Van Dyck

Just another WordPress site

Grisaille Sketches

Anthony van Dyck: The attribution of the grisaille oil sketches for his engraved prints

Anthony van Dyck’s portraits of great men and women in Gillis Hendricx’s edition of the Iconography comprises of a total of one hundred engraved prints. The process by which the majority of these engravings were produced consisted of three main stages; the first was a drawing, usually in black chalk, the second was a grisaille oil sketch and the third was the engraved print itself. This latter stage in turn consisted of printed impressions of partially completed proofs of an etched or engraved copper plate, which record an intermediate stage in the production of a print. Each stage of a print recorded in an impression is known as a state and any additional work on the production of a print constitutes a new state. If there was an existent oil portrait of the sitter by Van Dyck then the initial chalk drawing would have been taken from this.[1] These drawings are widely regarded as being by Van Dyck’s own hand; however the attribution of the grisailles is still contested today.[2]

A grisaille is a monochrome sketch in brown and white oil paint. These sketches are of a similar size to the engravings and would have been used as a guide in the production of the print.[3] The chalk drawing was the initial image used by the engraver, which suggests that it was the more significant stage of the process, rather than the grisaille. If the drawing was not completed with an ink wash then a grisaille was painted.[4] These grisailles were used in this process purely to inform the engraver and were not seen as an artwork in their own right. It is even unlikely that Van Dyck would have wanted these oil sketches to be seen by the wider public.

Image 1: Anthony van Dyck, Jean-Gaspard Gevaerts. Oil on panel. © Duke of Buccleuch and Queensberry, Boughton House, Northamptonshire.

The attribution of these grisailles has generated interesting debates among art scholars for many years. One example being the British art historian, and former Director of the National Portrait Gallery in London, Lionel Cust, who wrote of his difficulty in believing that Van Dyck had any role in this particular aspect of the process due to the simple matter of time constraints.[5] This was owing to the fact that, towards the end of his life, Van Dyck was inundated with a large amount of commissioned work.[6] Cust wished to assert his belief that the drawings were by the hand of Van Dyck, but after their creation they were given to one of his assistants to transform them into oil grisailles. The assistant, he claimed, trained for this purpose in the school of Rubens.[7] The grisaille and the drawing were then handed over to the engraver for the production of the print: ‘Many of these small grisaille portraits exist, most carefully finished and capital renderings of Van Dyck’s style, but it is not possible on any grounds to ascribe any of them, as has often been done, to the hand of the painter himself’.[8] Cust’s assertion places a significant amount of responsibility on Van Dyck’s assistants as the drawings had a tendency to be sparse in detail, or the grisaille could differ quite significantly from the drawing. For example the grisaille for Jean-Gaspard Gevaerts (1593-1666) in the Buccleuch Collection, Boughton House (Image 1), sees him wearing a different costume to the one depicted in the drawing in the Albertina, Vienna.[9] Gevaerts, spelt in Latin as Gevartius, is depicted wearing a collar trimmed with lace and a fur-lined cape slung partly over his left arm in the grisaille while he wears a plain collar and a different cloak in the drawing. Cust did suggest that Van Dyck could possibly have seen the grisailles before print production; however he believed it to be unlikely due to Van Dyck’s residence in England for the majority of the project.[10]

Conversely Arthur Mayger Hind, another British art history scholar, did believe that the grisailles were painted by Van Dyck. Ironically his theory for this attribution was due to the fact that the oil sketches would have been quick to produce, ‘an hour, or a few hours at the most’, therefore Van Dyck could have undoubtedly found the time.[11] When discussing the grisailles in his book, Van Dyck: His Original Etchings and His Iconography, Hind made reference to a particular group of thirty-eight, which reside in the art collection of the Duke of Buccleuch and Queensberry at Boughton House in Kettering, Northamptonshire and formerly in the now demolished Montagu House in Whitehall, London.[12] He believed the sketches had a distinguished and masterly approach: ‘Apart from their expressive power as portrait, they put the scheme of light and shade before the engraver with such conviction, that I am unable to conceive of the good assistant who could have accomplished the task with such brilliance as a mere intermediary’.[13] However, the artistic brilliance of these grisailles could be contested. If we take an example of a grisaille in the Buccleuch Collection at Boughton House, that of Karel van Mallery (1571-1635), we can study the grisaille alongside the drawing, in the Devonshire Collection at Chatsworth House in Bakewell, Derbyshire (Image 2), in order to ascertain the similarities or differences in skill and technique between the two sketches.[14] As it is assumed that Van Dyck created the chalk drawings himself, we can use this to understand the manner in which he worked. In the drawing Van Dyck’s line is confident and full of gesture. His use of subtle variations in pressure when applying the chalk to the paper adds to the fluidity of the piece, this aids the description of a fleeting moment and a suggestion of movement. There is an existing portrait in oil of Mallery by Van Dyck in the National Gallery (Nasjonalgalleriet), Oslo (Image 3), which he used to create this chalk drawing. There is a slight difference in the collar in the oil painting, with the addition of a ruff in the chalk sketch. Van Dyck is seen to be designing this ruff and the shoulders directly onto the paper as a vague outline shows it began at a higher point on the page, before he made the decision to move it down. The general gesture of the head remains integral to the original, but Van Dyck was willing to alter the costume of his sitter, this was a common occurrence throughout the prints. It is curious that the skilled approach in the drawing does not translate to the grisaille. Instead the grisaille has a static feeling and the pose looks forced and uncomfortable. The confident movement of the chalk seen in the drawing has altered in favour of what appears to be a far more laboured method with short, dithering brush marks. In terms of the attribution of the grisaille to the hand of Van Dyck, it is difficult to ascertain the masterly technique Hind was celebrating.

Image 2: Anthony van Dyck, Karel van Mallery, Black chalk on paper, The Duke of Devonshire, Chatsworth House, Derbyshire:;view=1up;seq=67

Image 3: Anthony van Dyck, Karel van Mallery. Oil on canvas, Nasjonalgalleriet, Oslo:

However, the debate over the skill seen in these grisailles cannot be concluded simply by studying the one image. To give an example of a grisaille with a more apparent painterly quality, we can look to the oil sketch for the Martin Ryckaert (1587-1631) print, which was sold by Sotheby’s, New York in the Weldon Collection sale on 22 April 2015 as lot 34.[15] It portrays a confidence in stoke and an understanding of painterly effects that were less forthcoming in the Mallery grisaille. The drawing for this print is not known of so we are unable to make a comparison between the two, yet when analysed with the original portrait of Ryckaert in the Prado Museum, Madrid (Image 4) we are able to ascertain the similarities or differences in both style and technique. The original portrait gives a skilful soft focus to the hair around the forehead and beard, whereas the grisaille shows far more detail in these areas. This has the effect of drawing attention away from what should be the primary focus, which in a portrait is typically the eyes. The reasoning for this extra detail could be to better inform the engraver. However, in terms of shape there are various differences between the two images. The forehead in the grisaille appears far larger than in the original and the right eyebrow looks to be on a greater scale in the sketch, which gives the illusion that the eye is at a steeper angle than in the original. Perhaps as an attempt to address this, the bottom eyelid has been marked too low. Nevertheless, this image displays a far greater understanding of technique and painterly skill than that seen in the Mallery grisaille. What is clear is that there were different levels of ability between each grisaille, which may suggest they were produced by a variety of different assistants.

Image 4: Anthony van Dyck, The painter Martin Ryckaert. Oil on panel,
Museo del Prado, Madrid:
Photo © Museo del Prado, Madrid.

From the two strands of debate put forward by Cust and Hind, the line of reasoning made by Cust does seem the most satisfactory. The straightforward time constraints placed on Van Dyck alone prove that he was under serious pressure without the added problem of creating his own project. This is particularly apparent when considering he had already created the chalk drawings for the piece, therefore the assumption that Van Dyck would create another sketch entirely by his own hand is questionable.

Aside from this perception of time limitations, another key aspect of Cust’s argument was based on the capability of Van Dyck’s assistants. When Van Dyck was painting it would not have been him alone who produced his oeuvre, he had a well-established and highly skilled workshop of assistants. These artists were employed to work on areas of Van Dyck’s painting, such as the clothing and drapery.[16] Cust also made reference to this in his book Anthony Van Dyck: A Further Study, where he lists a number of accomplished artists working with Van Dyck whilst he was in England, including names such as David Beck (1621-1656), William Dobson (1611-1646), James Gandy (1619-1689), and Peter Lely (1618-1680).[17] Cust described the volume of commissioned work Van Dyck had between April 1632 and December 1641, declaring that this time span and the immense number of portraits attributed to him meant that they cannot all be by the hand of the painter himself.[18] To further this argument, these artists were also skilled in making copies and replicas after Van Dyck’s original paintings, which were sold at a lower price from the studio and some artists were even employed chiefly as copyists.[19] Cust warns us that ‘the greatest care must, therefore, be exercised in accepting every painting attributed to Van Dyck in England as entirely by his own hand, or even the joint work of himself and his pupils’.[20] Van Dyck’s ability to give a significant amount of his portrait commissions to his assistants, on the understanding that their practice was of a sufficient quality, highlights the improbability of his inclination to produce these grisaille sketches himself. This is especially apparent when considering that they were to be used exclusively in the engraving process and not as a finished product in themselves.

Among Cust’s list of Van Dyck’s well-regarded assistants was David Beck, who became the painter for the Queen of Sweden.[21] Another noteworthy artist was William Dobson who went on to become the court painter for Charles I.[22] Dobson’s practice was of a high quality with expert craftsmanship. This expertise is evidenced in his self-portrait in the Earl of Jersey Collection at Radier Manor in Grouville, Jersey and currently on a ten-year loan to Osterley Park and House, Isleworth in west London.[23] The portrait has the masterly illusion of life within it. The soft brushstrokes in the hair, with particular reference to the sections falling around the face, are deliberately vague so as to suggest movement. The sharp highlight in the eye immediately acts to draw the viewer’s eye towards the main area of focus. The thin paint in the shadows aims to accent the thick paint in the lights, a technique which gives an overall breath to the portrait. The confidence seen in the production of this painting serves to encourage uncertainty when deliberating Hind’s suggestion that the presence of the masterly skill within these grisailles could only have been produced by Van Dyck’s hand.

Another claim made by Hind related to the fact that the engravings display the label Ant. Van Dyck Pinxit, which when translated from Latin reads ‘painted by Van Dyck.’ This statement, Hind asserted, could only refer to the grisailles as some of the prints were not produced after an original Van Dyck portrait. Hind believed that ‘this in itself is a strong argument for the authenticity of the grisailles’.[24] The difficulty in discerning any other meaning for the word pinxit leaves an area of uncertainty in the proposal that the grisailles were not painted by Van Dyck. The art historian Joaneath Spicer studied this opinion, going into great detail regarding this terminology in her essay Anthony van Dyck’s Iconography: An Overview of Its Preparation. She explained that the word pinxit on the prints in Hendricx’s edition is used on every plate apart from the set of ten that originated as Van Dyck etchings. These plates bare the inscription Ant. van Dyck fecit aqua forti, which translated as ‘etched by Van Dyck’. On the rare occasion that the plate was etched by Van Dyck after an oil painting also by him, for example the print of Frans Snyders, it would say Ant. van Dyck pinxit et fecit aqua forti, meaning that Van Dyck was both the painter and etcher.[25] This complicated explanation highlights the deliberation art scholars, such as Hind, were involved in. One major factor, however, which has mainly been overlooked, was Hendricx’s role as a publisher. Spicer briefly touched upon the notion, but did not enforce its importance.[26] Hendricx’s need to be successful in the selling of this book could have led to the potential embellishment of the truth, perhaps reinforced by the fact that it would have been far more difficult to prove that each print was not from an original painting by Van Dyck in the middle of the seventeenth century.

Worth Library material digitised by Digital Resources & Imaging Services, The Library of Trinity College Dublin

Image 5: Carlos II Coloma y de Saa, knight of Santiago, 1st Marquis of Espinar (1566-1637). Fig. 13 in book.

Another aspect of this debate is centred on the amount of control Van Dyck was willing to give to the engraver. For if he was willing to allow a significant amount of artistry to take place by the hand of the engraver; then it may be less likely that he would feel compelled to complete all other stages of production himself. One example of deviation within a print from both the grisaille, in the Buccleuch Collection at Boughton House, and the drawing, in the Fogg Museum, Harvard Art Museums in Cambridge, Massachusetts, can be seen in the engraving of Carlos Coloma (1566-1637) by Paulus Pontius (Image 5).[27] He produced his engraving so that it appears in the same direction as the two sketches; this means that Pontius would have had to reverse the image for the printmaking process. Why this is the case is unclear, but it does show an engraver working with the intention of using both the drawing and the grisaille as a rough guide rather than an exact copy. As well as this the drawing does not appear to show Coloma wearing armour, whereas the grisaille and the print do. This suggests that the plan evolved within the process of creating the print. Whether it was Van Dyck that had the final say or not is difficult to ascertain, however Cust did suggest that it was doubtful Van Dyck had even seen the grisailles, so his final role before the execution of the print would be in the drawing.[28] Perhaps Van Dyck gave more authority to other artists than scholars would like to admit.

In summation, the oil grisailles used strictly as a process for achieving an engraved print have caused considerable confusion in the world of art historians. Evidence with regard to the attribution of these images to Van Dyck alone is strong if you are willing to believe the words of a publisher wishing to sell books. On the other hand, when accounting for time constraints and the nature of artistic production in the seventeenth century, it seems less certain that Van Dyck had a hand in any of these images, or even witnessed their production. What is clear, however, is that this topic will continue to inspire debate for a long time to come.


Stijn Alsteens and Adam Eaker (eds.), Van Dyck: The Anatomy of Portraiture (New Haven and London, 2016).

Susan J. Barnes, Nora De Poorter, Oliver Millar, and Horst Vey, Van Dyck: A Complete Catalogue of the Paintings (New Haven and London, 2004).

Christopher Brown, Van Dyck Drawings (London, 1991).

Christopher Brown and Hans Vlieghe (eds.), Van Dyck 1599-1641 (London, 1999).

Lionel Cust, Anthony van Dyck: A Further Study (London, 1911).

Lionel Cust, Van Dyck (London, 1906).

Carl Depauw and Ger Luijten (eds.), Anthony van Dyck as a Printmaker (Antwerp, 1999).

Antony Griffiths, Prints and Printmaking: An Introduction to the History and Techniques, 2nd edn., (Berkeley and Los Angeles, 1996).

Michael Jaffé, ‘The Paintings and Drawings’, in Tessa Murdoch (ed.) Boughton House: The English Versailles (London, 1992), pp. 74-89.

Arthur Mayger Hind, Van Dyck: His Original Etchings and His Iconography (Boston and New York, 1915).

National Trust Collections, ‘Art history, family history | Treasure Hunt’, Treasure Hunt (27 February 2014). Available at:

Roger de Piles, The Principles of Painting (London, 1743).

Joaneath A. Spicer, ‘Anthony van Dyck’s Iconography: An Overview of its Preparation’, in Susan J. Barnes and Arthur K. Wheelock, Jr. (eds.), Van Dyck 350 (Washington, 1994), pp. 327-364.


[1] Joaneath Spicer, ‘Anthony van Dyck’s Iconography: An Overview of its Preparation’, in Susan J. Barnes and Arthur K. Wheelock, Jr. (eds.), Van Dyck 350 (Washington, 1994), p. 335.

[2] Susan J. Barnes, Nora De Poorter, Oliver Millar, and Horst Vey, Van Dyck: A Complete Catalogue of the Paintings (New Haven and London, 2004), p. 365.

[3] Ibid.

[4] Spicer, ‘Anthony van Dyck’s Iconography’, pp. 334-335.

[5] Arthur Mayger Hind, Van Dyck: His Original Etchings and His Iconography (Boston and New York, 1915), p. 42.

[6] Lionel Cust, Anthony van Dyck: A Further Study (London, 1911), p. 76.

[7] Lionel Cust, Van Dyck (London, 1906), p. 125.

[8] Ibid.

[9] Albertina, Vienna, ‘Anthonis van Dyck, Jan Gaspar Gevartius (Gevaerts), Inv. 17643’, ALBERTINA online – Datenbanksuche. Available at:[17643]&showtype=record.

[10] Cust, Van Dyck, p. 125.

[11] Hind, Van Dyck: His Original Etchings and His Iconography, p. 48.

[12] Ibid., p. 42.

[13] Ibid., p. 48.

[14] Barnes, De Poorter, Millar, and Vey, Van Dyck: A Complete Catalogue of the Paintings, p. 370.

[15] Sotheby’s, ’dyck dijck, sir anthony ||| portrait – male ||| sotheby’s n09335lot4d86pen’. Available at:

[16] Roger de Piles, The Principles of Painting (London, 1743), pp. 177-178.

[17] Cust, Anthony van Dyck: A Further Study, p. 76.

[18] Ibid.

[19] Ibid.

[20] Ibid., p. 77.

[21] Ibid., p. 76.

[22] Ibid., p. 77.

[23] National Trust Collections, ‘Art history, family history | Treasure Hunt’, Treasure Hunt (27 February 2014). Available at:

[24] Hind, Van Dyck: His Original Etchings and His Iconography, p. 63.

[25] Spicer, ‘Anthony van Dyck’s Iconography’, p. 330.

[26] Ibid.

[27] Harvard Art Museums, Cambridge, Massachusetts, ‘From the Harvard Art Museums’ collections Portrait of Don Carlos Coloma’. Available at:

[28] Cust, Van Dyck, p. 125.

Text: Ms. Hannah Baker, M. Phil. in Art History, ART+IRELAND, Department of History of Art and Architecture, Trinity College, Dublin.