Chapter 5
When making a judgement about the painting’s present appearance, the first factor to take into account is the panel’s condition. Robert Simon in the first chapter bears vivid witness to the damage, overpainting, conservation, and the revelations of passages of supreme quality. In this chapter I am not intending to reprise comprehensively what the conservation has revealed, but to pick out particular aspects that feed into our assessment of the visual qualities of the painting.
Like all objects of its age, it has suffered from the effects of time, though, mercifully, it avoided the fate of being transferred from panel to canvas as happened to many Renaissance paintings, not least those in the French royal collection. It has also avoided significant trimming, since the original lip of the gesso priming is visible around the top, bottom, and right edges, with a thin margin of largely unprimed wood beyond the paint layer. This indicates that the panel was prepared with an integral frame before it was primed and painted. The state of the left margin is more equivocal. The unprimed wood border has been trimmed in a slightly slanting manner, wider by about half a centimetre at the bottom, but this does not appear to have intruded on the painted surface, if Christ’s head was originally in the precise centre of the image.
The wood of the original panel is walnut, which could be taken as signalling its non-Florentine origin (Fig. 1.8). Poplar is the wood that was most commonly used by Florentine painters, including Leonardo. However, he listed walnut as one of his preferred woods for panels, and we cannot take poplar as his invariable choice in the city.1 The unfinished St Jerome, which I am still inclined to think dates from the same time as the Adoration of the Magi, is on walnut, as is the Buccleuch version of the Madonna of the Yarnwinder, which is securely documented as painted in Florence.2 The St John the Baptist is also on walnut. He seems to have liked it for smaller devotional paintings.
The image of the painting in its cleaned state—meticulously reassembled from two large pieces and five fragments after the removal of the backing board—is our main point of reference in assessing the qualities of the painter’s skills—in both observation and execution (Plate 1).
We have already noted one conspicuous pentimento, the ghost thumb of his right hand, which was once rather straighter (Plate 4). There are also some fugitive indications of changes in Christ’s costume, as outlined earlier, not least in the knot-work pattern on the bands and outside the margins of the central plaque on his chest, where it looks as if the knots originally occupied a wider area. Traces of earlier thoughts about the geometrical interlace within the bands suggest that the patterns were first freely sketched with yellow-white paint in a less tightly intertwined and curvaceous manner (Fig. 5.1), perhaps as reflected in the Yarborough version. The earlier scheme is notably close to the interlace on a drawing at Windsor (Fig. 5.2).
Fig. 5.1Macrophotograph and reconstruction of the underdrawing in the band to the right (Photo: Shan Kuang).
Fig. 5.2Leonardo da Vinci, Curved Interlace Pattern, Windsor, Royal Library, 12351, detail.
Yokes and Knots
The technique in which the knot-work is finally painted seems somewhat variable. The most brilliantly painted passages in the bands are visible nearest Christ’s right hand (Plate 11). Here, the threads are rendered in a fully plastic manner, using a white underpainting with modelling in dark reddish paint. The over-and-under motions of the threads are beautifully characterized. This quality is generally sustained in the crossed bands, while the knot-work of the horizontal band and its central plaque are more routine and less expressive of its structure—even allowing for more damage in this area. This might suggest some studio intervention, an impression which is sustained by the gathered rivulets of drapery below the band, which are less finely characterized than in the Mona Lisa. Any such intervention seems limited to the draperies.
Infrared reflectography confirms the surface observations of the earlier, freer sketching of the knot-work and reveals that the upper margins of the diagonal bands were once located higher. The crossing point of the upper margins of the bands appears at one point to have been somewhat right of centre, as in the Windsor drawing. Some delicate pentimenti are evident in the fingertips of Christ’s left hand. The double image of the heel of the hand visible through the sphere will be assessed below. The fold pattern above the lower section of the band to the left confirms the traces of the ‘omega’ fold that are more readily visible in the drawing and in other versions. The rhythmic structure of the shawl that passes diagonally over his left shoulder and behind the sphere is very evident. And, not least, the area of Christ’s forehead above his left eye, marked by a mesh of small ridges resulting from the pressing of parts of the artist’s hand into the moist underlayers of paint (Fig. 5.3), a technique observed extensively in Leonardo’s paintings and in the recently discovered profile portrait of a woman on vellum.3
Fig. 5.3Infrared reflectogam of the Salvator Mundi, detail of handprints above Christ’s left eye.
There are slight traces of the black dots of charcoal (spolveri), created during the transfer of the design from a pricked cartoon, along the top edge of Christ’s upper lip. A cartoon was used either for his head or for the whole of the composition. Elsewhere the elusive dots of charcoal powder have disappeared during the execution or simply do not show up in the examination. There are numerous, relatively fugitive signs of loose sketching with a darkish pigment, not least within the sphere. The general effect of the infrared reflectograms is of a fresh and spontaneous execution within the parameters of a design that was largely resolved in advance.
The quality of the final execution in the best-preserved sections is compelling. Particularly brilliant in execution are the curls of Christ’s hair (Plate 13), which hang in double helix tresses to the left of his head before separating into vortex strands of extraordinary vitality and mobility. The vivacious delicacy of the brushwork stands comparison with the watery turbulence of the vortices in the hair of Ginerva de’Benci in Washington (Fig. 5.4). Leonardo himself drew a famous analogy between the curling of hair and turbulence in water.4
Fig. 5.4Leonardo da Vinci, Portrait of Ginevra de’ Benci, detail of hair, National Gallery, Washington.
Observe the motion of the surface of water, which resembles the behaviour of hair, which has two motions, of which one depends on the weight of the strands, the other on its line of revolving. This water makes revolving eddies, one part of which depends on the principal current, and the other depends on the incident and reflected motions.
In the later painting, the structure becomes notably more deliberated and the motion less spontaneous, as characteristically happens in Leonardo’s later studies of natural forms. It is the structural and dynamic logic of Leonardo’s vortices that his imitators neither understood nor captured. The same applies to the miraculously fine swirls of glistening highlights on the more advanced curves of Christ’s hair.
The flesh tones are applied in thin layers with pigments finely dispersed in the oil binder, in a way that was typical of Leonardo and became a dominant feature of his technique after 1500, probably reflecting his contacts with Giovanni Bellini and other Venetian masters of the oil medium.5 Contours of the facial features, with the exception of the thin gap between the lips, and the edges of the eyelids, are consistently elusive, to a greater degree than the heads he painted before 1500. The obvious comparison is with the Mona Lisa.6 Christ’s eyes were painted with a melting and softness that transmuted their direct glance into something ineffable, to judge from his left eye, which has suffered less from abrasion and in which the original iris and pupil are apparent (Plate 14). The pupils are unusually small and surrounded by radiant and unusually pale irises, painted very thinly on the light priming. It is as if the ‘windows of the soul’, as Leonardo called our eyes, are emitting light as well admitting it—as if they are illustrative of the extramission theory of sight that Leonardo once countenanced and rejected. This radiance is in marked contrast to the starkly assertive stare that characterizes the darker eyes of the copies, with the exception of the etching by Hollar, who has captured the ineffable quality of the glance very effectively (see Figs. 1.12 and 1.3, respectively).
Does Christ have a beard, as is standard in images of the Salvator and in almost all the copies, in which his beard is often little more than a dark fringe? The Naples version (Fig. 5.5) is an exception in that it lacks a beard. A good number of commentators have assumed not. High-resolution imaging seems to reveal elusive traces of a soft, dark haze of brown pigment applied with a very fine brush. Such traces are very vulnerable to even slight abrasion. What seems to remain of the beard appears as if breathed on to the surface rather than painted. We know from the Mona Lisa that the very delicate brushwork with which Leonardo rendered such details as the fine and sparse hair of her eyebrows is very vulnerable to enthusiastic cleaning. There is a decent possibility that there was once a very fine and sparse beard in the picture, before it was abraded.
Fig. 5.5Follower of Leonardo, Salvator Mundi, Naples, San Domenico Maggiore.
Christ’s blessing hand is at once ideal in a classical sense and full of understated anatomical conviction. The joints of the fingers swell and diminish almost imperceptibly (Plate 11). The tips of the fingers and thumb of his other hand catch the light in a delicious manner, each illuminated to just the right degree in relation to the light that glances across them from the upper left (Plate 12). The characterization of light and shade in the fold under his little finger is notably subtle. This is the part of Leonardo’s own right hand that he pressed into his drying paint.
The cuff and outer sleeve of Christ’s raised arm exhibit the kind of structural logic that indicates close studies from real drapery, as the existing drawings at Windsor attest. The little rivulets of folds that cascade from the band across the top of his robe reflect the delicate energies of those in the corresponding red chalk study, though there is a possibility that the actual execution of the painted folds was undertaken by an assistant.
The ‘omega’ fold produced by the compressive forces of the right-to-left band is typical of Leonardo’s attention to the ‘physics’ of drapery. It is less visible than it once was. The iconographical function sometimes assigned to the shape of the fold, as a reference to the wound caused by the spear of Longinus after Christ’s death on the cross, is neither necessary nor convincing.
The other innovation in Christ’s costume was to clothe him entirely in blue, rather than the standard blue and red or just red. It is likely that Leonardo preferred the unified and radiant effect of the expensive ultramarine rather than breaking the tonal and colouristic unity with contrasting blue and red, or using the less luxurious red on its own.
The crossed yokes are notable for their intricate interlace designs of angular knots. Such patterns, generally in a more curvilinear form, had become a fashionable motive for Renaissance costumes, especially in the North Italian courts. An intricate interlace features in Leonardo’s portrait of Cecilia Gallerani in c. 1490. The decorative design of interlaced vinci came to be known as the fantasia dei vinci. The pattern was appropriated as a personal motif by Isabella d’Este, Marchioness of Mantua. In 1493 her sister Beatrice, who had married Ludovico Sforza, deemed it prudent to write from Milan seeking permission to use it on her own behalf.7
The rhythmic geometry of knots fascinated Leonardo as a kind of visual music, and he experimented insistently with alternative patterns. He could also exploit a pun on his own name, in which Vinci, his home town, puns with vinco (plural vinci), the flexible willow osier used in weaving baskets. The pun seems to have been central for the six printed knot designs for the ‘AC[H]ADEMIA LIONARDI VINCI’, datable to the 1490s (Fig. 5.6), which seem to signal his aspiration to promote an ‘academy’ in Milan.8 His name also gave humanist writers and poets the opportunity to play upon the verb vincere (to conquer).
Fig. 5.6After Leonardo da Vinci, Knot Design ‘Academia Leonardi Vin[ci]’, woodcut, National Gallery, Washington.
There is an ‘over-and-under’ rule that applies to the knot designs. That is to say if one thread passes over another, it should pass under the next thread it encounters, and so on. It is easy to lose track of this requirement and to make a mistake. Below the central plaque, it appears that the rule has not been obeyed, but the damage in this area makes it difficult to tease out the precise configuration of the threads. If one or more errors have been perpetrated, they might provisionally be attributed to an assistant who has not paid due attention to the rule. As we have noted, the skill with which the spatial and plastic properties of the interlace have been rendered is rather variable, suggesting studio intervention.
The final design of the interlace on the crossed band of the Salvator Mundi differs markedly in character from the curvaceous knot patterns that were paraded on the costumes of the Milanese ladies. The knots surrounding the central plaque are of the kind Leonardo favoured in Milan, while the pattern running along the horizontal border of Christ’s blue garment is a rectilinear variant of that in the Cecilia Gallerani. The basic geometry of the crossed bands is distinctively different, although signs of underpainting suggest that his original design might have been based on overlaid curves (Fig. 5.1). The heart of the finished design at the intersection of the two bands is generated from two overlaid squares, one of which is rotated by 45 degrees to create an eight-pointed star (Fig. 5.7). This generating figure is centred on the crossing of the two bands of the yoke. The resulting interlace as it passes along the bands uses angles of 90 degrees and 135 degrees (45o + 90o)—a simple means to achieve a complex effect. Hollar astutely picked up on the nature of the construction, if not precisely on how Leonardo had actually painted it.9
Fig. 5.7Pattern of two squares offset by 45 degrees.
A full study of Leonardo’s interlace designs and their chronology remains to be undertaken, but it may be possible provisionally to identify pre- and post-Venice patterns, that is to say before and after Leonardo’s brief stay in the maritime republic in 1500. Most of the earlier designs use curving loops in which rectilinear interstices result from the intersection of looped threads. Occasionally he substituted angles for the loops, as in the illustrated example of a printed knot design. The angular type also appears in a fragmentary interlace designs in MS A.10 In these angular knots the basic conception of the interlace remains the same.
The generating geometry of the interlace on Christ’s bands is different from even the most angular of the pre-1500 designs. It strongly recalls Islamic tile and other decorative patterns, which were readily available in Venice with its extensive Mediterranean trading links.11 We may well imagine that the notable sophistication of Islamic interlaced and tessellated designs would have caught Leonardo’s attention, not least given his interest in Islamic science. A group of interlace designs on folio 760r of the Codex Atlanticus will serve as a useful reference point since the sheet is datable (Fig. 5.8). The circular mechanisms in the lower left of the page belong to a series of analyses of perpetual motion devices, a number of which were triggered by his witnessing in Venice of a competition to invent mills that would operate in ‘dead’ water.12 The ones illustrated here use a series of hinged rods mounted on a wheel with weights at their free ends. The idea was that the fully extended arms on the left would generate sufficient downward torque to turn the wheel when the rods on the right were hanging limply. After his review of various apparently promising mechanisms, of which the hinged rod device seemed the most promising, he scathingly concluded that the search for perpetual motion was futile. A number of the sketched interlace designs to the upper right of the main device are of the wholly angular, ‘Islamic’ type.
Fig. 5.8Leonardo da Vinci, Designs for Interlaces, Perpetual Motion Machines, a Knot and Stars, with other Notes and Calculations, Milan, Bibliotheca Ambrosiana, CA 760r.
Although our present state of knowledge of the chronology of the knot and interlace designs is less comprehensive than we might like, it does appear that the pattern in the Salvator is characteristic of his post-Venice inventions.13 After 1500 the looped knot and ‘Islamic’ interlace schemes co-exist in his work. The presence in the painting of the looped design in the underdrawing for the angular interlace (Fig. 5.1) could be used to support a hypothesis that the painting was begun in Milan, but this seems to complicate things unnecessarily and is contradicted by other evidence, including the remarkable optical effects, to which we now turn.
Crystals and Optics
The jewel at the central crossing of the bands across Christ’s chest not only displays a sparkling highlight but also a delicate internal reflection around its lower margin to the right. It may well be of crystal not glass. A series of more complex effects are apparent in the lower portion of the large sphere in Christ’s left hand, an area that has not suffered from the abrasions that have reduced the definition across much of the sphere’s surface. It exhibits a complex interplay between the internal radiance near its margin, the skin tone of Christ’s palm and a series of scattered inclusions or ‘bubbles’ within the transparent material of the sphere (Plate 12). Some inclusions catch the light fully; others are half shaded; and yet others are predominantly in shadow. They are conjured up with little commas of light impasto and dark pigment. It looks as if he was practising such an economy of effect on a sheet that contains highly relevant knot designs of an angular kind, formerly in Christ Church, Oxford (Fig. 5.9).
Fig. 5.9Designs for Interlace Pattern and Sketches for Small Inclusions (?), formerly in Christ Church Picture Gallery, Oxford.
Those inclusions that coincide with the margin of the heel of Christ’s hand elegantly pick up blue refractions from his robe. The whole area is full of optical magic, recalling Leonardo’s persistent and complicated observations of light and shade on pebbles within water, some of which are directly illuminated while others are in shadow, to be discussed further below. Only Piero della Francesca had come close to the painting of such brilliantly observed effects in glass and crystal.14 Mary Magdalene’s precious ointment jar in Piero’s lofty fresco in the Cathedral at Arezzo is rendered with great optical brilliance (see Fig. 5.14).15
The sphere is definitely of crystal, probably quartz rock crystal rather than calcite. The inclusions visible in the region of Christ’s hand are not the kind of spherical bubbles that form in glass (Plate 12). Nor are they surface features. Their varied shape—ranging from relatively round to almost triangular—and the way they catch the light testifies that they are the kinds of internal gaps that can arise when a crystal is under formation.16 Often, irregular crystal formation results in undulating fault planes that run across the material. Flawless or reasonably clear rock crystal was highly prized from ancient Egypt onwards, particularly the rare large pieces, and was used to make artefacts of great beauty and price, as we will see.
Leonardo’s fascination with the minutiae of light effects within transparent substances is repeatedly clear in passages intended for his Treatise on Painting, not least since they lay outside the range of sculpture. Particularly telling is his account of what happens when pebbles are seen under water in mixed conditions of light and shade (Fig. 5.10).17 The accompanying note offers a long account that tries the powers of concentration of even the most dedicated reader. After a testing outline of the optical rules that come into play ‘when the images of dark or luminous objects are impressed on the dark and illuminated portions of those bodies located between the bed of the water and its surface’, he explains what happens in the specific set-up in the diagram:
Fig. 5.10After Leonardo da Vinci, The Effects of Light and Shade on Pebbles under Water, Vatican Library, MS Urb.lat.1270 f. 159r.
Let the pool nmtu have pebbles or plants or other opaque bodies on the bed of the clear water, which takes it light from the sun d. And let one area of the pebbles have over it the shadowy image which is reflected in the surface of this water, and let one area of the pebbles have over it the image of the air bcsm. I say the pebbles covered by the shadowy image will be more visible than the pebbles that are covered by the brightness of the light image, and reason is that the eye is overwhelmed and impaired by the illuminated portion of the water in which the air is reflected, and correspondingly the visual power is amplified by the dark portion of this water, and in this instance the pupil of the eye does not possess a constant power, because on one hand it is impaired by too much light and on the other is amplified by darkness.
As was typical in his later discussions of what we see in nature, he paid great attention to the behaviour of the eye and its elusive role in complex perceptions of light and shade in varying circumstances. The play of light on the inclusions in the crystal against the backgrounds of the pale flesh of Christ’s hand and his dark drapery requires a rationale no less complex than that involving the pebbles under the shaded water. Thus, not all the inclusions are seen as catching the light in the same way. Some are shaded as if they are solids, while a few are predominantly dark. Those set against the dark drapery are predominantly bright.
Finding a crystal sphere from the sixteenth century to match the one that Leonardo has painted is likely to be impossible, but the smaller sphere of 3.9 cm owned by John Tradescant in the seventeenth century and which passed with the contents of his ‘Ark’ to the University of Oxford in the Ashmolean Museum shares many of the same properties (Fig. 5.11).18 Spheres of good clarity and size were very rare and were much prized in such wunderkammern or cabinets of curiosities. One of the cherished exhibits in the famous Dresden kunstkammer of August of Saxony was the notably large seventeen-centimetre sphere presented by the Duke of Savoy, Emanuele Filiberto before 1580.19 Christ’s sphere seems to be larger than even the Dresden example. A manuscript was specifically devoted to ‘Description of the effects and powers of the crystal given by the Duke of Savoy to the Elector, Duke August of Saxony’—effects that typically blended optics and magic.20
Fig. 5.11Rock crystal sphere owned by John Tradescant, Museum of the History of Science, Oxford University.
Tradescant’s sphere is marked by numerous inclusions but not seriously marred by fault planes. Photographing it to present what can be seen with the eye is extremely difficult, but it was possible to capture the glow of internal reflection on the side opposite the light source and the varied effects of inclusions at various depths. A smaller ‘pebble’ of quartz contains notably dense arrays of inclusions that react to the light in notably complex and variable ways, some of which are comparable to those selected by Leonardo (Fig. 5.12).
Fig. 5.12Rock crystal pebble.
The photographs raise the obvious question of whether a shiny highlight or highlights featured on the surface of the painted sphere. Leonardo was much interested in defining the movement of ‘lustres’ as the eye moves. He also noted that a ‘lustre’ appeared particularly vivid against a dark background: ‘Among lustres on bodies of equal smoothness that will appear to have the greatest contrast with its background that is generated on the blackest surface’.21 Christ’s blue robe would thus have served to highlight any surface lustres on the crystal sphere, while the paler background of his hand plays more subtle games with the sparkle that emanates from the inclusions.
Hollar’s etching (Fig. 1.3) and the Yarborough version (Fig. 4.4) both display convincingly positioned but understated lustres, as does the version in Naples (Fig. 5.5).22 If they give an accurate indication of the original highlight or highlights, they suggest that Leonardo did not want too much of a distracting glare towards the bottom of his picture. The isolated, opaque white spots of varied sizes that are visible towards the left of the sphere might be remnants of a finely painted gleam or gleams. He knew from his studies of convex mirrors that the reflection of any light sources on a surface of a shiny sphere would be concentrated into small, glaring highlights rather than diffused in a wider manner. Rather strangely, the Yarborough picture also exhibits a centrally placed light spot, from which spidery rays emanate in a centrifugal aura. The centre of the orb in the Cook painting is marked by what seems to be a hole made by compasses.
We know that Leonardo was much interested in minerals that exhibited special optical properties. Variegated stones captured his attention. On the British Museum drawing for the London cartoon of the Virgin, Child, St Anne and St John, dateable to c. 1507, he reminded himself to ask ‘Paolo of Talvecchia [?] … to see the stains [macchie] in the German stones’.23 These German stones may make an appearance as the veined pebbles that appear at the foot of the Virgin in the Louvre painting. In the Buccleuch version of the Madonna of the Yarnwinder the fine beds of platy sandstone on the right below Christ are crossed by pretty undulating veins of coloured minerals.24 He went to some trouble to ape such effects artificially, compressing goose quills so that their crushed and skewed cross-sections resulted in vein-like patterns—which he called mistioni.25 He noted that ‘if in the transparent part exposed to the sun you will make with a small stylus a mixture of various colours … you can make very beautiful patterns and various small stains with contours like those in agate’.26
Leonardo was considered an expert on artefacts crafted from precious and semi-precious stones. In 1502 Isabella d’Este, writing from Mantua to her agent in Florence, Francesco de Malatesta, sought Leonardo’s opinion on vases of crystal, agate, jasper, and amethyst, previously in the hands of the exiled Medici.27 Francesco reports that Leonardo ‘greatly praised them all, but particularly that of crystal, which is all of one integral piece and very clear … That of amethyst or jasper … greatly pleased Leonardo for its being a novel thing and for the diversity of miraculous colours’. The vase of crystal was valued highest of all, at the considerable price of 350 ducats (more than any painting would have been valued). It may be that the ‘libro de’ diaspri’ (‘book of jaspers?) that Leonardo lent to Giovanni Benci around 1505 was linked to his study of these exceptional vessels.28 Both of the main lists of his books included a ‘Lapidario’, probably a manuscript version of Marbodius’s Liber lapidum, a well-known verse anthology of fact and fiction about gemstones, written by the Bishop of Rennnes in the eleventh century.29 He would also have been well aware of the lively discussion of precious minerals in the thirty-seventh and last book of Pliny’s Natural History.
There is also a reference to crystal in a rather enigmatic document in the Codex Atlanticus, not in his own hand but which seems to be a fragment of a dictated letter about his dispute with his half-brothers over his uncle Francesco’s inheritance in 1507–8. This letter speaks of ‘a load of crystal and a few other stones, such that there are three to make cameos and other similar stones for engraving’. One piece of beryl is ‘one braccio long [about 60 cm] and one fifth of a braccio thick, and is as clear as it could be, though it is true that it is split in the middle along its length’.30
About the same time he speaks of making a rectangular ‘occhiale di cristallo’ (eyeglass of crystal).31 The eyeglass consists of a small rectangular window of crystal mounted in a frame with a columnar handle (Fig. 5.13). Within the window, he writes, that it is ‘an eyepiece of crystal of a thickness at its edges of an ounce of an ounce’. An ‘ounce’ or ‘inch’ is one twelfth of a unit of measure. The thickness is thus 1/144th of a braccio, which is about 4 mm.
Fig. 5.13Design for a Rectangular Eyeglass on a Stand, MS F 25r, Paris, Institut de France.
The main text (which is tricky to interpret) tells us that the crystal should be
clean of blemishes and very clear … It should be of a thinness in the middle that accords with the vision of the user, that is to say according to the ratio of the eyeglasses that work well for him, and it should be contrived in the same configuration as his glasses.
Leonardo tells us that the window should be 10 mm high and 15 mm wide. It should be placed about 7 cm from the user’s eye and ‘as far again from the letter that is to be read’. He notes that,
if it is re-sited at a greater distance [moved away from the printed type?] the letter will appear larger. In effect a standard letter as printed will appear as if from a special box [of large type fonts?].
A marginal note indicates that ‘this eyeglass will be good if retained on the desk but if you want to locate it outside, make it 7.5 mm long and 5 mm high’.32 It is unclear why the outdoor version should be slightly smaller.
Whatever the problems caused by his description, it is clear that he is describing a small rectangular magnifying lens with a handle, which would have offered greater flexibility than standard spectacles with their fixed position on the nose. Beside the main note he places a cross, perhaps meaning that he was not satisfied with the scheme he had formulated.
All these references to the optical properties of minerals and semi-precious stones cluster in the period between 1502 and c.1508. His attention to rock crystal in the Salvator reinforces our growing conviction that the painting dates from the first decade of the new century.
Leonardo would have been well aware that a sphere of crystal (or of glass) would exercise a considerable effect on the forms behind it. Depending how far it was from the refracted surfaces, the spherical ‘lens’ would magnify and distort the folds of the drapery in a curvilinear manner, or at a greater distance would invert the image. No such effects are visible in the painting. We should remember that Leonardo was drawing on his documented knowledge of rock crystal to devise a large sphere for Christ to hold. He was not making a ‘portrait’ of an actual sphere held by a model. Indeed, he is most unlikely to have encountered a sphere of that size. In avoiding the extreme optical effects of the sphere, he was exercising decorum; that is to say ‘pictorial good manners’. He observed many visual effects in the real world, such the blur of very fast-moving objects, that he would never have incorporated into paintings. He said that such extreme effects belonged to the world of philosophical ‘speculators’ (speculatori) on natural phenomena rather than to the art of painting.33 He and other artists knew about refraction in water, but neither he nor other Renaissance painters would have portrayed Christ’s legs as optically bent in a Baptism.
Leonardo’s paintings re-make nature not only in accordance with natural law but also in obedience to the rules that governed functioning images for his patrons. He would not have disrupted the efficacy of the painting as a devotional image. His paintings were not raw demonstrations of optical science, any more than they were stark demonstrations of anatomy. In the Mona Lisa, for example, he does not follow the visual implications of looking at a woman inside a balcony against a bright landscape, although he knew precisely how what we see is affected by looking from a dark place into a bright one and vice versa. To have painted the full effect of magnification or inversion produced by such a sphere acting as a spherical lens would have been wholly detrimental to the functioning of the image.
As we noted, there is a distinct double effect visible in the heel of Christ’s hand at the base of the sphere (Plate 12). It is unclear if this double contour is a pentimento, if it results from damage or is deliberate. It might be taken as indicating Leonardo’s interest in the double refraction exhibited by calcite, which would not have been distinguished from rock crystal at the time. However, the angle of the double refraction in calcite is very slight, and the double contour is best recognized as a pentimento.
What Leonardo has done is to paint a rock crystal sphere on the basis of his study of the way that light reflects off and passes through quartz, how the inclusions catch the light, and how they were seen in the eye. He would include such effects only to the extent that they enhanced the efficacy of the painting in expressing its meaning. His procedures were essentially similar to those of Piero della Francesca, his predecessor as a master of reflection and refraction. The precious ointment jar in Piero’s frescoed Magdalene is clearly crafted from rock crystal. On the left, the jar brilliantly reflects the upright windows in the east wall of the cathedral, while on the other side reflection and refraction create a diffuse glimmer within the jar (Fig. 5.14).34 The significant difference is that Leonardo includes inclusions. Although he prized perfect examples of crystal, the sparkling inclusions may well have appealed to him to convey the effect of a cosmological sphere with its fixed stars, as we will see.
Fig. 5.14Piero della Francesca, Mary Magdalene, detail, Cathedral, Arezzo.
Leonardo is unlikely to have chosen a crystal sphere solely because he liked its optical characteristics. As a material of wondrous transparency, it had come to be associated with purity, virginity, innocence, and chastity.35 The most spectacular artefact that exploits the metaphorical purity of quartz is the Carolingian Lothair or Susanna Crystal, mounted in a fifteenth-century frame (Fig. 5.15). It is engraved with the story of Susanna, which conveys conjoined messages of chastity and justice.36 Susanna, accosted by lecherous elders after bathing, had resisted their advances, and was publicly accused of promiscuity, before their lies were unmasked in court. We need not think that Leonardo knew this magnificent object, but it does vividly testify to the prestige and connotations that rock crystal had come to carry. More generally crystal was widely used in reliquaries to form a transparent container or window for whatever holy shard might be the subject of devotion. Notable examples are the reliquary of the lance in the Vatican and the monstrance of the Castiglioni family in the Cathedral of Milan, both of which incorporate crystal of magnificent quality.37 The sixteenth-century Counter-Reformation theorist of religious images, Cardinal Gabriele Paleotti, specifically mentions the use of crystal to ‘sustain’ the mysteries of revered items in reliquaries: ‘before a sacred relic one places a subtle veil or transparent crystal so that the great secrets of eternal things should be sustained in all their majesty’.38
Fig. 5.15Lothair (or Susanna) Crystal, ninth century, British Museum, London.
Leonardo would have been alert to the well-known associations of rock crystal, but the way he has characterized the sphere, scintillating with inclusions, indicates that more specific, scientific allusions also played a significant role. In Ptolemaic cosmology, the fixed stars, beyond the orbits of the circling planets, were embedded in the crystalline sphere. This was portrayed in some detail by Raphael in his image of Astronomy on the ceiling of the Stanza della Segnatura (Fig. 5.16).39 The stars, with graphic traces of the signs of the zodiac, are scattered across the sphere, with a rotund earth visible at its centre. Leonardo’s inclusions do not appear to adopt the configuration of the constellations.40 In the fresco below Raphael’s Astronomy, the School of Athens, an astronomer (Zoroaster?) holds up a celestial sphere, while a geographer (Ptolemy?) displays a nicely characterized terrestrial globe (Fig. 5.17).
Fig. 5.16Raphael, Urania (or Astronomy), from the ceiling of the Stanza della Segnatura in the Vatican.
Fig. 5.17Raphael, Celestial and Terrestrial Globes, from the School of Athens in the Stanza della Segnatura, Vatican.
The earth is clearly characterized at the centre of Carpaccio’s and Gallego’s transparent spheres (Figs. 5.18 and 5.19), and a comparable orbis terrarum is visible at the centre of the crystalline orb held by Christ in a Leonardesque Salvator Mundi Surrounded by Angels, which may originate from the orbit of Luini.41 The date of the Carpaccio is not known, but the Gallego altarpiece is almost certainly earlier than Leonardo’s image and is an exceptionally precocious instance of the translation of the terra into the cosmos. It is very unlikely to have been known by Leonardo. His rather literal replication of the optics of a crystal sphere appears to be uniquely his.
Fig. 5.18After Leonardo, Salvator Mundi, detail of the sphere, formerly Yarborough Collection.
Fig. 5.19The Passage of Light from the Sun to the Earth and Moon and the Phenomenon of ‘Ashen Light’ on the Moon, Codex Leicester 2r.
The white circle with its diagonal aura at the centre of the Yarborough version (Fig. 5.18) may indicate that Leonardo had originally emphasized the terrestrial centre of his sphere with a small earth, where the compass hole is located. It may be in this respect as in others that the Yarborough picture reflects an early stage in Leonardo’s working on his composition, and that the central earth was not evident or was barely apparent in the final version. There is no obvious sign of it in the Hollar etching.
Leonardo’s interest in optical phenomena in the observable cosmos is reflected in his notable analysis of the lumen cinererum (‘ashen light’) on the shaded side of the moon on folio 2r in the Codex Leicester of c. 1506–8.42 He accounts, in an innovatory way, for the soft glimmer within the lunar shadow (Fig. 5.19) as the reflection of the sun’s light off the seas of the earth and hence to the moon, plotting the course of the rays from the orbiting moon and sun to demonstrate the optical validity of his theory. He also points out in a very novel way that the shaded portion of the moon looks lighter against the dark background of the void, and darker where it abuts the bright crescent—an effect arising in the eye of the observer. There is no doubt that the complexities of light on the starry inclusions, featuring shadows of both the ‘ashen’ and more assertive kinds, fit into his spectrum of interests in the passage of light through transparent media, its ‘percussion’ on the contours of bodies, and how we perceive such effects.
Whether or not Leonardo had contrived or had considered contriving an ‘earth’ at the centre of his rock crystal sphere, there can be little doubt that he was intending the transparent orb with its sparkling inclusions to stand for the crystalline sphere of the Ptolemaic heavens. His Saviour of the World might therefore be better described as the Saviour of the Cosmos.
Depth of Field?
There is also one other notable optical effect that does not quite appear in the same way in other Leonardo paintings. Even allowing for the varied condition of different areas of the painting, it seems evident that Christ’s hands are in clearer focus than his face.43 Within what photographers call a limited depth of field, Leonardo seems to have adjusted the focus to be sharpest on the objects nearest the picture plane. In MS D in the Institut de France, dating from around 1507, Leonardo had explored a whole range of issues that arose when looking at objects at various distances from the eye.44 As he became increasingly aware of mediaeval texts on optics based on those by the Arab philosopher, Ibn-al-Hatham (Alhazen), he worked on geometrical variations on complex arrays of light within the eye and devised ‘experiments’ to test which worked best. To rectify the inversion of the light within the aperture of the pupil, he devised a second inversion in the ‘crystalline sphere’ at the centre of the eye (Fig. 5.20). There is something rather nice about one ‘crystalline sphere’ facilitating the seeing of another.
Fig. 5.20Optics of the Eye, with a Proposed Experiment with a Glass Sphere and Bowl, Paris, Institut de France, MS D 3v.
This complex set-up meant that the array of light from images was registered in the eye across a surface of finite dimensions rather than focused at a single point.45 The system was designed to form an upright image at the termination of the optic nerve. The central ray that passed through the system without refraction gave the sharpest impression of the edge of a body but it was surrounded by other impressions of decreasing strength. The result was a firm contour surrounded by a zone of graded blur. At an optimum viewing distance the sharpness attained its highest level. He said that ‘the boundaries of things in the second plane will not be discerned like those in the first plane’.46 The relative visibility of objects diminished as forms become more distant, resulting from the weakening or tiring of the ‘species’ (images) as they passed across space and the diminution of the angle under which they are viewed—losing definition and colour. This uncertainty was enhanced by the diminishing angles under which more distant objects were viewed and the veiling effects of intervening atmosphere.
He knew that at very close ranges, of a few inches or less, that sharp perception was impossible—though he did not know that this was caused by the focal length of the lens. Christ’s hands are at the optimum distance—far enough away and yet close enough to be seen as sharply as the eye could manage. However, the ‘soft focus’ effect of more distant forms is not observed with complete consistency, since, unsurprisingly, the vivacious vortices of the hair are more crisply painted with little curves of delicate impasto, while the adjacent features of the face are painted more softly. Leonardo’s so-called sfumato is, as we shall see, selectively and instinctively tuned to act in concert with the desired pictorial effects.
Motions of the Soul
We tend to think of Leonardo as rebelling against the conventional formats of such established genres as Madonnas and portraits. However, he was also fascinated by what could be accomplished with traditional types, taking on the challenge of profile portraits, for example. He could have loosened the traditionally hierarchical format of the Salvator Mundi but chose not to do so. I think we can see why. He conducted a series of experiments with direct communication between a figure within the space of a painting and the spectator. These experiments came to a head in the period 1503–13. As portraits of women, both the early Ginevra de’ Benci and the Mona Lisa were unusually daring, even somewhat indecorous, with their unabashedly direct glances. The Mona Lisa is the more adventurous, since she not only looks at the viewer but also reacts to our presence—however enigmatically.47 This is taken a step further in the frontal stare of the so-called Nude Mona Lisa, which was devised if not painted by Leonardo.48 The Louvre St John announces the coming of the saviour directly to us. This form of communication is very much present in Giovanni Francesco Rustici’s St John Preaching, high above the north door of the Florentine Baptistery, in which the saint regards and speaks to the spectator in the vicinity of the column of S. Zenobio.49 Leonardo seems to have played a key role in the conception of Rustici’s group in 1507–8.
The Mona Lisa, St John, and the Salvator Mundi rely upon the direct and active engagement of the spectator, but it is a directness that is subverted by tantalizing ambiguity, given the intangibility of the subjects’ facial features. Christ’s head is in ‘softer focus’ than his hands. Leonardo’s blurring of surfaces is not only an optical technique, reflecting his opinion that ‘the eye does not know the edge of any body’, as he states in MS D, but a psychological device that presents us with a teasing latitude in how we read the expressions of the painted characters.50 His insistence that one of the key tasks for the painter was to show ‘the motions of the soul’ in his painted figures did not mean that these motions were always literal and explicit. Our mind is invited to read these ‘motions’ on our own terms.
The inviting engagement of the spectator’s imagination as we collaborate with this ambiguity is a special feature of Leonardo’s later paintings. He was a pioneer of what John Shearman has called ‘transitive’ painting, in which the spectator actively completes the communicative loop demanded by the image.51 Leonardo’s Salvator is at once insistent and elusive, assertive and mysterious. We are invited to make eye contact with Christ, but the soft definition of his irises and pupils ensure that his mode of looking remains spiritually ineffable and ultimately beyond our rational comprehension, as we saw in Chapter 2. Christ knows the secrets, but they are not accessible to us. All we can do is trust in his gentle stewardship of mankind and assume his sweet yoke.
Although we have described the Salvator as a smaller-scale devotional painting, it is in fact a good deal bigger than the small Madonnas. It is, in effect, a large small-scale painting. Christ’s head is only a little below life size as measured on the surface of the painting, which means that it effectively looks life-size located at its short distance behind the picture plane. Placed against a background of indeterminate depth Christ is endowed with a powerful presence in our own world, whilst ambiguously existing in a spiritual realm of ineffability. We know that Leonardo was totally averse to attempts to define the nature of things that lay outside the observable truths of the natural world, but he did openly recognize that revelation carried a very special value. We have seen that he revered the ‘sacred writings’ because ‘they are the supreme truth’.52 This corresponds very much to the differentiation that Cusanus had made between earthly sight and visions of the divine when he was informing the monks how to use the image of God’s face that he had sent them.
In the Salvator Leonardo was parading his mastery of optical truth, not least in the crystal sphere, and evoking the supreme mystery of a spiritual realm that lay beyond even the celestial globe that Christ holds in his hand.