Giovanni Girolamo Savoldo  c. 1480– c. 1548
NG 1031 
Mary Magdalene

Nicholas Penny, 2004

Extracted from:
Nicholas Penny, The Sixteenth Century Italian Paintings, Volume I: Paintings from Bergamo, Brescia and Cremona (London: National Gallery Company and Yale University Press, 2004).

c. 1535–40

Oil on canvas, 89.1 × 82.4 cm

Support

The dimensions given above are those of the stretcher. The canvas is of medium‐weight tabby weave, lined onto a tabby‐weave canvas of similar character. The stretcher is modern, of stained pine and without crossbars. The canvas is not rectangular, so some of the lining canvas is apparent along the upper right side. Although slightly trimmed, the canvas retains its original black painted border (now approximately 1.5 cm wide). It also retains tacking holes, a few of which may be original. There is pronounced cusping at the upper and lower edges.

Materials and Technique

The ground, which is a pinkish‐beige colour, has been analysed and found to consist of lead white with lead‐tin yellow, yellow earth and a small quantity of red lead and black pigment.¹ X‐radiography reveals that the mantle has been painted over areas originally intended as sky. It is indeed now apparent to the naked eye that clouds extend under it. A small adjustment was made to the left side of the Magdalen’s left hand. Arches seem to have been projected, but only sketchily, in the upper left side of the painting. The white vase was painted after the wall, since the division in the stone frame of the niche shows through it. The last fold of drapery on the right was also painted over the bricks. The silver‐grey of the shawl is painted with lamp black and lead white.

Some segmental lines of dark underdrawing are visible in the brick of the wall to the left, above the niche. As noted below, this thick brushed line of underdrawing is now also evident in the face. Its character is clearly apparent in infrared photography (fig. 1).

Conservation

The painting was cleaned lightly by Luigi Cavenaghi (1844–1918) under the direction of his master, Giuseppe Bertini, in Milan in November 1877. A note of 26 November from Bertini notified Frederic Burton, who was then Director of the Gallery, that Cavenaghi had merely removed some insignificant spots or marks (‘alcune macchie inconcludenti’), and assured him that for the rest the painting could be guaranteed to be in excellent health and to retain its original patina (‘lo posso garantire sanissimo ed averla ancora la sua patina originale’).² This is one of the earliest references in the Gallery’s archives to Cavenaghi, who soon afterwards became one of the most esteemed restorers in Europe, trusted even by the opponents of restoration.³

Burton may not have been impressed by the oxymoron of ‘original patina’. In any case, the picture was cleaned again after acquisition. The surface was ‘polished’ in January 1952. It was relined and cleaned in October 1960, and surface‐cleaned and revarnished in August 1994.

Condition

The surface is worn, probably from old abrasive cleaning. There are a few minor losses, one in the render of the wall to the right of the vase and some in Mary Magdalene’s left hand. Because of this, and because of the increased transparency of the paint, the underdrawing shows through with broad black lines outlining the eyelids and the ‘bags’ under the eyes as well as the lips and chin, and a line for the edge of the shawl, lower than it was actually painted.⁴ The effect of this was reduced, and the retouchings of the losses on the left hand made less obvious, when the painting was surface‐cleaned and revarnished in 1994. It is reasonable to suppose that the hair was originally more apparent in the shadow of the shawl. Also the forms of the ruin – its six tall arched recesses, its mingling of stone and old brick, its large trabeated opening crossed by a tree, with a stone arched opening beyond seen through the arch on the right – must all have once been easier to see. The greens of the trees and weeds in this area must also have darkened.

Attribution

The painting was recorded in the Avogadro and Fenaroli collections in Brescia from the mid‐eighteenth to the mid‐nineteenth century as by Titian.⁵ Otto Mündler in 1857 and Charles Eastlake in 1862 both recognised its relationship with the painting signed by Savoldo in the Berlin Museum;⁶ Cavalcaselle published it as an autograph variant in 1871,⁷ and it has been accepted as such ever since.

Subject and Interpretation

A woman with a shawl of white satin covering her head, shoulders and arms turns towards us. The crimson pleats of a full skirt are just apparent below the shawl. Her right hand is raised but completely concealed in the shawl. Her left hand presses under her right elbow. She may have been weeping against her covered hand. Behind her are ruined, perhaps sepulchral, buildings, and beside her there is a niche in a wall. In front of this niche, upon a ledge, there is a small vase of alabaster, or perhaps porcelain. Beyond the wall, dawn is depicted breaking over the Venetian lagoon. To the left on the near bank is a façade of a church with a campanile. Towers and domes animate the horizon.

The vase suggests that Mary Magdalene is intended, but this saint is usually represented showing her hair, which she employed to dry Christ’s feet, and in a state of undress sufficient to reveal her breasts. When this painting was considered to be by Titian it was sometimes known as ‘La Zingara’, the gypsy woman. Eastlake simply called it the ‘veiled woman’.⁸ The suggestion that it represented the Magdalen certainly had been made, but Mündler put this title as well as the name of Titian in inverted commas.⁹ One reason for his reluctance to accept the title without reservation must have been that the more famous, and signed, version of the painting in Berlin, which does not include a vase, was merely known as ‘La Veneziana’. Cavalcaselle in his draft notes of c. 1869 for his History of Painting in North Italy, although aware of the vase and the possibility that it was Savoldo’s painting of Mary Magdalene described by Ridolfi, still thought the title of ‘Zingara’ better suited to such a ‘figura di genere piacevole e piena di cochetteria’ (‘a subject from life, delightful and full of coquetry’).¹⁰ Burton at first called it a young woman in a white mantle, but then, noting its correspondence with the painting described as a Magdalen by Ridolfi, called it ‘Mary Magdalene approaching the Sepulchre’.¹¹

Some scholars have suggested that the painting is not so much an image of the Magdalen as a portrait of a Venetian courtesan indicating her penitence, or at least her profession, by means of the attribute. It has even been proposed that the yellow shawl in other versions of the painting should be taken as the garment which Venetian law ordained for prostitutes,¹² but the material in the paintings is a rich one (apparently satin) and the existence of one version in which the shawl is white provides a further obstacle to this theory.

On the other hand, as Michael Hirst has pointed out to me, support for the idea that a courtesan is intended is provided by one of the wood engravings cut by Cristoforo Guerra for Cesare Vecellio’s book of costume (fig. 2). This shows a courtesan ‘al tempo di Pio Quinto’, that is, during the pontificate of Pius V, wearing a ‘mezo velo’, that is, a shawl of similar length to that in Savoldo’s painting, covering ‘tutta la fronte’, all the forehead.¹³ There are some problems with this: Vecellio seems to be talking about the courtesans of Rome, not his native Venice; Pius V reigned from 1566 to 1572, some considerable time after Savoldo’s death; the shawl is said to have been white (not white or yellow), and it is ‘di cambrai’, of cambric, which is a very fine linen, rather than satin. However, the similarity between the images is very compelling.

Cecil Gould noted that the biblical source was the Gospel of Saint John, which alone describes Mary Magdalene as visiting the tomb unaccompanied (20:1): ‘The first day of the week cometh Mary Magdalene early, when it was yet dark, unto the sepulchre…’¹⁴ And that this was the subject was implied by the title given to the painting in all National Gallery catalogues before this one: ‘Saint Mary Magdalene approaching the sepulchre’. However, recently a persuasive case has been made by Mary Pardo that the painting is a highly original dramatic ‘close‐up’ of the meeting with Christ which follows Mary’s discovery of the empty tomb (John 20:14–16).¹⁵ Mary mistakes Christ for a gardener. He addresses her, ‘Woman, why weepest thou, whom seekest thou?’ She replies that if he has removed the body she wishes to know where it is. He then simply addresses her by her name and she knows him. ‘Jesus saith to her: Mary. She, turning, saith to him, Rabboni [which is to say, Master].’ In favour of this idea is the fact that the woman seems to have been weeping. The raised hand under the shawl may be supposed to have been pressed against her eyes – a gesture used by other artists at that date and earlier.¹⁶

Pardo also observes – brilliantly – that Mary appears to be illuminated as well as enlightened by the person towards whom she turns. She goes on to propose that Savoldo expected us to call to mind the use of the veil by the ancient Greek artist Timanthes to conceal the dismay on the face of Menelaus at the sacrifice of his daughter Iphigenia, a device praised by Pliny the Younger and often cited by Renaissance writers on [page 349][page 350]art.¹⁷ Here, by contrast, ‘the veil is lifted to reveal the passing of sorrow, yet continues to challenge us to look beyond’ and the painting’s subject thus becomes the nature of painting.¹⁸ This seems over‐ingenious.

If we accept the painting as a narrative (the same type of painting, that is, as Melone’s Christ carrying the Cross also catalogued here; see pp. 132–5), then it must be said that the woman’s expression does not – in any of the versions – suggest either the ecstatic acknowledgement or the deep grief that would be appropriate. And yet we only have to consider Savoldo’s Saint Jerome (NG 3092) to realise that he was a great painter of expression. It may be that more evidence of sorrow was originally present – this is perhaps suggested by the strong lines brushed below the eyes in the underdrawing. When the Berlin version was restored in 1989, the woman’s face was found to be considerably repainted.¹⁹ Had the woman looked as if she had been grieving, Crowe would not perhaps have found his encounter with her so erotic:The hour is sunset. The shape is that of a woman, young, warm, and impulsive. Her form half hid in a mantilla glides round the corner of a ruin. The light just tips her nose and leaves the rest of the face in gloom … the furtive archness of the glance, the twilight in which the scene is shrouded are full of mystery.²⁰

Nevertheless, something like ‘furtive archness’ may be in the other paintings and in the eighteenth‐century print, and it is impossible to believe that the expression was originally completely different. No doubt the painting refers to Mary Magdalene’s recognition of Christ, but it may be that it represents a woman casting herself in this role.

Two other points about the image should be mentioned. Creighton Gilbert noted that a choral figure with the same pose features in paintings which appear to be copied from the composition of The Continence of Scipio by Savoldo, which Michiel saw in Odoni’s collection in 1532.²¹ This might be an insertion by a copyist or a case of Savoldo trying to reuse an invention – more likely the former since it is very awkward. Frangi reports the observation of Marco Magnifico that the folds of the drapery around the raised hand form a mask. I do not agree that this is ‘certainly not accidental’.²²

Versions and Dating

Four versions of the painting are known, all of which have variations in setting and drapery folds and none of which is of a quality below that of Savoldo’s signed or documented work.²³ In addition, an engraving appears to record a fifth.

The version in the Gemäldegalerie, Berlin, measures 94.2 × 75.3 cm (fig. 3). It differs from the others in that it includes considerably more of the figure, gives more of an idea of her movement and also provides an architectural setting which closely relates to the figure’s pose. This painting does not include an ointment jar. It is the only version to be inscribed with Savoldo’s name.

Close to the Berlin version but larger – its dimensions are given as ‘alto palmi Rom.6. e once 6⅔. Largo palmi 4’, which would make it about 146 × 90 cm – was a painting which is now untraced but which was engraved by Lorenzo Lorenzi after a drawing by Giuseppe Zocchi (who died in 1767) when it was in the Gerini Collection in Florence in the eighteenth century (see fig. 4). It had an attribution to Veronese²⁴ (perhaps suggested by the figure’s similarity with a mourning woman in Veronese’s small Louvre Crucifixion²⁵). This version does include an ointment jar.

The three other painted versions are in the Contini‐Bonacossi Collection (formerly in the Giovanelli Collection, currently in Palazzo Pitti, Florence; fig. 5), in the J. Paul Getty Museum, Los Angeles (formerly with Koetser in Zürich and before that in Warwick Castle; fig. 6), and in the National Gallery. The first of these measures 83 × 76 cm and the second 99.7 × 76.2 cm. In these three the format and the figures are very similar, but the Getty picture has a blue sky like that in Berlin; whereas the Contini‐Bonacossi and London paintings are both given dawn settings. The London painting is the only one with a silver‐grey shawl. It would seem likely that the National Gallery and the Contini‐Bonacossi paintings were painted concurrently, or at least in close succession, for they are closest in design and detail. For example, unlike the Getty picture, the outlines of the shawl are curved against the sky and there is a small but telling glimpse of red cuff by the left hand. The Contini‐Bonacossi painting appears to be in better condition than all the other versions. The way that the brilliant yellow highlights of the drapery contrast with the brown and mauve in its shadows and the echo of these colours in the sky is a startling effect unmatched in the others – although the relationship between clouds and reflections in the distance and the lights in the drapery of the National Gallery’s painting are also remarkable and have considerable metaphorical resonance. Adolfo Venturi in a famous passage felt that the shawl itself resembled water stirred in the breeze as it reflected the moon.²⁶

Scholars have dated these paintings between about 1527 and ‘after 1540’. Boschetto followed Longhi in putting the National Gallery version early; Longhi also gave the Berlin painting a relatively late date.²⁷ Gilbert argued the opposite.²⁸ In support of the idea that the Berlin painting is later is the more graceful drapery, the greater sense of movement and the fact that the wall on the left in that painting was originally lower and thus once similar to that in the other versions.²⁹ In support of the idea that the London and Contini‐Bonacossi pictures are the last in the series is the stronger argument that the composition, which gains in impact from being cropped, also gains in drama from a setting at dawn and in realism from the painting of the drapery.

The paintings with a dawn setting are likely to be the last in the series: indeed it is hard to imagine the artist afterwards preferring a daytime setting. In the Getty, Contini‐Bonacossi and National Gallery paintings the shawl includes a seam [page 352]which immediately catches the eye. It emphasises the volume of the figure, and the puckering of the heavy fabric introduces a note of arresting familiarity into the poetic discourse. We feel that this was painted not just from drawings of drapery but from an actual sample of cloth. The white loincloth of Christ and the silk drapery held by Mary Magdalene in Savoldo’s Lamentation in the University Museum at Berkeley, California – almost certainly a later work – give something of the same impression. The more fluent rhythms and softer folds of the Berlin Magdalen are more reminiscent of the draperies worn by the angels in the Pesaro altarpiece, datable to the mid‐1520s (fig. 1, p. 337).³⁰

Previous Owners

In 1620 Ottavio Rossi described a painting by Savoldo in the house of ‘Doctor Lorenzo Averoldo’ in Brescia of ‘una bellissima Maddalena coperta da un pan bianco’, and Ridolfi in 1648 noted that this picture ‘in casa Averolda’ was ‘celebre’.³¹ The same painting has been found in the posthumous inventory of the possessions of Lorenzo’s father, Fausto, in 1611.³² He was a grandson of the Giovan Paolo Averoldi who commissioned a painting of Saint Jerome (perhaps NG 3092, catalogued here) from Savoldo in 1527. No other version of the painting has white (or silver) drapery, so it is nearly certain that this painting is the one now in the National Gallery.³³ When seen in the Fenaroli palace in Brescia in 1857 by Otto Mündler, who recognised it as related to the Savoldo in Berlin, it was noted by him as ‘A so‐called “Magdalen” by “Titian” ’.³⁴ It must therefore be the Magdalen believed to be by Titian which Odorici recorded in this collection in 1853,³⁵ and since the Fenaroli paintings come from the Avogadro Collection (see pp. 356–8) it must be the Magdalen by Titian recorded by Carboni in the Avogadro Palace in 1760.³⁶ It is not yet clear how or when the painting was acquired from the Averoldi by the Avogadri.

Acquisition by the Gallery

Otto Mündler, when he saw the painting in the Fenaroli Collection, considered it a repetition of the ‘original’ in Berlin so perhaps did not even regard it as an autograph work.³⁷ Eastlake five years later, on 10 October 1862 when he visited the collection, noted ‘veiled woman, Savoldo. The best at Berlin’.³⁸ This might be taken to imply that it was autograph, but he does not seem to have been greatly tempted by the picture. Giuseppe Baslini, the Milanese dealer (of Via Monte Napoleone II), extracted the picture, together with a ‘lotto di diversi oggetti’ including pictures of lesser merit, from Conte Fenaroli in the late summer of 1877 and then wrote to Frederic Burton from Merate on 16 September, mentioning it as a painting in which the previous director, William Boxall, had taken a keen interest. On 3 November he added more details and the price (£350) and on 5 November his nephew sent a photograph, which Burton acknowledged on 22 November in Italian: ‘Mi piace il quadro come egli si presenta nella fotografia, sebbene non tanto che quello bello (segnata) di Berlino, dello stesso soggetto, dove il modo di fare è più largo, i panni più graziosi, e il poso più fino’ (‘I like the painting as it appears in the photograph, although not as much as that beautiful (signed) one in Berlin, of the same subject, where the execution is grander, the draperies more gracious and the attitude more poised’). Burton agreed that it was a ‘bel pezzo’ (‘a fine thing’) and certainly no mere replica.³⁹ On 26 November Burton heard from Giuseppe Bertini that the picture had been restored by his pupil Cavenaghi (see above under Conservation). On the same day the Trustees agreed to let the Director acquire the picture if he saw fit,⁴⁰ and an agreement to make the purchase was arrived at by Burton in Milan on 19 December. The receipt for £350 is dated 6 March 1878, the delay being occasioned by Baslini’s domestic difficulties (his sister was on her deathbed and his wife in childbed).⁴¹ The money was found on 13 February 1878 by a ‘special grant out of the Civil Service Contingencies fund’.⁴²

Provenance

See above. Almost certainly with Fausto Averoldo in Brescia by 1611. Certainly in Palazzo Avogadro by 1760. By descent to the Counts Fenaroli. Sold to Baslini by September 1877 and by him to the National Gallery on 6 March 1878.

Exhibition

1990 Brescia, Monastero di S. Giulia, and Frankfurt am Main Frankfurt‐am‐Main , Schirn Kunsthalle (I.19).

Framing

The painting is in a carved and gilt frame with a bold reverse profile. The broad hollow is decorated with widely spaced flutes, the borders of which are interlinked. There is a narrow laurel leaf moulding on the inside of the hollow and a twisted ribbon on the outside. The ogee moulding at the outer edge is carved with a stylised leaf. This was one of a group of frames purchased for £500 on 13 March 1935 from Dr Kurt Cassirer, who described it as a Florentine sixteenth‐century frame. It may be a Tuscan seventeenth‐century frame but is more likely to be an imitation of about 1880. A rather similar frame is in the Galleria d’Arte Moderna, Florence.⁴³

Notes