In December 1504, after a hiatus of about four months, Leonardo begins work on the critical phase of the Anghiari fresco – the actual painting of it, from the cartoon, on to the wall of the Council Hall. On 31 December the works department of the Palazzo Vecchio pays suppliers for nails and cloth delivered that month to ‘cover the window where Leonardo da Vinci is working’, and for wax, sponges and turpentine ‘to wax the windows’. On 28 February 1505 there are further payments for iron, trivets and wheels ‘to make Leonardo’s carriage [carro], that is the platform’ – another custom-made mobile scaffolding, like the one he used in Santa Maria Novella, which has now been constructed in the Council Hall. The cost of this contraption is nearly 100 lire, as is noted by the parsimonious accountants, who on 14 March order that ‘the platform for Leonardo da Vinci’s picture, which is now being done in the Hall of the Great Council, and all the planks and panels which have been used to make it, should be returned and restored to the Department of Works when the said picture is completed.’ These entries suggest that Leonardo was equipped to begin painting the fresco around the end of February 1505 – precisely the date by which he had previously promised to have the whole thing completed.135
On 14 April, a new pupil: ‘1505 – Tuesday evening, 14 April, Lorenzo came to live with me; he says he is 17 years old.’ He is immediately set to work, for on 30 April the accounts include payment to ‘Lorenzo di Marco, journeyman, for 3½ sessions in the Council Hall on the picture which Leonardo da Vinci is doing’.136 He is paid at a basic rate of 9 soldi a day. Also disbursed during this month of intense activity are ‘5 gold florins to Ferrando Spagnolo, painter, and to Tommaso di Giovanni, who is grinding the colours.’ This ‘Ferrando the Spaniard’ is a shadowy painter to whom various Leonardesco works are attributed, among them the sultry Madonna and Child with a Lamb (Brera Gallery, Milan); a free version of the Madonna of the Yarnwinder (private collection); and even – according to some – the Uffizi Leda. He is probably Fernando (or Hernando) Yañez de la Almedina, who is documented in Spain after 1506; some of his works, such as the Epiphany and Pietà in Cuenca cathedral, show strong influence of Leonardo and Raphael.137 The other man mentioned, Tommaso di Giovanni the colour-grinder, is of course Zoroastro. In a later account-entry he is described as Leonardo’s ‘garzone’, a lowly status which reflects the incomprehension of the accountants.
To Lorenzo, Fernando and Tommaso may be added a certain Rafaello di Biagio, painter, whose work seems to have been less skilled since his rate of pay is only 2 soldi per diem, and another mysterious character who is not mentioned in the accounts but who is named by the Anonimo Gaddiano as one of Leonardo’s assistants on the fresco – ‘Il Riccio Fiorentino who lives at the Porta della +’ (i.e. the Porta alla Croce). He is presumably the same as the ‘Riccio, goldsmith’ who was on the David committee convened in January 1504. Another name missing from the accounts is Salai, whom one would expect to be involved. This is Leonardo’s team.
In a dramatic note headed ‘On Friday in June, at the 13th hour’, Leonardo writes:
On 6 June 1505, on Friday, at the stroke of the 13th hour [about 9.30 a.m.] I began to paint at the Palazzo. At the moment of putting down the paintbrush the weather changed for the worse, and the bell in the law-courts began to toll. The cartoon came loose. The water spilled as the jug which contained it broke. And suddenly the weather worsened, and the rain poured down till nightfall. And it was as dark as night.138
This extraordinary, almost apocalyptic memorandum has the usual ambiguities. When he says he ‘began to paint’ – or more precisely to apply colours (colorire) – does he simply mean he began the day’s painting at that time, or that this was the day when he first started to colour in the figures? And what quite does he mean by ‘il cartone straccò’? The verb straccare means to become tired or stale or worn-out; hence in this context to sag or come loose. How is this connected to the abrupt change of the weather? Is it a case of ominous synchronicity – the storm, the bells, the studio mishap – or has a sudden gust of storm-wind blown in through the veiled and waxed windows of the Council Hall, tearing the cartoon off its frame and knocking over the jug of water?
Time rolls on; the work moves slowly; there are mutterings in the Signoria as the money continues to be doled out. Vasari has a story which sums up the declining momentum of the project:
It is said that when he went to the bank for the salary which he was accustomed to receive from Piero Soderini every month, the cashier wanted to give it to him in piles of quattrini [small coins]. He did not want to take them, saying, ‘I am no penny painter!’ There were complaints about this behaviour, and Piero Soderini was turning against him. So Leonardo got many friends of his to gather up a whole pile of quattrini, and he took them to him to return the money: but Piero did not want to accept them.
This explosion of temperament is like the scandalo in Milan, when he had stormed out of the Duchess’s chambers – another unexpected storm from one who usually kept his emotions well-hidden.
This anecdote remains as the documents dwindle. The last recorded payments are dated 31 October 1505, though the work must have continued, perhaps fitfully, through till the following May, when Leonardo was very grudgingly given permission by the Signoria to leave Florence.
The Anghiari fresco is perhaps the most intensely documented of all Leonardo’s paintings – more so even than the Virgin of the Rocks, since the documentation concerns the actual execution of the painting, rather than just the contractual ramifications of it. We know the amount of paper he used for the cartoon, the price he paid for the paints, the quantity of wood used for the scaffolding, the names of his principal assistants and the rates they were paid, and possibly even the precise day and hour he began to ‘colour’ the painting on to the wall of the Council Hall.
The only thing we are missing is the work itself. The fresco was never completed, but a large central portion of it was. This has long since vanished from view: in fact it is not even certain which wall of the hall it was painted on – the consensus used to be that it was on the east wall, but now the west wall is favoured. Either way, if anything of it remains, it lies somewhere beneath the enormous fresco-cycle painted in the early 1560s by none other than Giorgio Vasari. It is hard to imagine that Vasari would have painted over an original Leonardo in good condition, so there are two inferences available – the pessimistic and the optimistic: that in Vasari’s view nothing remained which was worth saving, or that he took steps to protect what remained before covering it.
Our knowledge of what the fresco – or the fragment of it which was completed – actually looked like rests on various early copies.139 There is an anonymous copy in oils, painted on a wooden panel; it is called the Tavola Doria, as it was for a long time in the collection of Prince Doria d’Angri in Naples. It shows the dramatic collision of men and horses prefigured in the preparatory drawings; it also has some lacunae – blank gaps – which enhance the idea that it is a direct copy from the fresco. It is not in itself a very good painting, but it may accurately represent the original in an unfinished or deteriorated state. Another important copy is an engraving by Lorenzo Zacchia, dated 1558. This has more detail than the Tavola Doria, but it is not certain where this detail comes from. It could conceivably be based on the original cartoon, but Zacchia himself describes the work as ‘taken from a panel painted by Leonardo’ (ex tabella propria Leonardo Vincii manu picta). It may in fact be a version of the Tavola Doria, which Zacchia believed (or claimed to believe) was the work of Leonardo himself. If so, the additional details and filled-in lacunae would be his own interpolations rather than an alternative reporting of the original fresco.
Most of the details of the Zacchia engraving are also found in the superb watercolour version in the Louvre, attributed to Rubens (page 374). This is actually done on top of an earlier drawing. Beneath the Rubens additions of c. 1603 – the watercolour over-painting, the lead-white heightening, the glued-on extension on the right-hand side of the sheet – there is an Italian drawing datable to the mid sixteenth century, roughly contemporary with the Zacchia engraving. Though Rubens (born in 1577) could not have seen the original fresco, he has brilliantly caught the angry turmoil of the skirmish: seven figures – four on horseback, three on the ground – locked in combat, a pyramidal composition reaching to an apex of two clashing swords. The extraordinary, writhing horseman on the left (who represents Francesco Piccinino, son of the Milanese condottiere) has the attributes of Mars on his armour, and becomes in Rubens’s imaginative rendering a kind of archetype: a universal soldier.
There is plenty of anecdotal evidence to suggest that, as with the Last Supper, there were technical problems in painting the Anghiari fresco, which revealed themselves fairly quickly. Antonio Billi, writing sometime around 1520, speaks of it as abandoned:
He did… a cartoon of the war of the Florentines, when they defeated Niccolò Piccinino, the captain of the Duke of Milan. He began to work from this in the Sala del Consiglio, using a medium which did not stick [materia che non serrava], and so it remained unfinished. It was said that the cause of this was that he was deceived about the linseed oil he used, which was adulterated.
If Billi is right, we might lay the blame for this great artistic loss on a grocer named Francesco Nuti, who was paid on 31 August 1505 for ‘8 pounds of linseed oil supplied to Leonardo da Vinci for the picture’.140
The fresco was still visible on the wall in the 1540s. The Anonimo Gaddiano says of it, ‘anchora hoggi si vede, et con vernice’, which seems to mean that the work could still be seen, preserved under an application of varnish; and in 1549 Antonio Francesco Doni advised a friend, ‘Go up the stairs of the Sala Grande, and take a close look at a group of horses and men, a battle-study by Leonardo da Vinci, and you will see something miraculous.’141
A fresco visible and ‘miraculous’ in 1549, the optimists say, must surely have still been visible when Vasari began his redecoration of the hall some twelve years later – and may be there still. But to find a lost wall-painting you must first find the wall. Early accounts describe the fresco as being on one side of the tribune – the dais where the Signori and Gonfalonier sat when the council was in session – but the evidence is maddeningly ambiguous as to where the tribune was.142
In 1974 an American conservation scientist, H. Travers Newton, used Thermavision (‘an infra-red Vidicon system supercooled with liquid nitrogen’) to probe the walls.143 This machine produces a ‘thermal map’ of materials present beneath the surface (different materials absorb and emit heat at different rates). The first findings were exciting. The east wall revealed only normal architectural elements, but the west wall displayed what Newton described as an ‘anomalous layer’ beneath Vasari’s frescos. This was confirmed by ultrasound, which uses ultrasonic soundings to register variations of density; in 1974 this could ‘read’ the stratifications of a wall up to a depth of about 4 inches. From these twin probes, the ‘anomalous layer’ was defined as an area about 75 feet wide by 15 feet high. Next, core samples were removed from both walls. The east wall exhibited the normal sequence of layers one would expect from the Vasari fresco: a top layer of plaster, an underlayer of rendering, and then the supporting wall; some traces of colour were found, suggesting an underdrawing in some areas. The same was found on the west wall above and below the ‘anomalous layer’, but in that area itself things were different. All the core samples showed a layer of red pigment beneath Vasari’s intonaco, and some showed other pigments laid over this red ground. These included two suggestive of Leonardo’s practice – a green copper carbonate similar to that used in the Last Supper, for which Leonardo gives a recipe in the Trattato; and blue smalt, as found in the LouvreVirgin of the Rocks. Azurite was also found, which is not suitable for use in true fresco, so it suggests the anomalous area is not a conventional fresco.
This raised exciting possibilities, though Newton’s interpretation of the data is not universally accepted. There is currently a stasis. Any further investigation would be invasive, and the idea of damaging Vasari’s fresco and finding nothing but a giant blur beneath it is not attractive to the decision-makers: as the Florentine councillor Rosa di Giorgi put it in 2000, ‘Vasari may not be Leonardo, but he is still Vasari.’144 There is talk of adapting the ‘geo-radar’ system developed by NASA to map sub-surface features on the earth’s surface; a version of this is already used by archaeologists to map ruins before uncovering them, but it remains to be proved whether it can detect pigments on a flat surface.
Meanwhile Maurizio Seracini, whose investigations of the Adoration of the Magi caused such controversy, points back to the east wall,145 and to a tiny green flag which forms an apparently insignificant detail in the vast panorama of the Vasari fresco. On it, visible from the ground through binoculars, is a small inscription – two words, perhaps an inch high, written in white paint which Seracini’s chemical analysis shows to be contemporary with the rest of the fresco: words put there by Vasari. They read, ‘Cerca Trova’ – ‘Seek and you shall find.’