Anthony van Dyck (Antwerp 1599 – 1641 London)

Genoa circa 1621-25

The Virgin and Child

Oil on canvas (relined), 164 x 114 cm; 5 ft. 4 9/16 x 3 ft. 8 7/8 in

 I. DESCRIPTION

            Against a background of plus red velvet drapes with pompoms, the Virgin is attired in a silvery blue and violet taffeta overdress with golden sleeves. She holds the Child Jesus between her hands with a white cotton or linen fabric as he stands on a deep crimson tablecloth bordered with rich embroidery.

● PROVENANCE:

- Genoa circa 1621-1625 (Lebel documentation: black and white photograph, Sanchis, Valencia, before 1950);

- probably Jorge Diaz Martinez, Valencia, 1868[i];

- Mr. Lombard, Valencia, circa 1870 and by descent until the 1990s-2000s;

- Art market, Valencia, 2000;

- private collection since July 2001.

● RELATED WORKS:

- Antoine van Dyck, copy, Virgin and Child, oil on copper, 17 x 22 cm (Van Dyck en Espana, Matthias Diaz Padron, 2012, page 690, cat. A7) (Fig. 1)

- Antoine Van Dyck, The standing infant Jesus, black stone heightened with white chalk on paper, 20.3 x 11.1 cm (Boijmans Van Beuningen Museum, Rotterdam, inv. Van Dyck 12) (Fig. 2)

- Antoine van Dyck, Head of the Virgin, black stone and red chalk on paper, 19.5 x 14.3 cm (British Museum, London, Inv. 1847, 0326.14) (Fig. 3)

- Antoine van Dyck, after, Study of the Christ Child and the Hand of the Madonna, pen and brown ink, 18.7 x 13.9 cm (Metropolitan Museum, New York, Inv. 1975.131.177) (Fig. 4)

            The work described, Virgin and Child, to be included in Van Dyck's Italian period, was unpublished and known only through a copy on copper[ii] (Van Dyck en Espana. Matthias Diaz Padron, Prensa Iberica 2012, p. 690, cat. A7.) (Fig. 1). It would be both the largest known format and the source of the series representing this universally known model of the Virgin and Child.

In a distinctly Genoese style, the painting was executed well before the three works on the same subject from Van Dyck's second Antwerp period, now in museums in Cambridge, London, and Baltimore (see Cat, Barnes, De Poorter, Millar, Vey, Yale University Press, 2005)[iii] (Fig. 24, Fig. 25 & Fig. 26).

Fig. 1

Anthony van Dyck, copy

Virgin and Child

Oil on copper, 17 x 22 cm; 8 ¹¹/₁₆ x 6 ¹¹/₁₆ in

Van Dyck in Spain. Padron, Matthias Diaz, Madrid, Prensa Iberica (2012), p. 690, cat. A7

II. THE DRAWINGS

            The drawings The Infant Jesus standing (Boijmans Van Beuningen Museum, Rotterdam, inv. Van Dyck 12) (Vey, 1962, no.136) (Fig. 2) and Head of the Virgin (British Museum, London) (Vey, 1962, no. 135) (Fig. 3) are the only preparatory sources[iv] listed for this composition of the Virgin and Child.

Fig. 2

Anthony van DYCK

The Christ Child Standing
Black chalk with white chalk highlights on paper, 20.3 × 11.1 cm

Boijmans Van Beuningen Museum, Rotterdam, inv. Van Dyck 12

Fig. 3

Anthony van DYCK

C. 1621-27

Head of the Virgin
Black and red chalk on paper, 19.5 × 14.3 cm
British Museum, London, inv. 1847, 0326.14

            In their report of March 31, 2019[v], Dr. Nicholas Easthaugh & Dr. Jilleen Nadolny (Londres, Art Analysis & Research Inc) describe the alignment of the feet of the Infant Jesus in the Rotterdam drawing (fig. 2),  which bears a closer resemblance to our painting than to the three other works in the Fitzwilliam Museum, the Dulwich Picture Gallery and the Walters Art Museum, all painted by Van Dyck during his second Antwerp period. Some excerpts from page 10 of the Art Analysis & Research report :

« From a drawing, now in the Museum Boijmans van Beunigen, Van Dyck shows the child with small, delicate feet, placed parallel so that their bottoms form a single, straight line. In the painted versions, the Child’s feet become broader, more solid. The contrapposto of the pose is increased as well so that the Child stands with his full weight in his proper left foot, his right heel raised. The feet are no longer placed parallel, but rather the pose has become both more solid and more dynamic.

This positioning is visible in the earliest depictions of this composition; the right heel is raised in the earliest of the painted renditions although it shifts in later variants. The current version and the Museo Cerralbo painting are very close in this detail.

In working out the shape of the figure’s proper left foot, the painter has clearly made small adjustments to the toes, trying to place the Child so that the feet and legs are convincingly balanced. An image taken during cleaning reveals a shift in the position of the toes, while in both visible and IR images, the reserve above the foot is clearly visible. »

Van Dyck’s drawing depicting the Head of the Virgin (British Museum, London, inv. 1847, 0326.14) is preparatory for the Virgin and Child kept in the Fitzwilliam Museum, Cambridge, while at the same time recalling the artist’s initial intention in the compositional sketch of our painting, namely with a veil covering the Virgin’s head, as visible in the infrared images of the work (figs. 6 and 8).

Fig. 4 (front)

Anthony van Dyck, after

Study of the Infant Jesus and the hand of the Virgin Mary

Brown ink on paper, 18.7 × 13.9 cm; 7 3/8 x 5 1/2 in

Metropolitan Museum, New York, inv. 1975.131.177

The Virgin and Child leads to a third drawing to be related to the composition and catalogued as after Van Dyck: Study of the Infant Jesus and the Hand of the Virgin Marye, recto[vi], pen and brown ink, 18.7 x 13.9 cm (after Van Dyck,Metropolitan Museum, New York, inv. 1975.131.177) (Fig. 4). On the back a Poem and a sketch of a face possible with the other hand (?) (Fig. 5). The said drawing was classified in 1928 by A. Mos as a work by Peter Paul Rubens (A. Mos, Amsterdam, 7/8-11-1928, no. 520, PP Rubens), but was later correctly reassigned to after Van Dyck[vii].

There is every reason to question this sheet, now kept at the Metropolitan Museum of Art, about which so little is known, for all the obvious reasons that we cannot elaborate on here. How can not imagine a preparatory study? A thorough study of the drawing would likely provide an answer to this question.

Fig. 5 (reverse)

Anthony van Dyck, after

Poem and sketch of a face

Brown ink on paper, 18.7 × 13.9 cm; 7 3/8 x 5 1/2 in

Metropolitan Museum, New York, inv. 1975.131.177

III. SCIENTIFIC ANALYSES, REPENTANCES AND CHANGES IN COMPOSITION

There Virgin and Child was subjected to scientific examinations carried out by two laboratories, CSG Palladio, Vicenza and Art Analysis & Research Inc, London, the latter being considered an essential reference: radiography, infrared reflectography and chemical analyses of the stratigraphic layers, resulting in a final report by Dr Nicholas Eastaugh and Dr Jilleen Nadolny (Art Analysis & Research Inc, London) dated March 31, 2019.

As early as 2004, the first samples were taken from the painting's paint layer and studied by the CSG Palladio laboratory, and others were subsequently taken in 2013[viii] and 2019[ix] by Art Analysis & Research Inc, revealing materials and techniques consistent with many of Van Dyck's works painted in Italy, but also with others painted later in Antwerp and London; the analyses identified a double layer of brownish-orange preparation widely used by Genoese painters[x] in the 17th century with a top layer of grey colour called imprimatura. These techniques, practiced by Van Dyck in Italy, can be found in the Portrait of George Gage probably painted in Rome (National Gallery, London, inv. NG49) as well as in the Portrait of a Woman and Her Child painted in Antwerp or Genoa circa 1620-21 (National Gallery, London, inv. NG3011) (Fig. 22). Painted in Genoa, the Portrait of the Giustiniani Longo Children (National Gallery, London, inv. NG6502) (Fig. 20) on the other hand, feature a single ground layer in an intense brown color.

All the information described above is clearly presented in the Technical Bulletin N° 20 of the National Gallery in London[xi] (Painting in Antwerp and London: Rubens and Van Dyck, Ashok Roy, Yale University Press, 1999, pages 56-63). Aside from the works cited above, the combination of a brownish-orange ground layer with a gray upper layer has also been identified in other works by Van Dyck painted after his stay in Italy: Lady Elizabeth Thimbely and Dorothy, Viscountess of Andover, circa 1637, London (National Gallery, London, inv. NG 6437); Equestrian portrait of Charles I, 1637, London (National Gallery, London, inv. NG 1172).

The infrared images (Fig. 6) revealed no fewer than four main pentimenti, as well of a number of stages of development:

-         The most significant pentimento is that of the Virgin's head. In the first intention, the artist initially painted a veil instead of the hair, clearly visible in the infrared image (Fig. 8) (Ingas camera, Art Analysis & Research Inc). This pentimento is, moreover, similar to that in the Virgin and Child at the Cerralbo Museum  (Fig. 9), a work published in 2017, which could represent the artist’s original conception. The initial veil present in both compositions corresponds well to the version in the Fitzwilliam Museum in Cambridge, most likely painted at Antwerp in 1628 upon Van Dyck's return from Italy, suggesting that the artist may have been recalling the uncertainties of his early steps.

- The absence of strands of hair on the Virgin's right shoulder, initially apparent, is also highlighted by infrared image (Fig. 6 & 8) and thus reveals a later addition by the artist (fig.7).

Fig. 6.  Image IR         

                 Fig. 7. Normal light                                          Fig. 8. IR image                                       Fig. 9. IR image, detail (Madrid, Cerralbo Museum)

- Photograph of the left foot of the infant Jesus after cleaning (second cleaning carried out in 2002) (Fig. 10), highlights a psentimento on the thumb and extending to the instep, which clearly reveals the difficulty the artist had in positioning the foot; a compositional modification is also visible, with a second thumb appearing in transparency, which corresponds to a five-toed foot (see page 10, Art Analysis & Research Inc report), whereas the image counts six toes when including the pentimento.

Fig. 10. Normal light, Child's left foot, detail

- The IR image also reveals a modification of the initial position on the thumb of the child Jesus' left hand (Fig. 11), visible in a fine outline in black chalk or underlying fluid paint.

Fig. 11. IR image, Child's left hand, detail

- X-ray image of the painting (Fig. 12) provides information not visible in infrared images, particularly at the lower left of the composition: the original composition is revealed beneath the top layer of black paint, drapery is present, and there is a gap in the left corner (Fig. 13). The nature of the drapery differs from that observed in the three other versions of the composition painted later in Antwerp. In fact, they are mainly composed of smalt blue, a pigment regularly used by Van Dyck.

Fig. 12.  X-ray image

Fig. 13. X-ray image, detail

The enlargement of the drapery area (Fig. 13) is highly informative and, subsequent to the 2019 Art Analysis & Research report, allows us to identify a compositional modification not mentioned in said report: two lines of horizontal patterns are visible extending beyond the corner of the tablet.

Beyond the above, careful examination of the radiograph reveals numerous stages of modification, scattered throughout the composition but difficult to exploit. In addition to the above, a close examination of the X-ray image reveals numerous stages of alteration throughout the composition, though these are difficult to interpret.

In their report of March 31, 2019, Dr. Nicholas Estaugh and Dr. Jilleen Nadolny (London, Art Analysis & Research Inc.) describe on pages 5 - 12 the numerous pentimenti (excerpts from pages 10 - 12, and their conclusion):

" C.5 Other pentimenti:

In addition to the numerous adjustments noted in the hair, head drape and garment of the Virgin, there are other small pentimenti that may be noted in the picture, most especially in the development of the feet of the Christ child.

“Given more time, and comparable imaging in X-ray and IR, the series of paintings of the Virgin and Child illustrated above would make a fascinating study. The interrelationships within the group are revealed in the similarities and differences in the details, such as, for example, the relationship between the gaze of the Child in the Cerralbo and current picture. »

“The examination revealed that the visible painting is the end result of a number of stages of development, the different adjustments to the contours of the draperies and figures is far more suggestive of an artist working out a composition, not the development of a studio work or a copy. »

IV. DATING AND PLACE OF CREATION OF THE PAINTING

The dates of Anton van Dyck's trip to Italy:

- October 3, 1621, departure from Antwerp for Italy;

- November 20, 1621, arrival in Genoa;

- February 1622, departure from Genoa;

- 1622, February to August 2022, Rome;

- 1622, Florence;

- 1622, Bologna;

- 1622, Venice;

- 1622, December 1622, probably in Turin;

- 1623, February 1623 return to Rome;

- 1623, Mantua;

- 1623, return to Genoa;

- 1623, end of 1623 departure for Palermo;

- 1624, in the spring, arrival in Palermo;

- 1625, departure from Palermo, before November 16, 1625;

- 1625, Naples?;

- 1625, return to Genoa;

- July 4, 1625, departure for Aix-en-Provence;

- 1625, return to Genoa;

- 1627, autumn 1627 departure from Genoa, passage through France and Paris;

- 1628, early 1628 return to Antwerp.

            It is therefore now accepted that the Virgin and Child originates from the artist's Italian period, although the precise location remains uncertain and is documented only through the provenance in Robert Lebel archives. However, the combined study of the provenance, scientific information and stylistic comparisons, with all the painted works from 1621 to 1627, provides a strong set of indicators pointing naturally to Genoa.

For example, comparisons of the fabrics details covering the tablets of the Virgin and Child and the Portrait of the Young Man from the Spinola Family (Van Dyck, Genoa circa 1622, Palazzo Rosso) (Fig. 14; details fig 15, 16, 17 & 18) clearly reveals the same manner of execution. Similarly, the yellow right sleeve of our Virgin can be compared with the right sleeve of the youngest child in the portrait The Giustiniani Longo Children (Van Dyck, Genoa circa 1625, National Gallery, London)fig. 19, 20 & 21), further demonstrating a clear stylistic affinity with works painted in Genoa.

Fig. 14

Anthony van DYCK

Genoa, c. 1622

Portrait of a Young Man of the Spinola Family
Oil on canvas, 241 × 168 cm
Palazzo Rosso, Genoa (inv. Pr 115)

Fig. 15. Virgin and Child, detail - Fig. 16. Young man from the Spinola family, detail

Fig. 17. Virgin and Child, detail - Fig. 18. Young man from the Spinola family, detail

Fig. 19

Anthony van DYCK

Genoa circa 1625

The Giustiniani Longo Children (Previously The Balbi Children or De Franchi)

Oil on canvas, 219 × 151 cm

National Gallery, London (inv. 6502)

            And yet, the overall atmosphere and the spectacular character of the composition suggests a potentially major order from Rome. Comparison with the Portrait of Cardinal Bentivoglio (Florence, Palazzo Pitti) (Fig. 22) allows for a close parallel in conception; in both paintings, richly embroidered motifs appear on the fabrics draped over an imposing tablet. As with the stratigraphic analyses of the Virgin and Child prove to be very close to those of Portrait of George Gage (Van Dyck, probably Rome around 1622-25, London, National Gallery), so the hypothesis cannot be excluded. It is also worth emphasizing the manner in which the yellow and purple draperies of the Virgin’s right sleeve are painted, with a virtuosity almost unrivalled in Van Dyck’s work, thereby reinforcing the likelihood of a major commission.

The painting Virgin and Child should be included among the small select group of works bearing witness to Van Dyck’s genius in Italy, alongside the Portrait of Cardinal Bentivoglio, Rome circa 1623 (Palazzo Pitti, Florence) (Fig. 22) ; Portrait of the Young Man from the Spinola Family, Genoa circa 1622 (Palazzo Rosso, Genoa, inv. PR115) (Fig. 14) ; Portraits of Lord and Lady Shirley, Rome circa 1622 (Petworth House, The Egremont Collection, National Trust, London, inv. 38 & 39); The Giustiniani Longo Children (Balbi Or De Franchi[xii]), Genoa circa 1626-27 (National Gallery, London, inv. NG.6052) (fig. 19) and finally the Self-portrait of the artist, Rome circa 1622 (Hermitage, St Petersburg, inv. 548).

Fig. 22

Anthony van DYCK

Rome, 1623

Portrait of Cardinal Bentivoglio

Oil on canvas, 195 x 147 cm

Palazzo Pitti, Florence

Fig. 23. Portrait of Cardinal Bentivoglio, detail        

            The Virgin and Child can also be compared to other works by Van Dyck from the period 1621-22, painted in Antwerp before his departure for Italy, but also in Genoa: Portrait of a woman and her child[xiii], Antwerp or Genoa circa 1621 (Van Dyck, National Gallery, London, inv. NG.3011) (Fig. 24); Portrait of Luigia Cattaneo, Genoa circa 1622 (Museum of Fine Arts, Strasbourg, inv. 200) (Fig. 25). Van Dyck often used the same type of composition for the portraits he painted in Italy, with an imposing tablet covered in red fabric against a background of vermilion drapery. A stylistic parallel with the Virgin and Child can be observed in the execution of hands in paintings created between 1621 and 1627, all close to one another and quite different from what is known of Van Dyck's work in Antwerp, it therefore remains plausible that the Portrait of a Woman and Her Child (fig. 24) was painted in Genoa. The stylistic proximity between the Virgin and Child and the Portrait of a woman and her child indicates a close chronological relationship for the execution of these two paintings.

Fig. 24

Anthony van Dyck

Antwerp or Genoa, circa 1621

Portrait of a woman and her child

Oil on canvas, 131.5 x 136.2 cm

National Gallery, London, inv. NG.3011

Fig. 25

Anthony van DYCK

Genoa, c. 1622

Portrait of Luigia Cattaneo
Oil on canvas, 147 × 112 cm
Museum of Fine Arts, Strasbourg

V. REPRESENTATIONS OF THE VIRGIN AND CHILD IN VAN DYCK’S WORK

The three main versions of the Virgin and Child have long been catalogued and are described below (a, b, c):

a)       Considered the first version, the Virgin and Child (Fitzwilliam Museum, Cambridge) (fig. 26), painted by Van Dyck in Antwerp circa 1628, is arguably the most representative of the artist’s Italian journey.

Fig. 26

Anthony van Dyck

Antwerp, circa 1628

Virgin and Child

Oil on canvas, 146.8 × 109 cm; 4 ft. 9 ¹/₁₆ in. x 3 ft. 6 ¹⁵/₁₆ in

Fitzwilliam Museum, Cambridge, inv. PD 48-1976

This compositional formula, with a veil over the Virgin's head, is also the one that can be seen in the infrared images of our painting (Fig. 8), but also that of the Cerralbo Museum (Fig. 9), suggesting that these two versions represent the artist’s initial conception, later realized upon his return to Antwerp in other versions painted subsequently.

b)             The second version in chronological order, painted circa 1630, the Virgin and Child (Dulwich Picture Gallery, London) (Fig. 27), represents a model with a more Antwerp-like stylistic approach. In 1957, Ludovic Burchard considered the painting a probable workshop work, but it is now rightly catalogued as autograph by Van Dyck.

Fig. 27

Anthony van Dyck

Antwerp, circa 1630-32

Virgin and Child

Oil on canvas, 153.7 × 116.5 cm; 5 ft. ½ in. x 3 ft. 9 ⁷/₈ in

Dulwich Picture Gallery, London, inv. DPG090

c)           The third version, also painted circa 1630, the Virgin and Child (Walters Art Museum, Baltimore) (Fig. 28) is catalogued as Van Dyck and studio, yet stylistically very close to the painting in the Dulwich Picture Gallery considered to be by his own hand. As evidence, the drapery in the Dulwich painting is no more or less well painted than that in the Baltimore painting. Both works were intended to reflect the high level of quality sought by the artist in the continuation of earlier compositions. We now know that upon his return to Antwerp and later in London, Van Dyck rarely painted without the participation of his workshop. Also worth noting are the white drapery folds resting on the Virgin’s right forearm, which are reminiscent of those in our Virgin and Child painted in Italy, as well as the hesitation in the positioning of the infant Jesus’s left foot, with a clearly visible pentimento in exactly the same place as in our Virgin and Child. Van Dyck then reiterated his reservations about the design of the prototype in Italy.

Fig. 28

Antoine van Dyck and workshop

Antwerp, circa 1630

Virgin and Child

Oil on canvas, 126.1 × 114.6 cm; 4 ft. 1⁵/₈ in. x 3 ¹⁵/₁₆ in

Walters Art Museum, Baltimore, inv. MD 37.234

            In 2012, an unpublished version, the Virgin and Child (San Lorenzo Monastery, El Escorial) (fig. 29), was published by Matthias Diaz Padron[xiv], specified as never having been previously mentioned in the monographs.  

The examination of this painting by the author on site at El Escorial in March 2017 provided some insights. First of all, with regard to its condition, it confirmed what Matthias Díaz Padrón had described: a painting that has suffered greatly from previous restorations. Indeed, there is clearly extensive overpainting, which significantly diminishes the true quality of the work, and the painting would greatly benefit from a complete cleaning. Study of the painting on site at the monastery makes it possible to recognize the hand of Van Dyck. Beneath the overpainted areas, one discovers intact original passages that correspond to what is known of the artist’s manner in Italy; see, for example, the hair of the Christ Child, which can be compared to that in the Self-Portrait (Hermitage Museum, St Petersburg), to our Virgin and Child, to The Giustiniani Longo Children (National Gallery, London, inv. NG6052), as well as to the Portrait of the Jeweller Puccio and His Son (Palazzo Rosso, Genoa, inv. PR 50).

Fig. 29

Anthony van Dyck and workshop (?)

Antwerp around 1628

Virgin and Child

Oil on canvas, 131 × 117 cm; 4 ft. 3 ⁹/₁₆ in. x 3 ft. 10 ¹/₁₆ in

San Lorenzo Monastery, El Escorial, inv. 10014577

Nevertheless, and as its stylistic aspect confirms, it is impossible to place the Monastery of El Escorial San Lorenzo's painting within the artist's Italian period, but probably from Van Dyck's return to Antwerp. The thread count of the original canvas per centimetre[xv] corresponds to that of the canvases used by the artist in Antwerp[xvi]. This image of the painting appears to be the good link between the Cerralbo Museum's painting and the one in the Fitzwilliam Museum in Cambridge, in which case it could be regarded as a transition from the Italian period to the second Antwerp period. Thus, it is possible to date this painting to Van Dyck's return from Italy, circa 1628 (Padron, 2012, p. 220), and to situate it as preceding all other versions painted in Antwerp.

            In 2017, a probable fourth version was published, the Virgin and Child (Cerralbo Museum, Madrid) (Fig. 30)[xvii], this work challenges the established chronology and appears even earlier than our painting. Long mistakenly attributed to Mateo Cerezo, we have attempted to demonstrate that it is, in all likelihood, the first version of the composition painted in Genoa upon the artist's arrival circa 1621. The related scientific analyses have shown that the paint layers identified through sampling correspond to the techniques used by Van Dyck in Italy, as well as the canvas weave, whose warp consists of ten vertical threads and nine horizontal threads per square centimeter, matching the canvases employed by Van Dyck in Genoa (Roy, Technical Bulletin no. 20, 1999, London, National Gallery, Yale University Press)[xviii].

The comparison between the Cerralbo Museum’s painting and our Virgin and Child mainly concerns the resemblance of the Virgin’s face and the pose of the Child Jesuss. The Madrid painting is mysterious and raises questions, but ultimately how can one not see in it the artist's first steps in conceiving this composition of the Virgin and Child, suggesting a first attempt, a sketch with hesitant execution (for  information, the painting has been reduced at the bottom of the composition). There are so many pentimenti and modifications, some of which are even repeated in our Virgin and Child. The significantly lower quality of the Cerralbo Museum's painting suggests that, after much hesitation and numerous difficulties encountered in developing its composition, the artist did not complete his work, choosing instead to abandon and leave it unfinished. This could be explained by the cultural shock Van Dyck experienced upon his arrival in Genoa, he had to innovate following the passage of his prestigious predecessor Rubens, hence his numerous experiments. The significance of the work as a prototype should not be underestimated, everything suggests that we are looking at the source of the composition.

Fig. 30

Antoine van Dyck, workshop of (?)

Genoa, circa 1621

Virgin and Child

Oil on canvas, 98 × 84 cm; 3 ft. 2 ⁹/₁₆ in. x 2 ft. 9 ¹/₁₆ in

Museo Cerralbo, Madrid, inv. n. VH 0436

An execution by the workshop has also been strongly suggested for the Cerralbo Museum’s painting, but how could imagine it at such an early stage around 1621 - arguably the only possible date for a version executed by Van Dyck - when the creation of a workshop seems equally unlikely at a time when the artist still had everything to prove in Genoa.

In this context of Van Dyck’s prototypes, it is important to note the connection between the Virgin and Child in the Cerralbo Museum and a Virgin and Child (Metropolitan Museum, New York)[xix] (Fig. 31) painted in Antwerp circa 1620 (according to the author, an execution circa 1628 upon his return from Italy seems to correspond more closely with the painting's style), not because of any resemblance between these works - their stylistic approaches are quite different - but for their similar conception and the variations they generated. Reading the insightful catalogue entry for the Metropolitan painting (Anthony van Dyck - Virgin and Child - The Metropolitan Museum of Art), one can observe a characteristic shared by these two works: an initial tentative execution which subsequently gave rise to much more accomplished versions, no fewer than eight have been identified for the Metropolitan painting, by Van Dyck himself as well as by assistants in the workshop and certainly after 1628.

The text of his entry:  “At least eight versions of this tender image came out of Van Dyck's workshop, and this one is at once the finest and least finished example. The deftly fluid brushstrokes loosely follow an incised underdrawing, which must have been traced from a preparatory drawing. Van Dyck would have kept this panel in his studio as a model for more finished paintings by himself and by assistants.”

Fig. 31

Anthony van DYCK

Antwerp, c. 1620

Virgin and Child
Oil on panel, 64.1 × 49.5 cm; 25 1/4 x 19 1/2 in
Metropolitan Museum of Art, New York

     VI.  HISTORY, PROVENANCE

            The painting’s former provenance was identified in 2015 through consultation of the Robert Lebel archives. On the reverse of a black-and-white photograph dating from the first half of the 20th century, appear Robert Lebel’s personal annotations[xx]: Provenance: «M. Lombard, Valencia; cf. Bridgewater et Gênes(stamp : Fotografia F. Sanchis, Valencia) ». 

This Lombard provenance appears to correspond to Henri Lombard Gaujoux or one of his descendants. A French industrialist originally from Nîmes, he industrialized the town of Almoines in 1848 (municipality in the province of Valencia), where he established a silk-spinning mill. Fifty years later, three of his sons founded a bank, Maison Lombard Frères, and additional manufacturing centers were subsequently created. The Almoines factory employed around 500 workers between the 1950s and 1973s and finally closed in the 1990s. The reference on the back of the photograph to "Bridgewater provenance" raises questions, as this provenance is already referenced by two versions of the same subject, including the one at the Fitzwilliam Museum in Cambridge, described as follows in Mrs. Jameson's publication (1844, no. 264, page 156 of the catalogue): 

           « The Virgin, seen (three quarters) in a scarlet vest, with blue sleeves and a blue mantle, holds the enfant standing on her knee, supporting him with both hands ; he extends his right hand to her bosom ; her countenance, with a divine expression of hope and confidence, is raised towards heaven, as if mentally devoting her child « to do the will of his father ». Brought to England about 1790, by Bradshaw Pearson, Esq., who sold it to the Duke of Bridgwater. Van Dyck repeated this beautiful subject several times. There is one at Blenheim ; another at Dulwich ; another at Dresden. The picture here is allowed to be the finest of all. It is to be regretted that it has been painted on panel, and that it has split down the body of the Saviour, and through the drapery of the Virgin. P . 4 ft. 7 in by 3 ft. 4 in (Smith Cat. 263.). »

The painting thus described is indeed the one in the Fitzwilliam Museum, Cambridge (fig. 26). 

The Smith catalogue (1831), as well as the inventory of the Bridgewater Collection (1851, Marlborough Collection, cat. 263, p. 79), state:

« 263. The Virgin and Child. The composition represents the Virgin clothed in a scarlet vest, with blue sleeves and a blue mantle, holding the infant saviour erect at her left side, with both hands under his arm ; his countenance and attention is turned from his parent, while at the same time he bends slightly towards her, and extends his right hand to her bosom ; his left holds the white linen which surrounds his loins ; some blue drapery, thrown across the scroll of a couch, is under his feet ; her beautiful expressive countenance and eyes are raised to heaven. A portion of a pillow and an obscured sky from the back-ground.

4 ft. 1 in. By 3 ft. 9 in.--C.

A picture agreeing precisely with the above description, and indubitably the original, of superlative beauty, possessing the mellow richness and brilliancy of colour of Titian, with the most chaste and elegant design, and an expression in the countenance of the Virgin perfectly divine, is in the splendid collection of Lord Francis Leveson Gower, at Bridgewater House.

4 ft. 7 in by 3 ft. 4 in.—P.

A third picture, being a repetition of the preceding, is in the Dulwich Gallery.

4 ft. 8 in by by 3 ft. 5 in.—C.

The above subject and composition is engraved by Pontius, Carmona, Finden, and Salvador. The latter took his print from a picture then in the collection of the Count Vincii, in 1757. »

The second painting thus described in the Smith catalogue (1831), then in the collection of Lord Francis Leveson-Gower at Bridgewater House, corresponds to the work now held at the Walters Art Museum (fig. 28).

As for the third painting, mentioned for comparison, it is the version held at the Dulwich Picture Gallery  (fig. 27).

Could another version have been acquired by the Bridgewater Collection after 1851, the date of the second catalogue (Smith 1851), before making its way to Spain? This question arises when reading the back of Robert Lebel’s photograph, though it remains unclear whether this might simply be a case of confusion.

           The provenance prior to M. Lombard remains uncertain; a Van Dyck's Virgin and Child is recorded in 1868 in the collection of Jorge Diaz Martinez[xxi] in Valencia, likely our painting.

The active Mediterranean trade in the 17th century between the city of Genoa and the port of Valencia suggests that the work may have been transported to Spain via this natural maritime route, although no specific date can be proposed.

VII. LA VIE, LES FRÈRES DE WAEL ET LA VIERGE A L’ENFANT

             Literature on Van Dyck regularly refers to texts written in the form of a manuscript kept at the Louvre, entitled La Vie, les ouvrages et les élèves de Van Dyck (Anonymous, 1769–1791?), hereafter referred to as La Vie, which proves to be the most solid initial reference on the artist’s life. This Van Dyck biography manuscript was partly based on letters from Cornelis de Wael to Lucas van Uffel (La Vie, 1769–91, Louvre, folio 12), now lost. It also draws on Horace Walpole’s Anecdotes of Painting (c. 1762–1771), which were based on the Notebooks of the painter George Vertue (1684–1756), as well as accounts by Bellori, Campo Weyerman, De Piles, Descamps, Félibien, Houbraken, and Soprani. 

The brothers Lucas (Antwerp 1591 - 1661 Antwerp) and Cornelis de Wael (Antwerp 1592 -1667 Rome), sons of the painter Jan de Wael and Geertruide de Jode - herself the daughter of the engraver Gerard de Jode - were trained in various studios in their native city of Antwerp. Lucas apprenticed with his famous uncle, Jan Bruegel the Elder, while Cornelis worked with the battle painter Jan Snellinck. Both were engravers as well as art dealers. Cornelis specialized in battle scenes, both maritime and terrestrial, particularly sieges, whereas Lucas focused on landscapes and returned to Flanders in 1628 (Depauw & Luijten 1999, p. 21).

They arrived in Genoa in 1615, lived on the Strada Nuova (today Via Garibaldi) and, together with Jan Roos, formed the core of the expatriate Flemish community. They welcomed their colleague Anthony van Dyck upon his arrival in Genoa around the end of 1621.

 A close reading of the La Vie manuscript proves particularly insightful regarding the relationships between Van Dyck and the brothers Lucas and Cornelis de Wael in Italy. Preserved in two volumes in the Louvre archives, it includes both the draft and the final version. Although Larsen transcribed and annotated the complete manuscript in 1975, there remains a notable underutilization of the information it actually contains.

One passage concerning the Virgin and Child in the context of Van Dyck’s relationships with the De Wael brothers is particularly noteworthy, as it is absent from the published literature on the artist, aside from Larsen’s 1975 transcription. 

In this passage, it is stated that Anthony van Dyck supposedly presented three paintings to Lucas and Cornelis de Wael in 1625 before his departure from Genoa. In the Barnes et al. catalogue of 2005, only partial excerpts of this passage from the La Vie manuscript are cited, mentioning the Double Portrait of the Wael Brothers (Larsen, 1988, p. 142, cat. 347; Barnes et al., 2005, p. 210, II. 71) and Bacchus, paintings that have been identified and thus justify the manuscript citation.

The references to this Virgin and Child in the final version of the La Vie manuscript at the Louvre are therefore of crucial significance:

Draft, Vol. I, folios 26-28:

  « Ayant demeuré encore quelques mois à Gênes, il prit la résolution de retourner en sa patrie. Avant de partir il donna un beau témoignage de la reconnaissance qu’il devait à l’amitié des Frères de Wael. Il les peignit l’un et l’autre sur une même toile, et leur fit présent d’une Vierge qui tient entre ses bras l’Enfant Jésus, ainsi que d’un tableau représentant Bacchus dans sa jeunesse, ivre, monté sur une panthère et conduit par huit autres enfants. Le premier de ces beaux morceaux a été gravé par Hollar sur l’esquisse que Lucas avait rapportée à Anvers ; et le troisième, qui se voit encore à Gênes, au palais Gentile, le fut par D. Brunn. »

Final version, Vol. II, folios 46, 47:

   « Ayant demeuré encore quelques mois à Gênes, il prit la résolution de retourner en sa patrie. Avant de partir il donna un beau témoignage de la reconnaissance qu’il devait à l’amitié des Frères de Wael. Il les peignit l’un et l’autre sur une même toile, et leur fit présent d’une Vierge qui tient entre ses mains l’Enfant Jésus, ainsi que d’un tableau représentant Bacchus dans sa jeunesse ivre, monté sur une panthère et conduit par huit autres enfants. Le premier de ces beaux morceaux a été gravé par Hollar sur l’esquisse que Lucas avait rapportée à Anvers, et le troisième, qui se voit encore à Gênes au palais Gentile le fut par D. Brunn. »

The differences between the draft and the final version of the text are limited to a few punctuation marks and a single sentence: the position of the Christ Child, described as “in the arms” of the Virgin in the draft, but “in the hands” of the Virgin in the final version. The Virgin and Child presented by Van Dyck in 1625 to the De Wael brothers would, in all likelihood, correspond either to our painting or to the one at the Cerralbo Museum in Madrid.

Our painting, described in this dossier, depicts the Virgin holding the Christ Child “in her hands” rather than “in her arms” - the latter indicating a very different composition - and therefore bears the characteristics of the work Van Dyck offered in Genoa to the De Wael brothers.

Ger Luijten, referring to the La Vie manuscript at the Louvre, cites this Virgin and Child as “a Virgin holding the Christ Child in her arms”[xxii], thus referring only to the passage from the draft and not to that of the final manuscript.

VIII. CONCLUSION

             The Virgin and Child, most likely painted in Genoa circa 1621–25, appears to be the prototype of a series culminating in the Virgin and Child at the Walters Art Museum in Baltimore. All evidence suggests that the drawing The Christ Child Standing (Rotterdam, Museum Boijmans Van Beuningen) served as a preparatory study for our painting, rather than for the later versions painted in Antwerp. It is therefore possible that The Virgin and Child is the painting Van Dyck gave to the brothers Lucas and Cornelis de Wael when he left Genoa..

            The numerous pentimenti and compositional changes evident in The Virgin and Child attest to its early stage in the artist’s development. This painting, the largest in dimension, executed with costly pigments and distinguished by the richness of its fabric textures, the brilliance of golden-yellow sleeves against violets, the crimson of the draperies, as well as the expressive faces, constitutes a major work. It either indicates an important commission or, more likely, corresponds to the painting Van Dyck offered to the De Wael brothers.

              Its discovery marks an important advance in understanding the artist’s working methods during the 1620 - 1627 period, from initial conception to the gradual refinement of a theme that Van Dyck would later revisit in other compositions in Antwerp.     

Philippe Barnabé

Paris, le 12 février 2026 

 Free translation, Galerie Barnabé