Skip to content

The National Gallery’s Eighteenth Century French catalogue

4 February 2019


Humphrey Wine’s long-awaited catalogue of the French eighteenth century pictures has now appeared. I shall leave it to others better qualified than I to discuss broader aspects of the book’s achievement and limitations – indeed Wine’s own text is candid about the deficiencies and historical reasons for the NG’s coverage of the dix-huitième compared with its holdings in virtually all other schools and periods.[1]

In the hopes that the NG will soon put its wonderful series of catalogues online, I thought it might nevertheless be helpful to record a few minor observations and tangential remarks I made on a quick read through the book (concentrating on the chapters that interested me). Several years ago I was able to offer a number of comments (on the texts concerning the NG pastels) which the author has generously acknowledged, but further points have arisen since the text was completed (in May 2016 – the delay will confuse many readers, particularly as a few parts, such as the index, were compiled later). I’ve passed over minor typos. Some of the corrigenda below refer to entries by Wine’s contributors (which perhaps he was unable to review): 13 of the 72 entries, and 23 of the 32 artist essays are by others.


Page 16. Pillement returned to France in 1760, not 1763.

Page 27. Watteau’s Le Sommeil dangereux in Liotard’s sale: the 1773 catalogue lists 120 guineas as an asking price; it wasn’t sold then. It was finally sold in 1788 for “un vil prix”. The reference cited is Glorieux (who had not seen the 1773 catalogue and relied on Graves: there is a copy in the Frick, from which I prepared the entries in my exhibitions document in the online Dictionary of pastellists) but not the later discussion in Roethlisberger & Loche 2008, p. 153, and the sources cited there.

NG 1090 Boucher

Page 66 n.25. “Gaspard de Sireul”: this form, with the particle, is repeated in the index, although the biographical dates added, 1713–1781, were first published by me in November 2016 in my article on “Jean-Claude-Gaspard Sireul”.

Louis de Boullongne

Page 72. “The year 1722 also brought admission to the Ordre de Saint-Michel… In November 1724 Louis de Boullogne [sic] was ennobled.” However widely repeated, this sequence is not possible: the order can only be conferred on nobles. In Boullongne’s case the explanation is that he was already noble, having acquired the office of Conseiller du roi … en la Chancellerie près le parlement de Rouen in 1718. The 1724 letters so often cited merely made his status more ostensible.[2] Pedantic the point may be, but there is some historical interest in knowing whether the king was prepared to exempt his premier peintre from the chivalric rules everyone else had to observe, or indeed that a second application of savonnette à vilain could be required.

NG 1664 Chardin

Page 85. There are three, not two, pastel self-portraits in the Louvre.

Pages 90, 94. In an image where the cistern is 5 cm tall, the woman’s height is closer to 13 cm than the 17 cm used in this computation. This means the capacity of the cistern is significantly underestimated – at about half the correct number (the linear error is cubed).

Page 103 line 1 and p. 110. Jean-Jacques Lenoir, sgr de La Motte (1707–1796), the son of Alexandre Lenoir, marchand orfèvre à Paris, married Marie-Josèphe Rigo or Rigault in Paris, 12 février 1730 (AN mc/x/388), not 1731.

NG 6598 Danloux

Page 117. Danloux (whose first name was I think properly spelled Henry, not Henri) married Marie-Pierrette-Antoinette de Saint-Redan on 28 juillet 1787, not 1785 (AN mc/xxvi/761).

Page 123. Fig. 8 (J.257.109 in the online Dictionary of pastellists) is a copy (not in pastel but coloured chalk) of the lost Groult oval drawing (in chalk and watercolour) of an unknown lady.

Page 139 n.136. “Jeffares 2011” (my Gazette des Beaux-Arts article on Pommyer) is cited but not found in the bibliography; an update of that article can be found here.

NG 6495 David

Page 143. Exhibitions: The Dutch in Paris 1789–1914, Amsterdam, 2017–18.

The catalogue (in accordance with its rules) appears to make no mention of the other great David in the NG collection, from the nineteenth century.

NG 4253 Drouais Vaudreuil

Page 177. Marguerite Mathilde Slidell, Mme Erlanger died 18 février 1927, in Paris.

Page 183. Étienne-François, known as the duc de Choiseul (as he is correctly called on p. 191), was the younger, not older, brother of the comte de Stainville. (Stainville is omitted from the index, while Choiseul gets two entries.)

NG 6440 Drouais Pompadour

Page 187. The absence of information on frames is here particularly regretted, as this picture has had three frames recently: see n.12 to my post where I cite and expand Peter Schade’s note in The National Gallery Review of the Year 2009–2010, pp. 23ff, which itself is omitted from the bibliography.

Page 187. Related works: there are a number of omissions, three copies in pastel alone, nor I think is there mention of the version at Welbeck in 1936 (Goulding no. 254). It is unhelpful that the list on p. 187 has to be compared with p. 190 and then n.21 on p. 197 – which sends one to a different publication entirely. Neither the version sold Drouot, 10 June 1988, Lot 57 (and again at Neuilly, 14 December 1995, Lot 24; Drouot, 21 October 1999, Lot 124; Tajan, 18 December 2018, Lot 38) nor that at Monaco, Sotheby’s, 26 June 1983, Lot 494 seems to have been listed.

Page 190. Reference to the Valade pastel is to J.74.201 (and see my article), but the title is given erroneously as of “Madame Faventines de Fontenille”, daughter-in-law of the Mme Faventines intended.

La Tour’s famous pastel in the Louvre, which has so many relevant connections, is mentioned only in relation to the books it shows her to have read (p. 197, n.20).

Page 197 n.17. The picture sold in Paris, Sotheby’s, 25 June 2008, Lot 66 is not in pastel and is not the one in the inventory.

Page 197 n.26. Thiéry was “employé dans les vivres”. “Veuve Godefroi” deserves her own name: née Marie-Jacobe Van Merle ( –1764).


Page 200. The essay on Ducreux retains such myths as “possibly only pupil” of La Tour. Marie-Antoinette was never Louis XV’s “daughter-in-law”, nor was she Antoinette-Clémence’s godmother. (Georgette Lyon’s book is unreliable, as in the next sentence she transcribes the girl’s baptismal entry, where the marraine is her sister, not the queen of France.) The daughter who modelled for Greuze was not “Antoniette” as misspelled here but her sister Rose-Adélaïde. There were at least two more sons.

NG 1882 “French eighteenth century”

Page 216. This is probably a copy of the 1815 engraving by S. A. Oddy for Smollett’s England, which (unlike the Wille, Basan and Anker Smith prints suggested) is in the same sense as the miniature.


Page 220. The dauphine was Marie-Josèphe, not Marie-Louise, de Saxe (although she is indexed with the right dates). Severus’s nomen was Septimius, not Septimus.

NG 5584

On the NG website, the (supposed) sitter’s dates are still given as “about 1732 – 1795” but no dates seem to be given in the catalogue. I published (on Twitter, two years ago) her marriage record giving her age (13 at the time, 26 December 1748).

NG 1019

Page 229. Charles-François-René Mesnard de Clesle’s dates were 1732–1803, not as given.

The 1783 Vigée Le Brun portrait of La Reine en gaulle is elsewhere described as in Eichenzell (p. 526) or Kronberg (p. 518); the picture in the NGA Washington is a later copy.

Page 232. In this article the imperial pouce has been converted at the incorrect 2.54 cm instead of the correct 2.707 cm used in other parts of the book (so for example 14.5 pouces becomes 36.8 instead of 39.25 cm).

NG 1393 Lacroix

Page 245. Why would Grenier de La Croix be related to Charles-François de Lacroix, who had a different family name?

Page 245. Boyer de Fonscolombe was born 1716, not 1719, and died 1789, not 1788. Mrs Tarratt’s dates were 1813–1893 (she died in Cheltenham on 16 June 1893); her son Daniel predeceased her, in 1888, and so could not have inherited from her.

NG 6663 Lagrenée l’aîné

Page 252. Lagrenée’s year of birth is disputed – 1724 or 1725; but when he was appointed Conservateur et administrateur honoraire du Musée he was probably 79, not 80.

Page 253. The bibliography has been omitted. The Goncourt reprint of the Lagrenée list is conveniently available online (nor is there any reference to J. J. Luna’s Lagrenée article in Archivo español de arte, xlvi, 1973, pp. 35ff). The entry omits the price (3000 livres, notably high for two small pictures) of the two pendants sold to Lord Shelburne. Isn’t it quite possible that the painting in Lansdowne’s 25 February 1806 sale was bought in, and simply represented on 19–20 March 1806?

The Morellet–Lansdowne correspondence was published in 1898 (it’s been available online at EE since 2011), and includes references to Mme Geoffrin and Joseph Priestley, who noticed a resemblance between the children and his own; it deserves more extensive discussion, not least because it proves that the subject of the Shelburne pictures included children. The frames and transport costs are also mentioned.

The reading of the Mémoires secrets as interchangeable in hang is too contrived; it surely refers simply to the vagueness of the titles.

The 1806 purchaser, Mr Taylor, cannot be George Watson-Taylor, who only added “Taylor” to his name in 1815 (having married a Miss Taylor in 1810, four years after the sale where “Taylor” is recorded). As Taylor also bought pictures at the Lansdowne sales that year by Titian, Guercino, van Slingeland, van Gool and Pietro Fabris, he was probably a dealer. I suggest he was almost certainly Josiah Taylor (1771–1850). A colourful character who deserves a higher profile, he was the proprietor of the St James’s Gallery of paintings at 58 Pall Mall, where he had previously run a gaming house with Crockford and mixed with high society (the Duke of Wellington is described as “godfather” to Taylor’s son, baptised six months after his birth in 1817). The Lagrenée must have been sold before Taylor’s bankruptcy led to a series of London auctions between 1828 and 1837, when more than 3700 old master paintings were disposed of (including some 70 on copper).

NG 3883 Largillierre

Page 318. Mme de Souscarrière’s dates are known (1684–1733) – they were printed in the Mercure de France, 1733, p. 2089: she died 12 September 1733, aged 49. She married Bosc in January 1704, leaving two daughters.

NG 5118 La Tour, Dawkins

This is J.46.1612 in the online Dictionary of pastellists.


Page 340. This alludes to my observation (first published in 2016 here and reprinted here) that Liotard’s contract with Massé was of allouage, not apprentissage, but the contract did have a “pedagogical” element – “[Massé] promet montrer [Liotard] tout ce dont il se mêle et entremêle dans l’art de la peinture”. What was different was that it did not lead to maîtrise.

Page 340. The greatest works listed in the autobiography were four in number; the self-portrait described could be that in Geneva or the Uffizi picture. The Uffizi inscription spells Vienne correctly.

NG 4460 Liotard, Grand vizier

This is J.49.2425 in the online Dictionary of pastellists.

NG 5586 Nattier Balletti

Page 352. Exhibitions: omits Portraits anciens, Société artistique des amateurs, Galerie Jacques Seligmann, hôtel de Sagan, mars–avril 1933, no. 29, where the work was mentioned in all the contemporary reviews and reproduced in Vanderpyl, “Portraits d’autrefois”, Le Miroir du monde, 8 avril 1933, p. 6. (These were also overlooked by Xavier Salmon in his 1999 exhibition catalogue.) More recently the picture was in the Casanova: the seduction of Europe exhibition in Fort Worth, San Francisco and Boston, 2017–18.

The picture has been widely reproduced beyond the references cited.

Page 358 n.1. The Dictionary (J.46.2972) reproduces the La Tour pastel of Silvia said to be known only from the print. The provenance of the La Tour pastel and the Nattier oil are intertwined with the intricacies of the Balletti/Fortier pedigree, for which see here.

As to the argument that NG 5586 might have been commissioned by one of Manon’s suitors, the pedigree illustrates (as is suggested by the eight family portraits in the 1759 inventory) that the Balletti family had connections with a number of artists, starting with Rosalba Carriera (as I signalled on this blog and in my annotated transcription of her diaries): Margherita Balletti, the wife of composer Bononcini and Manon’s aunt, sought advice on miniature painting from Carriera, who in turn made a miniature of her brother-in-law, Luigi Riccoboni. Also active in Manon’s circle was the artist Balletti mentions in a later part of the 7 February 1760 letter to Casanova cited on p. 356. In this she refers to the amateur actor in her group (rather more talented as an artist), Louis-Michel Brun, dit Lebrun, peintre du roi, who is known as a miniaturist (although practically none of his work has survived) and nephew of Jean-Baptiste Van Loo. My recent research (see the article on his daughter Rosalie, Mme Vaïsse and genealogy) reveals that his wife was a Blondel, so the relationship may have been even closer than the letter suggests. He was probably the miniaturist Lebrun to whom the princesse de Talmont wished to lend La Tour’s pastel of Prince Charles Edward Stuart to be copied in 1759 (the suggestion – elsewhere – that this was Louis-Michel’s father Michel Brun, dit Lebrun cannot be right, as he died in 1753). Was he (or his mother – “de Lebrun” could be dame or dite) the Lebrun present at Tocqué’s marriage to Marie-Catherine-Pauline Nattier in 1747?

The group of black and red chalk drawings formerly attributed to Nattier, including the one of “Mlle Balevi”, remain problematic. Should we consider among the possible artists Nattier’s son-in-law, the painter Charles-Michel-Ange Challe, whose widow bequeathed a couple of pastels “ouvrage fait avec soin par Mr Challe quoi que ce ne fut point son genre”? To the numerous examples known (which go well beyond those cited on p. 358: Phyllis Hattis first started to list them in 1977) should now be added the double portrait[3] of Mlle Baron et sa mère, formerly in the Schwitter collection: its quality perhaps requires us to reconsider the attribution of others in the group – not least because the two figures in the drawing seem to show quite different techniques without being by different hands. Were we (going back to Xavier Salmon’s Apollo article in 1997, in the days when articles with 47 footnotes were published there) too hasty in dismissing Nattier? As noted by Philippe Renard (Nattier, 1999, p. 166f, where the six drawings from the Guérault sale are reproduced), Nattier’s daughter Mme Challe mentioned in her will “…un petit cadre noir et or renfermant sous verre six têtes de la famille dessinées par mon père” and “un grand cadre renfermant sous verre huit têtes de la famille et d’amis.”

NG 6435 Perronneau Cazotte

Page 370. The painting appeared in the recent Perronneau exhibition in Orléans in 2017 (no. 56).

Page 378. The suggestion that Perronneau asked Cazotte to wear a costume not his own strikes me as far-fetched. The wig is indeed the style Beaumont called “à la mousquetaire” – but indistinguishable from the ailes de pigeon style, the much commoner term, employed for Vaudreuil (p. 181) – which makes the idea of dating from it questionable.

Page 378. It is perhaps worth noting that Cazotte’s large collection included no pastel, providing a hint as to why Perronneau was commissioned to paint him in oil (the sitter’s choice).

NG 4063 Perronneau Legrix

This is J.582.1522 in the online Dictionary of pastellists. The pastel appeared in the recent Perronneau exhibition in Orléans in 2017 (no. 59).

Page 382. Marthe-Marie-Madeleine Legrix married Dublan not in 1759, but 19 juin 1760, in Talence.

NG 3588 “Perronneau” Girl with kitten

This is J.582.189 in the online Dictionary of pastellists. See the entry for a fuller bibliography, including for example Florence Ingersoll-Smouse.

Page 388. Provenance (p. 396 n.4: “There is no certain reference to NG 3588 in the Duveen Brothers records”): for an extended discussion of the information gleaned from the multiple references in the Duveen records, see my blog post.

Page 390. The presence of zinc and tin in recent examination of the “Perronneau” does raise questions which in the present state of our knowledge of materials used in pastel (whether pigments themselves, or the fillers or binders) cannot be definitively answered. But the occurrences in two Liotard pastels (both in his studio at the end of his life) are arguably less relevant than Chaperon’s manual of Paris procedures which Perronneau (who, unlike Liotard, was no maverick, and who had not travelled to the East) is more likely to have followed in 1743. The reasons for questioning the pastel’s authenticity are primarily connoisseurial.

Page 391. “Demoyel” was a misprint in Guiffrey’s 1869 reprint of the 1746 livret (and repeated elsewhere); the name was originally printed Desnoyel as I have at J.582.129. The child might be one of the sons of Charles Desnoyel ou Desnoyelles, a maître charcutier, rue Saint-Honoré; the cock then a reference to the family business.

Page 392. In the quote from me “[weaknesses]” is supplied where “pastiche elements” is intended.

Page 394. The “certain Brown” was Eric Brown, director of the National Gallery of Canada, with whom Ricketts had an extended correspondence.

Page 396 n.3. The “Two pastels by Perronneau of the same lady in different positions” in Duveen’s stock books were the pastels later sold to Edward Stotesbury (see my blog post).

NG 903 Rigaud studio

Page 433. “recently”: 2004.

NG 5588 After Roslin

Page 435. Numerous related works have been omitted, including an engraving by Gautier-Dagoty; see the list in for MV 6763, J.629.156 and the following items.

Page 438 n.15. François was indeed Fredou’s brother-in-law.

NG 4097 Tocqué

Page 472 & 603. “1920–1 London, Burlington Fine Arts Club”: it isn’t immediately obvious which exhibition this is. The painting is reproduced and discussed in The Connoisseur, lxii, 1922, pp. 199, 238, where the suggestion that it had been exhibited in the Burlington Fine Arts Club may have originated.

NG 3964 After Tocqué

Page 483 and n.30. The “hypothesis” I suggested was the identification of the Orléans sitter as Joseph-Thérèse, not his associé position.

Vigée Le Brun

Page 518. Did Mme Geoffrin and the duchesse de Chartres really commission Mlle Vigée c.1769? The claims of the latter are not verified, but at least link to her text (the earliest portrait known is 1778, and Vigée Le Brun often distorted dates); but Mme Geoffrin visited only as far as I know.

Jean-Baptiste-Pierre Le Brun was the great-great-nephew, not great-nephew, of Charles Le Brun.

NG 1653 Self-portrait

Page 519. The A. P. Pickering of 23 Queens Gardens was Arthur Proctor Pickering (1818–1902), a solicitor, and brother of the barrister Percival Andre Pickering QC (1810–1876) referred to on p. 530 n.1. The sale was not posthumous, nor apparently forced: A. P. Pickering’s estate was valued at nearly £25,000. The dealer identified here only as “S. T. Smith” was Samuel Theobald Smith (1842–1904), grandson of the famous John Smith who (in the 1820s) had handled the other (i.e. Rubens) Chapeau de paille now in the National Gallery.

Page 520. The only reason to assume that the self-portrait (lent by M. Péan de Saint-Gilles) from the 1891 exhibition of works from “the beginning of the [19th] century” should relate to the NG picture seems to be the description in Helm 1915, p. 207 (not cited): he must have seen it to offer the description, but he specifically separates it from the versions of the NG painting and suggests a later date. (It is however worth noting that Armand-Louis-Henri Péan de Saint-Gilles (1791–1860) was co-exécuteur of the will of Vigée Le Brun’s brother Louis-Jean-Baptiste-Étienne in 1820.)

Pages 527f. In addition to the numerous salon critiques cited, the original version is also discussed in the “Letter from a gentleman on a tour in Paris, to his friend in London”, which appeared in the Morning chronicle and London advertiser, 19 September 1783, but is omitted from standard bibliographies of salon criticism (e.g. McWilliam & al. 1991).

Page 530. Baillio’s 1987 letter, 31 years ago, should be viewed in the context of his decision to reproduce a print when the primary version of the painting was not available for the 2016 catalogue. Haroche-Bouzinac’s biography, of which the note on p. 552 is cited on p. 530, reproduced the NG on the cover of the first printing (January 2012, despite the copyright date of 2011) but replaced it with a reproduction of the original on the next impression (October 2013).

NG 5871 Mlle Brongniart

Page 538. Anne-Louise d’Aigremont was not Hazon’s granddaughter. Émilie’s brother Alexandre married, in 1800, Jeanne-Cécile Coquebert de Monbret (1782–1862), petite-fille de Michel-Barthélémy Hazon, who was only 19 years older than Mlle d’Aigremont. (The error also appears in Anne Lajoix’s article on Brongniart in The Sèvres Porcelain Manufactory exhibition catalogue, 1997.)

Page 539. I published the complete list of the petitioners here including a reproduction of Brongniart’s signature.

Page 543 n.2. Whether the “Adam” who bought the picture in 1897 was Charlotte Adam-Pichon or her mother is settled by consulting the 1901 exhibition catalogue, where the lender is explicitly mentioned as “Mme Adam, née Pichon”, i.e. the mother. The sources giving her death as 1896 are mistaken: online genealogies give it as “after 3 June 1896”, while consulting the État civil for Paris 8e reveals that in fact she died on 29 October 1929. Similarly (n.3) “one Mme Off” is a reference to Mme Adam’s other daughter, Rosalie-Josephine-Victoire (1870– ), who married, in 1896, one Georges Off.


[1] Perhaps a topic that may not be covered in the conventional reviews is that the NG’s aversion to the French eighteenth century in its permanent collection has been largely mirrored in its choice of temporary exhibitions: the pictures from Lille shown in 1993, Tradition & revolution in French art 1700–1880, tried to argue for a continuous tradition, a thesis anathema to any dix-huitiémiste (will the impending Boilly show be able to avoid this?), while the London leg of the Pompadour show in 2002 was such a pale version of the Versailles exhibition that it elicited scathing treatment from Richard Dorment and Alden Gordon – the latter blaming the then director rather than the NG curator.

[2] The numerous roturiers on whom it was conferred first had to buy ennobling offices before their admission. The small number of those (mostly foreigners) entitled by letter of the king to wear the decoration à titre honorifique were not members of the order: see Benoît Defauconpret, Les Preuves de noblesse au xviiie siècle, 1999, p. 86. Caix de Saint-Aymour’s 1919 monograph on Les Boullongne gives the clearest account of Boullongne’s case and reproduces in an appendix the 1722 document of admission to the order, including the 1718 proof of nobility which satisfied Clairambault.

[3] Lyon, Bérard Péron, 1.x.2016, Lot 10; subsequently shown by Talabardon & Gautier at the Salon du dessin in 2007, where it reportedly sold for a six-figure sum.




From → Art history

One Comment
  1. Yuriko Jackall permalink

    This is great Neil.
    I just sent to Apollo asking if we may reference your detailed list where I say how better editing would have avoided a number of relatively minor mistakes. Fingers crossed.
    All best

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: