3. Leonardo on Shadow in 1490–93

Baxandall, Michael

Save Cite Share

Free

Appendix: Three Notes on Leonardo and Early Renaissance Shadow: 3. Leonardo on Shadow in 1490–93

Michael Baxandall

Shadows and Enlightenment

Leonardo came out of the design ambience represented by the metalpoint and biacca drawings. He had probably gone into Verrocchio’s workshop in 1469 at the age of seventeen, was admitted to the Florentine painters’ confraternity in 1472 and was still living in Verrocchio’s house in 1476. There are works by him attributed to the 1470s – in particular, a group of studies of drapery, monochrome brush drawings in black or dark grey on mid-grey-toned linen with white heightening …

Author

Michael Baxandall

PublisherYale University Press

https://doi.org/10.37862/aaeportal.00111.011

Stable URL: https://aaeportal.com/?id=-13765

Chapter subject tags:Leonardo, da Vinci, 1452-1519Relief (Art)Shades and shadows in artDrawingVision--ResearchOptics and art

3. Leonardo on Shadow in 1490–93

Leonardo came out of the design ambience represented by the metalpoint and biacca drawings. He had probably gone into Verrocchio’s workshop in 1469 at the age of seventeen, was admitted to the Florentine painters’ confraternity in 1472 and was still living in Verrocchio’s house in 1476. There are works by him attributed to the 1470s – in particular, a group of studies of drapery, monochrome brush drawings in black or dark grey on mid-grey-toned linen with white heightening (fig. 51) – which seem both an implicit critique of metalpoint and biacca practice and an early stage in his own analysis of shadow.

~

Description: Drapery study by Leonardo da Vinci

Fig. 51 Leonardo da Vinci (1452–1519), Drapery study. Brush drawing on brown linen, 28.3 × 19.3. British Museum, London (1895–9–15–489).

In trying to understand Leonardo’s thought on a matter it is often an anxiety that, because of the massive quantity of notes written over decades, the whole process of thought is being telescoped and scrambled. The trouble is not so much apparent internal contradictions due to his thought evolving over time, as the loss of momentary internal coherences across his thinking about the matter and the loss of relation to his other activities of the moment. It is sometimes best to establish a base on a favourable moment – in the sense of a moment when his thought on the topic is cross-coherent, urgent, copious, securely dated and central to the total – and to set any different thoughts at other times in their relation to this moment. In the particular case of shadow this is an unforced and fairly easy course to take, and the benign moment happens to be early, 1490–93.

The years 1490–93 are fairly clearly the culmination of a period in which Leonardo was much interested in working out his views on vision, and on light and shadow. Among other things it seems the moment when he finally relinquished any doubts about intromission: that is, he puts aside residual hankerings after a visual ray emitted by the beholder’s eye as opposed to vision by light admitted to the eye (Lindberg, 1976, pp. 159–61). He had been in Milan since 1483, long enough to gain some distance from Florence. It does not seem an important moment in his active production of painting, lying between the first Virgin of the Rocks of the mid-1480s and the Last Supper of the mid-1490s: the great project of the time was sculpture, the aborted bronze equestrian monument of Francesco Sforza. Perhaps the work on vision was partly a surrogate for painting. In his technical drawings for anatomical and other such study there is a climax of fine diagonal rectilinear hatching representational of light on modelled surface. (In later years, it is generally held, this was to give way to more objective bracelet hatching following the contours of surfaces, a little like that of Dürer’s engravings, and at the same time his subjective or phenomenal drawing had a divergent development in such media as chalk.)

By 1490–93 Leonardo had already developed his concept of the ‘radiant pyramid’. That is, he believed that any object radiates material likenesses of itself in all directions and in straight lines, and that these likenesses diminish in size proportionately with their distance from the object. Formally the ‘radiant pyramid’ is the visual pyramid of vanishing-point perspective construction backwards. The medium of radiation is the atmosphere, filled with an infinite number of these progressively shrinking likenesses which, though they are material, are subtle enough to pass through each other on encounter. The atmosphere has to be activated for our visual perception of the likenesses, by light. Light also travels rectilinearly in rays.

Shadow is the absence of light and, in a very strong sense indeed, its opponent. Light is always accompanied by shadow. As there are luminous bodies emitting luminous rays there are also ‘umbrous’ (ombroso) or shadowing bodies emitting shadowing rays, by opposing light with denseness. Being dense is the opponent of being luminous. Leonardo, at this time, is even prepared to say shadow is stronger than light because it can entirely extinguish it, whereas light cannot entirely extinguish the shadow caused by dense shadowing bodies; but this is not an idea he develops.

Leonardo’s many particular propositions and demonstrations are carried out mainly with two straightforward operations: by tracing rectilinear rays from light sources to objects, and by establishing simple proportional relations between terms. For instance, extension of derived shadow is the product of the relation between the extension of light source and the occluding section of a dense shadowing object. Extension of light-source is a frequent element in the propositions partly because Leonardo is not yet fully tackling ambient, or ‘universal’, light; but he is moving towards this by studying lighting from very extended sources like windows with posited universal light outside (fig. 52).

~

Description: Light from a window on an umbrous sphere by Leonardo da Vinci

Fig. 52 Leonardo da Vinci (1452–1519), Light from a window on an umbrous sphere, with (from top) intermediate, primitive, derivative, and (on the surface at the bottom) cast shadow. Pen and wash over metalpoint on paper, 24 × 38. Bibliothèque de l’Institut de France, Paris (Ms. 2185; BN 2038, fol. 14.r.).

There are three kinds of shadow: (1) ‘original’ or ‘primitive’ shadow, which is formally but not functionally what we have been calling self-shadow, (2) ‘derived’ shadow and (3) cast shadow, which two are a redistribution of our projected shadow. There is also the greater or lesser illumination involved in our shading, sometimes called mezzano or (4) intermediate shadow.

(1) The main point about the original self-shadow is that it is not just the self-occluded surface of an object but is original or originating in the sense of being the source of derived shadow rather as an ‘original light’ is the source of rays of light: original shadow emits shadowing rays. Apart from its shadow-emitting property, there are two other peculiar characteristics about original shadow. Intrinsically it has the same value all over, unlike derived shadow and shading, which vary in intensity from here to there. And it clings to the object: move a sphere in light and the self-shadow may move on the surface of the object but will always be there.

(2) As for the derived shadow, Leonardo’s reference with the term is labile, but at this time he seems to use it less of shadow on surface than of the volume of atmosphere occupied by those shadowing rays. Derived shadow attenuates in intensity as it distances itself from its origin: this is seen as the expression of a general natural principle whereby things weaken as they leave their sources, rather than a result of the progressive intervention of reflected light, not mentioned at this point.

(3) Cast shadow is then specifically that surface product of derived shadow that is ‘surrounded by light’. Leonardo usually seems to visualise it as cast by something like a detached sphere (or rather, of course, by the self-shadow on the light-occluded side of that detached sphere). Cast shadow is interesting because it modifies the form (or rather, section) of derived shadow in response to the angle, as indeed to any form, of the surface on which it is cast – though Leonardo does not show the sciographer’s passion for working out case after complicated case of this, which indeed would have involved rather more elaborate geometry than he was using. One uses the eye, as his less formal advice that the painter adjust cast shadows to the shape of the surface they fall on suggests.

(4) Slant/tilt shading is firmly conceived of as a function of the angle of incidence of light of a lit surface, indeed of the force of impact resulting from that angle: throwing a ball at a wall at different, angles and so with different bounce is used as a formal analogy. (In relation to painting, Leonardo intermittently adopted the three-tone model without complaint.)

An important matter is the relative intensity of original self-shadow and the derived or cast shadow associated with it. On the one hand, the original self-shadow is intrinsically stronger because it is not subject to the attenuation with distance the derived shadow suffers. On the other hand, the original shadow is subject to interference by reflected light from illuminated surfaces around the cast shadow deposited by the derived shadow; whereas the cast shadow faces the darkness of the original self-shadow (fig. 53). Clearly it would depend on the angles of surfaces and the proportions of luminous and shadowing bodies, and also on the character and alignment of adjacent surfaces, but Leonardo seems intrigued by the case of the more intense cast shadow, and it was this aspect that Jombert-Cochin, Dandré-Bardon and so many others were later to pick up and simplify into a rule. Leonardo is generally alert to the role of reflected light in the intensities of shadow and within shadow.

~

Description: Light reflecting from around a cast shadow on to a self-shadow by Baxandall,...

Fig. 53 After Leonardo da Vinci. Light reflecting from around a cast shadow on to a self-shadow. A is light source, BC primitive self-shadow on intervening solid, NM cast shadow on surface DE. Light reflects back from DN and ME on to BC. Redrawn as a conflation of Vatican Library, Rome, Codex Urbinas Latinus, fol. 184 verso, and Bibliothèque de l’Institut de France, Paris, MS 2174, fol. 4 verso.

Finally, there is the matter of shadow edges. Leonardo describes the effect of contrast, that the edge of a shadow on a light surface will (specifically) ‘appear’ darker and also that on a dark surface it will appear less so. For painting he urges contriving the former, mainly to establish relief and distance. But he is already alert to the soft edge. Derived shadow edges soften as they distance themselves from the original shadow until the shadows imperceptibly disappear: it follows that cast shadow edges must soften with distance from the shadowing object. And, for painting, the balance is already towards the sfumato transition, for aesthetic reasons of grace. One should not sharpen the edges of indistinct or minor shadow or the painting will look wooden; contrive to paint faces in diffused light – and the rest of his well-known tips for setting up the working space.

There is some temptation to take the more idiosyncratic emphases of Leonardo’s account of objective and physical shadow – the shadowing rays emitted by shadowing surfaces, the derived shadow occupying volume with waning force, the dark cast shadow facing just the original shadow – as oblique or displaced propositions about the structure of shadow perception. One cannot quite do that because they have too many affinities with his general convictions about energy and proportion in physical nature.

In later notes on shadow (which will not be sorted out chronologically here) he does not so much revise as apply the theory of 1490–93 in expanded ways. In effect, he takes it out of the studio into the open and puts it to use. It is the frame for his meticulous observations of the complex light and shade of trees, and it is accommodated though not prominent in his accounts of atmospheric perspective. He works out cases of multiple light source. He extends his analysis of reflection to include more about hue and hued shadow – such as the blue shadows at sunset (fig. 37) – as well as intensity of shadow, but this does not involve any methodical innovation. The main development of method is the neat formula for working out the shading from universal light, diffused light out-of-doors. It is treated as a super-extensive light source in the form of a hemisphere enclosing the object, the hemisphere being reduced to a plane semicircle for the purpose of the geometry (fig. 54). This is an expansion of the arc used in his earlier window studies (fig. 52). The formula is particularly prominent in the seventeenth-century reduction of the sixteenth-century compilation known as the Codex Urbinas and was clearly familiar to Lambert (§28).

~

Description: Universal light and a house by Baxandall, Michael

Fig. 54 After Leonardo da Vinci. Universal light and a house. The hemispherical source is CMR. The relative irradiation on any point is determined by the angle of the arc to which it is open within CMR. For instance, BC is the proportion of universal light falling on the area AO under the eaves. Redrawn after Vatican Library, Rome, Codex Urbinas Latinus, fol. 201 recto.

(Texts: Leonardo had planned a systematic treatise on shadow in seven parts, which seems not to have been written; his observations are dispersed in notes of various periods. Something like half of them survived into the posthumous compilation from his papers by his heir, Francesco Melzi, of a ‘Treatise on Painting’, the Codex Urbinas Latinus 1270 in the Vatican Library; figure 1 is from this, and the French translation of 1651 (fig. 37) was done from a reduced manuscript version of it. The other half are scattered among several different autograph manuscripts.

There is much matter on shadow, since it is important to painting, in the Codex Urbinas. Some of it can be matched with material in two surviving notebooks of this period, ‘Libro G’ or MS C (Paris, Institut de France, MS 2174), twenty-eight folios, which is dated both 1490 and 1493, and Ashburnham II (now Paris, Institut de France, MS 2185), which consists of thirty-five folios abstracted from the Institut’s MS A and is attributable to 1492. These two contain much other work on shadow and relevant work on light not used in the Codex Urbinas and they constitute the basis for a view of Leonardo on shadow in 1490–93. Of the Codex Urbinas shadow matter not in them, much was taken from a lost notebook, ‘Libro W’, probably of considerably later date.

The Codex Urbinas is accessible in an edition with facsimile and English translation (Leonardo, 1956), with which light and shadow can be pursued in the excellent index. The most convenient selection of shadow observations in translation is now in Leonardo (1989), pp. 97–115 especially, which includes material not in the Codex Urbinas. The sharpest exposition of Leonardo on visual perception is still Lindberg (1976), pp. 154–68, with guidance through the earlier bibliography in nn. 34–98, and in general establishing the intellectual-historical frame. For shadow specifically, see also Keele (1983), pp. 49–60, Kemp (1990), pp. 267–9, and particularly Veltman (1986), pp. 326–37. Most studies of Leonardo have something on his shadow.)

Previous chapter

Base text size -

Chapter images ()