FR  |  EN

Rossella VODRET

Former director of the Galleria Nazionale d'Arte Antica, Palazzo Barberini, Rome

Report written by Rossella VODRET

Comparative study between the Judith and Holofernes from Toulouse and Caravaggio’s stylistic techniques

Thanks to the diagnostic analysis campaigns that were done from 2009 to 2012 on the twenty-two works by Caravaggio that are still held in Rome and on thirteen of the workslent to the exhibition Dentro Caravaggio, we now have comparable technical surveys available for thirty-five paintings by Caravaggio spread out through his career; that is to say about half of those that can be reliably attributed to him. On the basis of the studies conducted so far, it is possible to establish some comparisons between the painting technique used for Judith and that of the great Lombard master. These comparisons show how in the ensemble of the painting there are a number of elements of Caravaggio’s technique known to date.  

The twill weave canvas that comprises the upper part of the painting of Judith is used by Caravaggio in all of the known works that date from his first Neapolitan stay. The brown primer or ground, containing large grains of calcium carbonate, which give the surface a granular quality the better to make the light vibrate, is similar in color and composition to that of other works by Caravaggio. We can give as an example the same ground that we find in the David with the head of Goliath at the Galleria Borghese.  

The wide outline –reinforced by a finer subsequent additional line—which delineates the figures of Holofernes and the old servant, used to define the placement of the two figures and to enhance the contrast with the background as well as the painting’s three-dimensionality, is an element which can be found in various works by Caravaggio from the late Roman period as well as the first Neapolitan period. By way of example, this is visible in the Madonna and Child with Saint Anne (Dei Palafrenieri) in the Galleria Borghese and The Crowning with Thorns from the Kunsthistorisches Museum in Vienna.  

The delicate brush-stroke drawing is widely used by Caravaggio on the light ground that he uses in his early works.  As the preparatory ground becomes darker, the paintbrush drawing—becoming less visible—is lessened (but does not disappear), and is combined with the sketched and incised lines. Even in our picture, some lines of preliminary brushstrokes are visible on the head and the hands of the servant, as well as in Holofernes’ hands and in the outline of Judith’s face.  

In the canvas from Toulouse, the under drawing executed with a brush is there to a great extent, indicated by lighter or reddish brushstrokes, used by Caravaggio to set his composition on the darker ground. Sketched lines that are similar to those used by Caravaggio are to be found in the three figures of the Toulouse painting: a clear sketch relative to the initial position of Judith’s left eye has been found between the eyebrow and the eye, while reddish sketches can be seen on the beautiful hands of the servant as well as on Holofernes’ lips, cheekbone and ear. Analogous reddish lines can be found for instance on The Crucifixion of Saint Andrew in Cleveland, a work that refers to the painter’s first Neapolitan stay, and on the Flagellation of Christ of Capodimonte.  

There are numerous incised lines, drawn into the ground when it is more or less fresh, that define certain parts of the composition. These incised lines are visible only on the figures of Judith and Holofernes. There are two on Judith’s hand, one on her left eyebrow and another on her forehead to mark the start of her hairline. A wider one runs the length of Holofernes’ right arm. There are no such incised lines visible on the old servant (graphic of these incised lines). These are mainly short lines, similar in number and in size to those that Caravaggio used after fleeing Rome.  

The use of profili a risparmio, a technique of leaving the darker ground visible (or only slightly veiled) to delineate the areas of color and to construct the shadows, is widely used in this painting. This is a practice already used by Caravaggio in the last years of the Cinquecento, which has been finely honed by the time we find it used for the canvases of the Contarelli chapel of Saint-Louis-des-Français. The ground a risparmio is then constantly used in further works, to such an extreme in The Martyr of Saint Ursula that the painted surfaces come close to being dominated by the use of the exposed ground. In the canvas from Toulouse, we find that the ground is used for outlines as well as rendering shadows: in the shadowing of the eyes, nose and chin of Judith, in her hand, in Holofernes’ beard and mustache as well as in a detail of the old servant’s headdress. However there is no trace of the dark ground being used in the face of the servant where the shadows of the face as well as the wrinkles are rendered using paint glazes over a clear ground that had possibly already dried.  

In light of the above, it is clear that the most important of the characteristics that epitomize Caravaggio’s technique are present in this painting. If the appropriation of some of these traits that are specific to Caravaggio is known to have been done by some of his followers (above all the use of the incised line) at this point in the research to have so many of the same aspects of the artist’s technique has only been found in original works by the genius from Lombardy.  

The modifications of the composition

The most important technical fact that the diagnostic analysis brought out is that of the important number of modifications done to the compositional structure of the painting – not small adjustments, but substantial changes to the composition – which demonstrates that the Judith from Toulouse is an original painting and not a copy. In fact, by definition pentimenti is only found in original works. Thus, in my opinion, the theory of some scholars suggesting that the Judith from Toulouse is a copy of a lost original by Caravaggio, referred to in a letter from Pourbus (written on 25 September 1607, a little more than two months after the painter fled to Malta) has to be dismissed.  

As proven by the infrared reflectography and the X-Radiography, the most important compositional modifications concern the three figures that appear in the painting: Judith, Holofernes and the elderly servant.  

Judith: she is the key figure of the whole painting; her stylistic quality is the one closest to that of Caravaggio. The most significant pentimento concerns Judith’s eyes, which initially did not look towards the spectator but were turned downwards toward Holofernes, exactly like the Barberini Judith. Other interesting comparisons with the Roman masterpiece can be found in some morphological elements such as the three-quarter profile of her magnificent face: the relationship between the nose, the eyes and the ears of the Judith Barberini are very close, almost identical to that of the Toulouse Judith. 

Beyond the obvious stylistic differences, due to the time frame that separates them, the two faces coincide even in the way they were executed, for example the slight double-chin—illuminated by a beam of light—the lighting of Judith’s nose and lips, the position of her mouth, the reflection of light on her eyelids, the shadow between her chin and her lower lip and the light on her neck. Some differences can be noted, however, on the left side of the face of the Judith from Toulouse, which is rounder and illuminated differently compared to the Barberini Judith, whose lips are fuller. Such a coincidence is quite strange and presupposes, in my opinion, that the author of the painting from Toulouse had intimate knowledge of the Roman Judith painted for Ottavio Costa probably in 1602, given that he paints here the same eyes and their same expression. This detail is difficult to explain if we hypothesize that the Judith of Toulouse had been painted by an artist other than Caravaggio; especially given the particular attention that the banker Costa bestowed upon the painting, the masterpiece of his collection, which he kept hidden behind a curtain.  

Holofernes: the X-Radiography sheds light on various modifications done to his left hand. All of the fingers have been shortened, corrected and the position changed (fig. 6a and 6b). A modification very similar to the one we find in the fingers of Maria Cleofa in Caravaggio’s Deposition, today in the Vatican Museum. Another significant compositional modification is in the thumb of Holofernes left hand, which was originally much nearer to his thorax. This first thumb was subsequently taken out by integrating it into the figure’s thorax and covering it with a luminous brushstroke.  

The old servant: she is the most tormented and the most modified figure. There are various modifications, notably to the eyes, the hands, the veil and the sack. The most important one is to the eyes, which in the first version were truly impressive: completely wide open, reflecting the true aspect of thyroid disease and goiter. This detail was then reduced to the point of making it disappear. Other modifications concern the placement of the figure: it can be clearly seen in the X-Radiography that the servant was “invading” the space reserved for Judith with her shoulder and left arm. The unusually predominant role given to the central figure of the servant was, perhaps at a later moment, partially scaled down by the enlarging of the dark veil on Judith’s head and her left sleeve, which hide the servant’s shoulder and left arm. The enlarging of Judith’s veil and left sleeve is clearly visible because of its opacity (the red line delineates the part that was there before), and by the interruption of the original brushstrokes that gave light to the sleeve. This correction has in part re-balanced the spatial relationship between the two female figures but it remains confused and unresolved.  

The surprising and insistent pattern of wrinkles that is the stylistic and compositional element is the most alien to the rest of the painting and to the style of Caravaggio. The analyses not only show that the light colored wrinkles were rendered in a different color (pigment and binder) from that of the paint layer underneath, but that they are of a different composition; so much so that during the stratigraphy, the colored surface of the sample taken from the wrinkles cleanly detached itself from the one underneath. This leads us to believe that the wrinkles were painted at a later moment over a light colored base that had already dried. This also explains why, unlike the other two figures, there are no sketches or incised lines or use of the darker ground as outline or shadow on the servant’s face, which is logical given that the underlying paint layer there is lighter in tone

Conclusions on the Judith and Holofernes from Toulouse

From all of these elements, we can conclude that the figures of Judith and Holofernes were painted using the most significant aspects of Caravaggio’s technique (type of canvas, the ground, incised lines, sketches, outlines, under-drawings, use of the ground a risparmio in the shadows…etc). It would be strange to have all these elements together in the same painting if the work had not been painted by Caravaggio.

It still has to be considered that the work seems to have been “corrected” or completed by the enlargement of Judith’s dress and veil and by the wrinkles on the old servant, which are foreign to the stylistic practice of Caravaggio.  

My opinion is that Caravaggio, in leaving Naples for Malta (June-July 1607) perhaps left the Madonna of the Rosary and the Judith to be sold with Finson and Vinck, with whom he was quite evidently in some kind of relationship (he used their bottega in Naples). This would explain why, only two months later in September 1607, the two paintings were up for sale in Naples, where Pourbus saw them and gave a detailed description to Vincenzo Gonzaga. According to this hypothesis, it is possible that following Caravaggio’s departure for Malta, the Judith might have been “corrected” by another painter (Finson?) to redesign the space between the two female figures, to add the dense pattern of wrinkles to the old servant and maybe even to mitigate certain difficult aspects such as her wide-open eyes.  

The author of the “corrections” made to the Judith from Toulouse might well be the author of the modest copy in Naples, a painter of inferior quality, who in my opinion could be Louis Finson. The Neapolitan copy, although largely lacking the stylistic characteristics and above all the painterly technique of Caravaggio, seems to have closely followed the making of the Toulouse painting, as if the copyist had had the opportunity to see it while it was being painted and was thus able to repeat some of its technical details, that then became “invisible”, covered by the final paint layer (the size and the type of the canvas, the type of ground used, the dark outlines around the figures, the wide-open eyes of the old servant, etc). These details are too precise for them to be simple coincidence. A possible explanation is that the two Judiths (Toulouse and Naples) were painted simultaneously from two different painters in the same studio, probably Finson’s studio in Naples, which Caravaggio might have had at his disposition during his first Neapolitan stay. The provenance from the same studio might explain the use of the same strange type of canvas sewn together and used to paint the two paintings17. We can wonder if Caravaggio and Finson had come to an agreement by which Finson would have gotten authorization to copy the precious originals in exchange for allowing Caravaggio to have the studio at his disposal? This hypothesis could explain the existence of so many copies done by Finson of the works of the great Lombard painter.  

Sell in: