History

According to the legend, Voltaire commissioned Largillière’s portrait for his lover and protegée, Suzanne-Catherine Gravet de Corsembleu de Livry (1684–1778).1 The two met in 1716 at the Château de Sully-sur-Loire, where Livry was visiting her uncle and Voltaire was living in exile from Paris. During their stay, Voltaire gave Livry, who aspired to an acting career, lessons in declamation, and the two became romantically involved.2 After his return to Paris, Voltaire was imprisoned in the Bastille for a year.3 During Voltaire’s confinement, Livry began an affair with his friend Nicolas-Anne Le Febvre de La Faluère, known as Génonville (d. 1722)4 – a betrayal that Voltaire later alluded to in the poems À Monsieur de la Faluère de Génonville and Aux mânes de monsieur de Génonville, conseiller au Parlement et intime ami de l’auteur. Despite her infidelity, they appear to have reconciled. In 1719, Voltaire cast Livry as Jocaste in his first major success, Œdipe at the Comédie-Française.5 Alas, her performance was poorly received due to her strong provincial accent.6 Expelled from the Comédie, Livry soon joined a travelling troupe that took her to England.7

The nature and duration of the relationship between Voltaire and Livry after 1719 remain uncertain, but evidence suggests their connection endured. In 1722, Voltaire lamented Livry’s expulsion from the Comédie-Française;8 following her marriage to Charles Frédéric de La Tour Du Pin, marquis de Gouvernet (1694–1775), on 25 January 1727, he affectionately teased her in his Épître en vers à Mme de … quelque temps après son mariage, better known as Les "Vous" et les "Tu"; and in 1728, he sent her a copy of the new edition of La Henriade from London.9 Decades later, in a letter dated 2 February 1772, Voltaire asked Bernard Joseph Saurin (1706–1781) whether Livry, now the marquise de Gouvernet, was still alive.10

Indeed she was, and in 1778, when Voltaire returned to Paris, he visited her. According Charles Palissot de Montenoy (1730–1814), the author of the satirical play Les Philosophes(1760) and Voltaire’s biographer,11 he remarked during that visit: “I have just crossed from one bank of the Styx to the other, never in my life have I felt so old as I do today.”12 During this visit, Voltaire also noticed the portrait by Largillière and asked whether he might have it for his “fille adoptive”, the marquise de Villette (1757–1822). 13 Livry agreed and sent the painting to the hôtel de Villette shortly thereafter. 14

This, later, episode is corroborated by Voltaire’s secretary, Jean-Louis Wagnière (1739–1802). Defending Voltaire’s appearance in response to the past mockery of Jean-Baptiste Rousseau (1671–1741),15 Wagnière wrote:

That copy17 appears to have been executed by Catherine Lusurier (1752–1781) at Voltaire’s own request, though it only reached the Académie française after his death, as confirmed in the academy’s minutes.18

The portrait once owned by Livry (now at the Musée Carnavalet)19 seems to have been publicly displayed at the hôtel de Villette following Voltaire’s death. In a letter to an unspecified “jeune auteur”, Palissot, who himself owned a copy of the painting,20 recalled seeing it and encouraged the correspondent to go view it as well while it was on display:

Both Wagnière and Palissot date the portrait to 1718, a date also featured in the inscription to Lusurier’s copy: “FRANÇOİS · DE · VOLTAİRE · COPİÉ D’APRÈS · LE TABLEAU · FAİT · PAR · LARGİLİERE · EN · 1718 · ÉLU · EN · 1746”. Several scholars, however, have questioned this early dating. Dominique Brême has argued that the sitter appears older than 30 and that the portrait’s style aligns more closely with Largillière’s work from the 1720s, suggesting a more plausible date of around 1724–25.22 Similarly skeptical, Samuel Taylor proposed that the 1718 dating may have originated with an engraving of the portrait by Étienne Beisson (1759–1820), used as the frontispiece to the 1785 octavo Kehl edition of Voltaire’s complete works (VI9).23 That engraving bore the inscription “Agé de 24 ans?” – with a question mark that later was dropped, potentially cementing 1718 as the accepted date.24

To their observations I would add that it seems more plausible Voltaire would have commissioned a portrait from one of Paris’s leading portraitists after the publication of La Henriade in 1723, which firmly established his literary reputation, rather than during his initial success with Œdipe.25 However, the number of late eighteenth-century references and the chronology of Voltaire’s relationship with Livry, at its closest in 1716–19,26 make the 1718 dating the most likely, unless new evidence emerges. In either case, the portrait was unquestionably created before Voltaire’s departure for England in 1726.

Versions

The story of the portrait is further complicated by the existence of two versions, both attributed to Largillière and both regarded as possible originals: one now at the Château de Versailles (MV 8159) and the other at the Musée Carnavalet (P208). Unlike the other known copies, both are oval in format and measure approximately 80 x 65 cm. While the Versailles canvas is the more accomplished work from a technical standpoint, the Carnavalet version has an uncontested provenance – from being in the Villette family collection to its eventual acquisition by the museum.27

In attempt to resolve this apparent contradiction, scholars have proposed some questionable theories. For example, acknowledging the inferior quality of the Carnavalet canvas, the museum’s curators have suggested that Livry commissioned it as a copy, retaining the original (the Versailles version) for herself and sending the duplicate to the marquise de Villette.28

In truth, there is no real conflict between the Versailles portrait being the original and the Carnavalet portrait having belonged to Livry.29 Numerous instances are known in which Voltaire commissioned copies of his portraits, and this may well be another such case. The Versailles painting was almost certainly executed by Largillière himself, while the Carnavalet version, given its noticeably lower quality, was likely made after it. The most plausible scenario is that the Versailles portrait was produced with an eye to printed reproductions, whereas the Carnavalet copy was commissioned specifically for Livry.

The hypothesis that the Versailles portrait remained in Voltaire’s possession at least into the 1740s and was intended for use in frontispieces is supported by the fact that in 1742 it was engraved by Étienne Fessard (1714–1777) (VI15). That Fessard worked from the Versailles, rather than the Carnavalet version is evidenced by the size of the buttons on Voltaire’s justacorps, which are noticeably smaller in the Versailles painting. By contrast, all known copies after the Carnavalet portrait, such as Lusurier’s painting for the Académie, the anonymous copy produced for Palissot, Pujos’s drawing, and a 1790 painted copy by Pierre-Claude Gautherot (1769–1825),30 feature distinctly larger buttons.31

On 30 August 1738, Fessard wrote to Voltaire to complain that he was no longer to engrave his portrait after La Tour’s pastel (VI47), despite prior assurances from Voltaire’s Parisian agent, the Abbé Moussinot (d. 1771).32 In a subsequent letter to Moussinot, Voltaire wrote: “An engraver named Fessard has just written to me. I’d just as soon have him do my engraving as anyone else. Send him to Prault and let them get to work.”33 Fessard, however, did not ultimately engrave La Tour’s portrait.34 Instead, relatively soon afterwards, he engraved Largillière’s. His print (VI15) was used as a frontispiece to the 1742 Œuvres mélées de M. de Voltaire (Geneva: Chez Bousquet), an edition prepared at least in part under Voltaire’s supervision.35 It is tempting to speculate that Voltaire, in order to make amends for the earlier disappointment regarding the La Tour likeness, sent Fessard Largillière’s portrait to engrave on a subsequent occasion.

For a long time, Fessard’s frontispiece remained the only print based on Largillière’s painting. This changed only in 1784, when two new frontispieces appeared almost simultaneously: one by Charles Bertoni, after a drawing by André Pujos, for La Henriade de Voltaire. Nouvelle édition, la plus correcte qui ait encore paru, avec des remarques par M. Palissot (London: Mouatrd, 1784) (VI12), and what is perhaps the most significant frontispiece of all, that of the Kehl edition, engraved by Pierre Alexandre Tardieu (1756–1844) (VI10). Interestingly, the two engravings clearly derive from two different versions of the portrait – the former from the Carnavalet version, the latter from the Versailles version.36 The first case is easily explained: Palissot, who oversaw the edition, owned a copy of the Carnavalet painting and had it engraved for the publication.37 The provenance of the Kehl frontispiece, however, is more enigmatic. Voltaire, who had requested Livry’s copy in 1778, evidently no longer had access to the Versailles version by that time. How and where the painting resurfaced by 1784 therefore remains an open question.