Nicolas Poussin 1594–1665
NG 5763 
Landscape with a Man killed by a Snake

Humphrey Wine, 2001

Extracted from:
Humphrey Wine, The Seventeenth Century French Paintings (London: National Gallery Company and Yale University Press, 2001).

1648?

Oil on canvas, 118.2 × 197.8 cm

Provenance

Painted for the Parisian merchant Jean Pointel¹ (d.1660), perhaps by 31 August 1648;² recorded in his posthumous inventory of 20–22 December 1660 with an estimated value of 1000 livres;³ sold by his executors⁴ and bought by [Nicolas?] du Plessis‐Rambouillet ( c. 1576–1664) for 1800 livres;⁵ by October 1684 in the collection of Denis Moreau ( c. 1630–1707),⁶ who in 1689 became premier valet de chambre to the duke of Burgundy; thence, presumably by descent, to Moreau’s nephew, François‐Louis de Nyert, marquis de Gambais, seigneur de Neuville (d.1719);⁷ then to his son Louis de Nyert (d.1 736), among whose posts was governorship of the Louvre and whose posthumous inventory of 2 May 1736 records NG 5763 in his apartment there;⁸ then presumably to Louis’s son, Alexandre‐Denis Nyert (d. 1744), and by descent in the family;⁹ bought from the family (but not at the de Nyert sale, Paris, 30 March 1772) by the engraver and dealer Robert Strange;¹⁰ Robert Strange sale, Christie’s, 5–6 March 1773 (lot 113, £650 to Sir Watkin Williams‐Wynn, 4th baronet of Wynnstay, near Wrexham, and 20 St James’s Square, London);¹¹ presumably the painting recorded in 1885 in the Green Drawing Room, Wynnstay, as ‘Large Oil Painting in Gilt Frame Landscape with figures by Poussin’;¹² by descent to Sir H.L. Watkin Williams‐Wynn, 8th baronet at Wynnstay (where seen by E.K. Waterhouse in 1939),¹³ from whom purchased (through Horace Buttery) in 1947 for £6500.

Exhibitions

London(?) 1816, BI (60) (‘Landscape with Figures. N. Poussin. Sir W.W. Wynn, Bart.’);¹⁴ London(?) 1847 (2) (‘The Story of Phocion Sir W.W. Wynn, Bt., M.P.’);¹⁵ London(?) 1867, BI (18) (described as in the 1847 catalogue); Cardiff 1960, National Museum of Wales (54); Paris 1960, Louvre (83); Bologna 1962, Palazzo dell’Archiginnasio (74); London 1978, NG , The Artist’s Eye: Richard Hamilton (8); Paris 1994–5, Grand Palais (179); London 1995, RA (69); Los Angeles 2000, J. Paul Getty Museum (pl. 65).

Related Works

Paintings¹⁶

• (1) Dijon, Musée Magnin (no. 806),¹⁷100 × 173 cm, in which the drapery of the dead man is red and the seated woman wears a pale red dress under her blue tunic. A good quality copy, probably late seventeenth century. Photograph in NG dossier;¹⁸
• (2) Recorded in 1949 with John Hutton, Glasgow, 48 × 76 in., and said by E.K. Waterhouse to be a contemporary copy;¹⁹
• (3) According to Blunt,²⁰ a small copy, possibly after the engraving by Baudet (see below) and attributed to Richard Wilson, with Mrs Pascal, London;
• (4) Dufourny sale, Delaroche, Paris, 22 November 1819 and days following, lot 88 (22 × 30 in.), where claimed as that painted for Pointel;
• (5) A copy recorded in Lord Northwick’s collection in 1846²¹ in the drawing room at Thirlestane House, Cheltenham, but not in his posthumous sale of 1859, nor in A Catalogue of the Pictures, Works of Art, &c. At Northwick Park, 1864;
• (6) Sotheby’s, 2 March 1977 (lot 105) as ‘N. Poussin’, 122 × 193 cm. A copy;
• (7) Christie’s, 5 December 1989 (lot 76, sold for £3300 including premium) as ‘Circle of Joseph Goupy’ and described as bodycolour with touches of gum arabic on vellum, 46 × 60.3 cm;
• (8) New York, Sotheby’s (Arcade), 17 January 1990 (lot 122), as after Nicolas Poussin, 17½ × 21¼ in.

A painting in the Dal Pozzo collection, once thought to be a copy of NG 5763, is now identified with Poussin’s Man pursued by a Snake (Montreal, Museum of Fine Arts).²² A painting, Homme au serpent, was proposed as a purchase to the Louvre by Ozou in 1850, and another, L’effet de la terreur, by Hunes in 1865.²³

Drawings (by Poussin)

• (1) R.‐P. 335 (Dijon, Musée des Beaux‐Arts; inv. no. CA.872; fig. 5) is related to NG 5763, although considered by Rosenberg and Prat to be more probably connected with Poussin’s Landscape: a Calm (Malibu, Getty Museum of Art);²⁴
• (2) R.‐P. 334 (Paris, Louvre, Département des Arts Graphiques, inv. no. 32481). More probably connected with the Getty Museum painting than with NG 5763;
• (3) R.‐P. 351 (stolen in 1971–2, but formerly in the University Library, Uppsala). This drawing of a group of five trees may be connected with NG 5763 and/or Landscape with the Ashes of Phocion (Liverpool);
• (4) R.‐P. 348 (Düsseldorf, Kunstmuseum, Kupferstichkabinett; inv. no. FP 4699). Shearman noted that the whole of the composition on the Düsseldorf recto, with the exception of the blazing castle and motifs to the left of it, is ‘closely related in reverse to the Dijon drawing…for the Snake landscape…and appears to precede it…’;²⁵
• (5) A drawing of a landscape in the Musée Bonnat, Bayonne (inv. no. A1 1673; NI 48. R.‐P. 213 verso) has, as Rosenberg and Prat say, no connection with NG 5763. It has been suggested that it is connected with NG 6390, which it cannot be since the recto is several years later than that painting.

Other drawings

• (1) Some studies, not by Poussin, in the Louvre, Département des Arts Graphiques ( R.‐P. R856), have been connected with NG 5763;
• (2) A sketchy copy in watercolour and pen by J. M. W. Turner, c. 1798–9, London, British Museum, Dolbadern sketchbook, XLVI, p. 114a;²⁶
• (3) An anonymous variant drawn copy, Paris, Louvre, Département des Arts Graphiques.

Prints

• (1) By Etienne Baudet (1638–1711), one of four landscapes by Poussin engraved by Baudet, dedicated to Louis XIV and dated 1701 (Wildenstein 180). The engraving (fig. 1) bears an explanatory text: ‘Divers effets d’Horreur et de craintes sont [page 325][page 326]icy exprimés. Un jeune homme mort proche d’une fontaine a tout le corps enveloppé par un serpent d’une grandeur énorme. Cet aspect effroyable fait fuir un autre homme dont les regards troublez et les cheveux hérissez sur son front épouvantent une femme plus éloignée assise au bord du chemin. Et les cris de celle‐cy font encore plus loing tourner la teste à quelques pescheurs… L’on tient que le Poussin peignit ce tableau à l’occasion d’un accident semblable qui arriva de son temps aux environs de Rome’;²⁷
• (2) Plate 23 of P.M. Gault de Saint‐Germain, Vie de Nicolas Poussin, considéré comme chef de l’école françoise, Paris 1806, pp. 34–5 and 71, where called ‘L’Écho’;
• (3) by E.A. Réveil (1800–51) in Musée de Peinture et de Sculpture, ou Recueil des principaux tableaux, statues et bas‐reliefs des collections publiques et particulières de l’Europe, vol. 3, Paris and London 1828, no. 208.

Technical Notes

The painting is worn throughout, but particularly in the area of the woman’s left arm, part of which is illegible, the tree trunk at lower left and in the foreground. Much of the red‐brown ground is showing through. The foreground rocks have lost definition. The ultramarine in one of the middle ground figures has discoloured, and some of the foliage paint is similarly affected. Although there was some retouching in 1974, the last major cleaning (presumably by Buttery who was a restorer as well as a dealer) took place in 1947 shortly before the Gallery bought the painting.

The support is a fine twill canvas with a thread count of about 14 (warp) and 26 (weft) per cm, lined, probably in the early nineteenth century. The stretcher, though old, is not the original, of which the diagonal cross‐braces have left outline traces on the surface of the canvas. Baudet’s engraving (see fig. 1) suggests that the canvas has been cut at top and bottom some time after 1701, and the copy in Dijon suggests that it may have been cut by about 3 cm along the left and right sides also. However, examination of the cusping along the edges shows the canvas to have been cut by about 1 cm along the top and perhaps slightly down the right‐hand side. A label on the back of the lining, which may be in an eighteenth‐century hand, reads No. 1. F.D. Room, possibly referring to NG 5763’s location at Wynnstay (Front/First Drawing/Dining Room?). Another label (round) on the top stretcher reads HAB/48/E. The top stretcher is also marked No. 2 in black crayon(?), and has No. 3 gouged into the wood in a c. 1800 hand. The bottom stretcher is marked upside down in an old hand 1778 H(?). The mark No. 2 may refer to the painting’s number in the 1847 exhibition at the British Institution (see above), if indeed NG 5763 was exhibited there.

Analysis of paint in the sky and foliage shows that linseed oil was used as a medium. A pentimento shows that the boat was originally painted about 1 cm above and to the left of its present position. Both the fleeing man and the women were painted over the landscape without any areas being left in reserve. No material changes are shown by the X‐radiograph, save that the fleeing man’s head may once have been higher, but it is not easy to interpret in this respect.

Discussion

A number of interpretations, all focusing not unnaturally on the man being crushed by the snake, have been made of the subject matter of this painting. Of those which claim NG 5763 to show a mythological episode, none has been convincing. The first such interpretation was that published in 1806 by Gault de Saint‐Germain, who cited a passage from Apuleius’ Metamorphoses (I: 4), popularly known as The Golden Ass, ‘that noble serpent clinging in its slippery embrace to [Asclepius’] staff’. Gault de Saint‐Germain used this passage in connection with a more general point, that Poussin’s inspiration was based on Aristomenes’ account in the Metamorphoses of the effects of magic.²⁸ That account, however, is to do with magic as illusions, and, as one might expect, metamorphoses, none of which seems relevant to NG 5763.

Gow proposed that the painting illustrated another passage in Apuleius’ Metamorphoses (VIII: 19–21) in which a young slave was found ‘lying on his back…almost totally devoured by an enormous snake, which was leaning over him and eating him’.²⁹ However, as has been pointed out, first by Watson and then by Davies, these details do not accord with the painting.³⁰ Next, Tervarent proposed a passage in Euripides’ Phoenician Women (verses 638ff.), better known from Ovid’s Metamorphoses (III: 32ff.), which tells the story of Cadmus’ attendants slain by an enormous serpent.³¹ Again the details do not fit the picture, not least because in Ovid’s account the serpent, described as having a golden crest and eyes that flashed fire, clearly killed more than one man.³²

In 1967 Glózer proposed that the subject of NG 5763 was the death of Opheltes (also called Archemorus). Opheltes was killed by a snake after being left alone by his nurse, Hypsipyle, whom Glózer identified as the woman at the centre of the picture.³³ However, as is clear from the account of the incident by Apollodorus³⁴ and from the reference to it by Pausanius,³⁵ Opheltes was a child when he was killed, whereas the dead person in NG 5763 is fully grown. Bull’s suggestion that NG 5763 refers to the death of Aepytus as related by Pausanius is implausible, given the latter’s emphasis on the agent of death being the smallest kind of adder,³⁶ and his subsequent proposal, based on a story in a French translation of Nonnos’s Dionysiaca, that the painting be tentatively identified as Landscape with the Death of Tylos cannot be accepted, given the discrepancies between Nonnos’s account of Tylos’s death and what appears in the painting.³⁷

None of the above interpretations has been generally accepted. The second principal line of interpretation relates the subject matter to an actual event (without, however, going so far as to say that the painting represents it). The supposed incident is first mentioned in the inscription on Baudet’s engraving of 1701 (fig. 1): ‘It is said that Poussin painted this picture on the occurrence of a similar incident which happened during his day in the area of Rome.’³⁸ Then the catalogue of the 1773 Strange sale included the following in the description of the picture: ‘The scene of this picture represents a prospect of the ancient City of Terracina, in the Kingdom of Naples. It was in the neighbourhood of this city, in the morass of Pontius [Pontine Marshes], that the catastrophe which gave rise to the subject of this picture happened, [page 327] in the year 1641.’ As Blunt pointed out, the author of the sale catalogue quite possibly had this information from a (now lost) letter from Poussin to Pointel said to have been written in 1648 and referred to in the catalogue entry as then in the custody of ‘Mr Demaso’.³⁹ Blunt argued that the lake shown in NG 5763 was that of Fondi, the city which he claimed was in the background of NG 5763 and whose castello was on the hill at the left – albeit with many details altered. There is some similarity in the relationship between the lake, the city and the castello at Fondi and that between the same elements in the picture. Blunt’s argument was that Fondi was notorious in the seventeenth century for being plagued by snakes, and that it was near to Terracina, itself near to the abandoned ancient city of Amyclae. He also proposed that Poussin visited the area in 1647, possibly with Pointel.⁴⁰ One difficulty with this argument is that Fondi was abandoned in the seventeenth century,⁴¹ but there is no hint of abandonment in NG 5763: besides the principal figures, there are numerous smaller figures dotted around the landscape. The other difficulty is that the argument is an elaboration of an assertion (that the scene represents Terracina) derived from a source (Strange’s sale catalogue) that was not published until more than a century after Poussin’s death. Assuming, however, that there was indeed a letter from Poussin to Pointel, as seems likely, its contents may have amounted to no more than that Poussin had heard of a man killed by a snake in the Pontine marshes in 1641 (when Poussin himself was in Paris) and that a report of this incident provided him with an idea for a painting, which he chose to locate in an imagined landscape near Terracina (perhaps on account of both the city’s antiquity and its proximity to the notoriously infested Pontine marshes).⁴²

The third, and most venerable, interpretation of the picture is the one in the biography of Poussin written by Félibien, who had, after all, known both Poussin and Pointel.⁴³ Félibien first wrote of NG 5763 in his Sixième Entretien, published in 1679. There he referred to it simply as ‘a landscape’ in which ‘one sees a man who, wanting to approach a spring, stops quite scared on spotting a dead body surrounded by a snake: And further off a seated woman quite frightened, seeing with what fear this man flees’.⁴⁴ Félibien refers to no incident, mythical or actual, and it is clear from his subsequent discussion of the painting that what he saw in it was the effect of fear and horror on the human face. Writing subsequently of the painting in his Huitième Entretien, Félibien divides Poussin’s landscapes [page 328] into those in which there is placed some subject from history or myth, and others showing ‘some extraordinary actions which satisfy one’s intelligence and entertain one’s eye’.⁴⁵ It is clear from the context that for Félibien NG 5763 fell into the latter category and that it was about the depiction, facial and bodily, of horror and fear,⁴⁶ for which a snake had provided Poussin’s figures a similar agent for the expression of emotions some years earlier in the Man pursued by a Snake (Montreal, Museum of Fine Arts; see p. 345, fig. 3). At the time Félibien was writing, the expression of the passions was a key concern within the Académie de Peinture, whose doctrinal position he shared.⁴⁷ Given Félibien’s overarching project, to present Poussin as an exemplar in this respect, a reference to the Terracina incident would have been irrelevant, if not undermining.

It is in a similar context that the fictional dialogue between Leonardo da Vinci and Poussin written by Fénelon may be understood.⁴⁸ ‘Is it a story?’ Leonardo asks about NG 5763. ‘I don’t know it. Rather it’s a caprice.’ ‘It is a caprice,’ Poussin confirms, in the course of an explanation of the painting as a depiction of the different degrees of fear and surprise, and as being about figures in the background who go about their business in a peaceful landscape oblivious of, and in contrast to, the horror in the foreground.⁴⁹ There is nothing inherently inconsistent between the functions of the painting as perceived by Félibien and Fénelon (and by implication Poussin and Pointel) and knowledge of the Terracina incident, if it occurred, providing Poussin with a starting point for it.⁵⁰ On the contrary, such a tragedy would have served to increase the relevance of Poussin’s choice of incident. It is possible, as has been suggested, that Poussin may also, or alternatively, have had in mind a passage from Homer’s Iliad (III: 33ff.): ‘As a man who has come upon a snake in the mountain valley steps back, and the shivers come over his body, and he draws back and away, cheeks filled with a green pallor…’⁵¹ A visual source for the artist may have been relief sculptures on two antique funerary urns in the Barberini collection, which were the subject of engravings by, among others, Boissard in Romanae urbis topographiae (fig. 2).⁵²

NG 5763 was probably also intended to remind the viewer of the whims of fortune, a theme in which Poussin expressed an interest in a letter written to Chantelou in June 1648.⁵³ The theme of the snake in the grass as representative of the omnipresence of death in the midst of peaceable nature was a common one in antique texts likely to have been familiar to Poussin, and one which he had explored in other works,⁵⁴ and it would have been an especially appropriate manifestation of a trick of fortune in a landscape setting. Indeed, there is a passage in Apuleius’ Metamorphoses, which appears shortly after that cited by Gault de Saint‐Germain (see above), that seems peculiarly appropriate to NG 5763: ‘“Aristomenes” [Socrates] answered, “you just do not know the slippery windings and shifting attacks and alternating reversals of Fortune”.’⁵⁵ If Poussin had the passage in mind, then the snake alludes to ‘slippery windings’; the ‘shifting attacks’ and ‘alternating reversals’ are perhaps expressed by the respective attitudes of the figures, and their zig‐zag placement within the landscape, as well as by the alternating areas of light and dark.⁵⁶ The possibility that Poussin intended to show a trick of fortune is suggested both by the figures faintly visible to the left of the woman (fig. 3) who are playing the guessing game morra, then popular in Italy,⁵⁷ and by the association frequently made between the word ‘caprice’ (in Poussin’s ‘comment’ to Leonardo in Fénelon’s Dialogues: ‘It’s a caprice’) and the word ‘hasard’.⁵⁸

Thus, although NG 5763 shows no specific incident from mythology, the subject may have been inspired by an account of a contemporary tragedy and/or by the literary topos of the snake in the grass. Further, the painting may well have been intended to show a trick of fortune in a particular guise. Assuming this to be so, one may ask to what extent such a reading would be affected by the pendant to NG 5763, if indeed it had one? According to Félibien ‘[Poussin] executed two large landscapes for the same Pointel: in one of them there is a dead man encircled by a snake, and another frightened man fleeing’.⁵⁹ Félibien does not say that the two landscapes were intended to be a pair, but it would not be unreasonable to assume that that is what he meant.⁶⁰ Nor does he describe the second landscape, although at another point in the Entretiens he contrasts NG 5763 with the Landscape with Three Monks now in Belgrade, which for that reason has been proposed as having been NG 5763’s pendant.⁶¹ However, the discovery that the Louvre’s Landscape with Orpheus and Eurydice (fig. 4), which is close in size to NG 5763,⁶² was in Pointel’s collection⁶³[page 329] appears to support the suggestion, first made by Wild, that this was the pendant to the London picture.⁶⁴ As Rosenberg has pointed out, the likely dates of NG 5763 and the Orpheus are close; the one follows the other in the Pointel inventory; they share a number of motifs (the attitudes of the fleeing man and of Eurydice, the still waters of the lakes with their reflections, the incidental figures in or around the water, not to mention the snake); and they share a common theme, namely sudden death in an idyllic landscape.⁶⁵ Rosenberg concluded, however, that they were variations on a theme rather than pendants strictly speaking,⁶⁶ and that caution seems justified by a difference in the heights of the two paintings as recorded in the Pointel inventory, that of the Orpheus being some 16 cm greater than that of NG 5763.⁶⁷ The sizes of the two paintings as shown in Baudet’s engravings of 1701 appear extremely close, but this evidence cannot be trusted because the Orpheus had been reduced to its present height by 1685,⁶⁸ suggesting that Baudet had manipulated their format so that the suite of four engraved landscapes of which they were a part accorded in size. Furthermore, in none of Poussin’s other known pairings of landscapes did the artist mix a subject based on an identifiable literary source (the Orpheus) with one that was not (NG 5763). It nevertheless remains a possibility that, even if Poussin did not conceive the two paintings as a pair, Pointel decided to pair them by reason of the motifs and theme that they shared.⁶⁹

As Deperthes noted, both paintings heighten the viewer’s interest by locating violence in a tranquil setting.⁷⁰ However, whereas in the Orpheus there is a contrast between Eurydice’s horror and Orpheus’ blissful lack of awareness, both of which occur in the foreground, in NG 5763, as has often been stated, there is a gradation in the responses of the figures – from horror (the fleeing man) to fear (the woman) to mild curiosity [page 330] (the man in the boat), as they recede in zig‐zag fashion away from the source of concern into the landscape.⁷¹ These gradations of response are functions not just of distance but of what is revealed and hidden by the landscape, so that if the Orpheus is about nature in landscape, NG 5763 is also about the nature of landscape.⁷²

On the assumption that the drawings in Dijon (fig. 5), Paris and Bayonne (see Drawings above) were preparatory to NG 5763, Shearman proposed that the subject of the picture was of minor importance and no more than accessory to an already determined landscape.⁷³ Although the Bayonne and Paris drawings may not have had any connections with NG 5763, it is true that the fleeing man and the woman were painted over a pre‐existing landscape (see Technical Notes). This is not, however, true of the man killed by the snake. Consequently, it seems more likely that Poussin left the other figures to the end, not because the subject was undecided, but because their precise location or attitude had yet to be determined.⁷⁴ Were it the case, however, that Poussin decided on the subject at the last minute, it would support the view that there was no predetermined programme to pair NG 5763 and the Orpheus. As one would expect from Poussin, the positions and attitudes of the fleeing man and the kneeling woman have been carefully considered: a line drawn from the fleeing man’s left hand to the morra‐players through the woman’s right hand intersects, at the centre of her body, with a line drawn from the dead man’s head to the boatmen through the woman’s left hand. The kneeling woman therefore serves both a compositional purpose (uniting left and right, foreground and background) and a narrative purpose (indicating oppositions between work and play, the active and the passive, life and death).⁷⁵

It has generally been assumed that NG 5763 was painted in 1648 on the basis of the statement in the catalogue of the 1773 Strange sale that ‘Mr. Demaso has in his custody a series of letters wrote by the hand of Poussin to several of his friends, and amongst others is one to the above Pointel, dated from Rome the last of August 1648, and wherein he gives him advice of his having finished the above picture…’ The letter is lost. Since it is not known exactly what it said, or even that it mentioned a snake, it may have been referring to some other landscape painted for Pointel, of which there were several. Félibien was in Rome in the years 1647–9, and might have been expected to have been aware of this large and unusual painting. However, after a passage about a painting executed in 1651, he says that NG 5763 was painted ‘dans le même temps’. As Blunt pointed out, Félibien’s formulation is not as precise as ‘in the same year’, but in the continued absence of the Demaso letter, the possibility that NG 5763 was painted shortly after Félibien left Rome cannot be excluded.⁷⁶

The composition of NG 5763 was well known in France through copies and engravings and figured in early nineteenth‐century writings on landscape painting by J.B. Deperthes and Pierre Henri de Valenciennes.⁷⁷ Turner seems to have made a copy of the painting around 1798–9 when he was in Wales.⁷⁸ The subject of NG 5763, if not the composition, found an echo in paintings of the same period by Valenciennes, Girodet‐Trianon and Michallon.⁷⁹

General References

Gault de Saint‐Germain 1806, 21; Smith 1837, 308; Grautoff 1914, vol. II, p. 259; Friedländer 1914, p. 122; Bertin‐Mourot 1948, XXXVI; Davies 1957, pp. 179–82; Blunt 1966, 209; Thuillier 1974, 159; Wild 1980, 140; Wright 1985a, 155; Wright 1985b, p. 135; Mérot 1990, 236; Thuillier 1994, 178.

Notes