Skip to main content
Reading options

The Lavergne Family Breakfast

Catalogue entry

Jean‐Etienne Liotard
NG6685 
The Lavergne Family Breakfast

,

,

Extracted from:
The National Gallery; with entries by Emma Capron, Dillian Gordon, Sarah Herring, Mary McMahon, Letizia Treves and Francesca Whitlum‐Cooper; with technical contributions by Paul Ackroyd, Rachel Billinge, Lynne Harrison, Catherine Higgitt, Helen Howard, Larry Keith, Marta Melchiorre, Britta New, David Peggie, Marika Spring and Hayley Tomlinson, Online Entries (London: The National Gallery, 2024).

© The National Gallery, London

Pastel on paper mounted on canvas, 79.7 × 104.5 cm

Signed and dated on sheet of music in the table drawer: Liotard / a lion [in Lyon] / 1754

Provenance

Painted by Liotard in Lyon in 1754 and brought by him to London, where purchased in 1755 by William Ponsonby, Viscount Duncannon, later 2nd Earl of Bessborough (1704–1793), for 200 guineas; the sale of his collection, Christie’s, London, 5–7 February 1801, third day, lot *75, where either purchased by Bessborough’s son‐in‐law Aubrey Beauclerk, 5th Duke of St Albans (1740–1802), for 85 guineas, or bought‐in at that amount to disguise the failure to sell;1his posthumous sale, Christie’s, London, 27 March 1802, lot 78, where probably purchased by picture‐dealer Roch Jaubert (1756–1813) for 37 guineas (although again, it may have been bought in and Jaubert’s name used to disguise the fact); acquired by Luke Foreman (1757–1814), London, in April 1805, apparently from the 3rd Earl of Bessborough (1758–1844), which supports the hypothesis that the pastel had been repeatedly bought in;2 by inheritance to his widow, Mrs Mary Foreman (1763–1834), London; her posthumous sale, Christie’s, London, 19–20 March 1835, lot 51, where it was bought in at £31 10s. and subsequently passed to her nephew, Edward Greene (1797–1887) of Farnborough, Hampshire; by inheritance to his niece, Mary Isabel (‘Isabella’) Golding Palmer (1831–1916), of 36 Queen’s Gate, London;3 her posthumous sale, Christie’s, London, 28 July 1916, lot 5, for £1,260 to ‘Freeman’ on behalf of Asher Wertheimer (1844–1918); after whose death bought privately by Eugene Pinto (1854–1932) for £1,450; by descent via his son Major Richard Pinto (1892–1969) to George Richard Pinto (1929–2018); accepted in lieu of Inheritance Tax by HM Government from the estate of George Pinto and allocated to the National Gallery, 2019.

Exhibition

London 2023–2024.

Related Works

(1) Jean‐Etienne Liotard, The Lavergne Family Breakfast, 1773. Oil on canvas, 81.4 × 103.2 cm. Private collection.

Technical Notes

The work is in remarkably good condition given the fragility of the medium. While there may have been some fading of red, pink, yellow or orange tones, the colours largely remain vivid. The areas of impasto highlights, including the very raised details on the still‐life elements, are also well preserved. There is some wear to the pastel around the edges and corners of the painting and a few minor damages disrupting the pastel layers and paper below. There is also a series of near‐horizontal scratches in the background just above the table.4 Extending almost the full width of the painting, passing across the top of the woman’s forehead and descending, at a slight diagonal to the right edge above the child’s head, is a line of superficial damage. Two similar lines of damage, perpendicular to the more horizontal line, can be seen in the background on the left behind the woman’s head.5 The most highly lit areas of flesh in both sitters have a different surface texture to the surrounding regions. Detailed examination revealed no evidence of retouching in these areas and – while they may now appear more distinct than originally intended – their appearance is more likely related to Liotard’s technique rather than damage, as discussed below.

The pastel is on blue rag paper.6 This support consists of two large sheets of paper, slightly offset in their position but of approximately equal size, one for each figure, with four smaller pieces added to complete the final painting surface.7 The sheets of paper are pasted to what appears to be a single piece of fine weave canvas pulled taut and secured with metal canvas tacks to a simple four‐bar wooden perimeter strainer with one vertical central cross bar. The paper has been wrapped around the sides of the strainer, covering the tacks and extending onto the back, and does not seem to have been disturbed since.8 The method of preparation of the paper to receive the pastel is not clear but the pronounced paper fibres that can be seen suggest some treatment of the surface to produce a nap to retain the pastel crayon.9 The design was fully drawn onto the prepared paper using an infrared absorbing material, likely black chalk or possibly charcoal (see below).

Description

At their breakfast table, an elegantly dressed woman watches a little girl dip a piece of bread into a cup of milky coffee. We know that it is early morning as the girl wears paper curlers in her hair. On the table between them lies a sumptuous still life: a pewter coffee pot, a simple milk jug, a bowl of sugar and two delicate, beautifully decorated cups and saucers. The group is a symphony of reflective surfaces, from the gleam of the pewter and the tiny window visible in the glaze of the milk jug, to the black lacquer tray tabletop on which the ensemble sits and in whose reflection it is doubled. The objects are animated by a series of tenderly observed gestures, the fingers of the woman’s right hand curling perfectly around the milk jug’s handle, those of her left reaching out to steady the saucer into whose cup the bread is being dunked.

The Lavergne Family Breakfast is an exceptionally appealing picture – so appealing, in fact, that Liotard painted it twice. That in itself is not uncommon: artists (and their assistants) often made versions of important pictures, either at the request of the patron or to retain in their studio as an example to show future clients. It was unusual, however, for an artist to return to a work after almost two decades.10 The sheet of music that pokes out from the table’s open drawer confirms that Liotard painted the pastel in Lyon in 1754: Liotard / a lion [in Lyon] / 1754 (fig. 1). In London in 1773 he painted an almost identical version in oil on canvas (fig. 2).

Fig. 1

NG6685, detail of the sheet of music confirming that Liotard painted the pastel in Lyon in 1754. © The National Gallery, London

Fig. 2

Jean‐Etienne Liotard, The Lavergne Family Breakfast, 1773. Oil on canvas, 81.4 × 103 cm. Private collection. © courtesy the owner / photo: The National Gallery, London

Liotard was justly proud of The Lavergne Family Breakfast. He described it in his manuscript autobiography as one of his ‘principal works’ and devoted two passages of his Traité des principes et des règles de la peinture (1781) to the composition.11 Several eighteenth‐century biographies mention the high price of 200 guineas at which he sold it – information the artist must have been keen to circulate.12 The authors of the artist’s catalogue raisonné declared it ‘Liotard’s masterpiece’.13 Yet for all these accolades, neither the pastel nor the oil version of this composition is well known. Hidden away in British private collections for the past 250 years, they have only been exhibited once since Liotard’s lifetime and have only occasionally been published.14

The story of The Lavergne Family Breakfast begins in London. In early 1753, Liotard arrived in Britain for the first time, probably at the invitation of either William Ponsonby, Viscount Duncannon, later 2nd Earl of Bessborough, with whom he had travelled to Constantinople, or Sir Everard Fawkener (1694–1758), who had been ambassador in Constantinople during Liotard’s stay there. He took rooms in Golden Square at the heart of London’s artistic district, was presented (probably by Sir Everard) to the royal family, and quickly became a resounding success.15 By June, he was said to have ‘vast business at 25 guineas a head in crayons’.16 Some fifty finished works survive from Liotard’s two years in London in the 1750s, the majority portraits in pastel, including ‘one of the summits of his art’, the commission from Augusta, Princess of Wales (1719–1772) for pastel portraits of herself, her late husband and her nine children.17 Given the importance of this commission, it is peculiar that Liotard interrupted his work on the pastels in the summer of 1754 to visit family in Lyon. When he returned early the next year, it was with The Lavergne Family Breakfast.

Liotard left us some helpful clues about his stay in Lyon. Not only did he sign and date The Lavergne Family Breakfast with its location – ‘a lion’ – but he also advertised his return to London. In the Public Advertiser of 13 March 1755, he placed the following listing: MR. LIOTARD gives Notice that he is come back to London, chiefly in order to finish some Portraits he had begun before he went to France last Summer; and therefore does not intend to make here a longer Stay than will be required for that Purpose. He has brought over a Couple of large Conversation Pictures in Crayons, of his highest finishing.18 As he did not receive final payment for the royal pastels until August 1755, these are likely to be among the portraits that he returned to London to finish.19 Among the ‘large Conversation Pictures in Crayons’ was The Lavergne Family Breakfast. This pastel does indeed represent Liotard’s ‘highest finishing’, both in the sophistication of its forms and in his meticulous attention to detail. From the individual flyaway hairs that have come loose from their hairstyles to the pins holding the woman’s apron in place, no detail has been overlooked.

Yet for all the attention he gives to painting his figures, Liotard was deliberately lax in denoting who they actually were. Liotard brought The Lavergne Family Breakfast back to London as a ‘Conversation Piece’. It was never intended as a portrait, but rather as a generic depiction of a woman and child. He therefore sometimes referred to the figures as his ‘nieces’ and at other times as ‘a lady having in front of her a Chinese teaset, and giving a cup of coffee to her daughter.’20 In the Bessborough sale of 1801 the picture appears with various descriptions: first, as a manuscript addition, ‘Liotard. A Domestic Scene – a Drawing in Crayons of a Lady and a Little Girl at Breakfast’, then in print in the second edition of the catalogue as ‘a lady and child at breakfast, in crayons, an inimitable performance’.21 In the 1802 sale it was described as ‘portraits of a young lady and a gouvernante at breakfast’.22 By the time of the 1916 sale to Wertheimer, the figures were described in The Times as ‘a lady in striped dress serving chocolate to a young girl in a yellow frock.’23 The title by which the painting is now known is a twentieth‐century invention, reflecting our knowledge of where the picture was painted and who sat for it.

In 1713, Liotard’s elder sister Sara (1692–1757) had married François Lavergne (1678–1752), a businessman in Geneva who relocated to Lyon with his wife and their ten children in the early 1730s.24 An eleventh child would be born in Lyon. Three of Sara’s eight daughters had died by the time of Liotard’s 1754 visit to Lyon, but Neil Jeffares has convincingly argued that the adult figure depicted in The Lavergne Family Breakfast could be one of three of the artist’s surviving nieces: Catherine (1723–1757), Marguerite (1727–1789 or after) or Anne‐Andrienne (1728–1768), who would have been about 31, 27 or 26 respectively when Liotard painted his pastel.25 The little girl, who looks to be about five or six years old, may be Anne Delessert (1749–1802), the orphaned child of Sara’s daughter Jeanne (1720–1749).26 Ten years after The Lavergne Family Breakfast was painted we have an account from Jean‐Jacques Juventin, a Genevan who stayed with the family while passing through Lyon. By his description, they were warm and welcoming – ‘so united, so happy, so accommodating’ – their daughters the archetype of good and gracious Genevan ladies.27 Perhaps this is one of the reasons why Liotard made repeated visits to stay with them.

The Lavergne Family Breakfast was not the only work in which Liotard made use of his Lyonnais family. On the contrary, he seems to have counted on Lyon as a sort of entrepôt during his travels: a city in which he could pause, stay with family for extended periods, and use them as models for pictures to sell elsewhere. He had already done so in 1746 for a pastel known as La Liseuse (fig. 3): whether the girl reading the letter is Anne Lavergne (1717–1788) or Marie‐Anne Lavergne (1734–1809) – and her age remains a matter of contention – Liotard was very clear that he had used one of his nieces as a model. Liotard brought La belle liseuse with him to Paris and London, made several later versions of the composition, and was even commissioned by some of his British sitters to paint them in its guise.28 In late 1754 or early 1755 when he brought The Lavergne Family Breakfast to England, Liotard also had with him L’Ecriture, for which one of his nephews, Jacques‐Antoine Lavergne (1724–1781) or perhaps Hugues Lavergne (1732–1768), had sat (fig. 4). Although painted two years earlier in 1752, and not a pendant in the strictest sense, the compositional and tonal harmonies between this and The Lavergne Family Breakfast are striking.

Fig. 3

Jean‐Etienne Liotard, The Artist’s Niece, Marianne Lavergne (‘La Liseuse’), 1746. Pastel on vellum, 54.5 × 43 cm. Amsterdam, Rijksmuseum. © Rijksmuseum, Amsterdam

Fig. 4

Jean‐Etienne Liotard, L’Ecriture, 1752. Pastel on paper, 81 × 107 cm. Vienna, Kunsthistorisches Museum. © Heritage Images / Fine Art Images / akg‐images

Liotard’s Materials and Technique in The Lavergne Family Breakfast

Despite the numerous treatises about pastel painting written in the eighteenth century, the practical details of how an eighteenth‐century pastellist moved from a prepared sheet of paper to a fully finished work is less apparent. In terms of Liotard’s practice, his Traité des principes et règles de la peinture gives only key principles and rules rather than detailed descriptions of his materials or method.29 Instead, it is the evidence from the study of Liotard’s works themselves that provide the clearest insights into his likely working method,30 along with information in the papers and notes taken by Princess Karoline Luise von Hessen‐Darmstadt (1723–1783), later Markgräfin von Baden.31 The information about the materials and technique of The Lavergne Family Breakfast presented below is based on the non‐invasive study of the painting using stereomicroscopy, infrared reflectography and X‐ray fluorescence (XRF) scanning.32 However, it is the very essence of pastel painting that makes its technical study challenging: with multiple applications using numerous individual pastel crayons that have been so carefully blended, it can be difficult to interpret the application sequence and method of working. Amongst Liotard’s papers is a text describing the manufacture of a preparation of glue and ground pumice for application to the paper support,33 but it is unclear from visual inspection of the edges of the painting whether such a preparatory layer was used for The Lavergne Family Breakfast.

Infrared reflectography reveals a full underdrawing executed using a carbon‐based material, likely black chalk or possibly charcoal (fig. 5).34 Interpreting the infrared reflectograms is challenging because of the nature of pastels and how they are used, since types of marks made when applying darker colours in the final painting are easily confused with any preliminary drawing. However, it is clear that the whole composition was drawn in outline. The extent to which this outline was worked up to show areas of shadow is more difficult to be certain about, but the child definitely has areas of hatching, most clear in the shadow around the ear, and something similar may be present for shadows around the woman’s left shoulder and cleavage. It is also possible that the underdrawing was further worked up using chalks or pastels of other colours, such as red or white, which cannot be made visible with current infrared technology. No sketches or preparatory material survive for this composition, which is not unusual: indeed, hardly any such material by Liotard is known. Since pastel was his preferred medium, most of the preparatory work is likely to have been undertaken directly on the painting support and subsequently hidden by the finished picture.35 Here, the simple outlining for the figures has been followed carefully and only a few small changes have been found. The chairs were more freely sketched, that for the woman possibly starting as a different, simpler shape, so there are more underdrawn lines visible here that are not followed in the final composition.

Fig. 5

NG6685_5 Infrared reflectogram of NG6685. © The National Gallery, London

Once the drawing stage was complete, XRF scanning confirms that umber, vermilion and various white pigments were used in The Lavergne Family Breakfast to model the flesh. This corresponds with Karoline Luise’s notes taken during her lessons with Liotard, where she records that he recommended the use of vermilion, umber and white for the face, presumably using the blue of the paper that would still have been visible at this stage as a mid‐tone.36 The distribution of vermilion shown in the mercury XRF map does appear to be mainly related to the underlayers but it should be noted that many of these pigments were also used in more superficial layers, complicating interpretation of the XRF element maps (see fig. 6, fig. 7, fig. 8, fig. 9, fig. 10, fig. 11, fig. 12 and fig. 13).37 In addition to the hatching for shadows mentioned above, the infrared reflectograms show discrete dark areas where the pigment is more smoothly blended in some of the deeper shadows, for example those in the woman’s face and shoulder but also in the cups and saucers (fig. 5). These dark areas could be examples of the underpainting stage.

Fig. 6

Detail of the girl in The Lavergne Family Breakfast: NG6685, the pastel. © The National Gallery, London

Fig. 7

Detail of the girl in The Lavergne Family Breakfast: NG6685, XRF map for tin (Sn‐K line) showing areas containing tin white. © The National Gallery, London

Fig. 8

Detail of the girl in The Lavergne Family Breakfast: NG6685, XRF map for mercury (Hg‐L line) showing the distribution of vermilion in the underlayers and at the surface. © The National Gallery, London

Fig. 9

Detail of the girl in The Lavergne Family Breakfast: the oil painting (fig. 2). © courtesy the owner / photo: The National Gallery, London

Fig. 10

Detail of the woman in The Lavergne Family Breakfast: NG6685, the pastel. © The National Gallery, London

Fig. 11

Detail of the woman in The Lavergne Family Breakfast: NG6685, XRF map for tin (Sn‐K line) showing areas containing tin white. © The National Gallery, London

Fig. 12

Detail of the woman in The Lavergne Family Breakfast: NG6685, XRF map for mercury (Hg‐L line) showing the distribution of vermilion in the underlayers and at the surface. © The National Gallery, London

Fig. 13

Detail of the woman in The Lavergne Family Breakfast: the oil painting (fig. 2). © courtesy the owner / photo: The National Gallery, London

According to Karoline Luise’s notes, the main composition would then be developed through repeated application and blending of intermediate layers of colour, with application of ‘the beautiful colours and lights’ reserved until the end.38 Unlike working in oils, in pastel paintings the dark, strong colours are normally applied before the addition of lighter tones.39 According to Karoline Luise’s instructions, the backgrounds should be applied just after the underlayers, except when painting a profile, and should be applied smoothly, avoiding rubbing. As in The Lavergne Family Breakfast the application tends to be thinner in the background than in the figures, and the heads in particular.40 For the depiction of flesh, over the underlayer (worked with vermilion, umber and white pigments) Karoline Luise describes the addition of ‘half shadows with unburnt ochre, broken with blue or violet’ and the ‘lights with minium [red lead] shading’.41 XRF scanning broadly supports this description, with the most superficial palest pink passages of flesh being rich in lead, suggesting the predominant use of red lead (minium) and lead white rather than vermilion.42 Light‐sensitive red lake‐containing pastel crayons are also likely to have been used,43 and comparison of the pastel and oil paintings does suggest that there may have been some loss of colour in the flesh and particularly in the little girl’s face in the National Gallery picture.

In addition to the use of lead white and chalk as white pigments,44 XRF scanning indicates the presence of another highly unusual white pigment, tin white (tin (IV) oxide) (see fig. 14).45 Tin white is referred to occasionally in the documentary sources but it is only in pastel paintings by Liotard that its use has been confirmed in western European painting.46 It appears to have been deemed unsuitable for use in oil or distemper and was reputed to blacken in sunlight. It is more commonly referred to as being used in cosmetics, as a vitreous colour (particularly for enamelling) or as a pigment for watercolour or manuscript illumination. Liotard had initially trained as a miniature painter, which may have influenced his decision to use this pigment.47 The tin white has been applied in quite specific areas, for example in the bright white details such as the very edge of the girl’s proper right cuff, her paper curlers and the jewels in the woman’s hair. It also appears in the whitest parts of the woman’s apron and cuffs and many areas of impasto highlight on these fabrics and in the still life elements. In these areas, where the pigment protrudes from the surface the pastel must have been applied wet, probably as a paste created by mixing powdered pastel crayon with water, gum or fixative (fig. 15).48 Tin was also detected in specific regions of the flesh where the light catches the skin and in the pale pink parts of the stripes on the woman’s dress (see fig. 6, fig. 7, fig. 8, fig. 9, fig. 10, fig. 11, fig. 12, fig. 13 and fig. 14). In these particular areas of the flesh and the dress, the fibres of the paper support either cannot be seen or appear matted into the pastel layer (fig. 16). This is in contrast to most of the rest of the surface of The Lavergne Family Breakfast, where raised paper fibres are clearly visible, even where there has been blending of the dry pigment powder. Beneath the powdery pink pastel of the flesh tones (which contain lead‐based pigments), a whitish layer can be seen in the areas where tin is detected, which appears to be a tin white‐containing layer. That the paper fibres cannot be seen in these areas may simply result from this rather thicker layering of pastel, or perhaps suggests that – like the impasto highlights – the layer containing tin white was applied wet and worked with a brush or wetted finger.

Fig. 14

NG6685, XRF map for tin (Sn‐L line) showing the presence of tin white at or near the surface of the painting. © The National Gallery, London

Fig. 15

NG6685, detail under raking light illumination. © The National Gallery, London

Fig. 16

NG6685, detail showing the change in surface texture of the pastel painting in the girl’s décolletage region. The flesh areas show smooth passages as well as those where the more usual velvety quality of pastel is visible. © The National Gallery, London

Detailed examination of the surface of the painting combined with the observation that tin white is present in those areas of the pastel where the fibres of the paper support are less apparent, together with evidence from XRF for the use of vermilion in underlayers, suggests that to create the flesh Liotard started with the vermilion‐containing underlayers over which he added a layer rich in tin white, applied wet.49 Finally, dry applications of pastel were added over the tin white‐containing layer to create the flesh tones. The application of tin white was presumably intended to make the pale areas of the skin and silk dress particularly luminous, but whether the surface seen today is as Liotard originally intended is less certain. The Lavergne Family Breakfast and L’Ecriture (fig. 4, above) were his first large‐format paintings and because of their size they are executed on paper rather than parchment, the support he is perhaps more associated with. It is possible that the present rather unusual appearance of the flesh reflects the choice of support. Liotard may have been less familiar with working on paper, applying the pastel too thickly in an attempt to achieve the luminosity that is perhaps easier to obtain with parchment.50 It is also possible that the appearance of the surface has altered over time. For example, the pastel applied over the tin white underlayers may not have adhered so well, resulting in delamination and perhaps necessitating some intervention or reworking in the past.51

However, if the pastel and oil paintings are compared, the way the flesh is treated is very similar – the lightest areas appear as quite distinct paint applications, with pronounced pink brushstrokes seen over a paler or white layer (see fig. 6, fig. 7, fig. 8, fig. 9, fig. 10, fig. 11, fig. 12 and fig. 13), particularly obvious in the girl’s chest and the nape of the woman’s neck. This perhaps suggests that Liotard’s handling of the flesh was a deliberate choice, or maybe simply reflects a desire to mimic the pastel as closely as possible in the oil painting.52

Another unusual feature was also observed in the areas of pastel that lie over the wet applications of tin white. Numerous small inclusions that look translucent and glass‐like under magnification can be seen within the layers of pastel.53 Rather than being a component of the pink surface layers, these inclusions seem to be associated with either the tin white application or possibly a layer applied between the tin white and the final flesh layer, perhaps added to provide tooth for the overlying applications of colour.54

In addition to those already mentioned, other pigments used in The Lavergne Family Breakfast include a blue copper‐based pigment (probably azurite) and Prussian blue. The wide variety of yellow, orange and brown tones are based on various earth pigments, umber, vermilion and orpiment. In the black of the table and the stripes of the woman’s dress, ivory or bone black were used, likely in combination with more carbon‐rich black pigments (for example, vegetable blacks such as lamp black or charcoal or black chalks or earths).55 In addition to its use in the flesh colours, in some of the deep browns the presence of red lead (minium) is suspected, used in combination with black.

Although there was much debate at the time about the difficulties and relative merits of fixing pastel paintings, with many artists relying instead on preservation of their pastels under glass, the The Lavergne Family Breakfast seems likely to have been fixed to have survived as well as it has. In a letter to the 2nd Earl of Bessborough (who owned the Breakfast) written in 1763, Liotard recommended another Swiss artist, Sébastien Jurine (1722–1779), as a fixer of pastels and indicates that Jurine had fixed pastels for him and for one of his patrons.56 However, it remains unclear whether Jurine worked on this painting and there are no visual clues other than the condition to ascertain whether it was indeed fixed or not.

Depicting Taste: Tea, Coffee, Chocolate

Although in the end he sold the pastel version of The Lavergne Family Breakfast to his greatest patron, Viscount Duncannon, for the exceptionally high price of 200 guineas, Liotard painted the picture speculatively, in the hope rather than the knowledge that someone would buy it. This might seem surprising given the size and ambition of the composition but, in some ways, we can see Liotard making a calculated decision with The Lavergne Family Breakfast. In 1752 he had painted L’Ecriture, his largest pastel to date, and while it was admired, this piece – depicting a man at a writing desk and a young boy holding a lit candle – had yet to find a buyer. It would not be until 1762, almost a decade after it was painted, that the Empress Maria Theresa of Austria would purchase it. With The Lavergne Family Breakfast, Liotard may have been seeking a more immediate success. His choice of sitters and subject matter would certainly suggest so.

Throughout the eighteenth century, paintings of people drinking tea, coffee and chocolate became extremely popular. This was due in no small part to the fact that these luxurious, expensive beverages, which had to traverse half the globe to reach the breakfast table, enjoyed considerable prestige, and coffee most of all. That there were some 380 coffee vendors in Paris in 1723 and 2,000 on the eve of the French Revolution illustrates the enormous growth in coffee’s consumption in France, which was mirrored elsewhere.57 It may seem strange today to think of a six‐year‐old drinking coffee for breakfast, but it was deemed important in the eighteenth century to cultivate a child’s palate. Taste is what Liotard plays on in his composition. There are the literal tastes we are invited to imagine when looking at the painting – the milky coffee, the sweet sugar, the crunch of the bread – and there is aesthetic taste, the refined taste of those depicted in the picture and, by extension, the person who bought it. In choosing to depict a woman and a child drinking coffee, Liotard was deliberately painting an elite activity for an elite audience.

The growing popularity of these beverages saw an increase in pictorial representations of people enjoying them. One close example to The Lavergne Family Breakfast is A Lady taking Tea of 1735 by Jean‐Siméon Chardin (1699–1779) (fig. 17). Against a grey, minimally described interior, a woman in a striped dress stirs a spoon through a steaming cup of tea. Although the failures of Chardin’s perspective in this painting have been much commented upon, it is striking that his composition, like Liotard’s, depicts the back of a chair at far left, a fashionable striped gown and a lacquered tea table with the same motif of the open drawer. Is it possible that Liotard saw Chardin’s painting before taking on his own breakfast scene? A Lady taking Tea is signed on the back of the original canvas ‘ce tableau / a este fait en / fevrier 1735’, though it was not exhibited until the Salon of 1739 by which time Liotard was already in Constantinople.58 The painting was soon engraved by Pierre Filloeul (1696–after 1754), though the fact that both Chardin’s painting and Liotard’s pastel share the same orientation of the female figure suggests that Liotard had seen the painting rather than the engraving, where the image was reversed. The earliest known provenance for A Lady taking Tea is a sale in London on 26 February 1765, where it was acquired by Dr William Hunter (1718–1783),59 so it is just conceivable that Liotard saw the picture in London in the 1750s, prompting the compositional choices made in The Lavergne Family Breakfast. In any case, the differences between the two paintings are as notable as their compositional similarities. Chardin, the still‐life painter, depicts the decorations on the lacquer and porcelain in loose calligraphic strokes, seeking, as Marianne Roland Michel has noted, to evoke the type of object without describing it.60 It is Liotard, nominally the portraitist, who describes his still life in painstaking detail.

Fig. 17

Jean‐Baptiste‐Simeon Chardin, A Lady taking Tea, 1735. Oil on canvas, 81 × 99 cm. Glasgow, The Hunterian Museum and Art Gallery. © Hunterian Museum & Art Gallery / Bridgeman Images

The attention Liotard paid to the still life in The Lavergne Family Breakfast is further evidence of the good taste he was seeking to depict. We see here the juxtaposition of commonplace objects with extremely refined ones, a practice Liotard borrowed from the seventeenth‐century Dutch painters he so admired and later collected.61 The coffee pot on the table is of a standard mid‐century design, presumably made of pewter, not silver, since its tones are subtly different from those of the silver spoons. The milk pot is an undecorated, cream‐coloured faience‐ware, the type of everyday object commonly used across France. The table‐top, however, so shiny as to almost appear mirrored, is black Japanese lacquer finely decorated with gilded patterns. Such objects were popular in eighteenth‐century Europe but they were also as expensive as they were desirable. Since lacquer also originated in Asia, had a brilliant sheen and, for eighteenth‐century audiences, an almost magical aura to it, it earned the name ‘black porcelain’.62 This refined display is intended to tell us something about the figures we see before us. The tea cups and saucers continue the story.

The cups and saucers in The Lavergne Family Breakfast are a study in opposites. One is empty, while the other is so full as to almost overflow, the milky coffee within it a beautiful depiction of surface tension. On one saucer, we see the handle of a teaspoon; on the other, the bowl of the spoon is more prominent. What is perhaps less immediately obvious, however, is the fact that the two cups and saucers, for all the similarity of their bright blue and red designs, do not form a matching pair. For someone so meticulous in his attention to detail, and rigorous in demands to follow nature, this must have been a deliberate decision on Liotard’s part, even if faithful to the Lavergne family’s belongings. The cups and saucers we see here are Japanese Imari‐ware.63 They can be identified as such by their distinctive designs, which include red chrysanthemums and jagged blue rocks. Imari was a style of porcelain originating in Arita, Japan, exported to Europe in large quantities between the mid‐seventeenth and early eighteenth centuries by the Dutch East India Company.64 Their designs were later taken up by Chinese porcelain factories, which produced similar wares en masse for the European market, but Chinese Imari was less finely decorated than its Japanese counterparts and less delicate in its forms.

These cups and tea bowls can be identified as Japanese not only because of the gradation of blue tones in the decoration, the obvious thinness of their rims and the care taken with elements such as the fine hatched border and the scroll on the underside of the saucer, but also by comparing them to a set today in the Porzellansammlung in Dresden (fig. 18). We know that Europeans could tell the difference between Japanese and Chinese porcelain in the eighteenth century, and that original porcelains were highly prized.65 Liotard, who ‘like very few other artists … was able to render the different materials and surface qualities that constituted the charm of the novel import goods from Asia’, has chosen to paint objects that were not brand new.66 Rather, being perhaps thirty or even fifty years old and not part of a matching set, they seem to have been chosen for their refinement and rarity, Liotard again signalling not only the good taste of the sitters but also that of the painting’s future purchaser.

Fig. 18

Japanese Imari‐ware cup and saucer, about 1690–1720. Hard‐paste porcelain. Dresden, Staatliche Kunstsammlungen, Porzellansammlung. © Porzellansammlung, Staatliche Kunstsammlungen Dresden, photo: Adrian Sauer

Porcelain, and more specifically the accoutrements for drinking tea, coffee and chocolate, were a lifelong fascination for Liotard.67 Around 1744 he painted The Chocolate Girl, whose elegant profile and minutely depicted tray of beverages earned the accolade ‘the most beautiful pastel that one had ever seen’ (fig. 19).68 In 1752, he painted another breakfast scene, The Breakfast, which he would also sell to a British patron (fig. 20).69 Following his departure from London, he painted Dutch Girl at Breakfast, his love letter to Dutch art. Although the latter is painted in oil and on a much smaller scale, we nevertheless see the same minutely observed apparatus of coffee‐drinking, the same fascination with reflective surfaces (fig. 21). Towards the end of his life Liotard dispensed with the figures altogether in such works, painting a series of still lifes of tea sets without the people drinking from them, the finest of which is today in the J. Paul Getty Museum (fig. 22).70 If The Lavergne Family Breakfast has been called ‘a still life with coffee set and two humans in attendance’, the unpeopled Still Life: Tea Set is the distillation of this impulse, removing the figures altogether.71

Fig. 19

Jean‐Etienne Liotard, The Chocolate Girl, about 1744. Pastel on parchment, 82.5 × 52.5 cm. Dresden, Gemäldegallerie Alte Meister. © Gemäldegalerie Alte Meister, Staatliche Kunstsammlungen Dresden, Gal.‐Nr. P 161, Photo: Wolfgang Kreische

Fig. 20

Jean‐Etienne Liotard, The Breakfast, about 1752. Pastel on parchment, 67.3 × 54 cm. Munich, Alte Pinakothek. © Bayerische Staatsgemäldesammlungen ‐ Alte Pinakothek München

Fig. 21

Jean‐Etienne Liotard, Dutch Girl at Breakfast, about 1756. Oil on canvas, 46.8 × 39 cm. Amsterdam, Rijksmuseum. © Rijksmuseum, Amsterdam

Fig. 22

Jean‐Etienne Liotard, Still Life: Tea Set, about 1781–3. Oil on canvas, 37.8 × 51.6 cm. Los Angeles, J. Paul Getty Museum. Digital image courtesy of the Getty’s Open Content Program

Perhaps more than any other of his breakfast and beverage scenes, The Lavergne Family Breakfast is a virtuoso performance in illusionism, volume and reflective surfaces. It is impossible, however, to look at these images today without also taking into account that which is not visible. For each cup of chocolate, each cafetière, each bowl of sugar, each piece of Japanese porcelain, there was a whole world of global trade connections behind it; in the case of the chocolate, coffee and sugar, one reliant on the traffic of human beings. In 1769, the French writer Jacques‐Henri Bernardin de Saint‐Pierre noted: These beautiful pink and red colours in which our women dress themselves; the cotton with which they pad their skirts; the sugar, the coffee, the chocolate they have for breakfast, the rouge with which they relieve their paleness: the hands of miserable Black people have prepared all of this for them.72 Eighteenth‐century consumers would have been aware of these connections. For all that Liotard’s works attract a vocabulary of luminosity and brilliance, the reality behind his choice of subject matter was far darker.

Revisiting The Lavergne Family Breakfast

It was not clear, when Liotard left London in 1755, whether he expected to see The Lavergne Family Breakfast again. His stint in the British capital had been extremely profitable, so it was not out of the question that he would return in the future, but given the sizeable undertaking of such a journey and his marriage in the Netherlands relatively soon after departing, it cannot have been a given for him that he would ever again see this most highly prized work. The Lavergne Family Breakfast, however, certainly lingered long in Liotard’s imagination. He wrote of it in his autobiography in 1760, and this mention was taken up in many contemporary biographies of the artist.73 In late 1772, Liotard returned to London. In 1773, he returned to his painting.74

By the time of this second visit, the artist was 70 years old. He had long since shaved off the beard that had caused him such great renown, and the celebrity that he had enjoyed in the 1750s had diminished somewhat over the subsequent decades. This seems to be reflected in the relative number of works that survive from Liotard’s second British sojourn: while some fifty finished works are known from the period 1753–5, including the significant commission for the royal family, only about twenty from the period 1772–4 are recorded. The London art world to which he returned had changed. The Royal Academy had been founded; there was now a more established British school to speak of, and an institution celebrating that school and training its next generation of artists. There were now other figures working in Liotard’s preferred medium of pastel, such as John Russell (1745–1806). But the smaller number of commissions on this second London visit may also have been due to cultural animosity stoked by the Seven Years’ War against France and all things – including artists – French. Liotard’s return to The Lavergne Family Breakfast in 1773, and the decision to make a painstaking copy in oil paint, must be a result of finding fewer patrons.

The oil version of The Lavergne Family Breakfast measures 81.4 × 103.2 cm (see fig. 2). Although the figures in the oil painting sit slightly lower in the composition, when the two works are overlaid the figures, chairs and still life are almost identical (fig. 23). The extreme closeness between the two compositions is startling: even the complicated pattern of the cane backs of the chairs is in alignment. To achieve such proximity between the compositions Liotard must have used a tracing to duplicate his design: indeed, infrared reflectography of the oil painting reveals dotted lines in the underdrawing, suggesting the use of a pounced cartoon or similar approach for transfer (fig. 24). Pastels, after all, had to be glazed, so it would have been minimally disruptive to lay a translucent surface on top to make such a tracing. Liotard appears to have made use of a comparable traced drawing earlier in his career with The Chocolate Girl. An ink drawing in Geneva, executed across four sheets of paper, depicts a life‐size reproduction of the figure from this celebrated composition (fig. 25). As this drawing shows no physical signs of being used as a preparatory sketch, and as it remained in Liotard’s family until it entered the museum in Geneva, it is most likely a tracing made from the pastel, perhaps with a view to making a future replica of it. If so, it is the only tracing by Liotard known to survive, and the quality of its lines would seem to tally with the use of a tracing on either paper or possibly muslin.75

Fig. 23

Detail of the overlay of NG6855 and fig. 2 at 50% transparency. © The National Gallery, London

Fig. 24

Infrared reflectogram of fig. 2. Private collection. © The National Gallery, London

Fig. 25

Jean‐Etienne Liotard, The Chocolate Girl, about 1744. Brown ink on four sheets of paper, 83.8 × 58.5 cm. Geneva, Musée d’art et d’histoire, 1935‐0008. MAH Musée d'art et d'histoire, Ville de Genève. Don de la Société auxiliaire du Musée, 1934. © MAH Genève photographe Bettina Jacot‐Descombes

Liotard was not averse to making copies of his genre scenes, though the exactitude of the repetition in the oil version of The Lavergne Family Breakfast is unusual. Four versions exist of the Dame pensive sur un sofa, for example, although their differing dimensions – from 23.5 × 19 cm at smallest to 103.8 × 79.8 cm at largest – shows that this was a question of repeating rather than tracing a composition.76 Five autograph versions of Dame et sa servante au bain are known, four in pastel and one in oil, but while their dimensions are much closer, there are small differences between the compositions, such as the inclusion or exclusion of floor tiles and the differing distances between the two figures.77 The most pertinent examples for The Lavergne Family Breakfast are the two versions of the Femme au tambourin, one in pastel and one in oil, which demonstrate a comparable closeness – almost every detail has been replicated, save a few elements of the embroidered cushion – although there is no evidence to suggest such a great length of time between versions as with The Lavergne Family Breakfast.78

While no tracing or preparatory material survives for the National Gallery pastel, the closeness between the two versions makes the differences between them all the more interesting. There are two significant changes. The first is that the bright blue decorations of the Imari porcelain have, in the oil, turned brown. In the pastel, Liotard has used a crayon containing Prussian blue for the decoration, while in the oil painting he appears to have used smalt, a blue pigment that often loses its colour over time.79 Liotard clearly intended the decorations to match the bright blue of the pastel, as their reflections in the lacquer table, particularly those of the sugar bowl, are still blue (likely the result of having used a different pigment). The second change is that of the signature on the sheet of music, which now reads ‘Liotard / a londres / 1773’. There are other, more subtle differences in the oil painting, such as the greater amount of space left above the figures; the greater prominence of the shadow (this is visible in the pastel, and clearly seen with infrared reflectography, but much more subtle); and the absence of the filigree‐fine decoration on the inside rim of the tray beneath the little girl’s fingers. Yet what is extraordinary about the oil painting is how closely it emulates the pastel.

It is not merely a repetition of the composition: rather, it reads almost as an oil imitating a pastel. Every fold of fabric is replicated; every stroke of light and dark paint to create the reflective sheen of the woman’s dress mimics that of the pastel – even those areas of highlight in the flesh which read somewhat peculiarly today (see above). But where Liotard could drag the side of his pastel across the surface to create a mid‐tone between the pale pink and the red of the shadow, in oil he has to use three different tones: dark, medium and light (figs 26, 27). Perhaps most interesting are the moments when Liotard has acknowledged the difference between the two media by using reverse techniques to achieve the same effect. This can be seen in the wispy, flyaway hairs around the two figures’ heads. In his pastel, these hairs are drawn with a very fine, sharpened chalk on top of the background. In the oil, he leaves an unpainted reserve around the heads that he only partly fills to create the same impression (figs 28, 29). The hair itself is of a far more uniform colour in the oil, while looking closely at the pastel it is possible to see strokes of pink, blue, taupe and grey.

Fig. 26

Comparison of a detail of the woman’s bodice in the pastel (NG6885) and oil (fig. 27) versions of The Lavergne Family Breakfast. © The National Gallery, London

Fig. 27

Comparison of a detail of the woman’s bodice in the pastel (fig. 26, NG6885) and oil versions of The Lavergne Family Breakfast. © courtesy the owner / photo: The National Gallery, London

Fig. 28

Comparison of the woman’s hair in the pastel (NG6885) and oil (fig. 29) versions of The Lavergne Family Breakfast. © The National Gallery, London

Fig. 29

Comparison of the woman’s hair in the pastel (fig. 28, NG6885) and oil versions of The Lavergne Family Breakfast. © courtesy the owner / photo: The National Gallery, London

Pastel has long been associated with softness. This is certainly true in this comparison. The pastel medium brings a smooth, velvety quality to the flesh tones and fabrics. The picture appears warmer; the figures are animated by the fleur, which results in pastel’s diffuse reflection of light and which eighteenth‐century viewers so prized.80 It could be said, by contrast, that the oil has the upper hand in the smooth reflective surfaces. As an inherently shiny medium whose pigment is suspended in and seen through the clear, glossy oil, it is perfectly suited to depict the slick black lacquer. The pink dress fabric is paler in the oil, contributing to the painting’s overall sense of coolness: this may be the result of a slight shift in the colour of the oil paint over time, while the pastel has retained its original hue. Yet the girl’s yellow dress seems better preserved in oil, as some of the details that gave volume to its folds in the pastel version appear to have faded.81

If Liotard sought to make money from this repetition of The Lavergne Family Breakfast, he did not succeed. He placed the picture in a sale of his works with Christie’s on 15 April 1774 – ‘A lady and her daughter drinking coffee’ – but it did not sell.82 A letter sent from Liotard’s son to his mother in 1778 implies that Liotard had taken the picture with him on his final visit to Vienna, but he had again failed to sell it.83 It was included in a list of paintings Liotard offered for sale to the French royal collections in 1785, which were refused with some extremely derogatory comments.84 The painting would in fact remain with Liotard’s descendants for a century, until 1873. The failure to find a buyer must have been a blow, since it took much longer to make a copy than an original: on 10 December 1777, Liotard wrote to his wife that the copy he was making of his earlier portrait of Madame Necker ‘will take me a lot of time’; indeed, Liotard would work on it for six months between November 1777 and May 1778.85 1773 was far from the end of Liotard’s career; he would go on painting for another 15 years. But this second visit to London marked a turning point. In his seventies, he was no longer the celebrity he had been in his forties and fifties. Pastel was no longer the fashionable medium it had been in the 1750s. Not even Viscount Duncannon, by this point the 2nd Earl of Bessborough and a man who owned more than a dozen works by Liotard, was loyal enough to buy the same composition twice.

Yet if the second Lavergne Family Breakfast was not the golden ticket Liotard had perhaps hoped it to be, it did go on to have an unusual almost‐reunion with the pastel that it imitated. By the turn of the twentieth century, the oil was in the collection of Baron Edmond de Rothschild, in Paris. In 1913 his son, James A. de Rothschild married Dorothy (Dolly) Pinto. Five years later, Dorothy’s father, Eugene Pinto, purchased the pastel. While there is no evidence that the two works were ever placed together then – nor, indeed, that anyone other than Liotard and Bessborough and a handful of others would have seen them together when the oil was completed in 1773 – they did belong to the same extended family for the first half of the twentieth century. As far as we know, the only time the two have been brought together was in the National Gallery’s 2023 exhibition.

Notes

1 Manuscript annotations in two versions of the catalogue of the Beauclerk sale (Christie’s, London, 27 March 1802) state that the picture was bought in at the Bessborough sale. (Back to text.)

2 For the Foreman inventory, see Sebag‐Montefiore 2016, pp. 151 and 169. (Back to text.)

3 Her niece, Mrs Margaret Grace Mary Williams, later placed an advertisement in Country Life seeking the whereabouts of the picture, see ‘Whereabouts of a picture’, letter, Country life, 12 September 1952, p. 770. (Back to text.)

4 Detailed examination of the surface of the pastel under magnification revealed some small fragments of glass (up to about 0.3 mm) in this area of the background, suggesting that at some stage there was a breakage or damage to a sheet of glass covering the pastel. (Back to text.)

5 The form of the damage may suggest that it resulted from slippage of a sheet of glass covering the pastel. (Back to text.)

6 At the edges of the pastel the support paper is visible, and under magnification blue‐coloured fibres are visible within the paper structure. X‐ray fluorescence (XRF) scanning also suggests there are small quantities of cobalt and nickel everywhere across the painting, suggesting smalt may have been added to the rag pulp to give a more intense and even colour. (Back to text.)

7 On the left, the main sheet is extended by the addition of two smaller pieces of paper, one above and one below. These two joins appear to be butt joins. On the right, two smaller pieces of paper are positioned above the main sheet. The nature of the join between the main sheet and the additional piece directly above it is unclear, but the smallest, uppermost piece appears to extend below the other additional sheet suggesting an overlay join. There is a long vertical join down the centre of the support connecting the right‐ and left‐hand assemblages but it is unclear whether this represents a butt or overlay join. (Back to text.)

8 Several layers of old, non‐original brittle brown gummed paper tape now cover the back and sides of the strainer bars which may at some point have extended in places onto the front turnover edge. However, where this later addition of tape has detached at the back of the painting revealing the edges of the blue rag paper, the blue paper appears to have been adhered to the strainer, beyond the extent of the canvas, suggesting the painting retains its original construction. (Back to text.)

9 For a general description of methods to prepare supports for pastel paintings see Shelley 2011. See also Jeffares 2016a, pp. 15–18. (Back to text.)

10 The only other documented instance of Liotard making a second version of a picture directly from the original after such a significant gap of time is his portrait of Suzanne Curchod, later Madame Necker (Schönbrunn Palace, Vienna). Painted in Geneva in 1761, this pastel was sold to the Empress Maria Theresa during Liotard’s second visit to Vienna in 1762. Returning to Vienna for a third time in 1777–8, he made a copy of the picture for himself (now lost). See Jeffares 2018a. (Back to text.)

11 Roethlisberger and Loche 2008, vol. 1, p. 62; Liotard 1781, pp. 48–9, 57–8. (Back to text.)

12 For a summary of contemporary biographies of Liotard, see Roethlisberger and Loche 2008, vol. 1, pp. 71–92. (Back to text.)

13 Walpole 1786, p. 194; Roethlisberger and Loche 2008, vol. 1, p. 465, no. 299. (Back to text.)

14 See Whitlum‐Cooper 2023, passim and figs 1 and 2, and Jeffares 2018b. (Back to text.)

15 We know Liotard’s address from various notices he placed advertising the exhibition of his works (see the Public Advertiser, 21 November 1753 and 11 January 1754). In 2015, Neil Jeffares identified Liotard as the ‘Turkish Gentleman, lately arrived here, who is very eminent in Portrait Painting … and remarkable by his Beard being long, curiously sharped [sic] and curled’ (see Jeffares 2015). (Back to text.)

16 Joshua Reynolds, in Northcote 1818, vol. 1, p. 60. (Back to text.)

17 See Roethlisberger and Loche 2008, vol. 1, pp. 431–2. (Back to text.)

18 The Public Advertiser, 13 March 1755. (Back to text.)

20 ‘Autobiographie de 1760’, reproduced in Roethlisberger and Loche 2008, vol. 1, p. 62; Liotard 1781, p. 48. (Back to text.)

21 London, Christie’s, 5–7 February 1801, 3rd day, lot *75. See copies at the Société Royale des Antiquaires / Koninklijk Oudheidkundig Genootschap, Rijksmuseum, Amsterdam and the Centre National de Documentation pour l’Histoire de l’Art et l’Iconographie / Rijksbureau voor Kunsthistorische Documentatie, The Hague. (Back to text.)

22 London, Christie’s, 27 March 1802, lot 78. (Back to text.)

23 ‘£6105 for two Gainsboroughs’, The Times, 29 July 1916, p. 5. (Back to text.)

24 For a full genealogy of the Lavergne family, see Jeffares 2018b, p. 11. I am indebted to Neil Jeffares’s ongoing and comprehensive investigations into the genealogical records, which have set straight many confusions in previous scholarship. (Back to text.)

25 See Jeffares 2018b, pp. 13–18 (Back to text.)

26 Ibid. , p. 11. (Back to text.)

27 Jean‐Jacques Juventin to his uncle, 22 October 1764, transcribed and published by Neil Jeffares, in Jeffares 2018b, p. 14. (Back to text.)

28 I am very grateful to Francis Russell for sharing with me his thoughts on Liotard’s reuse of this composition and facilitating visits to some of these works. See Roethlisberger and Loche 2008, vol. 1, nos 159, 160, 292–4. (Back to text.)

29 Other treatises on pastels and pastellists written in the seventeenth and eighteenth centuries are also a valuable source of information. For the principal texts of relevance see Jeffares 2016b. (Back to text.)

30 For other technical studies of works by Liotard see Sauvage and Gombaud 2015; Gombaud and Sauvage 2016; Wallert, Vaz Pedroso and Iperen 2016. (Back to text.)

31 The princess’s notes, dated 1745–6, are available online at Landesarchiv Baden‐Württemberg n.d. and are transcribed in the original German in Jeffares 2016b, pp. 21–2 and transcribed and translated into French in Roethlisberger and Loche 2008, vol. 1, pp. 110–11. Among her papers there is also a shade card sent to her in 1746 with samples of 90 pastels from the Swiss pastel master Bernard Augustin Stoupan, a supplier also used by Liotard. These swatches have been analysed by micro‐XRF spectroscopy and/or polarised light microscopy: see Gombaud and Sauvage 2016. (Back to text.)

32 When the painting was reframed and the plate glass replaced with low‐reflective laminated glass there was an opportunity to undertake high resolution and raking light photography, and record images of the surface under magnification. Infrared reflectography was carried out using an Apollo digital infrared camera with an indium gallium arsenide (InGaAs) array sensor; XRF scanning was also undertaken using a Bruker M6 Jetstream XRF at a spatial resolution of 580 microns with dwell time of 10 ms. While XRF scanning was helpful in identifying many of the pigments used, it should be noted that the technique cannot detect elements below phosphorus in the periodic table. Thus, compounds based on elements of low atomic number such as kaolin (potentially used as a filler in the pastel sticks) or any organic colorants (such as indigo, lake pigments and most black pigments) which may have been used to tint the pastels cannot be detected. (Back to text.)

33 See Roethlisberger and Loche 2008, vol. 1, pp. 112–13. For a broader discussion of supports for pastels see Jeffares 2016a, pp. 15–18. (Back to text.)

34 In her notes (see note 31, above), Karoline Luise suggests using charcoal for preparatory drawings (and then fixing the final design with red chalk and dusting off the black lines). In the case of this finished pastel, the black underdrawing can still be observed which may suggest the use of black chalk rather than the more friable charcoal medium. (Back to text.)

35 Unfinished pastels by Liotard, such as the Portrait of Archduke Maximiliaan Franz van Oostenrijk, 1778 (SK‐A‐1199) and the Portrait of Joseph II of Austria, 1778 (SK‐A‐1198) in the Rijksmuseum and the possibly unfinished, possibly damaged Portrait of a Grand Vizir, or of a European dressed as one at the National Gallery (NG4460), reveal some of the underdrawing and underpainting (or undermodelling) that must have gone into Liotard’s finished pastels. (Back to text.)

36 See no. 16 in Roethlisberger and Loche 2008, vol. 1, p. 110. (Back to text.)

37 For comparison see Sauvage and Gombaud 2015 and Gombaud and Sauvage 2016 where transmitted light photography was used to visualise vermilion‐containing underlayers. (Back to text.)

38 See no. 16 in Roethlisberger and Loche 2008, vol. 1, p. 110. Translations are the authors’ own. (Back to text.)

41 See no. 5 in Roethlisberger and Loche 2008, vol. 1, p. 110. Translations are the authors’ own. (Back to text.)

42 It is possible that other lead‐containing pigments such as the yellow pigment massicot (lead (II) oxide) may also have been used in the flesh, and the surface layers probably also contain some vermilion, orpiment, tin white and chalk (either as a white pigment or a filler in pastel crayons). To depict the pale, translucent skin of the girl, a pastel crayon containing a blue copper‐based pigment was laid in before the application of the surface flesh layers. See also Karoline Luise’s notes in Roethlisberger and Loche 2008, vol. 1, pp. 110–11 for use of blue in flesh paint. (Back to text.)

43 In her notes, Karoline Luise refers to carmine and violet lake pigments for example, and a range of other pastels based on organic colorants are referred to in contemporary treatises; see Jeffares 2016b. (Back to text.)

44 Calcium is detected throughout the painting and in many areas is probably related to the use of calcium carbonate or sulphate (usually in the forms of chalk or gypsum) as inert fillers in pastel crayons, mixed with pure pigment(s) to create a range of lighter hues. However, chalk has also been used as a white pigment, for example in the slightly off‐white fabric of the little girl’s cuffs. (Back to text.)

45 With certain elements, including tin, it is possible to create XRF element distribution maps using more than one characteristic element line. The XRF element map created using the less energetic tin L (Sn‐L) element line is shown in fig. 14 and reveals the use of tin white near or at the surface. A map can also be created with the higher energy tin K (Sn‐K) line which can reveal the presence of tin white lower in the stratigraphy and details of this map are shown in figs 6 and 7. (Back to text.)

46 Tin white has been identified in other works by Liotard in the Rijksmuseum, reported to be the first confirmed use of this pigment in western European painting. In these other works, tin white has not been used in the flesh tones but appears elsewhere in the paintings. The use of bismuth white in the flesh tones and as a white pigment has been noted in other works by Liotard but does not appear to have been used in the National Gallery pastel. See Sauvage and Gombaud 2015; Gombaud and Sauvage 2016; Wallert, Vaz Pedroso and Iperen 2016. (Back to text.)

47 For more information about tin white as a pigment see Harley 2001, pp. 172–3. Some of the problems reported with tin white as a pigment are likely to have been less pronounced in pastel painting where the amount of organic binder is minimal, exposure to light is generally limited and the presence of glass over the pastel may offer some protection from poor environmental conditions. (Back to text.)

48 The results of XRF scanning show the very raised highlights on both the pewter and ceramic pots contain tin and lead, probably present in the forms of tin white and lead white. It is unclear whether they were combined within a single pastel crayon that was then used wet or if powder from several crayons was combined. For further discussion of the use of wet pastel, see Burns 2007, Shelley 2011 and Jeffares 2016a, p. 37. (Back to text.)

49 The tin‐rich passages seen in the raking light images correlate more closely with the tin XRF map created using the Sn‐K element line than that created with the Sn‐L line, indicating that the tin white application associated with the change in surface texture is not necessarily in the most superficial layers in all cases but can lie deeper in the stratigraphy in some areas. See also note 45, above. (Back to text.)

50 Neil Jeffares, personal communication, 2023. (Back to text.)

51 Occasional brushmarks are visible in the pink pastel application at the surface of the painting in some areas of the flesh, suggesting working into wet paint or a wet application at some stage. (Back to text.)

52 It should be noted that in the oil painting the flesh of the woman’s proper left arm is also painted in this manner but interestingly in the pastel painting the tin white layer is not present on this arm and the altered surface texture is not seen. (Back to text.)

53 The particles are up to about 0.2 mm in size. XRF did not assist in identifying the nature of these particles and sampling was impossible. (Back to text.)

54 It is also possible that these inclusions are sand or fragments of ground glass employed as an abrasive, possibly worked with a pumice stone or similar object and perhaps used to create tooth to level or even erase areas of pastel. Neil Jeffares, personal communication, 2023. (Back to text.)

55 The presence of ivory or bone black is based on the co‐location of calcium and phosphorus by XRF scanning. While ivory or bone black contains some elemental carbon, in areas that look particularly dark in the infrared reflectogram, particularly if calcium and phosphorus are not detected, the presence of black pigments that are richer in carbon are suspected. It should also be noted that various different black pigments might be combined within a single pastel crayon. (Back to text.)

56 See Anderson 1994, pp. 23–5. See also Jeffares 2006a and Jeffares 2014. (Back to text.)

58 Rosenberg et al. 1999, p. 216, no. 46. (Back to text.)

61 Following his departure from London in 1755, Liotard settled in the Low Countries, where he acquired a number of seventeenth‐century Dutch paintings. For Liotard as a collector, see Roethlisberger and Loche 2008, vol. 1, pp. 120–3. (Back to text.)

63 The identification of these pieces of porcelain would not have been possible without conversations with many colleagues in the decorative arts who have shared their knowledge so generously, especially Mia Jackson, Patricia Ferguson, Denise Campbell, Errol Manners and Julia Weber. (Back to text.)

64 For an in‐depth look at Imari, see Nagatake 2003 and Schiffer 2000. (Back to text.)

66 Weber 2018, pp. 68–9. (Back to text.)

67 See Moon 2023, pp. 82–99. (Back to text.)

69 The Breakfast, now in the Alte Pinakothek, Munich, was owned until 1786 by the 2nd Earl of Harrington. (Back to text.)

70 For a close study of this work, see Lippincott 1985. (Back to text.)

73 See Roethlisberger and Loche 2008, vol. 1, pp. 71–91. (Back to text.)

74 Neil Jeffares discovered a mention of Liotard in the minutes of the Society of Arts in November 1772, thus confirming his return to London earlier than had previously been noted (Roethlisberger and Loche 2008 give 1773). See Jeffares 2006b. (Back to text.)

75 With many thanks to Marie‐Eve Celio and her colleagues at the Musée d’art et d’histoire, Geneva, for sharing their thoughts on the tracing process with us. (Back to text.)

76 Roethlisberger and Loche 2008, vol. 1, nos 190, 295, 296 and 348. (Back to text.)

77 Ibid. , nos 67, 68, 69, 297 and 298. See also Smentek 2018. (Back to text.)

78 Roethlisberger and Loche 2008, vol. 1, nos 65 and 66. (Back to text.)

79 There is possibly a little copper‐containing blue or green pigment in localised areas of the cup nearest the girl. In the oil version of The Lavergne Family Breakfast the blue details on the porcelain now appear brown and it is assumed that they must have been painted using smalt, but there is no evidence for the use of smalt for the blue details on the pastel painting. For the deterioration of smalt in oil paint see Spring et al. 2005. (Back to text.)

80 Shelley 2011, p. 15. (Back to text.)

81 A range of pigments for pastel crayons based on organic colorants that would be prone to fading are referred to in treatises written in the seventeenth and eighteenth centuries, including ‘stil de grain’, a yellow lake made from unripe buckthorn berries. See Jeffares 2016b and note 43, above. In addition to the fading of lake pigments, loss of colour could also be associated with the use of orpiment which degrades to the white compound arsenolite. (Back to text.)

82 Christie’s, London, 15 April 1774, lot 76. (Back to text.)

83 Letter from Liotard’s eldest son to his mother, sent from Delft, 24 August 1778, cited in Roethlisberger and Loche 2008, vol. 1, no. 495, p. 621. (Back to text.)

84 Jean‐Baptiste Pierre (1713–1789), first painter to the king, called Liotard a charlatan. See Roethlisberger and Loche 2008, vol. 1, pp. 128–9. (Back to text.)

85 Jean‐Etienne Liotard to his wife, 10 December 1777, reprinted in Roethlisberger and Loche 2008, vol. 2, pp. 745–6. See also Jeffares 2018a, pp. 13–14. (Back to text.)

List of references cited

Anderson 1994
AndersonJ., ‘Fixing Pastels: A Letter from Liotard to the 2nd Earl of Bessborough in 1763’, The Burlington Magazine, January 1994, 136109023–5
Burns 2007
BurnsT.The Invention of Pastel PaintingLondon 2007
Ferguson 2016
FergusonP.F., ‘“Japan China”, Taste and Elite Ceramic Consumption in 18th‐Century England: Revising the Narrative’, in The Country House: Material Culture and Consumption, eds J. Stobart and A. HannSwindon 2016, 113–22
Gombaud and Sauvage 2016
GombaudC. and L. Sauvage, ‘Liotard, Stoupan and the Colours Available to 18th‐Century European Artists’, in Sources on Art Technology: Back to Basics, eds S. Eyb‐Greenet al.London 2016, 115–23
Harley 2001
HarleyRosamond D.Artists’ Pigments c.1600–1835: A Study in English Documentary Sources, 2nd revised edn, London 2001
Herda‐Mousseaux et al. 2015
Herda‐MousseauxR.M.et al.Thé, café ou chocolat? Les boissons exotiques à Paris au XVIIIe siècle (exh. cat. Musée Cognacq‐Jay, Paris), Paris 2015
Jeffares 2006a
JeffaresNeil, ‘Jurine, Sébastien’, in Dictionary of Pastellists before 1800http://www.pastellists.com/Articles/Jurine.pdf, accessed August 2023, 2006 (online edn in, Pastels & Pastellists, updated 2021)
Jeffares 2006b
JeffaresNeil, ‘Liotard, Jean‐Étienne’, in Dictionary of Pastellists before 1800http://www.pastellists.com/Articles/Liotard.pdf, accessed January 2024, 2006 (online edn in, Pastels & Pastellists, updated 2023)
Jeffares 2014
JeffaresNeil, ‘Loriot, Pellechet, Jurine: The Secrets of Pastel’, Pastels & Pastellistshttp://www.pastellists.com/Essays/Loriot.pdf, accessed January 2024, 2014 (updated, 2023)
Jeffares 2015
JeffaresNeil, ‘Liotard and the Medium of Pastel’, in Jean‐Etienne LiotardM. StevensC. Bakeret al. (exh. cat. Scottish National Gallery, Edinburgh; Royal Academy of Arts, London, 2015–16), London 2015, 27–33
Jeffares 2016a
JeffaresNeil, ‘Prolegomena to Pastels & Pastellists’, Pastels & Pastellistshttp://www.pastellists.com/Misc/Prolegomena.pdf, accessed 2023, 2016 (updated, 2019; updated, 2023)
Jeffares 2016b
JeffaresNeil, ‘Treatise & Other Historical Texts Related to Pastels & Pastellists’, Pastels & Pastellistshttp://www.pastellists.com/Misc/Treatises.pdf, accessed 2023, 2016 (updated, 2023)
Jeffares 2018a
JeffaresNeil, ‘Liotardiana’, Pastels & Pastellistshttp://www.pastellists.com/Essays/Liotardiana.pdf, accessed 2023, 2018
Jeffares 2018b
JeffaresNeil, ‘Liotard’s Le Déjeuner Lavergne’, Pastels & Pastellistshttp://www.pastellists.com/Essays/Liotard_Dejeuner_Lavergne.pdf, accessed 2023, 2018 (updated, 2019; updated, 2023)
Koja and Enke 2018
KojaStephen and Roland Enke, eds, ‘The most beautiful pastel ever seen’: The Chocolate Girl by Jean‐Étienne Liotard (exh. cat. Staatliche Kunstsammlungen Dresden, Gemäldegalerie Alte Meister, 2018–19), Munich 2018
Kopplin and Aken‐Fehmers 2004
KopplinM. and M.S. van Aken‐Fehmers, eds, Schwartz Porcelain: The Lacquer Craze and its Impact on European Porcelain (exh. cat. Museum für Lackkunst, Münster; Schloss Favorite, Rastatt), Munich 2004
Landesarchiv Baden‐Württemberg n.d.
Landesarchiv Baden‐Württemberg, ‘FA Nr. 5 A Corr 96, 35: Generalregeln für die Erstellung eines Pastellbilds nach Liotard von Karoline Luises Hand’, Bildexplorerhttp://www.landesarchiv-bw.de/plink/?f=4-4553199-1, accessed 2023, Karlsruhe n.d.
Liotard 1753
LiotardJean‐Étienne, in Public Advertiser, 21 November 1753
Liotard 1754
LiotardJean‐Étienne, in Public Advertiser, 11 January 1754
Liotard 1755
LiotardJean‐Étienne, in Public Advertiser, 13 March 1755
Liotard 1781
LiotardJ.E.Traité des principes et des règles de la peintureGeneva 1781
Lippincott 1985
LippincottL., ‘Liotard’s “China Painting”’, The J. Paul Getty Museum Journal, 1985, 13121–30
Moon 2023
MoonI., ‘Of Pastels and Porcelain’, in Discover Liotard & the Lavergne Family BreakfastF. Whitlum‐Cooper (exh. cat. The National Gallery, London, 2023–4), London 2023, 82–99
Nagatake 2003
NagatakeT.Classic Japanese Porcelain: Imari and KakiemonTokyo 2003
Northcote 1818
NorthcoteJ.The Life of Sir Joshua Reynolds, LL. D., F.R.S., F.S.A. &c., Late President of the Royal Academy: Comprising Original Anecdotes of Many Distinguished Persons, his Contemporaries: and a Brief Analysis of his Discourses2 volsLondon 1818
Roethlisberger and Loche 2008
RoethlisbergerMarcelRenée Lochewith Bodo Hofstetter and Hans BoeckhLiotard: catalogue, sources et correspondance2 volsDoornspijk 2008
Roland Michel 1994
Roland MichelMarianneChardinParis 1994
Rosenberg et al. 1999
RosenbergP.et al.Chardin (exh. cat. Galeries nationales du Grand Palais, Paris; Kunstmuseum im Ehrenhof, Düsseldorf; Royal Academy of Arts, London; The Metropolitan Museum of Art, New York, 1999–2000), London 1999
Sauvage and Gombaud 2015
SauvageL. and C. Gombaud, ‘Liotard's Pastels: Techniques of an 18th‐Century Pastellist’, in Studying 18th‐Century Paintings and Works of Art on Paper, eds H. Goodman and K. MuirLondon 2015, 31–45
Schiffer 2000
SchifferN.Imari, Satsuma, and Other Japanese Export Ceramics, revised and expanded second edn, Atglen 2000
Sebag‐Montefiore 2016
Sebag‐MontefioreCharles, ‘A Regency Collection: Luke Foreman (1757–1814) and his wife Mary (1764?–1834)’, Furniture History, 2016, 52143–80
Shelley 2011
ShelleyM., ‘Painting in the Dry Manner: The Flourishing of Pastel in 18th‐Century Europe’, in Pastel Portraits: Images of Eighteenth‐Century Europe, eds K. Baetjer and M. Shelley (exh. cat. Metropolitan Museum of Art, New York), 2011, 5–56 (The Metropolitan Museum of Art Bulletin, Spring 2011, 6845–56)
Smentek 2023
SmentekK., ‘Jean Etienne Liotard, A Lady in Turkish Dress and her Servant, about 1750’, in French Paintings and Pastels, 1600–1945: The Collections of the Nelson‐Atkins Museum of Art, ed. Aimee Marcereau DeGalanhttps://doi.org/10.37764/78973.5.322, accessed 2023, online edn, April 2023
Spring et al. 2005
SpringM.et al., ‘Investigation of Pigment‐Medium Interaction Processes in Oil Paint containing Degraded Smalt’, National Gallery Technical Bulletin, 2005, 2656–70
Times 29 July 1916
£6105 for two Gainsboroughs’, The Times, 29 July 1916, 5
Wallert, Vaz Pedroso and Iperen 2016
WallertA.J. Vaz Pedroso and J. Iperen, ‘Remarkable White Pigments on Liotard’s Pastel Paintings’, in Painting Techniques: History, Materials and Studio Practice, ed. A. Wallert (Rijksmuseum symposium, 2013), Amsterdam 2016, 168–73
Walpole 1786
WalpoleH. and G. VertueAnecdotes of Painting in England: with some account of the principal artists; and incidental notes on other arts4 vols (1765–71), third edn, London 1786, 4
Weber 2018
WeberJ., ‘What struck Algarotti as Chinese in Liotard’s Chocolate Girl’, in ‘The most beautiful pastel ever seen’: The Chocolate Girl by Jean‐Étienne Liotard, eds Stephen Koja and Roland Enke (exh. cat. Staatliche Kunstsammlungen Dresden, Gemäldegalerie Alte Meister, 2018–19), Munich 2018, 68–75
Whitlum‐Cooper 2023
Whitlum‐CooperF.Discover Liotard & the Lavergne Family Breakfast (exh. cat. The National Gallery, London, 2023–4), London 2023
Williams 1952
WilliamsMargaret Grace MaryMrs, ‘Whereabouts of a picture’, Country Life, 12 September 1952, 770

List of exhibitions cited

London 2023–2024
London, National Gallery, Discover Liotard & The Lavergne Family Breakfast, 16 November 2023–3 March 2024

About this version

Version 1, generated from files NG_2024__16.xml dated 30/08/2024 and database__16.xml dated 20/09/2024 using stylesheet 16_teiToHtml_externalDb.xsl dated 30/07/2024. Document created from design‐ready Word document and prepared for digital publication.

Cite this entry

Permalink (this version)
https://data.ng.ac.uk/01A7-000B-0000-0000
Permalink (latest version)
https://data.ng.ac.uk/01A5-000B-0000-0000
Chicago style
Whitlum‐Cooper, Francesca, Catherine Higgitt, Lynne Harrison and Rachel Billinge. "NG6685, The Lavergne Family Breakfast". Version 1, September 20, 2024. https://data.ng.ac.uk/01A7-000B-0000-0000.
Harvard style
Whitlum‐Cooper, Francesca, Higgitt, Catherine, Harrison, Lynne and Billinge, Rachel (2024) NG6685, The Lavergne Family Breakfast. Version 1. Available at: https://data.ng.ac.uk/01A7-000B-0000-0000 (Accessed: 27 September 2024).
MHRA style
Whitlum‐Cooper, Francesca, Catherine Higgitt, Lynne Harrison and Rachel Billinge, NG6685, The Lavergne Family Breakfast, version 1 (2024) <https://data.ng.ac.uk/01A7-000B-0000-0000> [accessed: 27 September 2024]