Before restoration work began, the painting underwent a campaign of scientific investigation aimed at characterising its constituent materials, its technique of execution and its condition.¹ The painting is executed on a single piece of plain-weave canvas with a density of approximately 12 × 13 threads/cm², as can be observed in the small losses in the paint film in the X-ray image [fig. 1] and from direct observation of the reverse of the canvas after the removal of the lining canvas during restoration treatment.

Along the right edge of the painting, the radiographic image revealed the characteristic garland-shape deformations induced by the stretching of the canvas onto the original stretcher [fig. 3]. Approximately 6 cm from the edge, also visible in the X-radiograph, is the imprint of the stretcher bars produced on the ground when this was laid down on the canvas. The stretcher bars must therefore have been approximately 6 cm wide. Along the upper and lower margins of the canvas, the weaker evidence of canvas deformation and the narrower imprint of the stretcher suggest that these margins were slightly trimmed in the past. The situation at the left margin is more complex: neither the wave-like deformation of the canvas nor the imprint of the original stretcher are detectable, except, perhaps, at the very edge of the support. This would suggest that along that side the canvas was trimmed by about 6 cm. However, this assessment does not agree with the observation made during the restoration, which, on the portion of the canvas folded and nailed to the left side of the stretcher, revealed no traces of the paint film, suggesting that the original limit of the painting must have coincided with the current edge.

The presence of damage to the paint film aligned along the vertical edges of the painting are to be referred to the folding of the canvas and its nailing on a stretcher, smaller than the original one, which would have reduced the width of the painting to about 78 cm, with the left margin touching Pilate’s ear and the right edge touching the tip of Christ’s sceptre. Fortunately, on that occasion, the excess portions of the painting were not cut off, and in a subsequent restoration these were recovered, and made visible again, replacing the stretcher. In conclusion, the height of the painting would originally have been only slightly greater than at present, and 86 or 92 cm wide; at a later date it was reduced to a width of 78 cm, and finally the canvas recovered its current dimensions of 111 × 86 cm.

The ground of the painting is applied in two layers, the lower of which has a yellowish-brown colour, while the upper layer is reddish-brown. The inner layer consists of earth pigments, including umber, with the addition of lead white and a carbon black pigment, mixed with a calcium-based compound [fig. 2]. The calcium compound can be identified as calcite, in the form of translucent, whitish particles that sometimes reach dimensions of 100 microns, and these are mainly responsible for the coarse granulometry of the layer, which appears irregular and rough on the surface, with a texture which not even the second preparatory layer completely hides. The last layer of ground is largely composed of earth pigments, mainly red and brown earths, as well as smaller quantities of lead white and particles of a black pigment.

The ground is intentionally left in sight on the surface and used as a midtone in the fleshe tones. It can be seen, for example, in the shadows of Pilate’s face, particularly below the right eye [fig. 4], the nose and just above the beard, in the palm of his left hand and in the construction of Christ’s face, particularly in the eye sockets and on the left side of the nose. The same ground layer is also visible in the shaded area of the face and on the back of the left hand of the figure behind Christ. The cleaning of the painting then revealed how, in the darker and cooler parts of the shadows of the fleshtones, the ground was instead glazed with a thin dark application of paint.

IR-reflectography shows how the composition was put in place with a few preparatory brush-strokes, used to define the outlines of the figures as also the facial features, or the fingers of the hands [fig. 6] or, very clearly, the entire outline of Christ’s left arm. At the same time, radiography and raking light have made it possible to detect numerous incisions used to delineate the boundaries or make marks useful for establishing the anatomy of the figures [fig. 5]. Some of these incisions, for example, mark the outlines of the nose, the axis of the eyes and the overall dimensions of the face of Christ and of the man placing the red cloak on his shoulders. In the latter figure, both the reference-marks for the positioning of the eyes and those for the nose, and indeed the entire profile of the face where it is in shadow, are not followed at the painting stage, which renders the head which originally reclined towards the left shoulder, much more upright. In the figure of Christ, an incision indicates the right outline of the neck, until it crosses an almost horizontal line which, together with the line leading to the left shoulder, seems to indicate the position of the collarbones. The incisions and the preparatory brush drawing complement one another, as is evident, for example, in the formulation of the intertwining of Christ’s and Pilate’s hands. Christ’s right wrist is incised, both the upper and lower outlines, several centimetres higher than it was finally painted [fig. 7]. A similar shift also seems to affect Pilate’s left hand, originally probably conceived with the fingers extended upwards in the direction of the soldier’s left hand. Pilate’s right hand, on the other hand, appears more faithful to the incised design, from which it deviates only by a slight variation in the thumb.

The placing of these intertwining hands, certainly the most compositionally and perspectivally complex part of the painting, is also based on the use of brush drawing, which at times seems to follow the incisions, correcting their position or marking with precision the boundaries that will be painted, such as the upper outline of Christ’s wrist or that of Pilate’s first metacarpal and left thumb. The soldier’s left hand is almost completely defined by incisions, which also indicate the boundary between light and shadow that runs immediately behind the knuckles. The face of the Roman prefect, on the other hand, seems to be painted in the absence of any graphic reference-lines, except for that of the shoulder, which is incised, as well as the three short strokes that seem to define the placement of the head-dress, the forehead and the mouth. A few incisions define the red cloak, probably to mark the position of some of the folds, as well as the reed sceptre held by Jesus [fig. 8].

In the concluding stages of the painting process, perhaps using the back end of the brush handle, other incisions were made: in these instances the artist removed the freshly applied paint so as to bring into view the ground layer beneath, in order to achieve certain details more quickly, such as Pilate’s eyebrows, or the locks of hair that fall on the lit part of his forehead, and Christ’s hair where it overlaps his right ear [fig. 9]. Lastly, and still at the preparatory stage of the laying in of the composition, in particular related to defining the areas of light and shadow, are the zig-zag brushstrokes revealed in X-radiograph in the areas of maximum light incidence in Christ’s chest and shoulder [fig. 10], as well as in Pilate’s right temple.

In addition to the corrections already described and revealed by the comparison among incisions and painted layer, few revisions seem to have occurred while painting, with the exception of a diagonal mark, visible in the X-radiograph, which rises up to the Saviour’s right shoulder from the knot of the rope at his wrists, touching Pilate’s bearded profile [figs. 11a-b]. This stretch, comparable in width to that of the cane, could be related to an earlier position of the sceptre, which originally crossed the arms at the height of the knot and was directed towards the upper left-hand corner of the painting; or perhaps it could be related to a rope, held with the right hand by the executioner behind Christ, or to a shorter ligature that may have supported the red mantle around the chest. Other adjustments during painting relate to Pilate’s head-dress, partly superimposed over the red cloak held by the soldier with his right hand, as well as his black robe, which partially overlaps the balustrade in the foreground which had already been painted in full when the robe was executed and, at that point, was wider than at present, as revealed in the X-radiograph.

More difficult to interpret was the in- formation regarding the folds of the robes worn by Pilate and the soldier behind Christ. At the time of the investigation, in fact, the legibility of the painting was somewhat compromised because of the thickness, inhomogeneity and alteration of the layers of varnish, as also can be seen in the UV-induced fluorescence photograph [figs. 12a-b], and the microscopy documentation of the cross-sections. The infrared reflectography [fig. 13] and the X-radiographs indicated modelling in the drapery that was no longer visible at that time, but which has now been recovered thanks to the cleaning of the painting.

The palette was investigated non-destructively by means of X-ray fluorescence analysis [fig. 14, table 1] and through the examination of the three samples taken from the painting. The palette is mostly composed of lead white and earth pigments, selected and mixed in different tonalities and proportions according to the effect desired in each area. These pigments can be found in brown areas, both in the light and dark, such as for instance Pilate’s robe, the crown of thorns, the reed and the balustrade in the foreground, and in the construction of the flesh tones (accompanied by small amounts of red lake pigment only in the pinkish tones of Christ’s left arm). For particularly dark layers such as the hair, Pilate’s beard, the inside of the mouths or the areas in shadow of the balustrade and the faces, XRF analysis also indicated the use of umber. The red paint of the rim of Pilate’s headdress, the cloak held on Christ’s shoulders, and the drops of blood on his body are painted with vermilion where the tonality is brightest, and with vermilion and a red lake pigment in the darker areas and those of shadow. The traces of lead-tin yellow found in one of the red points analysed possibly refers to its use to achieve a more orange hue in some of the light areas of the cloak.

As to the condition of the painting, the X-radiograph indicated, in addition to the damage caused by the aforementioned nailing along the vertical edges, losses in the paint layer in the upper part of the painting, especially close to the upper left corner. However, the most important parts of the composition appear to be free of significant losses in the original paint film. A few losses are present in Pilate’s robe on the wrist and the on the back of his right hand, and in the lower section of the red cloak, near the lower right-hand corner of the canvas. An inverted “L”-shaped tear, approximately 5 cm in length along each side, is present in the support corresponding to the top section of the sceptre, while in the figure of Christ, there is a small horizontal cut on the right side of the neck, and a slightly longer but thin and vertical damage running down from the Adam’s apple to the sternum. Finally, losses, of no more than a few square millimetres are present in the shadowed part of Christ’s face and hair, and in Pilate’s forehead and robe.

Examination of the painting under UV, carried out prior to the restoration, recorded almost exclusively a very intense fluorescence of the thick layers of varnish applied throughout the painting, reaching a considerable thickness in the area of Pilate’s robe, as evident in the sample taken from this area. This same analysis, however, also documented how a selective thinning of the old varnish had been carried out in the past in the flesh tones and areas of light, which were then revarnished with the application of a substance, probably acrylic, spread unevenly and with the presence of drips. This procedure was particularly evident around Pilate’s beard and cloak, in the area around the right elbow, where the black hue of the original layer has re-emerged, compatible with what was found in the cross-section, the perception of which had been strongly altered into green by the optical interaction with the overlapping layers of yellowed varnish. Finally, the raking light photographs, prior to the restoration, had indicated the presence of numerous and widespread areas of lifting in the paint layer, the result of the loss of adhesion between the ground layer and the canvas, and responsible for the most recent losses, such as the one in the shadow of Christ’s face, beneath his left eye. The elements relative to the technique of execution derived from the diagnostic campaign can be compared with what has emerged from similar diagnostic campaigns conducted on numerous paintings by Caravaggio.²

As far as the support is concerned, it should be noted that a plain, canvas weave with 12 × 13 threads per/cm2, albeit with var- iations in density, is the one most used by Caravaggio during his time in Rome. The use of this type of canvas becomes almost exclusive during the Maltese, Sicilian and in the second Neapolitan periods, while for the large altarpieces of the first Neapolitan sojourn, the artist seems to have preferred twill-weave canvases. With regard to the width of the piece of canvas used, it should be noted that in Rome, Naples and Malta Caravaggio used canvases some of which were extensive in width, while during his Si cilian sojourn, with the exception of the lost Palermo Nativity (which scholars today tend to anticipate to the artist’s late Roman phase), the maximum width of the canvas available to the painter seems not to have exceeded 95 cm. The width of the canvas used for the Ecce Homo thus appears to be compatible with those of the Roman paintings, as well as with those painted in Malta, Sicily and during the artist’s second Neapolitan period.³

The dark ground, applied in several layers, is a recurring characteristic in Caravaggio’s paintings and is applied in this manner in order to achieve exactly the colour and tonality that would allow the artist to leave it directly visible in the shadows, using the a risparmio (in reserve) technique. The granulometry of the preparatory layers is usually quite fine in paintings executed before the flight from Rome in 1606, and generally becomes coarser later. Among the paintings executed in Rome on dark grounds, made up of several layers, and with a granulometry similar to that of the first layer applied to the canvas depicting Ecce Homo, is Judith with the Head of Holophernes now in Palazzo Barberini.⁴ Among his later works, similar conditions are to be found in the Seven Acts of Mercy in the Pio Monte di Misericordia in Naples,⁵ the Saint Jerome in Malta,⁶ the Adoration of the Shepherds in Messina⁷ and the Martyrdom of Saint Ursula from his second stay in Naples.⁸ The appearance, colour and composition of the second ground layer in the Ecce Homo finds affinities with one of the preparatory layers of the Saint John the Baptist in Palazzo Corsini (Gallerie Nazionali di Arte Antica)⁹ and with a layer, possibly a ground layer, of the aforementioned Saint Jerome in Malta.

The practice of leaving the ground layer visible along the contours between areas of colour or as mid-tone in the flesh tones and draperies, generally referred to as the a risparmio (in reserve) technique, and used in this work at least in part in the shadows of the faces and hands, already appears in Caravaggio’s early works, and is a technique he would then use increasingly extensively as his work progressed. In essence, from an extremely rarified use of this technique in the paintings he produced up to 1599, the artist began to extend its application in the lateral paintings for the Contarelli Chapel, and maintained a fundamental but still restrained use of it throughout the Roman period, finally resorting to it in an increasingly unrestrained manner during his Maltese and Sicilian sojourns and in the final Neapolitan period. In the shadows of the figures in the Ecce Homo, the ground is often glazed, either to cool its tonality or to achieve the darkest shadows, according to practice not unfamiliar to Caravaggio, especially in the Roman paintings and in any case for those prior to his Sicilian sojourn, when his technique tends to make extreme use of the colour of the ground, limiting the superpositions to the few brush-strokes required for the rendering of the deepest shadows.¹⁰

Turning now to the laying-in of the composition, the coexistence of dark brush-drawing, incisions and abbozzo as found in the Ecce Homo is anything but extraneous to Caravaggio’s practice.11 In the course of his production, the place given to brush-drawing is reduced from the moment that he begins to systematically paint on a dark ground, but the use of the brush (and in some cases charcoal) will be maintained to define certain details right up to his later works, including those executed in Sicily and his second sojourn in Naples. The abbozzo brush-strokes used at times to mark the outlines and more frequently the areas of maximum light incidence, seem to be absent in the early works, but they gradually assume a more prominent role as early as the lateral paintings in the Contarelli Chapel. Examples of zig-zag sketching brushwork similar to that found in our painting can be seen, for example, on the arm of the angel of the Sacrifice of Isaac in the Uffizi,¹² in the Penitent Saint Jerome in Montserrat¹³ or in the Saint Jerome writing in the Galleria Borghese.¹⁴

As far as the incisions are concerned, the significant development in their compositional role coincides with the turn of the century, and their use as a tool for the graphic laying in of the composition is almost entirely confined to the artist’s mature Roman production, for reasons that have over the years been linked to Caravaggio’s extensive use of posed models and/or a particular configuration of the studio in which he worked. There is no doubt that their role diminished after his flight from Rome and, although he occasionally returned to their not inconsiderable use, perhaps only for certain figures, the incisions no longer held the fundamental role they had had in Rome. The most striking similarities for the role and extent of the use of incisions identified in the Ecce Homo seems precisely to be those found in the paintings executed during the Roman period.

In this regard, one should note the use of incised strokes to mark the axis of the eyes in the figure of Isaac in the Sacrifice in the Uffizi,¹⁵ in one of the executioners of the Crucifixion of Saint Peter in Santa Maria del Popolo,¹⁶ in the Child of the Madonna dei Palafrenieri¹⁷ and in the Saint Jerome writing in the Galleria Borghese.¹⁸ The expedient of outlining the contours of the nose and the ovals of the faces with broken strokes can instead be seen in the Holophernes of the Judith in Palazzo Barberini,¹⁹ in the aforementioned face of Isaac in the Sacrifice in the Uffizi and in the Saint John the Baptist of Kansas City,²⁰ up to the almost identical way of marking the boundary between light and shade in the knuckles on the back of the hand, found in the left hand of the soldier behind Christ, and observable in the construction of the left hand of the Saint Jerome writing in the Galleria Borghese.²¹

After the flight from Rome, a similar way of tracing the eyes and contours can probably only be found in the figure of the presumed donor in the Madonna of the Rosary which is now in Vienna.²² The characteristic way of scratching through the fresh paint film, producing incisions that reveal the ground or underlaying layers, is also particularly frequent in Roman works. These are observable for instance, in the chips and teeth of the comb on the table of Martha and Mary Magdalene in the painting now in Detroit,²³ the pilgrim’s hair on the bonnet in the Madonna of Loreto²⁴ or those on the neck of the soldier on the far left of the Conversion of Saulo in the Odescalchi collection.²⁵ The palette, based on the use of a rather limited range of pigments and without complex mixtures to obtain particular hues, is consistent with what is found in Caravaggio’s works painted on a dark ground.