron House or the 'Tin-pot', that his old friend G.F. Watts had built in the grounds of what would soon be his new home in Holland Park. There it languished for many years until Burne-Jones reclaimed it, but it was never seen in public, Philip describing it as 'exhibited for the first time' when it went on show in Birmingham in the 1920s. Perhaps the artist's failure to complete it was a tacit acknowledgement that he had created something alien to his deepest artistic convitions. At all events by 1872 a prospect beckoned that was much more in keeping with his true objectives, the painting of independent versions of 'Troy' compositions.

Other versions

The Triptych's three main compositions were never to be recycled. They are known today only from Rooke's outlines on the unfinished Story of Troy and a handful of preparatory drawings. As for the other narrative subject, The Feast of Peleus (fig. 7), we have suggested that this was begun as a panel for the revised Triptych, although as the great enterprise faded it certainly acquired an autonomous status. Eventually completed in 1881, the panel was exhibited at the Grosvenor Gallery the following year and entered the collection of Burne-Jones's greatest patron, William Graham. No sooner was it finished, moreover, than the artist embarked on another version, this time on canvas and colossal in scale. Laid in by assistants, it was subsequently abandoned and remained unfinished at Burne-Jones's death in 1898. Having lurked for decades in the bowels of the Victoria and Albert Museum, it has recently been restored and hung on the Museum's East Stairs (Staircase M).

But it was the allegorical subjects to either side of Peleus that had the most powerful hold on Burne-Jones's imagination. In a sense they had always taken precedence over the narrative scenes, and one of them may even have pre-dated the conception of the Triptych itself. A small watercolour of Fortune in the Carlisle Art Gallery (fig. 8) dates stylistically from about 1870, and could represent an existing design that was adapted for the Triptych rather than one specially devised for it. Certainly the composition is quite different to that of our panel or any of the later versions. Executed in blue monochrome, it is somewhat in the manner of G.F. Watts, who had such an influence on Burne-Jones in the 1860s, and it may be no coincidence that he owned it for many years.

If Burne-Jones was already thinking in such terms before the Triptych was conceived, it is perhaps not surprising that he designed the allegorical parts of the predella before turning his thoughts to Peleus, or that the predella was the area on which he focused in The Story of Troy. Such a consistent emphasis on the allegorical element reflects the influence of his mentor John Ruskin, who had been actively shaping his development since the late 1850s. Ruskin believed that his protégé had the makings of a great exponent of allegory, a painter of what the critic called 'ideal grotesques', images drawn from myth and legend that were replete with moral, spiritual and intellectual meaning.

All six of the predella's allegories were begun, as Burne-Jones put it, 'in large' in that year of unprecedented productivity, 1872. If Peleus was conceivably begun for the revised Triptych, these imposing canvases certainly had nothing to do with the project except their design. Neither Venus Concordia nor Venus Discordia was ever finished, and the latter (National Museum of Wales, Cardiff) remains very much a work of the 1870s. Venus Concordia (Plymouth Art Gallery), however, was taken up again in 1895, three years before Burne-Jones's death, and has all the hallmarks of his most mannered late style.

As for the four single-figure allegories, these were all commenced on canvases nearly 6 feet high. Fame, Oblivion, and Love remained unfinished, and were given to Birmingham in 1922, together with The Story of Troy. Rooke, who was once again responsible for the lay-in, described them in the 1930 catalogue as having been 'only touched on by the originator with suggestions for after-work, which, resulting from the change of scale, they never got.' Fortune was separated from her companions and is probably the painting now in the National Gallery of Victoria, Melbourne (fig. 9). This is only fractionally larger than the three unfinished canvases at Birmingham, and when Burne-Jones completed it in 1885 he described it as 'the first' version of the subject, which has 'stood unfinished for many years'. However, there were two more versions with similar dimensions, the unfinished oil that exists in the National Museum of Wales, Cardiff (fig. 10) and a painting formerly in the collection of the first Lord Leverhulme that now seems to be missing.

Fortune was undoubtedly Burne-Jones's favourite among the Troy compositions. Indeed, according to his son it was the design of which he was fondest in his entire oeuvre. There were at least two more versions in addition to the five already mentioned (ours, the watercolour at Carlisle, and the Melbourne, Cardiff and Leverhulme paintings). One, a medium-sized watercolour on canvas, part of the Cecil French Bequest belonging to Hammersmith and Fulham Public Libraries (on loan to Leighton House), was painted in 1875 for Sir Charles Dilke. The other is the definitive version, in oil and larger than any of the others, in the Musée d'Orsay, Paris (fig. 11). Also begun in 1875, this was not completed and exhibited at the Grosvenor Gallery until 1883, when it was warmly received even by formerly sceptical critics and did as much as anything to establish the artist's reputation before the famous King Cophetua (Tate Britain) was shown the following year. Possessing an authority that none of the other versions quite have, it is still considered one of his greatest achievements.

The subsequent versions of Fortune (or The Wheel of Fortune as the design became known) all differ in one important respect from the image in our painting. Whereas here Fortune is shown blindfolded and has drapery flying about her head, in the later accounts she merely lowers or even closes her eyes, indifferent to the fate of the revolving figures, while her headgear is changed to a neat coif or bonnet. The effect of this change, particularly in the climactic masterpiece, is to make the composition significantly more Michelangelesque. The male nudes bound to the wheel had betrayed from the outset a debt to Michelangelo's Captives in the Accademia, Florence, which Burne-Jones recorded in a sketchbook during his visit to Italy in 1871, and perhaps even more to the so-called Dying Slave in the Louvre, of which he owned a small plaster copy; but in the versions subsequent to ours Fortune herself grows closer to the Sibyls on the Sistine Chapel ceiling, both in terms of her awesome implacability and the details of her dress. This was another source that he studied intently in 1871, lying on the Chapel's floor and looking up through opera glasses in order to 'read the ceiling from beginning to end'.

Burne-Jones's fondness for the composition must have stemmed partly from a sense that it was a worthy tribute to an artist he passionately admired. It was also a work that he felt fulfilled the Ruskinian concept of meaningful allegory by constituting a profound reflection on the human condition. 'My Fortune's Wheel is a true image', he wrote in later life, 'and we take our turn at it, and are broken upon it.' Nor did it fail to please Ruskin himself, despite the sharp difference of opinion between himself and Burne-Jones on the subject of Michelangelo. When Ruskin lectured on his old protégé at Oxford in 1883, he singled out the design as an example of Burne-Jones's 'deeply interesting function' as an artist, namely to convey 'the spiritual truth of myths'. Perhaps not surprisingly, the definitive version was acquired by Arthur Balfour, who was brought to Burne-Jones's studio in 1875, the year the picture was started, and, he recalled, 'at once fell prey to both the man and his art'. The rising Tory politician and future Prime Minister was a keen amateur philosopher, publishing his Defence of Philosophic Doubt four years later.

A neo-Renaissance concept

So much for the identity of our panels and their place in the sequence of versions. What of their wider context?

The great epic of the Fall of Troy, though classical in theme, had fascinated artists and writers in the Middle Ages; and the medievalists among the Pre-Raphaelites regarded it as almost mandatory to follow suit. So far as Burne-Jones is concerned, there could hardly be a better illustration of Henry James's perceptive comment that his art is a 'reminiscence of Oxford, a luxury of culture'. He would have read Homer and Virgil as a boy at King Edward's Grammar School in Birmingham, where the highest educational standards were set by the headmaster, James Prince Lee, a brilliant classical scholar who had taught under Arnold at Rugby. At Oxford, too, he and his friend William Morris not only studied these authors as part of the orthodox curriculum but encountered the 'Trojan' story of Troilus and Cressida in its Chaucerian and Shakespearian versions. Within a few years Morris had embarked on a series of never-to-be-completed poems entitled 'Scenes from the Fall of Troy', while Burne-Jones was planning to paint 'subjects from Troy... and a great ship carrying Greek heroes' on the staircase at Morris's newly-built Red House.

It was probably the friends' enthusiasm for the epic that led D.G. Rossetti to handle it in the 1860s: in the pen-and-ink drawing Cassandra (1861; British Museum), the painting Helen of Troy (1863; Hamburg), and the poem 'Troy Town', composed at Penkill in 1869. Rossetti in turn influenced Frederick Sandys, who painted his own accounts of these heroines during that decade. Nor had Morris by any means finished with the stories, publishing translations of the Aeneid in 1875 and of the Odyssey in 1887. Even in the 1890s he and Burne-Jones were still as hooked on the subject as ever, 'talking hard all morning', as Burne-Jones recalled of one of their Sunday colloquies, about 'why the medieval world was always on the side of the Trojans'.

But the Triptych was not only a major contribution to this genre; it also expressed more personal preoccupations. One of these is betrayed by the artist's choice of form, his decision to paint what Stephen Wildman has described as 'a modern secular version of a Renaissance polyptych.' From the mid-1860s Burne-Jones had increasingly felt the influence of the Italian masters, not the 'primitives' and the great Venetians that he had studied earlier under Ruskin's tutelage, but the artists of the early and high Renaissance, mainly active in Florence. The trend would reach a climax in the early 1870s. It was then that he paid his last two visits to Italy (1871 and 1873), focussing his attention on Florence and Rome, and brought to fruition a style that speaks eloquently of his reverence for Botticelli, Mantegna, Michelangelo and their peers.

The Troy Triptych comes relatively early in this process, its conception at least pre-dating the Italian tours, but it is nonetheless one of its supreme manifestations. Burne-Jones has opted for a format that was often a showcase for Renaissance pictorial values, using it, like the old masters, to create an ensemble of designs closely integrated on both formal and thematic levels. The classical design of the frame is a further sign of Renaissance influence, as would have been those 'chaplets of jewels in the Crivelli manner' if Burne-Jones had ever got around to painting or hanging them across the main panels. Significant, too, is the formal language of the compositions, heavily dependent on the nude and openly acknowledging a debt to Pollaiuolo, Michelangelo and others. Our four allegories are good examples, and we have already discussed the increasingly Michelangelesque character of Fortune as the design progressed through successive versions.

Even the preliminary studies are indicative of Burne-Jones's thinking, being in the 'Florentine' idiom he adopted in the 1870s (fig. 12). Discarding the red chalk or soft pencil that he had favoured hitherto, he drew almost exclusively at this period with a hard pencil that approximated to silverpoint.

Only the iconographical programme differs. Whereas the triptych had traditionally been a means of representing sacred legends and figures of patron saints as aids to public or private devotion, Burne-Jones uses it as a vehicle for one of those narrative cycles of which he was so fond, the pictorial story-telling he had already attempted in the 1860s in his illustrations to Morris's Earthly Paradise and would return to later in the Perseus and Briar Rose series. Yet the way he narrates the story of Troy still owes much to sacred prototypes. In earlier times the predella had often been reserved for scenes from the lives of the saints portrayed full-length in the upper panels, thus giving the faithful a more nuanced image of these spiritual heroes. Burne-Jones does something very similar, devoting his predella to allegories which spell out the moral and philosophical implications of the events represented above.

A love-allegory

The priority that Burne-Jones consistently gave to the allegorical element in his retelling of the Troy story is not only attributable to Ruskin's plans for his development. It was also the product of his obsession with Italian art. 'The Triumph of Love', the subject of our panels and the collective title they would eventually acquire, had been the theme of a well-known poem by Petrarch. It was written in the form of an allegorical vision or dream, a convention Burne-Jones had long since encountered in the work of Chaucer and which always moved him deeply. But on this occasion there was the additional attraction that the poem had been a potent source of inspiration for Renaissance artists and engravers.

Not that Burne-Jones was prepared to accept these images without question. In 1871, the very year that he painted our panels, he evolved a design, known as The Car of Love or Love's Wayfaring, that echoed but in his view improved on its antecedents. As he said when he went back to the design in the 1890s, it was 'not a new subject. It has been done before, but not as I like it.' Something of the same high-handedness is found in our Triumph of Love. Petrarch's poem was one of a series of Trionfi in which Love is among many abstract values (Chastity, Death, Time, etc.) that are successively annihilated, only God finally remaining. In making Love his ultimate victor, Burne-Jones, having secularised the polyptych, does the same for one of his principal literary sources.

However, if this seems to encapsulate what Henry James called 'Oxford', we should be wary of dismissing it as some donnish intellectual indulgence. Burne-Jones often used literary conceits to cloak autobiographical references, and The Triumph of Love is very much a case in point. The picture was painted at the height of his affair with the Greek beauty Maria Zambaco (fig. 13), the undoubted climax of his emotional life. Wayward and headstrong, artistically talented and ravishingly attractive, Maria was a cousin of his friends and patrons the Ionides. Born Maria Cassavetti in 1843, she had married Demetrius Zambaco, a doctor to the Greek community in Paris, in 1861, and borne him two children, but in 1866 she had left him and returned to London. There, at the age of twenty-three, she was introduced to Burne-Jones, her senior by ten years, and he had fallen head over heels in love. It was hardly surprising. Never before in his life of studious application, driven first by Tractarian and then by Ruskinian idealism, had he encountered such a passionate and elemental creature, while the contrast with his plain, high-minded wife Georgiana (who handled the crisis with characteristic dignity and restraint), could hardly have been more startling. The emotional strain was a severe test of his always delicate health, and the affair took a tragi-comic turn in January 1869, when Maria tried to commit suicide in the Regent's Canal. Restrained by her lover, she failed, and the relationship continued well into the 1870s. Some suspected that it went on even longer.

Maria posed frequently for Burne-Jones during these years, and her features, unlike those of so many of his models, are easily recognisable in his current paintings and drawings. Many of the pictures in which she appears, such as Phyllis and Demophoön (1870; Birmingham), Love among the Ruins (1870-73; private collection) and The Beguiling of Merlin (1873-77; Port Sunlight), have subjects which clearly reflect or comment on the lovers' predicament, and must have been chosen at least partly as expressions of the artist's feelings. In the Troy Triptych the references are more oblique, but nonetheless present on several levels. His revived interest in the Greek epic was almost certainly a tribute to Maria's nationality; indeed, we know that he associated her with the story of Troy, drawing her as Cassandra, King Priam's daughter who foretold her nation's fate (Ionides Collection, Victoria and Albert Museum).

Our panels, however, point to a more profound correspondence between his affair with Maria and the way he interprets the legend as a demonstration of the redemptive power of love. In other paintings she inspired the mood is sombre. In Phyllis and Demophoön she leans from an almond tree to forgive her faithless lover. In Love among the Ruins the couple's world has crumbled and the future they face is bleak, while in The Beguiling of Merlin Maria herself has become the destructive force. Only in The Triumph of Love does Burne-Jones strike a more optimistic note, suggesting that their love will endure, whatever trials and tribulations they have to suffer.

Exhibition at the Dudley Gallery

In discussing our panels' relationship to the mysterious painting at Birmingham (fig. 4), we have already noted that they were included in the Dudley Gallery's eighth 'General Exhibition of Water Colour Drawings', which opened at the beginning of February 1872. The four small canvases had been stuck side by side on muslin-lined paper stretched over a wooden board, and then placed together in a single frame, almost certainly the one that is still on the picture today. The title given in the catalogue was The Triumph of Fortune, Fame, Oblivion, and Love, a clumsier version of the one later adopted that made sense on one level but nonsense on another.

The Dudley held its exhibition in the Egyptian Hall, Piccadilly, a building situated opposite the Burlington Arcade that is now occupied by a block of offices and shops known as Egyptian House. Constructed in 1812, the Hall had seen some of Benjamin Robert Haydon's greatest triumphs, as well as his ultimate undoing; but it was better known for displays of monstrous freaks than for its exhibitions of fine art. The Dudley belied this slightly sinister ambience. Since its opening show in 1865, it had promoted a group of young artists whose work heralded the advent of the Aesthetic movement, including several followers of Burne-Jones. Like their mentor, they tended to favour watercolour, while the poetic approach to landscape which was a hallmark of their style owed much to the backgrounds he had given to such early works as The Merciful Knight and Green Summer.

Burne-Jones himself had not showed there before, preferring to support the Old Water-Colour Society, to which he had been elected in 1864. But his work had never found favour with the more conservative members, and in 1870, when objections were raised to the nude male figure in Phyllis and Demophoön and his colleagues requested its removal, he resigned. There followed what he called 'the seven blissfullest years of work I ever had; no fuss, no publicity, no teasing about exhibitions, no getting pictures done against time'. Showing hardly any works in public, he lived in retirement in Fulham, relying for financial support on such devoted patrons as F.R. Leyland and William Graham. The period was to end with a bang in 1877, when he astonished London with the eight large works he sent to the inaugural exhibition at the Grosvenor Gallery.

The exceptions to this rule of non-exhibition were his three appearances at the Dudley Gallery in 1872-3. Having made his debut with The Triumph of Love in the spring of 1872, he contributed two drawings, one of them the highly-finished study for Venus Concordia (fig. 2), to the 'black and white' exhibition held later that year. Finally, he sent two much larger works, Love among the Ruins and The Hesperides (Hamburg), to the Gallery's ninth show of watercolours in 1873.

The advent of Burne-Jones, mentor to so many of the other contributors, in 1872 seems to have caused some excitement. Though modest in scale, the Triumph was 'placed in the best position', and it was widely and copiously reviewed in the press. Tom Taylor, writing in the Times, complained that it was impossible to do justice in a single article to a show in which 384 artists were represented, and then went on to devote four column inches to The Triumph of Love.

But this did not mean that the critics liked the picture. On the contrary, they had always had problems with Burne-Jones, often attacking his contributions to the OWCS, and a different venue did not make them change their minds. True, they could see that he had great 'facility of design' and a 'fine faculty as a colourist'. William Bell Scott, writing in the Academy, went so far as to refer to his 'transcendent powers' and his ability to 'answer the desires of the heart' with his 'lovely inventions'. But to a man they deplored what they saw as faults of drawing, attributing them to 'defective training'. Nor could they believe that he was more than a plagiarist, apeing the 'mannerisms, deformities and disproportions' of the old masters 'as well as their beauties'.

In fact it is hard to resist a suspicion that the critics discussed Burne-Jones at such length partly in order to parade their art-historical knowledge. The variety of influences they identified is almost ludicrous, or would be if many of them were not parallelled by copies in the contemporary sketchbooks of this most erudite and source-conscious of artists. For Bell Scott (who, incidentally, was a fellow exhibitor) Burne-Jones's 'drawing' was 'that of Signorelli'. Another critic with close Pre-Raphaelite connections, F.G. Stephens, observed in his usual column in the Athenaeum that 'these designs reproduce the feeling as well as the peculiar manner of the Venetian masters who dealt with allegories of this sort.' Meanwhile, the Art Journal evoked the names of Michelangelo and Giulio Romano, and the Illustrated London News saw the artist 'reproducing and exaggerating the greatest peculiarities of form of the Peruginesque schools'. As for the Times, it detected echoes of early German and Italian engravings, and while it admired the artist for capturing 'something of the grandeur of Mantegna and the vehement action of Signorelli', it deplored his fondness for 'the meagre wrists and ankles and pigmy hands and feet of Pintoricchio and other (early) mannerists.'