Hans Holbein the Younger 1497/8–1543
NG 1314 
Jean de Dinteville and Georges de Selve (‘The Ambassadors’)

Susan Foister, 2024

Extracted from:
Susan Foister; with Rachel Billinge, Marika Spring and Lea Viehweger; and contributions by Lorne Campbell and Allison Goudie, The German Paintings before 1800 (London: National Gallery Global and Yale University Press, 2024).

Oil on wood (oak, identified), 207 × 209.5 cm

Inscriptions

Signed lower left on the floor: IOANNES/HOLBEIN/PINGEBAT/1533

Inscribed on the dagger of the left‐hand sitter, Jean de Dinteville: AET. SVAE/29; on the book on which rests the right arm of the sitter on the right, Georges de Selve: AETAT/IS SVAE. 25.

Inscribed on the celestial globe, left to right, top to bottom (upper half): CORONA, BOOTE[S], DRACO, VRS[A MIN]O[R], [URSA]MAIO[R], AGI[TATOR], HER/CVLES, VVLTVR/CADENS, [CAS]SIOPEIA, GALACIA, CEPHEVS, PERSEVS, ANDROMEDA; (lower half): VVLTVR VOL[ANS], [DEL]PHINVS, [...]OR, EQVVS / PEGASVS, PISCES, 350, CETV[S]

On the terrestrial globe are the following inscriptions, on the grid squares from left to right, following the red lines of latitude from top to bottom (between the first and second lines of latitude): MELINDA, ANGRA DAS DAS, MANGI/CONGO, RIO[…]/GA[…], S THO/ME, TROP[?ICUS] CAPRACORNI, ÆQVINOCCIALIS CIRCVL[VS], BRISILICI R; (between the second and third lines of latitude): OM[…], A[…], […], […]AMA[…], ADEN, ARABIC[…] SIN[VS], […]/LA, SCILA, TROGI.O/D[…], HABESCH/P[R]ESBITER/[JOA]NNES, MEROE [N]IL[VS] FL[VMEN], ETHOPIA/SVB EGIPTO, NVBIE/REGNV[M], RIO DE CA/MAREOS, ORGVENE/REGNV[M], ETHIOPIA, MELLI/REGNV[M], AFRICA, SENEGA, •C•VIRIDE, GINOIÆ, C: DE PALMIS, •S•/IACOBI, INSVLE HESPERIDV[M], TROPIC CANCRI, •S•ANTHO/NI, LINEA DIVISONIS CASTELLANORV[M] ET PORTVGALLEN, ANTIGLIE INSVLA, […]SVLA ES[…]A; (between the third and fifth lines of latitude, first grid square from the right): DESERTVM M[…], NA[…]/[…], […], BISERMINAR/VM TERRA, TARTAR[IA]/CORASINE, TARTARIA/TORQVESTA[M], CHALDEA, PERSARV[M]/REG:, CAMBE[…] […] […]; (between the third and fifth lines of latitude, third grid square): LAPPIA, RVSSIA, TARTARIA/CVMANIE, LIVO/NIA, MOSCO/VIA, NEPER FL[UMEN], CREMA/NIA, TANAIS, BVLGARIA, NAGAI, PONT[VS] EVXIN[VS], IBE/RIA, GEORGIA, HIRCANV[M]/ET CASPIV[M] MA:, TVR/CHIA, ARME/NIA, MOSA/LIA, RODIS, CIPRA, SIRIA, MESOPO/TANIA, PERSIC[VS] ET/BESSIVA SL, IEROSOLIMA, ALEXA[N], EGIPT[VS], ARABIA ST[E]RILIS; (between the third and fifth lines of latitude, fourth grid square): NORBEGIA, SCANIA, SVETIA, DACIA, PRVSSIA, SAXONIA, POLONIA, GERMANIA, RUSSIA, DANVBIVS, NVRE[M]BERGA, SERVIA, CO[N]STA[N]Ti, GENVA, VEN/ECIE, CORSICA, [S]ARDINIA, ROMA, ITALIA, SICILIA, SINV[S] ADRIATICVS, DALMACIA, GRECI, MORIA, CRETA, LIBIA INTERIOR, LIBIE DESERTVM, MARMARI/CA; (between the third and fourth lines of latitude, fifth grid square): ISLAN, IBERNIA, SCOC/IA, ANGLI, EVROPA, POLI[S]Y, B[A]RIS, BVRGVND, NORMA[N] PRITTAN/IA, LEON, AVERN, FRANCIA, BAION, LANGVEDOC, MINORICA, MAIORICA, NAVAR, ARAGON, HISPANIA, GRANA, CASTIL, PORTVGAL, FESSA, CANARIE, BARBARIA, GETVLIA; (between the third and fourth lines of latitude, final grid squares): ACCORES, MEDERA, ARCT

The text visible on the arithmetic book is as follows:

Dividirt ein ander Ex/rürtte Art⧸als 8 1 6 4 8 in
      1̷  1̷
   9̷  2̷  8̷
8  1  6  4  8
─────────────────────────
   0  5  6  2     5
         0  2     5
         0  1     2  5
         0  0     6  2  5
            0     1  2  5
─────────────────────────
Facit 5  6  7     0  0  0
Bleibt bey der rechten/strich/so bedeuts ein halbes/deuts ¼ /125 bedeuts ⅛ /Als im nachgesetze[n] exe
8  1  7  2  0
─────────────────────────
   0  5. 6  2     5
         0  5
─────────────────────────
Facit […]
Dividirt […]
[…]    0  0  0  0
─────────────────────────
[…] 1  2  5
    0  0  6
─────────────────────────
[…]    3  […]

On the hymnal the text is as follows (left‐hand page): XIX / Kom heiliger geyst herregott⧸erfüll mit deiner gnaden [gut] / deiner gleubge[n] hertz mut un[d] sin⧸dein bru[n]stig lib entzu[n]d in ihn / O herr durch deines lichtes glast / zu dem glaube[n] versamlet hast⧸das volck aller welt zunge[n]⧸ [das] s[e]i dir herzu lob gesungen gesungen; (right‐hand page): MEnsch wiltu leben seliglich und bei Gott blibene [e]wig]lich / Soltu halten die zehen gebot die uns gebeut unser Gott unser Gott / Diiii

On the torquetum the text reads LINEA SODIACI

Provenance

The painting is first recorded in the inventory dated 21–24 January 1589 of Louise de Rochechouart (died 1589), widow of Guillaume de Dinteville (1505–1559) in the great hall of the Dinteville family chateau of Polisy, near Bar‐sur‐Seine in the Champagne region of France.¹ Guillaume de Dinteville was the younger brother of Jean de Dinteville (1504–1555), seigneur de Polisy, who commissioned the painting (see further below and also The Sitters) and must have brought NG 1314 back from England to France, probably in November 1533, at the end of his posting as ambassador that year, although three further visits to England took place in 1535, in 1536 and in 1537.² Jean de Dinteville had no children, so on his death Polisy passed to his brother and then to his widow, whose [page 387] [page 388]death occasioned the taking of the inventory. The painting is recorded as ‘Vng Grand tableau ou sont en paintz le feuz Sieurs de Polisy & dauxerre’ (‘a large picture in which the late Lords of Polisy and Auxerre are painted’). This describes a portrait of Jean de Dinteville and his brother François de Dinteville, Bishop of Auxerre (1498–1554), rather than Georges de Selve, Bishop of Lavaur (1508/9–1541), who is identified as Jean de Dinteville’s companion in the seventeenth‐century notice of the painting at Polisy (see further below); however, the misidentification is understandable in view of the fact that the Bishop of Auxerre had also occupied part of the chateau of Polisy. The allegorical family portrait with the subject of Moses and Aaron (now in the Metropolitan Museum of Art, New York, fig. 1) also hung at Polisy over the fireplace in the chamber occupied by the Bishop of Auxerre.³ As these are the only large paintings mentioned here, and as NG 1314 and the Moses and Aaron descended together in family collections over the following two centuries (see further below), there can be little doubt that the portrait described must be NG 1314.

On the death of Louise de Rochechouart in 1589 the painting passed to Claude de Dinteville (died 1619), Louise’s daughter with Guillaume de Dinteville, who had married François I de Cazillac, marquis de Cessac (died 1593) in 1562 (fig. 2).⁴ The painting presumably passed next to their elder son, Charles (died 1633), and then to his elder son, François II de Cazillac, marquis de Cessac (died 1679), who married Marie de Choiseul (died 1661).⁵ François appears to have removed the painting from Polisy in 1653, as mentioned in three evidently related contemporary documents. One of these, a fragment of a document now in the National Gallery Archive, dated 1653 (fig. 3), is annotated ‘Remarques sur le sujet d’un tableau excellent [page 389] des Srs d’Intevile Polizy et de George de Selve Evesque de Lauuour contenant leurs employs, et tems deleur deceds’ (‘remarks on the subject of an excellent picture of the seigneurs of Dinteville Polisy and Georges de Selve, Bishop of Lavaur, containing their deeds and dates of their deaths’) and describes the painting thus: En ce tableau est represente au naturel Mestre Jean de Dinteville, chevalier sieur de Polisy pres de Bar sur seyne, Bailly de Troyes, qui fut Ambassadeur en Angleterre pour le Roy Francois premier en années 1532 & 1533 & de puis Gouverneur de Monsieur charles de France second Files diceluy Roy, lequel Charles mourut a forest monslier en l’an 1543, & le dict Sr de D’Intevile en lan 1555 Sepulture en Leglise du dict Polisy. & fort vertueux & qui fut Ambassadeur pres de Lempereur Charles cinquiesme, le dict Evesque Fils de Messire Iean de Selve premier president au parlement de Paris iceluy Sr Evesque decede en l’an 1541 ayant des la susdicte année 1532 ou 1533 passe en Angleterere par permission du Roy pour visiter le susdict sieur de Dintevile son intime amy &de toute sa famile, & eux deux ayant recontre en Angleterre un excellent peinctre holandais lemployerent pour faire iceluy tableau qui a esté soigneusement conserue au mesme lieu de Polisy iusques en lan 1653.⁶ The document not only identifies the sitters and gives their biographical details but also states that the painting had been carefully preserved at Polisy until 1653.

Very similar information is given in two further documents. The first, a memorandum concerning the contents of three letters relating to ‘l’excellent tableau par un peintre holandois, Holben’ (‘the excellent picture by a Dutch painter, Holbein’), was sent to the Godefroy brothers (Theodore 1580–1649 and Jacques 1587–1642), French historians of the seventeenth century, by Nicolas Camusat (1575–1655), a canon of Troyes and historian who had become interested in the Dinteville family history. The second document is a copy dated 1654 of a memorandum written by Camusat himself, also referring to the painting ‘fait de la main d’un Hollandois’ (‘made by the hand of a Dutchman’) and which, he says, ‘est estimée la plus riche et mieux travaillée qui soit en France’ (‘considered to be the richest and best worked in France’).⁷ Camusat was acquainted with Claude de Dinteville, marquise de Cazillac, and obtained details of the Dinteville family history from her.⁸ Both documents corroborate the parchment’s identification of the sitters in the painting and the circumstances under which it was painted (de Selve’s visit to England). Both also record that the painting had been removed from the chateau of Polisy and taken to Paris by François de Cazillac, marquis de Cessac: the copy of the memorandum by Camusat dated 1654 states that it is ‘à present a Paris, au logis de M. de Sessac’ (‘at present in Paris, at the residence of M. de Sessac’), while the memorandum concerning the letters sent by Camusat says of it ‘comme appartenant au seigneur du lieu, Sr. de Sessac, jusques en l’année 1653 qu’il la faict transporter à Paris, en sa maison proche la parroisse de St. Sulpice’ (‘as belonging to the lord of the place [i.e. Polisy], seigneur de Sessac, until the year 1653, when he had it transported to Paris, to his house near the parish of St Sulpice’).⁹ François II de Cazillac sold Polisy and evidently settled in the chateau of Milhars in the Tarn, restored by his father, to which he may perhaps have taken NG 1314.¹⁰

NG 1314 presumably passed to the daughter of François II de Cazillac, Charlotte‐Marie de Cazillac, baronne de Cazillac, who in 1651 married Charles le Genevois, marquis de Blaigny, and whose only child, Marie Renée le Genevois (died 1721), married François Voisin in 1684. Marie Renée le Genevois Voisin made her will of 27 April 1716 at Albi. She had no children and sold the barony of Cazillac in 1689 to the duc de Bouillon, Viscount Turenne, but left the marquisat of [page 390]Milhars to her niece, Marie Jeanne Voisin (died 1727), who in 1674 married Chrétien François de Lamoignan, marquis de Basville (1644–1709).¹¹ NG 1314 evidently thus came into the possession of the de Lamoignon family, since it is next recorded in the inventory of Chrétien François de Lamoignan’s daughter‐in‐law, Louise Gon de Bergonne, on 30 January 1728, either at Milhars or at Paris, where she died. Louise was wife of the lawyer Chrétien II de Lamoignon, marquis de Basville et de Milhars (1676–1729), Avocat du roi and a high‐ranking member of the French Parliament. In the inventory attached to her will dated 30 January 1728, the painting is listed as ‘deux personages anciens avec globes et instruments de musique’ (‘two historic figures with globes and musical instruments’); it was one of a pair of portraits valued at the high sum of 3,000 livres, along with ‘l’endurcissement de Pharaon’ (‘the hardening of Pharaoh’s heart’).¹² It was subsequently listed in 1759 in the inventory of her son Chrétien Guillaume de Lamoignon (1712–1759), minister of Louis XVI, President of the French Parliament, as ‘deux ambassadeurs de France en Angleterre avec les attributes des différents arts’ (‘two French ambassadors in England with the attributes of different arts’), along with ‘Moïse devant le Pharaon’ (‘Moses before Pharaoh’), both paintings still hung in the same room where they were recorded in 1728, probably at Milhars, but in the latter case are likely to have been removed to Paris when Milhars was sold in 1765.¹³ Chrétien Guillaume’s son Chrétien François II de Lamoignon, marquis de Basville (1735–1789), Conseiller au parlement, acted as executor to Nicolas Beaujon (1718–1786), banker to Louis XV, and he appears to have used the opportunity of the sale of Beaujon’s possessions to dispose of the two paintings he owned himself.¹⁴ On 25 April 1787 NG 1314 was offered for sale at Paris with Beaujon’s effects (Hôtel d’Evreux, Rémy and Juliot fils, 25 April to 4 May, lot 15/6 bis): Un autre tableau de quatre pieds et demi ou environ de hauteur sur près de huit pieds de large; il représente deux ambassadeurs, MM de Selve er d’Avaux, l’un Ambassadeur à Venise, et l’autre dans les pays du Nord, avec le costume des nations chez lesquelles ils étoient envoyés; et les attributes des arts qu’ils aimoient. On voit aussi une tête de mort en perspective, à prendre de l’angle gauche du tableau, et qui a l’agrément de rassembler en face à un grand poisson. Le tableau est du même Holbein, mais la date de l’année n’y est pas. Il est du règne de François I. ou de Henri II.¹⁵ This was offered together with the Moses and Aaron, lot 16 (as ‘La Cour de François II … by Holbein’).¹⁶

NG 1314 was bought at the Beaujon sale by the dealer Jean‐Baptiste‐Pierre Lebrun (1748–1813), who commissioned the engraving published in 1792 for his three‐volume history, Galerie des Peintres Flamands, Hollandais et Allemands, said to show ‘les portraits de MM. De Selve et d’Avaux, l’un fut Ambassadeur à Venise, l’autre fut dans le nord’ (‘the portraits of MM. de Selve and d’Avaux, one was Ambassador to Venice, the other was in the north’ [fig. 4], see Engravings below). According to the text accompanying the engraving, by 1792 the painting had been sold in England.¹⁷ It was bought from the dealer Buchanan by Jacob Pleydell‐Bouverie, 2nd Earl of Radnor (1749/50–1828), in 1808–9.¹⁸ The painting hung in the first‐floor gallery at Longford Castle, Wiltshire: in 1814 it was described by John Britton as ‘Two full‐length portraits of a philosopher and a mathematician, with several musical, astronomical, and mathematical instruments, by Holbein’.¹⁹ It was sold to the National Gallery in 1890 by William Pleydell‐Bouverie, 5th Earl of Radnor (1841–1900), for £55,000 along with two other paintings, one by Moroni (NG 1316), then attributed to Titian, and one by del Mazo (NG 1315), then attributed to Velázquez, and purchased by a special government grant of £25,000 and with gifts of £10,000 each from Charles Cotes (1846–1903), Sir E. Guinness Bt, Lord Iveagh (1847–1927) and Lord Rothschild (1842–1918).²⁰

Copies and Versions

A copy with variations of the upper half of the figure of Dinteville was in a London sale on 6 February 1952 (lot 61).

Engravings

Etching and engraving by J.‐A. Pierron was published in J.‐B. Lebrun, Galerie des Peintres Flamands, Hollandais et Allemands, vol. 1, Paris 1792 (fig. 4).

Related Works

Two drawings of the order of Saint Michael (fig. 20) may be preparatory studies (see further below).²¹

Exhibitions

RA 1873 (114); NG 1974; NG 1975 (19); NG 1983 (not numbered); NG 1997–8 (29); NG 2008–9 (18); NG 2014b; NG 2015 (not numbered).

Technical Notes²²

The painting was cleaned and restored in 1891, soon after its acquisition, and in 1993–6.²³ It is generally in quite good condition, although the left half differs markedly from the right half in this respect, as can be seen in the photograph taken after cleaning and before restoration in 1994 (see fig. 6). The right half has suffered relatively little paint loss, except in small areas along the vertical joins in the panel and at the bottom edge. Most of the losses are in the left half of the picture and mainly in the lower part, with the area most affected being the lower left part of Jean de Dinteville’s costume. The skull has suffered some paint loss, notably in the nasal bone (fig. 7), as has the floor below it at the very bottom of the panel. The present medallion of Saint Michael is the work of a restorer in an area of total paint loss; the damaged area extends slightly beyond and below the medallion (fig. 8). It was left in place during the most recent cleaning, as was [page 392]another old restoration across most of the end of the handle of the terrestrial globe. The objects painted in the centre of the composition are mostly very well preserved; the worst area of loss is in part of the arithmetic book and part of the table in the same area (fig. 26). The pigments have survived well, although there has been some fading of red lakes.

The support consists of ten vertically grained oak boards, butt‐joined originally with dowels. The wood originates from the Baltic‐Polish region. Six of the boards could be dated by dendrochronology; the youngest heartwood ring was formed in the year 1515.²⁴ The panel was thinned to about 6–7 mm thickness and cradled in 1891. The current non‐original strips of wood around the edges replace earlier edging strips that were also not original.

The ground is natural chalk (calcium carbonate) bound in animal glue, applied in two distinct layers; vegetable fibres reinforcing the structure are present in the lowest of these, visible in some samples. There is a mid‐grey priming layer composed of lead white and a relatively fine black that may be lamp black, leanly bound in linseed oil. Both the preparatory layers and the paint extend to the edges of the panel, except at the top edge where there is a narrow strip of unpainted wood.

Underdrawing, consisting of simple outlines, mostly followed closely during painting, can be seen in infrared reflectograms despite the relatively dark priming and the many areas of thick dark paint that make it difficult to produce good images (figs 9–11 (a, b, c)). The underdrawing is black and lies over the priming, the texture of which causes the lines to break up and appear as though made with a dry material, although they are probably liquid. The figures’ faces, hands and costumes, and the objects on the table, have been drawn, as has some basic structure for the curtain. Where the underdrawing has not been followed exactly it is easier to identify. For example, a first position for the chain across Dinteville’s chest was indicated as parallel lines visible in infrared images to the right of the painted chain (fig. 10), while the tuning pegs on the lute were initially drawn in higher positions and at different angles (fig. 12).

Analysis of a number of samples by GC–MS identified the medium, in every case, to be based on linseed oil.²⁵ In samples from the white fur edging of Dinteville’s gown, from the grey paint beneath the crucifix and the blue of the celestial globe, the oil had been heat pre‐treated to thicken the medium and enhance the drying properties. Heat‐bodied linseed oil combined with some pine resin was detected in areas where greater gloss and depth of colour were required, such as in Dinteville’s black costume, his pink doublet and the background green curtain. A non‐heat‐bodied linseed oil binder was identified in paint from Dinteville’s hand.²⁶

The flesh paints used for the two figures are composed of lead white mixed with variable quantities of red and yellow earths, vermilion, black and azurite of exceptionally small particle size. A brownish‐pink tone was used in the area where the beards are painted, with the individual hairs then applied on top in brown and black paint.

Dinteville’s black garments are painted with a single layer containing a black pigment that seems to be lamp black, [page 393][page 394]which is blended with a little white for the dark grey that delineates some of the folds. The fur lining consists of white with a translucent brownish‐black earth, and some black pigment. The pink doublet is underpainted in a strong orange red composed mainly of vermilion, with some lead white and red lake. Pink paint mixtures of lead white and red lake then build up the structure of the folds. The use of rather pale hues in the lighter areas at this stage, sometimes only thinly covered by the final translucent red lake‐containing layers that further develop the modelling, creates the impression of a shiny satin fabric. HPLC analysis identified the presence of dyestuffs extracted from three sources; the lac and kermes insects, Kerria lacca (Kerr, 1782) and Kermes vermilio Planchon respectively, and the roots of the madder plant (Rubia tinctorum L.).²⁷ It seems likely that at least two separate red lake pigments are present: a more orange‐red one based on madder, most probably in the lower layers, and one or two based on the insect dyestuffs lac and kermes to give the rich crimson hue of the final glazes, although it was not possible to confirm this from the paint cross‐sections.²⁸ The paint of de Selve’s purple‐brown robe consists of black mixed with red earth and varying amounts of white, according to the pattern of the textile. While this mixture would itself account for the slightly purple hue, analysis showed that a small amount of a copper‐containing material is also present, either as a drier or as tiny particles of azurite (as in the flesh) that would have contributed to the colour. In some areas there is a more intense red layer composed of red earth with a little lead white and black beneath the uppermost purple‐brown paint, and while this might simply relate to the build‐up of the drapery, it is also possible that it indicates a change in original intention in this area.

Paint cross‐sections from the green background curtain show multiple lighter or darker green paint layers, the tone and sequence depending on the location of the sample in the patterned design or modelling of the folds. The layers are composed of varying amounts of verdigris, lead white, yellow ochre, lead‐tin yellow and black pigment.

The tiled pavement seems to have been laid in first with the pale greenish‐grey used for the lightest areas of the stone, a mixture of lead white with small amounts of verdigris and black. The outlines of the design were delineated in black and the paint of the coloured stones was then applied. Lead white tinted with a little yellow and red earth was used for the pink and yellow veined marble circles, and stronger shades of green containing more verdigris for the serpentine borders, with highlight flecks composed of lead‐tin yellow.

The strong colours of the knots in the rug are laid on as small square slabs of paint over a very dark grey, virtually black underpaint, visible at the junctions of these touches where they do not quite meet. The pattern of the design is painted in red (vermilion and red lake), dark blue (azurite), golden yellow (yellow earth) and grey (white with black).

The lute case is painted with a mixture of white, black, translucent brown earth and a small amount of red earth. The wood frame of the quadrant is painted with white, brownish black and a trace of red earth, with more red earth in the shadowed parts; black paint was used for the linear surface designs. The celestial globe is painted with azurite and white in the lights and pure azurite in the shadows.

A small amount of metallic gold is used for the decoration and highlights on the chain worn by Dinteville (not all of it original), the toggles connecting the slashing of the sleeves [page 395] of his cloak and those on his hat, his dagger sheath (fig. 13) and the gold threads in its tassel (fig. 21). The gold leaf was applied on a dull yellow linseed oil‐based adhesive or mordant containing calcium carbonate, lead‐tin yellow and possibly lead white, a little yellow earth and a zinc‐containing compound, possibly zinc vitriol, probably added as a drier.

Paint cross‐sections show that beneath the paint of the green curtain to the left of the figure of Dinteville is an extensive yellow‐brown paint layer, similar in colour and composition to the paint used for the wooden shelves and scientific instruments. This may have been intended as a background of wood panelling (see below).²⁹ The crucifix at the far left edge and the grey background around it lie on top of the paint of the green curtain, so may have been a late addition.

Infrared reflectography has also revealed a number of changes, among which are those made during painting to the contours of Dinteville’s hat and those during both drawing and painting to the lengths and positions of the flutes near the lute (fig. 14) and to the positions of the pegs of the lute (fig. 12). In addition, the dial at the top of the celestial globe and the torquetum have been painted over the curtain (fig. 15).

The Sitters

The two sitters can be identified as Jean de Dinteville, [page 396]ambassador to England in 1533, the date of the painting, and his friend Georges de Selve, Bishop of Lavaur. They are named in three seventeenth‐century documents published in 1900 by the historian M.F.S. Hervey (see Provenance).³⁰ The identification of Jean de Dinteville as the left‐hand figure is supported by the fact that the location of his family chateau, at Polisy in Champagne, is to be found on the terrestrial globe represented adjacent to him on the left of the painting (fig. 16).³¹ Two letters survive in which Dinteville mentioned de Selve’s visit to London in May 1533. In the first, dated 23 May and written by Dinteville in England to his brother François he mentioned a secret visit paid him in April by Georges de Selve, Bishop of Lavaur: ‘Monsr’ de Lavor ma fait les honneur grand de me venir veoir qui ne ma ester peu plaisir’ (‘Monsieur de Lavaur did me the great honour of coming to see me, which was no small pleasure to me’).³² In the second letter, dated 4 June, Dinteville refers again to de Selve’s visit and states that he has left England: ‘Monsr’ de Lavor mest venu veoir quj sen Retourne’ (‘Monsieur de Lavaur came to see me but has gone away again’).³³

The earliest reference to the picture in the inventory of 1589 of the contents of the chateau of Polisy, the Dinteville family seat, describes the painting as a portrait of Jean de Dinteville and his brother the Bishop of Auxerre (see Provenance). The seventeenth‐century identifications of the sitters as Jean de Dinteville and Georges de Selve were not passed on in subsequent inventory references in the eighteenth century, by which period the painting had passed from Dinteville’s descendants and was located far away from Polisy (see Provenance). By the time the painting had arrived in England from France at the beginning of the nineteenth century as a work by Holbein, the subjects were unknown. Since the artist spent much of his career in England, and the dress of the left‐hand figure must have appeared similar to that worn by Tudor courtiers, various English identities for the sitters were proposed, including Sir Thomas Wyatt (1503–1542) the poet; Richard Pate (died 1565), Archdeacon of Lincoln; or Sir Hugh Askew (died 1562), Yeoman of the Royal Cellar, and even a foreign one, as the counts Otto Heinrich (1502–1559), Count Palatine of Palatinate‐Neuburg, or his brother Phillip (1480–1541); none was wholly convincing or susceptible to proof.³⁴ Hervey’s discovery of the seventeenth‐century documents with the identifications, as well as documentary proof that de Selve visited England in 1533, settled the matter. De Selve’s visit is likely to have inspired the commissioning of the portrait.

Jean de Dinteville wears a black satin gown lined or edged with lynx fur over a black tunic or jerkin, probably of velvet, trimmed at the lower edges with two rows of a black textured material (fig. 18). The wearing of lynx was restricted to the nobility in French sumptuary law.³⁵ The folds of the tunic might originally have revealed a codpiece as worn by Henry VIII in the cartoon for Holbein’s Whitehall mural (fig. 17) but damage to this area makes its existence difficult to ascertain.³⁶ Under his tunic Dinteville wears a pink satin doublet and a white shirt with a high neck, evidently of very fine linen, [page [397]][page 398]embroidered with white thread and gathered to create a small frill. The pink doublet is slashed in the upper part of the chest and in the sleeves, through which the folds of the shirt protrude. The upper parts of the sleeves of the black gown are slashed and held together with paired gold tags. Pink and black was a fashionable colour combination for clothing in the sixteenth century.³⁷ Dinteville’s stockings are black, as are his shoes, the toes of which are slashed. He wears two garters around his left leg. His hair and beard are brown and he has brown eyes.

Around his neck Dinteville wears a medallion suspended from a gold chain that shows Saint Michael standing over a dragon. The medallion is not original (see Technical Notes) but presumably replaces a similar one. The French order of Saint Michael was founded by Louis XI in 1469. Dinteville is recorded as a member of the order, as were his father and brother Guillaume.³⁸ The members of the order wore a collar of cockle shells with a pendant showing Saint Michael in a medallion; on the accession of Francis I the form of the collar was altered to incorporate Franciscan knots.³⁹ According to the statutes of 1512, wearing the pendant without the collar was permitted in private if there was no ‘assemblée de gens d’état’, an assembly of people of rank: a chain or silk ribbon might be substituted.⁴⁰ The earliest surviving example of a chain and pendant of the order dates from the end of the sixteenth century and is now in the Rijksmuseum, Amsterdam (fig. 19); it has an openwork pendant but representations of the pendant from the early sixteenth century show an oval medallion similar to the reconstruction in NG 1314.⁴¹ Two alternative designs by Holbein preserved at Basel with others from his time in England (fig. 20) show Saint Michael within a border of clouds, presumably an openwork design: they may have been made in connection with the painting, although if Holbein was recording Dinteville’s own pendant, the reason for the presentation of an alternative design is unclear. Perhaps Dinteville had not brought his order with him, although this seems unlikely, but the Dukes of Norfolk and Suffolk had been presented with the order, so similar pendants would have been available to borrow or copy.⁴² The relationship between the personal ornaments represented in Holbein’s portraits and in his small sketches for such ornaments is not always clear: for example, in a drawing of Lady Meutas (died 1577; London, Royal Collection) she is shown wearing a medallion with Mary Magdalene, but Holbein’s sketch with a similar medallion shows two alternatives, raising the possibility that he was providing designs or embellishing what he saw.⁴³

Around Dinteville’s waist is a knotted belt of ribbon, and on the right the steel pommel and hilt of a sword or rapier are visible; its sheathed tip protrudes below the hem of his tunic, between his legs. On the left a dagger is suspended from a substantial tassel of gold and blue silk. The tassel consists of two larger fringed tassels, the upper one partly obscured; from the lower tassel smaller knotted tassels are suspended from golden rings (fig. 21). The golden sheath of his dagger is inscribed with his age, which is given as in his twenty‐ninth year (see Inscriptions). The dagger is of a French type, with a disc pommel, but appears to have a cross guard visible above the sheath rather than a disc guard. Wearing ‘la dague et l’epée’ (‘the dagger and sword’) was evidently de rigueur among ‘gens de robe courte’ (‘people wearing short robes’, that is laity rather than clerics).⁴⁴ The pommel and hilt of the dagger are not well aligned to the sheath, which should logically be moved about a centimetre downwards from its present position, but the position of the [page [399]][page 400]hand obscures this miscalculation. The pommel and hilt are covered with rows of decorative patterns including acanthus leaves scrolls. Surviving drawings by Holbein show a variety of designs for pommels, including some with scrolling leaf patterns, but there are no precise parallels for the type of handle seen here.⁴⁵ The scrolling leaves of the base of the sheath show some resemblance to the leaves overlapping the top of one of the designs Holbein made for a small decorative book cover probably for Sir Thomas Wyatt.⁴⁶ The male caryatid in the sheath is an adaptation of a form that appears in other decorative drawings by Holbein, for example on the bowl of the Jupiter cup and Anne Boleyn fountain in the Kunstmuseum, Basel.⁴⁷ Similar leaf forms to the base of the sheath and a female caryatid appear on the stem of a bowl (Basel, Kunstmuseum).⁴⁸ An offset of a sheath design (also Kunstmuseum, Basel) shows a much more complex arrangement of motifs than those painted here, analogous to the sword design in the British Museum.⁴⁹ The type and disposition of the embossed design has some similarities to Holbein’s clock design of 1543 for Sir Anthony Denny (1501–1549).⁵⁰ As mentioned above in relation to the medallion of Saint Michael, the objects represented in Holbein’s portraits cannot be assumed to have a direct relationship to the reality of what his sitters wore.

Dinteville’s black hat appears to be trimmed with a black textured or fringed braid, probably a silk velvet trimming similar to that on the lower edge of his coat. It is decorated with paired gold tags or ‘aglets’ similar to those of his coat and attached to the front is a small hat badge featuring a skull (fig. 22). The flat cap embellished in this way was ‘the most dashing form of male headgear’, fashionable from the second decade of the sixteenth century when Henry VIII came to the throne; it is worn by Francis I of France in a portrait by Clouet of about 1525.⁵¹ There is no evidence that Dinteville used a skull as a personal device: such badges were not uncommon and skulls were frequently worn in the sixteenth century as badges or other macabre forms of jewellery. Dürer records in his diary that when he was in Cologne in 1520 he paid a gulden for an ivory skull, which was probably an ornament of some kind, and some portraits by artists such as Hans Eworth (active 1540–after 1578) show sitters wearing rings set with small skulls.⁵² The Marquess of Northampton (1513–1571) was sketched by Holbein wearing jewellery including a pendant inscribed with the word ‘MORS’, death (London, Royal Collection).⁵³

Georges de Selve, on the right, is more austerely dressed, although his long, deep brown silk robe lined with brown sable would have been costly (fig. 23). The robe is ‘smart informal outerwear’ (to quote Monnas) of a type worn during the day by both clerics and laity.⁵⁴ Since de Selve was not consecrated as a bishop until 1534, he could not have been represented with the mitre and crosier that were the usual attributes of a bishop, but his dress would have indicated that he was a cleric. The robe has an exceptionally large damask‐weave pattern, of a type that appears to have originated in Venice, where de Selve was ambassador in 1529 (see below). The pattern, only a portion of which is visible, is made up of stylised vegetation and outsized friars’ knots, an extremely popular motif in textiles of the early sixteenth century: symbolising the virtues of the Franciscan order, poverty, chastity and obedience, they were especially appropriate for a cleric.⁵⁵ The pattern might also have been interpreted as a sign of allegiance to Francis I, who chose to include the Franciscan knotted cord belt in the collar of the order of Saint Michael.⁵⁶ Under the robe de Selve wears a plain black garment, possibly a clerical cassock.⁵⁷ He also wears a white shirt, closed around his throat by two white ribbons, evidently of coarser linen than Dinteville’s fine shirt and therefore suitable for a cleric, evoking a sense of penitence.⁵⁸ His flat, square hat is black, a type of headgear by 1533 no longer fashionable and worn only by scholars, professional men and clerics; it is probably made from blocked felt.⁵⁹ In his right hand he holds a pair of brown gloves. His eyes, beard and hair are brown. Under his right arm is a book, the pages of which are inscribed with his age, his twenty‐fifth year.

The Setting

Linking the two full‐length figures is a piece of furniture with two shelves made of light brown wood, the grain of which is evident and may be intended as oak, on which are arrayed a large number of different types of object. The piece of furniture is of the buffet type, a kind of sideboard or cupboard made from oak or from walnut, with shelves that often stood at the side of a dining room or refectory loaded with plates and drinking vessels of precious metal.⁶⁰ At banquets or festivities such as the Field of the Cloth of Gold in 1520, when Henry VIII and Francis I met at Calais, plates and drinking [page [401]][page 402]vessels of precious metal were laid out on wooden shelves, not for use in the feasting but, as tapestries were hung on the walls, to create a vivid impression of the magnificence of their owner.⁶¹

A brightly patterned Oriental or Turkish carpet hangs over the edge of the top shelf; the short fringe made up of the end of the warp thread is just visible on the left. The carpet has a palmette border, within which is a pattern consisting of octagons within two large rectangles on a field of smaller octagonal medallions. This pattern is now known as a ‘large Holbein’ design (or Holbein Type III), although the carpets within this group include designs other than this, the most familiar type.⁶² Carpets with this design, exported from Anatolia, are shown in western European paintings from the mid‐fifteenth century onwards. It was usual for the rugs to be quite small, comprising just one rectangle in the width and two in the length.⁶³ The carpet is draped so that the edges of the two large rectangles running along its length are visible, and a part of the octagons within them can be glimpsed next to the hands of both sitters.

Similar carpets were displayed as table coverings at this period, as here, as well as being laid on floors for ceremonial occasions. For example, ‘Item Seventene greate carpettes of Turquey makinge of soundrye fashions serving for Tables and Footcarpettes whereof’ are mentioned in the inventory of Henry VIII, which includes 437 carpets ‘of Turkey making’; at least some of those rugs came from the collection of 60 acquired by Cardinal Wolsey in 1520.⁶⁴ The carpet in NG 1314 would evidently have been classified as a large carpet in the context of Henry VIII’s collection, since the smaller Turkey carpets were only about 60 cm long and 30 cm wide (2 × 1 ft), while large carpets were 150–180 cm (5–6 ft) long and 60 cm (2 ft) in width. To judge by the comparative size of the two Frenchmen in the painting, the carpet Holbein portrays is of similar dimensions to the latter, but he may have used the typical design of a smaller carpet and then exaggerated its size to fit the composition.⁶⁵ The carpet design shown here is not repeated in any other painting by Holbein, or in any copy: the copy of the Whitehall mural (London, Royal Collection) shows a star Ushak type, while the Barber‐Surgeons Company painting shows a carpet with a single row of large octagons and a ‘kufic’ border.⁶⁶ The portrait of Georg Gisze (1497–1562; Berlin, Gemäldegalerie) again shows a carpet of different type (fig. 58). Although Holbein had represented an Oriental carpet previously in his ‘Meyer Madonna’ painting (Würth Collection) – again of a different type – and may have had notes of other types, he presumably could have based his representation in the Whitehall mural on carpets in the royal collection and could conceivably have had access to them at the time of painting NG 1314 (see biography).

The shelves between the two sitters offer space to display a collection of significant objects (figs 24, 25). On the carpet on the top shelf are a celestial globe and a series of instruments for telling the time, measuring altitudes and finding the positions of all the celestial bodies. From left to right these are: a cylindrical or shepherd’s dial; a quadrant with a white face; a disassembled equinoctial dial with, in front of it a small circular dial; a polyhedral sundial; and a torquetum (see below).

On the lower shelf are a terrestial globe, a pair of dividers, a set square or folding rule, an arithmetic book (Peter Apian’s Eyn newe unnd wohlgegründte underweysung aller Kauffmanss Rechnung, first published in Ingoldstadt in 1527, Book III, Q8 verso; figs 26, 27); a lute with a broken string; a case of flutes; and a German hymnal (fig. 28), an edition of Johann Walther’s Geystliche Gesangbüchlein, first published in Wittenberg in 1524 (figs 29, 30). Under the shelf is the lute case.

On the far left of the upper shelf is a celestial globe supported by a stand ornamented with brass rams’ heads (fig. 31). The celestial globe allows the replication of the positions of the stars at any given latitude, time or day of the year. The blue heavens are shown with the constellations depicted as the figures of the mythological beings from which they took their names: Hercules, Pegasus the winged horse and the lyre bird (Vultur cadens), for example (see Inscriptions); the stars [page [403]][page 404]are gilded.⁶⁷ The ring supported by rams’ heads represents the horizon ring, while the globe itself is marked out with celestial longitude lines and other divisions, including the celestial equator. The globe is also equipped with vertical brass meridian ring, marking the circle that passes through the north and south celestial poles. This has a scale that is numbered twice, once for the angular distance from the North Pole and once for latitude; near the top is a turning brass hour ring with its pointer. The globe has, additionally, on the upper left side, an unmarked semicircular arm hinged where the horizon and meridian rings meet with another marked quarter ring.⁶⁸ To use the globe to examine the heavens it was first necessary to turn it to the correct latitude, otherwise the relative positions of the stars would not be correctly shown. The globe could then be used to show the positions of the constellations visible from the chosen latitude and to take measurements of the stars. [page 405]This was effected by using the ecliptic to find the current day, then turning the globe to locate the meridian; the pointer on the hour ring was then adjusted to find noon and the globe was turned to the correct time. The globe could also be used to find the times of the rising and setting of the sun throughout the year and, with the aid of the intersecting marked or unmarked arms, the starting points for the 12 houses of a horoscope.⁶⁹

Such a heavenly globe would be a luxurious object, one which might well have been owned by a noble courtier such as Jean de Dinteville. His brother Gaucher de Dinteville (1509–1550) owned a sphere: whether heavenly or earthly is uncertain, but it is listed in the inventory of his effects of 1550.⁷⁰ It is more questionable whether Jean would have brought such a large and precious object with him to England but similar pieces may have been seen at the English court: Henry VIII owned such globes.⁷¹ In its representation of the figures of the constellations, Holbein’s globe is extremely close to woodcut gores (flat patterns which could be glued to a sphere) designed by the Nuremberg astronomer Johannes Schöner (1477–1547) who is known to have produced woodcut globes from 1515 onwards; a few of these gores survive attached to spheres on stands dated 1534 and 1535, while another is today in the collection of the Royal Astronomical Society, London (on loan to the Science Museum; fig. 32).⁷² Holbein’s representation differs from Schöner’s earlier globes in his use of capital letters and from all the known examples in the addition of a number of locations, raising the question of whether he followed the known gores and made his own variations, or whether, as Dekker and Lippincott have suggested, there was another model he used, which is no longer extant.⁷³

The combination of the latitude setting used to position the globe and the time shown, along with one of the star positions, can indicate a date. The time in Holbein’s painting is set for around 2.40 pm and the date has been calculated as 10 July.⁷⁴ The intersection of the horizon ring with the scale of the meridian ring indicates that the latitude shown is 42 (or possibly 43) degrees north, the latitude of Rome, rather than the 48 degrees that would be the latitude of Paris and of Polisy, or 52 degrees, the latitude of London.⁷⁵ However, Nuechterlein has observed that the meridian scale measuring latitude is very distorted, such that the ten degree divisions have become substantially elongated in the distances between them; comparison with a virtual model has suggested that Holbein deliberately changed the latitude in the painting from a model in which latitude had originally been set at 52 degrees, that of London.⁷⁶ She also draws attention to the large size of Holbein’s hour ring and finds that the positioning of the hours on the right is slightly out of alignment with the pointer.⁷⁷

On the far left of the top shelf is the cylindrical shepherd’s [page 406]dial or pillar dial, a common scientific instrument in the sixteenth century. Many such instruments survive and Holbein himself represented another in a drawing now in the British Museum, London (fig. 34).⁷⁸ The cylindrical dial was used as an easily portable solar clock. It consisted of a gnomon or pointer, a triangular‐shaped piece of metal used to cast a shadow from the sun, and two sets of markings on the body of the instrument which were necessary to produce a reading. The cylinder was normally hollow and the bell‐shaped cap could be removed, so that the pointer could be detached and slipped into the body and the cap replaced to keep the pointer safe. Here there is a white carrying cord tied to a ring at the top. The dial has a brown metal pointer but, like most of the instruments displayed in the painting, it appears to be made out of a yellow‐brown wood rather than the metal that would denote a more precious object. All the pillar dials listed in the inventories of Henry VIII are made of bone, except for one of gold: ‘Item a Litell pillor of Astronomye of golde’.⁷⁹

In Holbein’s portrait painted in 1528 of the German‐born astronomer to Henry VIII, Nikolaus Kratzer (1487–about 1550; fig. 33), a larger wooden shepherd’s dial is shown on a shelf, along with a quadrant and a semicircular dial that resemble those shown in NG 1314. Kratzer is shown making a polyhedral sundial extremely similar to that in The Ambassadors; on the table in front of him are also a small dial with an upright needle pointer, a rule and a pair of dividers, all resembling those in NG 1314.⁸⁰ Kratzer and Holbein had a close association during both periods of Holbein’s residence in London: in 1527 they worked together on a design showing the heavens for the ceiling of a temporary royal theatre at Greenwich, and they also collaborated in 1528 on the ‘Canones horoptri’, an illuminated book of instructions for the use of an astronomical instrument presented as a New Year’s gift to Henry VIII.⁸¹ In the 1530s Kratzer and Holbein were instructed to send German Protestant books to Thomas Cromwell. The possible relationship between the representations of the instruments in the two paintings is considered further below.

The markings arranged in two concentric rows at the base of the shepherd’s dial represented the sun’s path along [page 407]the zodiac using the year, while the curved lines across the body of the dial mimic the course of the sun’s shadow during the day according to the different times of the year. At the base of the dial the signs of the zodiac were shown with vertical lines forming a grid dividing the representation of each month. Each row represented six months of the year. Although a symbol for each month was present, in practice the instrument could not distinguish between the two halves of the year on either side of each equinox: the readings representing the first portion of Aries and Virgo in spring and autumn would be identical. The signs of the zodiac were therefore paired and set one above the other. The sequence of signs of the zodiac was set in opposing directions, with Capricorn moving from left to right and Cancer moving from right to left, reflecting the fact that the movement of the sun from summer to winter is the reverse of that from winter to summer.⁸² The spacing of the months gave an accuracy to within two days. Sometimes the zodiac was represented figuratively on the body of the dial in relation to the lines representing the movement of the sun during the day according to the time of year (as in the small drawing by Holbein of a similar dial in the British Museum [fig. 34] with lettering for the months).⁸³ These lines would give the reading of the time of day when the shadow cast by the gnomon fell across them.

To use the instrument it was suspended by its cord or placed on a flat surface. The cap at the top was used to turn the metal gnomon or pointer to a position over the appropriate time of the zodiacal month. The dial was then turned until the gnomon pointed at the sun and the shadow became vertical and absolutely parallel with the body of the instrument. The time could be read from the position of the tip of the shadow against the lines spiralling around the cylinder. The reading would not distinguish between morning and afternoon, as it related only to the height of the sun on either side of noon. The whole object is designed for a specific latitude. As in the case of other instruments using a gnomon, the triangular piece of metal casting the shadow, in order to obtain an accurate reading the angle and height of the gnomon needed to reflect the latitude of the user: for example, a northerly latitude for London but a different, more southerly latitude for Paris.

In the painting the gnomon points to a time that could be interpreted as close to either 9 am or 3 pm, but the shadow is represented as falling obliquely across the body of the dial, with a slight curve, which is impossible: in any case there is no light source within the painting, which is set indoors.⁸⁴ The date appears to be set to read the time on about 11 April or 11 August.⁸⁵ Nuechterlein notes that the zodiac and months are not in complete alignment (although they are properly aligned in Holbein’s rendering of the larger dial in his portrait of Kratzer).⁸⁶ The only sign now visible in the upper row of the instrument, owing to damage to the paint surface, is that for Aries. In the lower row there are only two signs: the one below Aries should be Virgo but resembles part of that for Aquarius, while further to the right is a sign resembling Leo, which would be correctly positioned.⁸⁷ The months are indicated in the right sequence: the letters read ‘[M]A AP M’ and inverted below, ‘IVL AU S’ (figs 35, 36). To the left, above the indications of the months and the zodiac signs, Holbein has depicted in outline figures that illustrate them: a ram with its curling horns, facing left, for Aries, and below, placed directly above the sign for Aries, a naked, kneeling woman facing right, with a flower adjacent that she perhaps once held, presumably representing Virgo. By contrast, Holbein has represented the sequence of images, signs and months correctly in his British Museum drawing for a similar cylinder dial, despite its tiny scale (fig. 34).

The semicircular instrument in the centre of the top shelf is an equinoctial dial, which could be adjusted to be usable at all latitudes.⁸⁸ This is not a standard instrument, and scholars have tried to decipher its working from Holbein’s depiction. A somewhat similar instrument also appears in Holbein’s portrait of Nikolaus Kratzer (fig. 33), shown from the other side. Kratzer drew a related instrument in his notebook and it has been suggested he might have invented it.⁸⁹ The instrument has a plumb line and two scales marked out, which serve to adjust the dial to a particular geographical latitude. The time was found by the shadow from the appropriate edge of the square arm of the dial plate. In order to function, the circular piece slides on to the pivoted square arm of the main body and intersects with the curving arc of its scale. The sloping arm bears another zodiac, with the signs here depicted in the proper order.

The dial in NG 1314 is disassembled: the circular dial itself is lying flat in front of the instrument with a spike through it.⁹⁰ A similar small dial with an upright needle pointer appears on the table in Holbein’s portrait of Kratzer. Dekker and Lippincott suggested that it is disassembled to show it being employed in the laying out of the other dials, such as the pillar dial and horary quadrant.⁹¹ Again, whereas in the Kratzer portrait the calibrations are correct, here the hollow semicircular arc is calibrated into 90 degrees but the marks begin 5 degrees too low and there is a space of 5 degrees uncalibrated at the top.⁹² Furthermore, the scale is divided into 5 degrees, which is insufficient, and the zodiac on the side of the swinging arm is too spread out; the appearance of the small circular disc is also distorted.⁹³ No information about time is available from the instrument in its disassembled state.⁹⁴

Behind the universal equinoctial dial is a white quadrant. Quadrants were so called because they take the shape of a quarter circle. The horary quadrant with straight lines, or horarium bilimbatum, is shown here, an instrument used for observation rather than calculation.⁹⁵ Drinkwater has identified this as a two‐rimmed quadrant such as that published by Oronce Fine (1494–1555) in 1532, although Holbein’s example is closer to one depicted in the treatise on the astrolabe by Johann Stöffler (1452–1531).⁹⁶ This type of quadrant was developed during the first quarter of the sixteenth century and used for determining the angle and altitude of the sun from the horizon; it was made for a specific latitude and at that latitude can be used to tell the time. Quadrants were therefore much used by mariners at sea, as well as by makers of maps and marine charts.⁹⁷ Kratzer sketched a similar quadrant in his notebook.⁹⁸

The quadrant in NG 1314 is placed on its side. It has two sighting vanes on its vertical side and a graduated border. Hour lines are marked by Arabic numerals along the inner limb and there are Roman numerals along the outer limb; inside the inner border is a so‐called ‘shadow square’; the side of the square is labelled ‘VMBRA VERSA’. The white face might perhaps be intended as ivory but is more probably the much cheaper alternative of paper glued to the wooden body of the instrument. The quadrant would normally carry a plumb line, a small lead or brass weight on a string with a moveable bead which would fall straight onto the marked segment when the instrument was raised towards the sun. The position of the bead among the hour lines gave the hour. Stebbins noted that the plumb line is not represented here, although the bead on the plumb line was needed to adjust for the sun’s declination on the date of observation; the horizontal edge of the quadrant lying on the shelf could be sighted towards the sun.⁹⁹ Such a dial also required a date [page 409]scale for setting the bead. This could be along the curved edge but is not here, so the other position would be along the edge that is hidden. This type of quadrant needs to be used in connection with a specific latittude. Although the hour lines are correctly drawn in a general way, the termination points of the visible hour lines are not identical or consistent with any one latitude (some suggest one latitude, others another) and therefore it is not possible to determine for which latitude the instrument was intended.¹⁰⁰ It would not be possible to use the shadow square: although it has an unusual number of divisions in order to look precise, the scale is wrong.¹⁰¹

The quadrant as shown seems to have its sights or sighting vanes (of which only one is evident) on the ‘wrong’ side, the straight side from which the scale begins numbering with 0 degrees. In order to use the scales and hour lines on this face, the sights would have to be on the other straight side, the side on which the instrument rests. It is possible that the sights were intended for use with a set of hour lines and scales on the reverse side of the quadrant.¹⁰² Several contemporaries advocated using both sides.

A little further along the top shelf is a polyhedral sundial, clearly made of wood, with, on top, a compass inset under a clear pane, probably a crystal (fig. 40). The instrument has two gnomons with a pin gnomon on the smaller, square face and could be used to tell the time with the aid of the sun’s rays. Such multifaceted sundials were intended to show off the skills of their makers.¹⁰³ Holbein shows Nikolaus Kratzer making a similar dial in his 1528 portrait (fig. 33). Kratzer’s notebook also includes a drawing of a polyhedral dial with an inset compass similar to that in Holbein’s pictures (fig. 37). A similar brass polyhedral dial survives today which has some claim to be Kratzer’s work.¹⁰⁴

Small cylindrical and polyhedral sundials were not only relatively inexpensive, they were meant to be easily portable objects and it is likely that Dinteville owned one, if not several of these. In his letters from London to his brother the bishop François, Jean mentions a ‘compas’: in his letter of 28 May he requests François to ‘send me the pattern of the oval compass’ (‘compas auval’), which is to be understood as a small oval portable dial including a compass, as here. He writes that he is puzzled by the way in which it is made.¹⁰⁵

The compass was used to place the dial in the correct north–south position for telling the time in a particular latitude. The sun would then cast a shadow from the gnomon, which would fall on the surface of the scale. The dial is shown with shadows giving two different readings of time for three different faces: 9.30 am on the left‐hand face and 10.30 am on the other two faces. The dials tell the time in equal hours (horae communae). The smaller, square surface shown upright at the front of the instrument has markings indicating that it needs to be aligned with the equinoctial plane; then the other faces are also read in that orientation (so the pin gnomon is aligned with the north‐south polar axis). However, the compass shows an east–west orientation, not a north–south position, so it is not possible for it to be telling the right time.¹⁰⁶ Moreover, the splay of hour lines on the side planes are wrongly grouped so the dial would not actually work.¹⁰⁷ Compared to the Kratzer portrait it is even more illogical.¹⁰⁸

On the far right of the shelf is by far the most elaborate and sophisticated instrument, a construction made up of differently angled wooden platforms topped by a large circular dial, evidently of brass, with a viewing device (fig. 40). This is a torquetum, an instrument that could be used to observe and measure a heavenly body in any of three different ways using three different coordinate systems (local horizon, celestial, ecliptic), or to convert a measurement in one system to another, an early form of calculator.¹⁰⁹ There are two descriptions of such an instrument from the thirteenth century.¹¹⁰ In the fifteenth and sixteenth centuries there was interest in designing versions of the torquetum, only a very few examples of which survive. One such belonged to Nicholas of Cusa (1401–1464), which he acquired in 1444, another was donated to the Jagellionian University, Cracow, in 1493.¹¹¹ In 1524 Johannes Schöner built a torquetum and used it to observe Halley’s comet in 1531.¹¹² Peter Apian (1495–1552) of Nuremberg, one of the leading instrument makers in the early sixteenth century, published his design for a torquetum in his treatise of 1533 and used his instrument to observe comets in 1531, 1532 and 1533.¹¹³ Apian included the same design for a torquetum in his Astronomicum Caesareum in 1540, a copy of which was presented to Henry VIII (fig. 39).¹¹⁴ According to Dekker and Lippincott, ‘From a patron’s perspective, having an up‐to‐date torquetum in his portrait was an indication of his first‐hand acquaintance with the cutting edge of European [page 410]scientific thought’.¹¹⁵

The lower part of Holbein’s instrument is formed of a box aligned with north to the right for its use. It is made up of two triangular supports hinged to a square plate with a second plate hinged to its edge. When hinged upright the plate lies in the plane of the celestial equator. It is marked with an hour scale calibrated in 24 equal hours with noon at the top and midnight at the bottom (the hours from 5 to 11 are visible, along with 2). On the circle are two hinged discs. The lower disc, which could be rotated, represents the celestial ecliptical plane and has a zodiac marked on its edge. The upper disc, shown in the fully open position elevated by a triangular support, has a second, evenly spaced zodiac scale with some pictorial representation of the signs, such as Capricorn. At the top is the circular apparatus needed for making observations, set vertically. The alidade, the device used for sighting, is placed along the diameter of the scale. Attached to the alidade is a pendant semicircular flap on which horizon altitudes are measured against a plumb line.¹¹⁶ The construction of the torquetum allowed it to be adjusted for the daily movements of the night sky in the local horizon.¹¹⁷ To take a reading, the current date was located on the zodiac scale and the correct time was set by turning the lower disc once; this would align the upper zodiac disc with the celestial ecliptical plane. The circular apparatus on top could then be used to find the ecliptic latitude and longitude of an observed star.

This construction can be clearly seen in Holbein’s painting, and closely resembles the woodcut shown in Apian’s treatise (fig. 39).¹¹⁸ A more schematic representation of a torquetum appears in Nikolaus Kratzer’s notebook (fig. 38) and may suggest that he and Apian used the same source to devise the instrument.¹¹⁹ Kratzer himself may have made such an instrument, using Apian’s woodcut as a template.¹²⁰ Holbein’s representation may derive from Apian or from an instrument owned by Kratzer.¹²¹ Whatever his source, the representation does not offer readings of date, time or latitude.¹²² Some deficiencies in what is shown have been noticed: the degree scales on the upper flap are incorrectly positioned, so that the plumb line reading is irrational; the zodiac signs on the lower circle (only Sagittarius and Pisces are shown) are misplaced, whereas those on the scale above are clearly shown in greater number and are positioned one place to the right of those below.¹²³ Other aspects of the depiction also show departures from Apian that would stand in the way of the instrument operating correctly.¹²⁴

The left‐hand object on the lower shelf, the terrestrial globe, lies at an angle so that one turned wooden handle is visible; such a globe would usually have had two handles by which it would have been held. Its positioning shows Europe to the fore and France in particular is prominent. The inscriptions ‘Baris’ for Paris and ‘Pritannia’ for Brittany are assumed to reflect Holbein’s own pronunciation of these French names.¹²⁵ A central position has been given to Polisy, the very small hamlet south‐east of Paris, in which Jean Dinteville’s chateau was located (fig. 16). In order to facilitate this placing and its labelling, Polisy is in fact shown to the north‐east rather than the south‐east of Paris.¹²⁶ Beyond Europe, parts of Asia and Africa are visible, as well as a land mass around the North Pole (marked by the handle) and parts of the Americas. The globe is marked out with lines of longitude and lines denoting the Equator, the Tropics of Cancer and Capricorn and the Arctic Circle in red, as well as a red line dividing the Spanish and Portuguese possessions according to the Treaty of Tordesillas of 1494.¹²⁷

It has been suggested that the map of the world depicted on the globe has its origins in printed gores (shaped sections of maps made to be applied to spheres made of papier mâché to form globes) associated with a description by Johannes Schöner of Nuremberg of a globe he made in 1523. However, there are a number of differences between the gores (now in New York Public Library) and Holbein’s representation, and it is possible that these gores are not themselves authentic.¹²⁸ Holbein’s globe also bears some resemblance to gores thought to have been designed by Peter Apian probably after 1527.¹²⁹ However, Nuechterlein has pointed out a number of discrepancies with globes and maps of the period: the handle attached to north; the absence of any markings for latitude, meaning that no proper measurements could be taken; the extreme proximity of Brazil and West Africa; and the presence of a political boundary in the line marking the Treaty of Tordesillas, absent from authentic maps of the time, its position in any case not generally agreed.¹³⁰

Beside the globe is a pair of dividers or compasses, which could be used to measure distances on it, as well as a folding rule arranged like a set square.¹³¹ These were standard types of instrument in the period. On the table in Holbein’s portrait of Kratzer (fig. 33) are a rule and a pair of dividers resembling those on the lower shelf of NG 1314.

In front of the globe is a partly open arithmetic book, its red cover evidently intended to represent a fine silk textile.¹³² The book has only a single set of clasps to fasten it but the painted surface is extremely damaged in the area where the second set of clasps would have been represented and, since no evidence of these remained, the edges of the book were reconstructed without clasps during conservation (see Technical Notes). Enough can be seen of its pages to identify it as Peter Apian’s Eyn newe unnd wohlgegründte underweysung aller Kauffmanss Rechnung (A new and reliable instruction book of calculation for merchants) first published in 1527 (fig. 27). Although this was published as a merchant’s arithmetic book, it might have been used by others as a guide to calculation. The arithmetic book is open at a page beginning with the words ‘Dividirt’ (‘let division be made’). Nuechterlein points out that the page gives three examples of an unusual method of division involving decimal fractions.¹³³

The other book on the lower shelf is a Lutheran hymnal shown open, its blue silk fabric ties undone, with what appear to be three other parts stacked up underneath it, their ties also visible (fig. 41).¹³⁴ The first edition of Johann Walther’s Geystliche Gesangbüchlein, first published in Wittenberg in 1524 (Holy Hymnbook) was published at Wittenberg in 1524; Holbein adapted the 1525 edition for his representation (figs 29, 30).¹³⁵ The first verses of two hymns are given in German [page 411]here on facing pages, both in the tenor part of a four‐part setting. On the left is ‘Komm heiliger Geyst’, Luther’s version of the ‘Veni Sancte Spiritus’ (‘Come Holy Spirit’) and on the right, ‘Mensch wiltu leben seliglich’, Luther’s hymn on the Ten Commandments. Both were traditional anthems of the Catholic Church but Lutheran Wittenberg published German versions, following the publication of the Bible and other church texts in German. The book shown in the painting seems, however, to be a carefully doctored version of the original, for the two hymns are not consecutive in Walther’s publication of 1525.¹³⁶ The ‘Veni Sancte Spiritus’ there is the second piece, here number 19, and occupies the left‐hand page, but the hymn of the Ten Commandments is number 19 in the 1525 edition and also a left‐hand page. Holbein cannot therefore have copied a book in which the intervening pages had been lost, making it likely that these two hymns have been specifically chosen for display: a whole opening of this book is clearly visible and readable, which is not the case with the arithmetic book. Moreover, as has been pointed out, the endings of the verses have been amended in both cases, omitting respectively ‘Alleluia’ and ‘Kyrie Eleison’ and replacing them with repetitions of the preceding words, ‘unser gott’ and ‘gesungen’; the amendments conflict with the musical notation.¹³⁷

There are also musical instruments on the lower shelf. On the right is a lute, its dark case, upturned on the floor below, apparently lined in red. Few sixteenth‐century lutes survive but this one appears to be portrayed very accurately, although the grain of the wood of the soundboard on top, which would have been of spruce, is invisible. The neck and ribs would have been of close‐grained sycamore, the wood’s typical appearance with its slightly darker streaks is depicted here.¹³⁸ The lute has six courses (a single short string on the left and five paired strings) and eight frets. Each paired fatter and thinner string in the fourth, fifth and sixth courses is shown, as is the stringing of the pegs in the pegbox. The patterned rosette fretwork of the centre is also visible.¹³⁹ The lute would be eminently playable were it not for the single broken string, the sixth string.¹⁴⁰

To the right are four flutes in a leather case, which fastens with a lock and key. The flutes are of differing diameters, those with the widest bore making the deepest sound. Some similar instruments, as well as the remains of a leather case that may have been of this type, have survived from the wreck of the Tudor ship Mary Rose in 1545.¹⁴¹ The flute with the largest bore seems to contain a cork, which was necessary for playing it. One flute seems to be missing. The case, which has wooden ribs, seems to be of leather but not lined; it was not a luxury item.¹⁴² Flutes like these were associated with soldiers but flute consorts were also becoming common in social music performance by this time, although traditionally the playing of wind instruments was avoided by gentlemen.¹⁴³ Similar cases are recorded in the inventory of Henry VIII: ‘Item v cases with flutes and in euerie of iiij of the saide Cases iiij flutes and in the vth three Flutes’.¹⁴⁴

The subjects are arranged on a stone pavement. The pattern consists of a six‐pointed star within a central circle within a central square, arranged with its point towards the viewer. Within the square is at least one smaller circle, nearest to the viewer; presumably circles in the other three points of the square are implied. There are also two large circles to left and right of the square in the foreground, one, on the left, running under the square, the other, on the right, running over it, the whole linked and evidently forming a [page 412]double quincunx.¹⁴⁵ The quincunx pattern was typically used in ‘Cosmati work’ paving (named after the Italian family of marble workers from the region of Rome who made this their speciality in the thirteenth century). Holbein’s pattern resembles, but is not identical to, the floor of Cosmati mosaic work in the sanctuary at Westminster Abbey (fig. 42), laid in 1268 under Abbot Richard Ware (abbot 1250–83) and used as ‘an arena for the consecration of English kings’.¹⁴⁶ The Abbey floor, like others of this type, is composed of an elaborate series of circular and geometric patterns created from coloured stones and marble, the circular forms of which were obtained by slicing through ancient Roman columns, as was commonly done to create such effects. It includes a Latin inscription with references to the duration of the world or the primum mobile and to the spherical globe as the ‘archetypal macrocosm’, indicating that its circular patterns were intended to symbolise the cosmos: ‘If the reader considers all that is set down, he will find here the end of the primum mobile. The hedge [lives] three years; add dogs and horses and men, stags and ravens, eagles, huge sea serpents, the world: whatever follows triples the years of the foregoing. The spherical globe[?] shows the archetypal macrocosm.’¹⁴⁷ When calculated, the sum gives 19,683 years to the end of the world and the Last Judgement.

The details of Holbein’s patterned borders differ from the Westminster floor: the large circles on either side of the foreground are filled with pinkish stone rather than further patterns, as is also the case with the background to the main patterns. While Holbein may have taken inspiration from the Westminster floor, he appears to have varied and simplified its design. It has been suggested that he took this type of patterned floor not from the Abbey but from the palace of Greenwich, where, later in the decade, it is recorded that there was a painted decoration with circular patterns. However, the description of ‘antique work’, a classically inspired pattern, on the Greenwich Palace floor makes it clear that this could not have resembled the floor in Holbein’s painting.¹⁴⁸

The signature on NG 1314, set out in perspective on the left‐hand side of the floor, is unusually full and extensive (fig. 43). Holbein signed his work infrequently and the extent of the signature in this painting is comparable only to that of his 1523 portrait of Erasmus (Longford Castle collection, on loan to the National Gallery), a Latin text presumably composed by Erasmus, which makes a punning reference to the difficulty of imitating Holbein’s art.¹⁴⁹ The lettering of the signature is comparable to that employed in the dating and inscriptions of portraits of the 1530s and 1540s.

It was not immediately apparent when the painting was in England in the nineteenth century that it included a distorted skull (fig. 44). At the time Wornum referred to ‘a singular object which looks like the bones of some fish’.¹⁵⁰ However, although such a distortion occurs nowhere else in Holbein’s work, the way in which the skull is concealed, using the distortion known as anamorphosis, was not unusual in the sixteenth century and appears to have been somewhat fashionable. Anamorphic woodcut portraits of the Emperors Charles V and Ferdinand, for example, were produced by the German artist Erhard Schön (about 1491–1542): long and thin, they were restored to normal proportions when viewed sideways (fig. 45).¹⁵¹ The most famous example of this painted witticism in sixteenth‐century England was the portrait of Edward VI in Whitehall Palace, today in the National Portrait Gallery (fig. 46). It once bore the signature of the artist Guillim Scrots (active 1537–53) and the date 1546, the year before Edward became king. Visitors to Whitehall Palace in the late sixteenth century described it: it was viewed from the right‐hand side with the aid of a telescopic device that had at its end a metal plate with a pinhole: looking through the pinhole, a clear, undistorted view of the painting was obtained (fig. 47).¹⁵² This device is now lost but a notch cut in the right‐hand side of the original frame served to allow an unobstructed view of the picture plane and the reverse [page 413]of the frame still clearly shows where the apparatus was originally attached.¹⁵³ Jean de Dinteville may have known of the anamorphic experiments produced by Leonardo da Vinci (1452–1519) during his years at the French court.¹⁵⁴ Conceivably de Selve saw examples of the anamorphoses of Schön or other Germans when he was at Speyer in 1529.

The existence in northern Europe of anamorphic prints and paintings from the first half of the sixteenth century shows that methods for making anamorphoses existed long before they were published. Handbooks giving instructions to artists on how exactly to make an anamorphosis survive from the late sixteenth century onwards.¹⁵⁵ A simple method of producing a distortion that would become legible on viewing the image from one side, as the portrait of Edward VI and others were to be viewed, involved transferring an image square by square to an elongated regular rectangular grid. However, when Holbein’s image was refashioned in this way [page 414]with the aid of computerised image processing techniques, the resulting picture of a skull appeared unconvincing, which seemed to indicate that Holbein cannot have constructed his anamorphosis using this simplified technique.¹⁵⁶

A more sophisticated method used intervals in the squaring process spaced further and further apart, with a trapezoid grid to take account of the angle of viewing and achieve a more convincing resolution of the distortion in perspective (fig. 48). When Holbein’s elongated skull is subjected to manipulations on the computer imitating this process in reverse, the result is a perfectly drawn image of a skull, the image with which he must have begun. He may have made a drawing from life of a real skull, or possibly copied a print or drawing by another artist. He was then ready to transfer the outlines to the graduated trapezoid grid. However, the skull is a complex three‐dimensional object, strongly lit from the right, and it would have been a difficult task to transfer all its details to this type of grid and maintain such an accurate depiction. It is possible that another method was used, namely projecting the outlines of a drawing of the skull by means of shining a light through a pinhole onto the drawing at an angle, and then onto a wall or other surface, a method that would result in a perspective distortion.¹⁵⁷ The outlines of the resulting distortion could then be traced and used to prepare the outline of the skull as shown in the painting.

Since the anamorphosis of the skull was prepared with a system in which the focal point for the perspective distortion lies to the right side of the painting, the image of the skull is more easily resolved when viewed from the right than from the left. There is an optimum viewing point when standing to the right, in a position at a right angle 12 cm away from the wall surface, 10.4 cm from the bottom of the picture and some 79 cm from the right edge of the painting.¹⁵⁸ No artificial viewing device is involved. Although the picture’s original frame is long lost, so that it is no longer possible to establish whether it had any viewing device attached (of the type seen on the portrait of Edward VI at Whitehall Palace mentioned above), it is difficult to imagine that it would have been viewed in the same manner as a small peepshow‐type painting. Holbein had evidently long been interested in ways of utilising changes in the beholder’s viewpoint, and particularly in exploiting the effects of an oblique viewpoint, and his use of this in The Ambassadors might be seen as the culmination of a series of exercises of this kind.¹⁵⁹

An alternative method of viewing the anamorphic skull involves holding a glass cylinder at a slight diagonal in front of the picture. The image of the skull appears reflected in it, with only a little distortion to the left and right, which can be corrected by moving the cylinder slightly; an excellent image of the undistorted skull may be obtained in this way (fig. 49).¹⁶⁰ The precise diameter of cylinder required is unimportant; indeed, any ordinary solid glass object may be used. Such a viewing of the picture can be carried out standing directly in front of it and involves neither bending nor standing at right angles close to the wall. This method is not documented in the sixteenth century, although in the seventeenth century [page 415]anamorphoses of a different kind, designed for viewing with cylindrical mirrors, were very common.¹⁶¹ Nuechterlein has suggested a third viewing method, employing a curved mirror of a type described by Cornelius Agrippa (1486–1535). She has noted that Nicholas of Cusa wrote about apprehending the invisible through a convex or concave surface of beryl, a stone prized for its clarity, superior to that of glass in this period.¹⁶²

Behind the sitters is a green curtain of silk damask consisting of several drops of silk with two repeats to a width. The pattern is of stylised vegetation within an undulating trellis of branches bearing large, tear‐shaped leaves and including large pine cones and a diamond ring with three pendant pomegranates, and/or a register of pomegranates containing a single diamond ring, the latter especially visible above the celestial globe and polyhedral sundial on the top shelf.¹⁶³ The silk is probably of Florentine origin: although a ring was also a device of the Venetian republic (the ducal cap of Doge Marco Barbarigo [ruled 1485–6] or Agostino Barbarigo [ruled 1486–1501] is made up of a textile woven with a ring device), the diamond ring had been adopted as an emblem by Piero di Cosimo de’ Medici (ruled about 1464–9). Silk damasks in use from around 1512 to the 1560s with a proven Medici connection incorporate the ring motif and are similar to those shown in the painting. The green damask curtain in Holbein’s portrait of Archbishop Warham (about 1450–1532) is similarly but not identically patterned (Paris, Louvre).¹⁶⁴ The portrait of Thomas Cromwell (fig. 56), which can be dated to 1532–3, shows a textile adjacent to wooden panelling of a type that was evidently planned for NG 1314 (see changes discussed under Technical Notes). The curtain is shown with a number of folds, mostly suggested by the subtle illusionistic use of shadows and highlights on what would otherwise be a perfectly flat surface on which the pattern continues uninterrupted. Where the curtain turns back on the left‐hand side, Holbein has precisely reproduced ‘the reversal of tonal values which would naturally occur on the reverse of a silk damask weave’.¹⁶⁵

Curtains, or ‘traverses’, were much used at court to mark out personal space, particularly in relation to royal worship. Henry VIII possessed a number of such traverses made of sarsenet (plain, lightweight silk also used for curtains drawn across pictures). These were drawn across his closets, adjoining his bedchamber or his chapel, or set up on a frame in a church when he wished to carry out his devotions, marking out a division between the temporal and spiritual.¹⁶⁶ As Lisa Monnas has suggested, the presence of the curtain in the painting, with its glimpse of the crucifix behind, might have evoked such conventions in the minds of contemporary viewers.¹⁶⁷

The curtain is folded back at top left to allow the view of a silver or monochrome crucifix hung at right angles, in darkness; the first two letters of the familiar inscription ‘INRI’ are indicated on the plaque above Christ’s head, but only the letter ‘I’ seems readable. Christ’s feet are attached by means of a single nail and his right arm is shown angled upwards. The figure of Christ can be compared to that of the crucified Christ in the Passion Altarpiece in the Kunstmuseum, Basel, which Holbein first painted with arms in an upward direction and with a single nail through the feet, before changing the representation to the present view showing arms horizontal and feet side by side (fig. 50). Conceivably the artist had with him in England a drawing that had served for the first version of the crucified Christ in the altarpiece and based his depiction in NG 1314 on this.¹⁶⁸

Jean de Dinteville and his Family

The Dinteville family held estates in Champagne and Burgundy. Their origins can be traced to the thirteenth century but by the end of the fifteenth they had established themselves in Champagne, to the south of Troyes, itself south‐east of Paris, and not far from the royal hunting lodge of Fontainebleau (see fig. 2).¹⁶⁹ Jean de Dinteville’s grandfather Claude was killed in 1477 at Nancy, in the same battle as Charles the Bold, Duke of Burgundy, whose Surintendant des finances he had been. Jean’s father, Gaucher (1459–1531), seventh of the nine sons of Claude de Dinteville, served in the households of Louis XII (ruled 1498–1515) and Francis I (ruled 1515–47). He was Francis I’s lieutenant in Italy and was awarded the order of Saint Michael. His youngest brother, François, became Bishop of Auxerre in 1514.

Gaucher de Dinteville married Anne du Plessis (1480/1–1546) in 1496; their family consisted of five surviving sons and two daughters (nine children were born but a son and [page 416]daughter both died in infancy). The eldest, François II de Dinteville (1498–1554), became Almoner to Louise of Savoy, Bishop of Riez in 1527 and in 1530 was appointed Bishop of Auxerre, the cathedral town south‐west of Troyes, in succession to his uncle. In 1531 he was accused of the murder of a gamekeeper but was pardoned by Pope Clement VII in 1531 and sent as ambassador to Rome.¹⁷⁰ François was abbot of Montier‐en‐Der in 1530–8 and of Montier‐la‐Celle and Montiéramey in 1538–54, giving up these benefices briefly in 1542. In 1541 he was replaced as Bishop of Auxerre but restored the following year.¹⁷¹ He served as French ambassador to Rome from 1531 to 1533 and was in Italy again in 1539–42, returning to France in 1533.¹⁷² The second brother, Louis, a Knight of Saint John, died at Malta in 1531 aged 28.¹⁷³ The remaining three sons were Jean, seigneur or lord of Polisy, the subject of NG 1314; Guillaume, seigneur of Échenay; and Gaucher, seigneur of Vanlay.¹⁷⁴

The Dinteville family occupied positions of high importance at the court of Francis I. Gaucher de Dinteville was made ‘gouverneur’, with responsibility for the three royal princes, and his sons Jean, Guillaume and Gaucher de Dinteville were appointed to posts in attendance on them, each responsible for one of the princes. Guillaume, Master of the Horse, attended the dauphin François (died 1536) and Gaucher was an enfant d’honneur (page) and then from 1519 until about 1552 a pannetier attending the future Henri II, who succeeded in 1547.¹⁷⁵ Guillaume later served as a royal envoy and ambassador and was in the king’s military service in Italy in 1536 and 1537, while Gaucher raised troops for the king in Italy in 1536 and 1537.¹⁷⁶ In the late 1530s the Dinteville family fortunes underwent a dramatic reverse as a result of plots against them by the king’s mistress, Madame d’Etaples, together with Anne, duc de Montmorency (1493–1567), the king’s chamberlain.¹⁷⁷ In 1536 Guillaume was accused of poisoning the dauphin but was cleared of this charge; in 1538 his brother Gaucher was accused of sodomy. As a result he fled from court and Guillaume followed soon after. The brothers exiled themselves in Italy and, after further intrigue at court, François II de Dinteville lost his bishopric of Auxerre. In 1542 Francis I visited Polisy and the family fortunes were restored.¹⁷⁸ Gaucher and Guillaume both served in Italy in 1542. Jean de Dinteville again took up a post in the royal household, followed shortly afterwards by Guillaume, and their brother’s bishopric was restored to him, although it was not until 1551, under Henri II, for long a supporter of the Dinteville cause, that full restitution was made.¹⁷⁹ Jean, Gaucher and Guillaume subsequently became members of the household of Henri II.¹⁸⁰

Jean de Dinteville, who was born on 21 September 1504, joined the royal household in 1524, first as an ‘échanson’ (cupbearer) to the royal children, and then in 1531 as a ‘gentilhomme’.¹⁸¹ His individual responsibility was for the youngest prince, Charles, duc d’Angoulême, and from 1536 duc d’Orléans (1522–1545). The prince’s tutor was Jacques Lefèvre d’Etaples (about 1455–1536), humanist and translator of the Bible into French, and Jean de Dinteville may have known him well.¹⁸² In 1527 Jean de Dinteville was made Governor of Bar sur Seine, by which time he had also been appointed bailly de Troyes, in succession to his father, with responsibility for administering justice in the city.¹⁸³ In 1531, on his father’s death, he may have become a member of the order of Saint Michael (see discussion of his pendant of the order under The Sitters).¹⁸⁴

Jean de Dinteville was sent as ambassador to England on five occasions: in 1531, 1533, 1535, 1536 and 1537. In autumn 1531 he made a first, brief visit and left on 3 December.¹⁸⁵ He departed for England again in 1533, leaving Paris on 18 January for a stay that was expected to last six months.¹⁸⁶ The main diplomatic concern of this mission was Henry VIII’s marriage to Anne Boleyn, which took place on 25 May, followed by her coronation on 1 June and on 7 September the birth of her child, Elizabeth, to whom Francis I was to stand godfather; the christening took place on 19 September. These concerns can be followed through the surviving correspondence of Dinteville written from London.¹⁸⁷ On 23 May 1533 he writes to his brother the Bishop of Auxerre of having tertian fever, a form of malaria (J’ay eu la fiebvre tierce’) and goes on to say: Je suis le plus melancolique fasche et fascheux ambassaduer que vistez oncques…. Je vous prie menvoyer le portraict du compass auval duqul maez escript/car je suis bien empesche a comprandre la facon de laquelle il est fait. Je commence bien a me fascher en ce pays icy en attendant la fin des six moys de lesquelz esche[v?]irent? le vingt deuxieme juillet¹⁸⁸ In the same letter he also reports the visit of ‘M de Lavor’: ‘Monsr’ de Lavor ma fait les honneur grand de me venir veoir qui ne ma ester peu plaisir’ (‘Monsieur de Lavaur did me the great honour of coming to see me, which was no small pleasure to me’) and complains that the coronation will be expensive – ambassadors were expected to expend their own money on such events: ‘Il me fault faire une grosse depense pour se couronement’ (‘I have to lay out a huge sum for this coronation’). On 31 May Dinteville’s 12 servants processed from the Tower to Westminster, dressed in blue velvet with yellow and blue sleeves and white plumes in their hats, and on horses with trappings of blue silk with white crosses.¹⁸⁹ On 4 June Dinteville wrote to his brother that his recent illness was not serious, but ‘Vous scauez comme Iayme ce pays Icy. Ie croy bien que tant que Ie y seray / Ie ne y seray pas trop sain’ (‘you know how I love this country. I truly believe that as long as I am here I shall never be entirely healthy’).¹⁹⁰ He reports that ‘Le couronnement est fait / ou Iay fait grant despense en habiollemens / et en mangeaille et ancores nensuis dehors ny ne seray de huict Iours / Il ma faillu doubler ma table’ (‘the coronation is over, for which I have spent very heavily on clothing and comestibles, and I am still not out of it and shan’t be for another week; I had to double my usual table’).¹⁹¹ In the same letter he mentions that the queen is heavily pregnant and that M. de Lavaur, who came to see him, has left.¹⁹² Dinteville was repaid at [page 417]court the sum of 1,740 livres the following year.¹⁹³ The dates and circumstances of de Selve’s mission are discussed further below.

In September 1535 Jean de Dinteville was sent to England once more to make Henry VIII aware that Pope Paul III was demanding that Francis I should be prepared to declare war as a result of Henry’s execution of Sir Thomas More and Bishop John Fisher in that year.¹⁹⁴ In 1536 he was sent again, leaving on 23 April.¹⁹⁵ In 1537 he made his last visit to England, where he was instructed to make known the Pope’s intention of sending Cardinal Pole (1500–1558) to foment rebellion in favour of the Catholic Church.¹⁹⁶

In 1542 Jean de Dinteville took up a post again in the royal household, being readmitted to the household of Charles, son of Francis I.¹⁹⁷ He also had to fend off attempts by his brother Guillaume to take property but was made ‘gentilhomme ordinaire’ of the chamber of Henri II in 1549.¹⁹⁸ He became paralysed by illness and in 1552 it was said he was no longer able to leave home.¹⁹⁹ He died, unmarried, in 1555.

Portraits of Jean de Dinteville

A drawing by Jean Clouet (about 1485–1540/1), the portraitist of the French court, is presumably a study for a lost painting (fig. 51).²⁰⁰ A seated full‐length portrait of a young man wearing classical armour in the guise of Saint George, the conquered dragon at his feet, has been said to represent Jean de Dinteville and has been attributed to the Italian artist and stuccoist Francesco Primaticcio (1504–1570) and to Domenico del Barbiere (about 1506–about 1570), who worked with him. However, the identification rests on no more than a physical similarity and, like the attribution, is of recent date.²⁰¹ Another portrait is mentioned in correspondence: on 11 March 1551 or 1552 Primaticcio wrote to the Bishop of Auxerre saying that he would draw Jean de Dinteville for the duc de Guise and would colour it with his hand so that the bishop would find it less ugly than the first portrait.²⁰²

The Dinteville Family as Artistic Patrons

In the two letters from England sent to his brother François II, Bishop of Auxerre, in 1533, quoted above, Jean de Dinteville mentions pictures on both occasions. In the first, of 23 May, he wants to know his brother’s news and ‘que direz de la tour et des tableaux’ (‘what you will say about the tower and the paintings’). In the second letter, of 4 June, he adds at the end: ‘Ie vous prie mandez moy sy auez trouue le painctures bien fctez’ (‘I pray you tell me if you have found the pictures well made’).²⁰³ These presumably refer to embellishments of the Dinteville chateau of Polisy, and to paintings that cannot now be identified rather than to NG 1314.

Several surviving works can be associated with the patronage of the Dinteville family, and with François II de Dinteville, Bishop of Auxerre, in particular. The large and distinctive painting of Moses and Aaron by an unidentified, likely Netherlandish, artist (probably Bartholomeus Pons) now in the Metropolitan Museum of Art, New York, which bears the date 1537, shows all four surviving Dinteville brothers, identified by name on the borders of their costumes (fig. 1).²⁰⁴ Aaron, in the garb and headdress of a Jewish priest, is François II and Moses is Jean de Dinteville: they are shown in the Old Testament story asking Pharoah to allow the Israelites to leave their exile in Egypt. To demonstrate God’s power, Aaron turns his staff into a serpent, which appears in the picture as a magical crystalline creature. Elizabeth Brown has suggested that the painting was commissioned in 1539 to express the Dinteville brothers’ desire for revenge on their enemy Pierre de Mareuil (died 1556), as a result of whose plotting the previous year three of the brothers found themselves exiled in Italy.²⁰⁵ According to this theory, the date 1537 would refer to the last period in which the family was in favour, rather than the date of the painting’s execution.²⁰⁶ The fact that the painting is recorded in 1589 at the chateau of Polisy in the bishop’s apartments strongly suggests that he commissioned it, whatever the motivations for the choice of subject.²⁰⁷

François II de Dinteville is identifiable as patron of a number of works of art. He may have developed his tastes during a period in Rome in 1531–3.²⁰⁸ In August 1533 Jean de Dinteville sent him casts of the faces of the dauphin and his brother Henri, perhaps so that he could have portraits made from them.²⁰⁹ In France the bishop undertook building work at the ecclesiastical foundations with which he was associated. In his church at Varzy is the Martyrdom of Saint [page 418]Eugenia, including a portrait of the bishop and his motto, ‘Virtuti Fortuna Comes’ (‘Fortune the companion of Virtue’), a work he donated in 1537; the artist has been identified as the Dutch painter Bartholomeus Pons (active 1518–41), who may also have depicted another member of the Dinteville family in a portrait now in the Louvre.²¹⁰ A small picture in the Städelsches Kunstinstitut, Frankfurt, The Descent into the Cellar, depicting men in a cellar with barrels of beer also bears the Dinteville coat of arms and may be by the same artist.²¹¹ An engraving of around 1538 of the stoning of Saint Stephen, in which the saint bears the Dinteville arms, was made by Domenico Fiorentino (Domenico del Barbiere, also known as Dominique Florentin), a Florentine sculptor and stuccoist; it has been suggested that it may be based on a lost painting, perhaps an altarpiece, for one of the bishop’s churches.²¹² Domenico Fiorentino is described as an imageur and also as a painter, but there are no documented painted works by him.²¹³ The bishop seems also to have commissioned an altarpiece at the abbey of Montiéramey, north of Polisy, from Domenico Fiorentino, presumably a carved altarpiece.²¹⁴ For the southern rose window of the cathedral at Auxerre the bishop commissioned stained glass depicting the life of Moses, dated 1550, and a painting of the same date showing the stoning of Saint Stephen, in which he appears as a Hebrew leader.²¹⁵ The painted decoration of the Dinteville Hours in the Bibliothèque Nationale bears the arms of the Bishop of Auxerre and is the work of the leading French court illuminator, known as the Master of the Hours of Henri II.²¹⁶ His correspondence with Primaticcio, who was making him portraits of Jean de Dinteville, is referred to above.

The Chateau of Polisy

Jean de Dinteville inherited the chateau of Polisy on his father’s death in 1531; the estate was shared by Dinteville and his brother the Bishop of Auxerre.²¹⁷ The site of the chateau, on which a building was first erected in 1390, borders both the river Seine and the river Laigne, which flows south from it. A drawing, now lost, apparently dating from 1629 and published in a lithograph in 1864 (reproduced by Hervey [1900] and others) shows a plain gabled building with four towers or turrets of varying dimensions overlooking the river Laigne to the east (fig. 52), with a drawbridge; evidently there was previously a full moat surrounding the chateau.²¹⁸ The building shares its general aspect with other similar small French chateaux dating from the fifteenth or early sixteenth century. The chateau of Polisy is relatively small, with five windows on the ground floor of the main part of the building and five windows above. The reproduced drawing shows a mill further down the river, and a cross and a building immediately adjacent, which Hervey supposed was a chapel, but it does not appear to show the ancillary buildings near to the house – a barn, lodging, dairy and stable – also mentioned by Hervey, which survive and appear to be of early date.²¹⁹ Substantial alterations were made to the chateau in the eighteenth and early nineteenth centuries; it was stuccoed and given the appearance of an Italianate villa, with a loggia apparently completed in the early twentieth century. The building, only registered as a historic monument in 2011, is now in a poor and damaged state but aspects of the interior were assiduously recorded by local archaeologists in the nineteenth century (figs 53, 54). Informative accounts of aspects of the chateau are given in publications and photographs of the nineteenth and early twentieth centuries as well as by Hervey, who evidently visited Polisy herself.²²⁰ This information can be supplemented by that contained in the inventory of the chateau dating from 1589, in which NG 1314 is first recorded.

Jean de Dinteville’s remarks in the letters of 23 May and 4 June 1533 cited above to his brother the bishop, concerning a tower and paintings, suggests that he first carried out some building work or renovation just after he inherited the chateau in 1531.²²¹ After Jean became seriously unwell in the 1540s, his cousin Jean de Mergey (born about 1537), taken to live with him from the age of 8 until the age of 14 or 15, later wrote that because of Jean’s illness that forced him to leave the court, ‘s’etant retire chez soy, se mis pour son plaisir et exercise à bastir ceste belle maison de Polisy’ (‘retreating home, he set about building for his pleasure and exercise this beautiful house of Polisy’).²²² Extensive building work appears to have been carried out at the chateau in 1544–5. Hervey recorded an inscription on a foundation stone in the cellar dated 1544, referring to the completion of a lower courtyard (basse cour) by Jean de Dinteville.²²³ Tiles in a room on the upper floor were dated 1545 (see further below) and a stained‐glass window on a subsidiary building was dated 1545. Hervey also recorded a small building dated 1545 that had the device of a winged column over a window, which she associated with the Bishop of Auxerre and his motto.²²⁴ Finally she documented a chapel with tombs and a cross that stood on a hill.²²⁵

The inventory of the chateau of Polisy, dating from 1589, states that The Ambassadors hung in the great hall above the courtyard of the old building (see note 1), suggesting that it was located on the west side, away from the river on the eastern side.²²⁶ It was probably always the most important room of the older part of the chateau, dating from the early sixteenth century or before. In 1589 its contents included two large old copper andirons, a painted wooden table, a large oak bench, a large dresser, a grey stone basin on an oak base, three old oak chairs and a wooden chandelier, painted green, hanging in the middle of the room; there was also a painting of Saint Sebastian.²²⁷ There were adjacent towers and a ‘grand chambre’ belonging to the ‘dame’, that is, Louise de Rochechouart, widow of Guillaume de Dinteville, the younger brother of Jean de Dinteville and his brother the bishop, as well as smaller rooms. The inventory also refers to accommodation in the upper part of the ‘logis neuf’ (new building), evidently the result of work carried out in the 1540s. This reference occurs after a description of further rooms, including the ‘chambre’ of the Bishop of Auxerre with a painted and gilded wooden bed, a walnut table and dresser and the painting of Moses and Aaron hung over the fireplace, ‘Vng Grand Tableau estant au [page 419]dessus de la cheminee ou est painte lhistoire de Pharao Roy deGipte’ (described as ‘Pharoah King of Egypt’), as well as a room in the ‘high tower’. This suggests that the bishop’s accommodation was to be found in the newly renovated part of the building.

The names of both Domenico del Barbiere and Primaticcio have been mentioned in connection with the renovation of Polisy during Jean de Dinteville’s lifetime, but this remains highly speculative. Primaticcio was in contact with Dinteville’s brother, the Bishop of Auxerre in 1551/2 (see above), and while working at Fontainebleau Primaticcio was made abbot of Saint‐Martin‐ès‐Aires at Troyes, not far from Polisy, on 15 December 1544. Domenico Fiorentino and Hubert Julyot stood proxy for him and a procuration was issued to Jean Thienot, vicaire generale of the bishop of Troyes, by Primaticcio at Polisy. This ecclesiastical record has led to the assumption that these artists were present at Polisy to carry out work there.²²⁸ Domenico Fiorentino had been working with Primaticcio at Fontainebleau as a stuccoist but had moved to Troyes, just north of Polisy, by 1540. It is certainly possible that he could have played some role in the building or decorative design work at Polisy, and even that he made some stucco work for the chateau: he is recorded at Troyes until 1564, where he made the screen for the church of Saint Etienne at Troyes. He also continued to work for the court as a sculptor, making a grotto and providing the base for the casket holding the heart of Henri II. Engravings by Fiorentino, with elaborate scrolled borders similar to the borders made in stucco for the Galerie François I at Fontainebleau, suggest the kind of decoration he could have carried out.²²⁹

An elaborate tiled floor (fig. 55) dated 1545 and measuring 28 square metres, now in the collection of the Musée national de la Renaissance at Ecouen, is all that remains of the Renaissance‐style renovation of Polisy: in 1900 Hervey recorded its presence in a room on the first floor of the chateau of Polisy.²³⁰ A few further tiles survive in the museum at Troyes, including examples dated 1545 and 1549.²³¹ Made of brightly coloured ceramic, in which yellow and light green are prominent, the floor tiles are arranged in a pattern featuring crosses and octagons: in one, displaying the motto ‘Virtuti Fortuna Comes’, mermaids hold up a shield with the arms of the bishop. Three further octagons include images of Spes, Fides and Charitas and inscriptions in Greek. The prominence of the coat of arms of the Bishop of Auxerre suggests that the room in which this floor was laid may have been his ‘chambre’, in which the 1589 Moses and Aaron hung, perhaps also embellished with stucco decorations by [page 420]Domenico Fiorentino. Such tiled floors were highly fashionable at the time: others to be seen in the period when Polisy was being renovated included those at the chateaux of Ecouen and Oiron, commissioned by, respectively, Anne de Montmorency, the king’s Grand Maître, and Claude Gouffier (1501–1570), his Master of the Horse.²³²

Although The Ambassadors and the Moses and Aaron, evidently painted four years later, seem to have been regarded as a pair later in their history, and were kept together until Holbein’s picture came to England, in the 1589 inventory they are recorded in different rooms. Unfortunately both pictures have been cut slightly, so that it is impossible to establish with precision the original dimensions of either picture. However, they are now extremely close in size and must also have been so originally: the Moses and Aaron was perhaps designed to match the dimensions of the Holbein.²³³ When The Ambassadors was brought back from England (presumably in 1533), Polisy may still have been undergoing the renovations mentioned by Jean in his letters of that year, which perhaps provided a particular setting for it that he may have borne in mind when commissioning the painting. The proportions of both paintings may have been devised to follow the ambient architecture of the great hall of the chateau before the renovations of the mid‐1540s. Comparison with other paintings in French chateaux of the time would suggest that the paintings are likely to have been installed as part of a room decoration rather than hanging free on a wall, perhaps either side of a fireplace: according to the 1589 inventory the Moses and Aaron was displayed over a fireplace. However, it is possible the two pictures were not intended as pendants and never hung in the same room. It is impossible to be certain whether the ‘great hall’ in which The Ambassadors hung in 1589 was the room in which it was originally placed but, if so, the painting was conceived for a different, more public display in the major room of the chateau at that time than the room in the newer part of the chateau in which the Moses and Aaron was recorded in 1589, which appears to have been the bishop’s own private chamber.²³⁴

The 1589 inventory does not provide any information on the height at which The Ambassadors was then hung or any elucidation of how the room was reached. The ‘grand salle’ in which it was displayed adjoined small rooms and it is likely that there would have been an entrance at each end.²³⁵ The room would presumably also have been lit with windows along one side (probably facing west, as noted above). Entering a room with doors to right and left of the painting would have facilitated the viewing of the skull from the correct perspective, conceivably reducing the element of concealment and surprise that is part of the experience of viewing today, when the painting is usually approached from the front, with the discovery of the skull as part of a secondary act of viewing. It is less probable that there would have been a third entrance to the room, with the picture hung opposite.²³⁶ It has been suggested that the distorted skull may have been viewed from a staircase but it was rare to find grand staircases with straight flights of steps in buildings in the first half of the sixteenth century and a spiral staircase would have been usual in a chateau such as Polisy.²³⁷ According to the engraving of the chateau (fig. 52), there were a number of turrets (perhaps one being the tower mentioned by Jean de Dinteville in his letter of 1533), which would have incorporated such spiral staircases.²³⁸ A spiral staircase would have made a difficult viewing place and an impossible hanging place for a large painting such as The Ambassadors.

Georges de Selve and his Family

Georges de Selve was the second of the six sons of Jean de Selve (1475–1529), who rose from a merchant family to high judicial and diplomatic office under Louis XII and Francis I, and of Cécile de Buxi (about 1490– about 1557).²³⁹ Jean de Selve was one of the leading figures of the French parlement in the early sixteenth century, a conseiller in the parlement of Toulouse and successively Premier président of the parlements of Rouen, Bordeaux and Paris. He was vice‐chancellor of the duchy of Milan during the French occupation and played important roles in the negotiations for the marriage of Louis XII to Mary Tudor, sister of Henry [page 421]VIII, and for the release of Francis I after his capture at Pavia. Five of Jean de Selve’s sons were active in diplomacy and state affairs. His second son, Georges, and third son, Jean‐Paul (died 1569), both became bishops, the latter bishop of Saint‐Flour.

Georges de Selve was educated by the French humanist and future lecteur royal Pierre Danès (1497–1577), to whose instruction in Greek and Latin he refers in his translation of Plutarch’s Lives, which he made by order of the king.²⁴⁰ Georges began his career as a clerc tonsuré of the diocese of Rouen. He was made a protonotaire apostolique and later a canon in the cathedral church at Chartres.²⁴¹ In May 1526 Pope Clement VII issued a licence in favour of Georges de Selve, son of Premier président Jean de Selve, allowing him to hold several benefices, despite the fact that he was aged only 17.²⁴² At the end of October 1526 the see of Lavaur near Toulouse became vacant on the death of the bishop Pierre de Buxi, a relative of Georges de Selve’s mother. Francis I, grateful for the role his father had played in securing his release from captivity with the Treaty of Madrid, nominated the young Georges to the see. He took the oath of fidelity for the temporalities of the see of Lavaur in May 1528, although because of his age he could not be consecrated until 1534, the year after he was painted by Holbein.²⁴³ Georges de Selve spent much of his life as a diplomat in Rome, Venice and at the court of Charles V in Brussels. In 1529 he attended the Diet of Speyer, at which Charles V attempted reconciliation between the Protestants and Catholics. De Selve’s Remonstrances … aux dicts Alemans, a plea to the German Protestants to reunite Christian Europe, was evidently prepared as address to the Diet.²⁴⁴ It is not known how Georges de Selve became acquainted with Jean de Dinteville but the latter was constantly at court in the 1520s and early 1530s. Georges de Selve also had the possibility of contact with the court through his father as well as through his tutor, Pierre Danès, and his elder brother, Lazare, who was a page and later Varlet de la chambre ordinaire at court.²⁴⁵ The contemporary evidence for de Selve’s visit to England is contained in the two letters written by Jean de Dinteville referred to above (see The Sitters): the first, on 23 May 1533, mentioning the pleasure and personal cheer that the visit gave him (as opposed to the melancholy he was suffering) and the second, dated 4 June, making it evident that he has gone away. The reference to Montmorency and the necessity for secrecy makes it clear that the purpose of the visit was political, but nothing more is known.

There is no contemporary evidence for the starting date of de Selve’s visit. Hervey provided a date of before Easter 1533 (i.e. 13 April) from an interpretation of the 1653 document recording the history of the picture (see Provenance), which states that the visit took place in 1532 or 1533, rather than 1533, as Dinteville’s letters attest (‘ayant des la susdicte année 1532 ou 1533 passe en Angleterere par permission du Roy pour visiter le susdict sieur de Dintevile son intime amy’).²⁴⁶ Hervey also published a copy dated 1654 of a memorandum on the subject of the de Selve family stating that de Selve’s visit took place in 1532. As the Old Style French year began at Easter, Hervey assumed that de Selve must have arrived in England before the new year 1533 began, that is, before 13 April, and therefore in 1532 (Old Style). Dinteville’s first two visits to England took place in the latter part of 1531 (Old Style calendar 1531) and from January 1533 (Old Style 1532) until autumn 1533 (Old Style 1533) (see biography). It is likely that the dates of Dinteville’s embassies were known in the family and that the first two would therefore have been recorded as 1531 and 1532, or possibly 1532–3 (more likely the latter, as recorded in the document). Therefore the seventeenth‐century documents are probably based on an accurate record of the dates of Dinteville’s embassies, rather than erroneously introducing the idea of an embassy in 1532 (New Style). However, conceivably the references in Hervey’s documents to de Selve’s visit in 1532, at the distance of over a century, are based solely on the record of Dinteville’s mission to England, rather than on specific knowledge of the precise months when de Selve was visiting (covering both years), and therefore the possibility remains that he came to England after the start of the French New Year at Easter, 13 April 1533, rather than before. This would still have allowed time for him to arrive and depart before 23 May, particularly if the mission was a secret one and he was anxious not to make his presence known. The 1653 document emphasises the friendship between the two men, and this interpretation of their relationship is supported by Jean de Dinteville’s comment in his letter (quoted above) on the pleasure the visit gave him.²⁴⁷

De Selve’s subsequent mission to Venice following his consecration in 1534 lasted over three years; he was sent for six months, beginning on 12 December 1533, and in 1536/7 he went to Rome as ambassador to the Pope and in 1539 to the Low Countries.²⁴⁸ In 1540 Georges de Selve responded to an invitation from Charles V to deliver another address on the subject of peace between the King of France and the Emperor Charles V.²⁴⁹ In 1540, while ambassador to the court of Charles V at Brussels, he fell ill and died at Lavaur on 12 April 1541.²⁵⁰

Portraits of Georges de Selve

There is a later sixteenth‐century portrait of de Selve but no other contemporary likenesses are known.²⁵¹

The Making of the Portrait

There are no identifiable references to NG 1314 in the surviving letters that Jean de Dinteville sent back to France in 1533. As mentioned above, he refers to pictures on two occasions in the course of the year. In the first, a letter of 23 May, he wants to know his brother’s news and ‘what you will say about the tower and the paintings’. In the second, of 4 June 1533, he adds at the end: ‘I pray you tell me if you have found the pictures well made’.²⁵² These references are presumably to paintings which cannot now be identified, acquired in France for the chateau of Polisy, rather than to the London commission, which was probably at this stage still in the making – or not yet begun.

As noted above, the signature on the painting, ‘Holbein pingebat’, is unique in Holbein’s work and might suggest that a considerable part of the year 1533 was spent on the painting, although there are other portraits from this year. The use of the Latin imperfect may be a deliberate reference to classical antiquity.²⁵³ The date when it began to be planned or painted presumably bears some relationship to the dates of de Selve’s visit, which may have started in mid‐April, as mentioned above (but was perhaps planned earlier in the year), and which was over by 4 June and possibly a little earlier. The economical fashion in which it is painted (see Technical Notes) indicates that the work may not have taken as long as the size and complexity of the composition might otherwise suggest; Holbein was also painting and designing other works that year.²⁵⁴ The portrait may have been ready when Dinteville left for France in November 1533: presumably he took it back to France with him then or on one of his later visits, or perhaps it was transported there as soon as it was ready.

No drawings for The Ambassadors survive, other than the study of Saint Michael (fig. 20) discussed above, which may have been made in this connection. Holbein would presumably also have availed himself of the opportunity to sketch clothing, jewels and weapons. No drawings survive representing objects in any portrait by Holbein but it is likely that such drawings were made. In particular it seems likely that the royal astronomer Nikolaus Kratzer would have had instruments that Holbein could have studied for the double portrait. As noted above, Holbein’s portrait of Kratzer of 1528 includes several instruments that resemble those shown in The Ambassadors. It is possible that Holbein merely reused sketches that he had made for Kratzer’s portrait, especially as in some cases what is shown in The Ambassadors appears to adapt the detail of what is shown in the portrait; or even that he relied on Kratzer’s drawings or copies of them, or, in the case of the torquetum, adapted a woodcut (see above).²⁵⁵ The implications of these possibilities are discussed further below. The representations of the arithmetic book and the hymnal are based on the study of specific examples that must have been made available to the artist. The somewhat wooden appearance of the face of Georges de Selve suggests that Holbein did not have very much opportunity to acquaint himself with his features, but he presumably made a drawing according to his usual practice. It is reasonable to suppose that Holbein would have had more opportunity to take Dinteville’s likeness but no drawings of either man survive.²⁵⁶

It is also possible that larger compositional studies were made. A compositional sketch with annotations and amendments survives for Holbein’s lost full‐size group portrait of the family of Sir Thomas More of 1526–8.²⁵⁷ Another surviving working drawing takes the form of the left‐hand part of a cartoon made to transfer the composition for Holbein’s lost Whitehall wall painting of 1537, with full‐size, full‐length figures of Henry VIII and Henry VII.²⁵⁸ It is possible that similar sketches, and even a cartoon, were used for The Ambassadors. However, the underdrawing has not revealed any evidence, such as pouncing dots, which would indicate the use of a cartoon, but only roughly indicated placement lines for such features as Jean de Dinteville’s chain (fig. 10).

As only a little of the portrait’s underdrawing can be discerned (see Technical Notes) indicating minor changes and adjustments, it is not known whether Holbein had established the compositional structure through, for example, perspectival or geometrical contruction. The outlining of the instruments indicates their positioning rather than making significant changes: presumably Holbein must have referred to other drawings to establish their details before making slight adjustments, for example to the top of the cylinder dial, to the top of the torquetum, to the position of the pegs on the lute and to the positions of the flutes, one of which has been added. There are, however, some changes that were made in the course of painting. One of these was to the colour of Georges de Selve’s robe, which appears to have been planned as a brighter red, perhaps to be modified by glazes. The other change is to the background, which was originally to have included what may be interpreted as wooden panelling, at least in the lower left‐hand part of the composition, perhaps an indication of a background similar to that seen in Holbein’s portrait of Thomas Cromwell of the same period, 1532–3 (fig. 56), or the portraits of unidentified female sitters at Detroit and Winterthur.²⁵⁹ Any interpretation of the painting’s meaning must acknowledge that the composition was subject to at least some change during painting, which may indicate that a certain elasticity of meaning was foreseen, rather than a rigid schema being planned.

Interpretation of the Composition

The composition of The Ambassadors is unique in Holbein’s work. He evidently painted very few full‐length portraits: apart from NG 1314, the known examples consist entirely of royal commissions, the full‐length portrait of Christina of Denmark (NG 2475) and the lost Whitehall wall painting, for which the full‐length cartoon portrait of Henry VIII and Henry VII survives. The Ambassadors seems to have directly inspired this portrait, dated 1537 in a copy made in 1667 by Remigius van Leemput: two full‐length figures stand on either side of a stone tablet, equivalent to the shelves in NG 1314, and the pose of Georges de Selve appears to have inspired that of Henry VII, in reverse (fig. 57).²⁶⁰ The paired composition in turn may have derived from earlier book title‐page designs in which Holbein placed standing apostles and other figures on either side of the title information.²⁶¹ The arrangement of two full‐length standing figures also reflects other similar compositions of Holbein’s, which include paired saints, such as those for stained glass, organ shutters and altarpieces; he would presumably have drawn on such material and perhaps followed the same methods of establishing a compositional structure for the painting. The extent of the elaboration of objects is unlike any other of Holbein’s portraits, although the half‐length portrait of the Hanseatic merchant Georg Gisze of 1532 also incorporates a number of still‐life details, the meaning of which has been debated (fig. 58). As discussed above, the portrait of the astronomer Nikolaus Kratzer of 1528 also includes many instruments similar to those in the larger painting.²⁶²

The references in the correspondence of Jean de Dinteville to Georges de Selve’s visit in London strongly suggest that the two men were friends, despite the probable political motivation for the visit. Full‐length portraiture was a well‐established genre in the sixteenth century, particularly for dynastic or marriage portraits; although many earlier examples have undoubtedly been lost, both formal and more informal full‐length portraits were evidently in use.²⁶³ There is little surviving evidence of the extent to which two male friends may have commissioned full‐length portraits of themselves but this may not have been exceptional: for example, a Bruges miniature of about 1470 (fig. 59) depicts two men contemplating a nearly life‐size full‐length double portrait of themselves, while a French manuscript of 1512–15 (fig. 60) based on the adages of Desiderius Erasmus (1466/9–1536) shows two young men standing at full length with a third, representing Animus, the soul, standing on their shoulders.²⁶⁴ It is a symbolic illustration of the virtue of friendship, ‘amicitia equalitas, amicus alter ipse’ (‘equality in friendship, the friend another self’), evidently demonstrating the topos of the friend as second self, as well as friendship enduring in spite of fortune and death.²⁶⁵ It is difficult to judge the extent to which conventions of friendship portraiture may have developed in northern Europe by the early sixteenth century. However, it is clear that not only was friendship of great significance in this period, coloured in particular for educated men like Dinteville and de Selve by knowledge of Cicero’s De Amicitia, it also motivated the production and exchange of individual portraits.²⁶⁶ Holbein himself was the author of such an exchange when he painted the portrait of Archbishop Warham in 1527 (Paris, [page 424]Louvre) on the model of the portrait of Erasmus he had made in 1523 (London, National Gallery, on loan from the Longford Castle collection).²⁶⁷

Perhaps inevitably, particularly given the contrasting secular and ecclesiastical spheres of the two sitters, NG 1314 seems to adopt some of the conventions of portraits of married couples, with Jean de Dinteville appearing in the active male role on the left, the position of heraldic right normally adopted by male sitters in such portraits.²⁶⁸ Placed closer to the viewer and more prominent by virtue of his more flamboyant and colourful dress, Dinteville is thus given superior status in the portrait, as befits his role as commissioner of the portrait, despite the suggestion of symmetry in the manner in which both men are arranged with their arms resting on the carpet over the shelves. This mirrored pose, Bomford has proposed, may conceivably suggest an allusion to the classical topos of friend as second self.²⁶⁹ The choice of a design for the green damask curtain, which includes a ring (see above), might allude to the diamond ring as a symbol of constancy and fidelity in marriage, and by extension to friendship more generally: diamond rings were regarded as symbolic of matrimony and associated with fidelity and constancy because of the durability of the stone; they were also given as tokens of friendship between men.²⁷⁰ Although the portrait is elaborate in the details of its setting, it may suggest the intimacy of friendship within a private world by representing the two subjects without the formality of dress and perhaps also pose that would be considered essential in portraits addressed to the outside world of public position and activity. In these the sitters would present themselves in their official roles as ambassadors, courtier and bishop: for example, Dinteville wears the order of Saint Michael without the chain and with the lesser version of the pendant (see above) – although de Selve was not yet consecrated and so could not wear the dress and mitre of a bishop.²⁷¹ The portrait may have hung in the relatively formal setting of the ‘grand salle’ of the chateau of Polisy, but Dinteville would presumably have closely controlled access to his family seat and to the viewing of the portrait.

Whatever the manner in which those who saw the portrait in its original setting in France approached and viewed it (see above), they would undoubtedly have experienced the way in which the distorted skull intrudes and overlays the portrait of the two young men. They would presumably also have noted the presence of the partly hidden crucifix placed at right angles in the top left‐hand corner of the painting. The interpolation of the partly concealed skull and crucifix within such a double portrait is exceptional. However, the combination of paired full‐length figures and skull occurs in Holbein’s Dance of Death woodcut series: at the end of the book – but before the Last Judgement – the triumph of death is complete and a skull is presented on a shield, with two elegantly dressed supporters, a man and woman, standing on either side (fig. 61). Showing a skull on a shield was no novelty: for example, Dürer had depicted such a shield in an engraving in which two lovers are approached by death, but the similarity between Holbein’s woodcut design and the composition of The Ambassadors is striking.²⁷² The Dance of Death woodcuts were not published until 1537 but were probably designed in Basel before Holbein first left for England in 1526.²⁷³ He may have been able to show this last design to Dinteville, who sports a small skull [page 425]hat badge in the painting. The similarities between the two compositions are evident: in the painting several instruments, some of which can be associated with telling the time, rather than a single hourglass, occupy the top shelf between the two figures, and the skull, rather than being overtly displayed as on the heraldic shield in the woodcut, is concealed through the device of anamorphosis. The fact that its shadow falls in the opposite direction to other items in the painting may serve to indicate the presence of an object from a dimension beyond the temporal.²⁷⁴

The painting includes a highly significant third element that is lacking in the woodcut, namely the semi‐hidden crucifix. Thus, rather than the triumph of death over time presented in the woodcut, the painting seems to offer the Christian promise of salvation and everlasting life. As Bomford has noted, this sentiment might also have been appropriate for a portrait of two friends. The emblem book published by Alciati (Andrea Alciato, 1492–1550) refers to ‘Amicitia etiam post mortem durans’, friendship enduring beyond death, suggesting that the portrait may depict the triumph of two Christian men of virtue over death.²⁷⁵ The likelihood that Dinteville knew Alciati or was familiar with his emblem book, which circulated in manuscript at the French court before its publication in 1531, and that his emblems might have influenced the presentation of the lute with its broken string, is considered below.

Skulls commonly intruded to complete the meaning of religious paintings and portraits, and skeletons were frequently represented in tomb sculpture, sometimes underneath the effigy of the departed.²⁷⁶ The presence of skulls intimating the nearness of death in life was common in portraiture of the fifteenth and sixteenth centuries and many examples can be cited, for instance the skull on the reverse of the diptych painted for Jean Carondelet (1469–1544) by Jan Gossaert (died 1532) early in the sixteenth century, or the portrait in the National Gallery by a follower of Jan van Scorel in which the sitter is posed with his hand on a skull holding a bunch of pansies: the name of the flower in French is also the word for thought, pensées, so the painting suggests a meditation on death.²⁷⁷ In some individual portraits, such as the double portrait of a young couple in the Cleveland Museum, the skull – or even a decaying corpse in some German examples – was only to be discovered when the painting was turned over.²⁷⁸ In a painting at Vienna of 1529 the artist Lucas Furtenagel (1505–1546) depicted the painter Hans Burgkmair (about 1473–1531) and his wife looking into a mirror to see two skulls (fig. 63).²⁷⁹

Holbein’s English portraits are in general remarkable for their lack of concern with mortality. It is notable, for example, that his several portraits of the Hanseatic merchants in London include inscriptions and sometimes specific still‐life details of a kind that are not present in portraits of English sitters. The background to the portrait of Georg Gisze (fig. 58) includes a number of details that may contain references to mortality, such as the flowers in a vase and the scales on the shelves, the latter perhaps alluding to the Last Judgement.²⁸⁰ The only portrait of an English sitter to show such a preoccupation with mortality is that of Sir Brian Tuke (1471–1545), Henry VIII’s Treasurer and Postmaster, who wears a prominent crucifix with a badge of the Five Wounds [page 426]of Christ; the portrait includes a text drawing attention to the brevity of Tuke’s lifespan.²⁸¹

Jean de Dinteville and Georges de Selve stand in a shallow space in which the accoutrements of the study are on shelves between them, and in which the principal axes of meaning of the composition are a skull and a crucifix. The development of the image of the scholar‐saint Jerome in his study into a meditation on mortality that took place in northern Europe in the early sixteenth century might well have influenced the content and composition of Holbein’s Ambassadors. Paintings produced by Netherlandish artists such as Joos van Cleve and Marinus van Reymerswaele and their followers, and inspired by Dürer in particular, show Saint Jerome seated in his study with a skull, contemplating a crucifix, clearly meditating on mortality, as indicated by inscriptions and other objects on a desk such as books and clocks.²⁸² Jerome was associated with a vision of the end of the world, and images of the Last Judgement were sometimes included in depictions of the saint.²⁸³ Contemporary secular figures such as Erasmus, painted by Holbein and Quinten Massys, were shown implicitly as Jerome, while Cardinal Albrecht of Brandenburg (1490–1545), a churchman, was shown explicitly as the saint.²⁸⁴ A painting by a Netherlandish artist in Amsterdam dated 1541 shows a couple, a man and wife, half‐length rather than full‐length, but with the familiar objects – clock and candle – and with the skull and crucifix seen in the background through a window (fig. 64).²⁸⁵ The two men in Holbein’s painting are not depicted in the guise of Saint Jerome but, like him, they are juxtaposed with the symbols of death and redemption. Other portraits by Holbein seem to show him strongly influenced by contemporary Saint Jerome imagery: in his earlier portraits of Erasmus in his study, for example, the scholar is connected with Jerome both through the setting and through his own work of editing the saint’s texts. He might well have seen copies of Massys’s portrait when passing through Antwerp in 1526 and would certainly have seen the original in the house of Sir Thomas More.

In Northern paintings of the sixteenth century and especially the seventeenth, objects such as the books and musical instruments in NG 1314 are frequently shown with skulls, hourglasses or snuffed candles, indications that all earthly endeavour is vain illusion and soon turns to dust.²⁸⁶ Symbols of the passing of time (candles, clocks, hourglasses) are juxtaposed with symbols of earthly glory or learning (books, globes, coins). The hourglass in Holbein’s Dance of Death woodcut (fig. 61), measuring the shortness of life, has an obvious relationship to this tradition, frequently related to the Old Testament book of Ecclesiastes, which lamented that ‘all is vanity’.²⁸⁷ It has been suggested that a source for The Ambassadors may lie in the writings on the vanity of the world by the magus Cornelius Agrippa, who was at the French court in the early 1530s and whom Dinteville may have known, but this seems unnecessary to account for the content of the painting: perhaps more relevant are Dinteville’s complaints of his melancholy and his miserable situation in London.²⁸⁸

As noted above, Holbein’s patterned floor bears some [page 427]similarity to the Cosmati stone floor of the sanctuary of Westminster Abbey, which Dinteville might have seen, notably at the coronation of Anne Boleyn on 1 June 1533.²⁸⁹ It is possible that Holbein was able somehow to copy and adapt the complex and rich pattern of the floor, and conceivable that the pattern was understood to symbolise the cosmos. If so, it would have a clear connection to the terrestial and celestial globes of the painting and the evident division of the objects on the shelves into earthly and heavenly realms (discussed below). However, it cannot be ascertained to what extent the cosmic meanings expressed in its composition and inscription were still clearly understood in Holbein’s time.²⁹⁰

Meanings of the Objects

If the objects shown in the picture are not traditional indications of the vanity of the world, they might be viewed simply as pieces that Jean de Dinteville and Georges de Selve themselves owned or would like to have owned and with which they wished to be depicted, luxurious items to be admired along with the splendour of the garments they wear. Yet in their number and their range, as well as in their disposition and the way in which the two men are positioned to either side so that the objects take centre stage, the items represented on the shelves go far beyond that which might be found in Saint Jerome or vanitas compositions, and have therefore seemed to demand some interpretation.

The way in which the objects are arranged on two shelves appears to be specific and precise rather than random, in particular their allocation to upper or lower shelf; their arrangement to the left‐ or right‐hand side may also be significant. The objects are clearly organised into those relating to the heavens (the astronomical globe and instruments) on the top shelf and the terrestrial sphere (books, instruments and compasses) on the lower shelf. Those on the top shelf can be related to the study of heavenly bodies, the subject of astronomy, while those on the lower shelf include a book for the study of arithmetic, two instruments useful in geometrical calculations (compass and square) and the musical instruments – lute and flutes – beside a hymnal. The choice and organisation of the objects have been related to the classical tradition of the organisation of learning into the two branches of three and four scholarly disciplines, namely the trivium and the quadrivium. The latter comprised geometry, arithmetic, astronomy and music, all of which were studied as branches of mathematical knowledge rather than for any practical application.²⁹¹ All the objects on the shelves can clearly be related to the quadrivium. In the small marginal illustration Holbein provided to Erasmus’s Praise of Folly in 1515 he showed an astronomer holding an armillary sphere and with other objects placed at his feet, symbolising the folly of the mathematician (fig. 65).²⁹² These include a harp, a globe, an arithmetical table and a compass, which again may symbolise the quadrivium. The objects in The Ambassadors, displayed between the two men as though representing the keys to the understanding of the world, might in this way be associated with the worldly vanities over which death, in the form of the skull, will triumph, in a manner that has some similarity to Holbein’s earlier images in the Dance of Death woodcuts, one of which shows an astronomer (fig. 62). It has been suggested that the ambassadors themselves represent the skills of the trivium: grammar, rhetoric and dialectic.²⁹³ It is notable that among the objects on the top shelf there is no armillary sphere, an instrument used to represent planetary motion and the usual accompaniment to representations of astronomers in Holbein’s time (although possibly the torquetum takes its place, along with some of the other instruments). Certainly the balance is uneven, with so many instruments concerned with heavenly bodies ranged along the top shelf.

The objects on the shelves may also be considered in relation to the sitters to whom they are adjacent. Those on the left side might be grouped as attributes of Jean de Dinteville. It is clearly intentional that Jean de Dinteville should stand next to the terrestrial globe that includes a prominent reference to his chateau of Polisy, which is not to be found on contemporary globes otherwise similar to this. This might be seen as a witty substitute for the inclusion of an inscription giving the name of the sitter. It is less clear whether any of the other objects adjacent to him should be [page 428]interpreted as having personal reference: the globe is next to the arithmetic book, folding rule and set square.

The dividers and set square may have been recognised as symbolic of geometry in the quadrivium. The image of the compasses was also used symbolically in the sixteenth century: images of prudence wielding a compass were well known in France as a symbol of good governance, and were used as such in a book made for the young Francis I.²⁹⁴ Indeed, Holbein designed a small badge or pendant (fig. 67) with a pair of compasses and a French text: ‘Par prudence et par compas incontinent tu viendras’ (‘You will come immediately in a prudent and measured way’), ‘compas’ here suggesting an element of plotting or machination. This badge has been associated with Anne Boleyn, who was well acquainted with French, although other courtiers had also spent time in France and owned books in French.²⁹⁵ However, the device could conceivably have been designed for Dinteville, for it might be read as a suitable motto for a diplomatic mission – and might suggest that his prudence is implied here. Geometrical instruments including the compass and the set square were also sometimes depicted as attributes of the planet Saturn, a malign, chaotic influence, and were linked with melancholy in Dürer’s famous engraving of 1514, in which the personification, Melancolia, holds a pair of compasses and is seated by a polyhedron.²⁹⁶ It seems conceivable that these instruments could have been included in Holbein’s painting for their association with disorder and the melancholy of which Jean de Dinteville complained.

The fact that the arithmetic book on Dinteville’s side is open at a page stating ‘Dividirt’ and describing a method for arithmetical division (see above) has been interpreted as a further indication that some of the objects in the painting highlight a divided world, although Nuechterlein has noted that there are more emphatic headings for division elsewhere [page [429]]in the book, and has drawn attention to what may seem an unusual choice of calculation.²⁹⁷

On the right‐hand side Georges de Selve stands next to the specifically Lutheran hymnal, a deliberately selective representation that includes two pages which do not appear consecutively in the original (see above), and to the lute with a broken string, which thus appears to indicate something more than an interest in music or the representation of this element of the quadrivium. Less obvious at first sight, one of the flutes is missing (see above). Musical instruments were common symbols of harmony, and heavenly harmonies were imagined when the spheres in the cosmos moved about the earth; lack of harmony, discord, might be a metaphor for worldly conflict.²⁹⁸

It has been suggested that the image of the broken lute here is taken directly from Alciati’s emblem book published in 1531 (fig. 68). Alciati was an Italian who had spent much time at the court of France, and whom Dinteville is likely to have known.²⁹⁹ His book also included the Latin line ‘Virtutis Fortuna Comes’ (‘Fortune the companion of Virtue’), which members of the Dinteville family had taken up as their motto, in slightly adapted form (see above). Alciati’s emblems, some of which were published in 1529, and many of which had been in circulation for some time, consisted – in their published form – of a woodcut image, an enigmatic motto and some verses; courtiers such as Dinteville often wore such images as badges, although the skull badge represented in the painting does not have an obvious relationship to any of Alciati’s emblems.³⁰⁰ One of Alciati’s emblems, first published in 1531, shows a lute (fig. 69) and the text refers to discord brought about by a string being out of tune or broken: but it is addressed to the Duke of Milan and intended to refer specifically to breaking of alliances between Italian states; moreover, importantly, the published image does not actually show a broken string.³⁰¹ Yet since Dinteville’s embassy took place in the context of pan‐European divisions and disharmonies, with which de Selve was also deeply concerned, it seems possible that the inclusion of the broken string of the lute was chosen to highlight their troubled world.

The presence of the flutes may also be intended to underline this sense of lack of harmony, for one of them is missing from the case. Moreover, the music of flutes is associated with war, and it has been suggested that the function of the flutes might therefore be to underline the presence of conflict and dissension in the world of the two sitters.³⁰² Flutes like these were associated with soldiers, as mentioned above, and can be seen, for example, with a lute in a border designed by Holbein. It has been suggested that both the musical instruments and the books in NG 1314 make specific reference to a lack of harmony and division and that, by contrast, the presence of the pomegranate motif in the curtain might be a reference to the unity of the Church, which was at this time under threat.³⁰³

The presence of the Lutheran hymnal might be thought unexpected in a portrait of two orthodox Catholic Frenchmen. It seems highly unlikely that this was chosen at random, and similarly unlikely that the two non‐consecutive pages that are shown adjacent to each other were not deliberately chosen for representation (see above). A Lutheran text in German, to [page 430]conservative opinion a highly contentious book, could hardly have been included in the portrait without the express wish of the sitters: it would have been controversial to possess such a book in England or France at this time.³⁰⁴ The crisis in the Church caused by the advent of Protestant movements was affecting France significantly; Christian disunity was the most important issue in Europe in 1533. Neither Jean de Dinteville nor Georges de Selve was sympathetic to the Reformers’ cause, although Dinteville evidently tended to liberal and Erasmian views and his bishop brother had appointed the reformer Jacques Lefèvre d’Etaples as tutor to son of Francis I.³⁰⁵ In 1529 de Selve had composed his Remonstrances aux Alemans, in which he had urged the German nation to leave behind their differences and once more unify the whole of Christendom. The inclusion in Luther’s version of the traditional ‘Come Holy Spirit’ and the ‘Ten Commandments’, texts of universal Christian significance, might have been intended to be understood as part of de Selve’s desire for Church unity and harmony, an end to current division and a reinforcement of his message in his own book. It has been plausibly suggested that the combination of the two might further allude to the Lutheran opposition of Law and Grace, represented at this time by artists including Holbein.³⁰⁶ It may well be that Georges de Selve, on whose side of the picture the book is placed, wished it to be included.

The inclusion of the Lutheran hymnal brings the content of the painting directly into relation with secular politics in 1533, with issues with which Dinteville and de Selve would certainly have been negotiating between the English and French court at that time, bringing a more specific significance than a general allusion to the human condition within a framework of Christianity.³⁰⁷ While some have argued that to view the painting as an expression of concern with English political events is too narrowly Anglocentric,³⁰⁸ others have proposed that the painting is focused on the political turmoil of 1533 on which the ambassador Dinteville was to report, and that this is supported by references to specific dates contained within the painting.³⁰⁹ 1533 was undeniably a momentous year for the reign of Henry VIII, as well as for relations between England and the rest of Christendom. Jean de Dinteville’s embassy covered the period during which Henry VIII divorced Katherine of Aragon and married Anne Boleyn, who gave birth to the future Elizabeth I shortly after; to achieve this the English Church was established in the ‘break with Rome’. Dinteville was observing the disintegration of the Catholic Church in England during his own embassy, and in France there were deep divisions over which Francis I was maintaining unity with some difficulty.³¹⁰ Anne Boleyn herself was to receive French Protestant exiles at the English court, one of whom, the poet Nicholas Bourbon (about 1503–1549/50), had his portrait drawn by Holbein.³¹¹

Some interpretations of the painting have linked the date of de Selve’s visit to London and those of external events, religious and political, with dates and times read from the instruments in the painting.³¹² However, as discussed above, the dates of de Selve’s visit are not proven: it probably began before Easter, but this is not certain: the evidence is drawn from an interpretation of seventeenth‐century records rather than contemporary documents. Dinteville’s letters provide evidence only of his departure, probably before the end of May and certainly by 4 June. It has been asserted that the date of 10 or 11 April can be read on the cylindrical pillar dial, and it has been presumed that Georges de Selve was in England on this date; the conclusion has been drawn that the painting therefore commemorates his visit. It has also been suggested that this date might have political significance for Dinteville’s mission.³¹³ Yet, as explained above, the date provided by reading the pillar dial is ambiguous and unclear: since the dial cannot distinguish between dates in early April and mid‐August (or times in morning or afternoon), it does not offer the certainty that would surely be required if the date were to be of key significance for the painting.³¹⁴ Furthermore, although all the instruments were at least theoretically capable of giving readings of the time, the only other instrument capable of displaying a date is the celestial globe, and this is set for 2.40 pm on 11 or 12 July.³¹⁵ The three faces of the polyhedral dial show different times in the morning, 9.30 and 10.30 am. The readings taken from the instruments in the painting therefore cannot support references to specific dates or times and do not appear to relate to the little that is known of de Selve’s visit.

The ambiguity and inconsistency in the dates and times shown by the instruments might be the random product of Holbein’s approach to the generation of the composition: he may have used a range of studies, with readings taken from instruments on different days and at different times; consistency of dates and times might not have been considered significant by the artist or, more importantly, by his patron. The instruments may be intended as no [page 431]more than an elaborate backdrop, an extended illustration or visual exposition of astronomy in the quadrivium, for which accuracy of depiction and congruent times might have been unnecessary. Alternatively, if the instruments are intended to evoke the measurement and passing of time, their inconsistency might reflect deliberate intent, one perhaps according with the discord or lack of harmony signalled by the broken lute string and missing flute, suggesting that the times are somehow ‘out of joint’.³¹⁶

As often noted, the instruments do not simply show different dates and times, their representation includes significant inaccuracies. This might be to some extent the result of Holbein’s putative reuse of studies that he had made for his 1528 portrait of Nikolaus Kratzer, which he then combined to produce superficially convincing depictions of the instruments from differing points of view. The extent to which Kratzer may have been involved in the planning and representation of the instruments is a significant consideration here: would Holbein have needed his assistance for the inclusion of the torquetum – or did he rely on the Apian woodcut? Did Kratzer oversee and check the representations? Any professional astronomer would have known how a quadrant should be displayed, for example. Some commentators have suggested that the inaccuracy is not intentional but a product of Kratzer’s lack of knowledge, and even that another German advisor was involved in devising the programme of the painting.³¹⁷ Nuechterlein has demonstrated that the instruments in Holbein’s 1528 portrait of Kratzer are more accurately rendered than their counterparts here, and has argued that the inaccuracies are not careless or ignorant but specific, considered and deliberate: in her view the artist went to considerable trouble to distort and obscure the presentation of the instruments, as well as some of the other objects, and Holbein and Kratzer must have collaborated closely to create such inaccuracy.³¹⁸

Nuechterlein’s interpretation of the painting focuses on the concept of incommensurability, suggesting that the representation of impossible readings, inaccuracies and distortions in the instruments is deliberate and represents the limitations of human knowledge of the universe as it is structured through forms of mathematical enquiry exemplified in the quadrivium and presented in the philosophy of Nicholas of Cusa and Jacques Lefèvre d’Etaples.³¹⁹ According to Cusa, precise mathematical concepts could never be fully grasped by the human mind, nor could a perfect circle be represented (it is notable that Holbein’s rendering in perspective of the circles or ellipses seen on the tiled floor are imperfect).³²⁰ Dinteville and de Selve were both highly educated men. Dinteville in particular was certainly acquainted with Lefèvre d’Etaples owing to his role as tutor to the son of Francis I.³²¹ De Selve may have been influenced by Lefèvre’s views, which were sympathetic to the Lutherans. It is plausible that Dinteville commissioned a painting expounding incommensurability.

From what we know of Dinteville from his correspondence, many of his concerns appear personal and universal rather than specifically political, despite his diplomatic mission. The intellectual curiosity reflected in his letter to his brother of 28 May 1533 concerning the unusual oval compass might well find its counterpart in the panoply of instruments represented in the painting. His family interest in elaborate paintings and portraits with meanings obscure today is certainly reflected in his commission of 1533. We also know that Dinteville was miserable and melancholy during much of his time in England in 1533, and that he was also frequently ill: his letters to his brother emphasise a preoccupation with his health, homesickness and the melancholy of his position in England rather than concern for political events (‘You know how much I love this country’, he wrote, with feeling irony); he wanted the coronation and royal birth to be over as quickly as possible so he could return to France. It is highly probable that he often thought of death and, as a Christian, of judgement and redemption. What we can glean from Dinteville’s correspondence may perhaps reflect the thinking that inspired the commissioning of the painting. His own remarks are more closely related to thoughts of vanitas and death than to political events, and align with ideas concerning the unknowability of the universe, such as Nicholas of Cusa’s injunction ‘do not seek God through human knowledge’. As noted above, Nicholas of Cusa also wrote about apprehending the invisible through a transparent convex or concave surface such as beryl.³²²

The whole painting may be read as a meditation on Dinteville’s melancholy, and perhaps also on de Selve’s despair at the condition of Europe. Standing on a floor that may possibly allude to the cosmos, and placed between objects including astronomical instruments, perhaps arranged to simulate heaven and earth, and which certainly allude to a world of chaos, in which such instruments of learning frustrate the encounter with God, both men think of the brevity of life and their end but also of the hope of the life to come. The counter to death, symbolised by the skull, was resurrection, and it is clearly to this end that the crucifix is placed in the top left‐hand corner, at right angles so that, like the skull, it is deliberately difficult to discern and must be discovered (fig. 70). It is probably the last thing in the picture that the viewer will find: only after admiring the two splendidly dressed ambassadors will they observe the sign of their, and our, mortality, and finally the hope of resurrection appears. Death is not triumphant here, as in Holbein’s woodcut; rather, as Bomford has argued, the portrait suggests the triumph of two Christian men of virtue over death.³²³

Together with Holbein, the sitters may have constructed a pictorial meditation on their mortality. However, whatever the motives of Jean de Dinteville in commissioning this double portrait, the ways in which the artist has conveyed meanings through a painted composition should be recognised as exceptional. The manner in which Holbein can be discerned and understood to have used his skill in representation to convey meaning extends the experience of viewing The Ambassadors far beyond the normal range of portraiture. Although it is undoubtedly structured compositionally to express Christian religious beliefs (heaven above, divided discordant earth below, death in front and salvation behind), the spatial manipulation of temporal and physical dimensions [page [432]]in the painting undermines and makes it difficult to assert specific and conventional meanings, such as the triumph of Christian virtue, and perhaps deliberately so. Rather than rendering exactly specific existing objects, as at first appears, Holbein uses his extraordinary pictorial imagination: he is deceptive and elusive, for what he represents evades us as we look. The painting becomes more illogical the more closely it is studied. The space in which the two men are set is more like a theatrical set than any attempt to evoke a room with windows and doors. The instruments on the top shelf are depicted with a clarity that encourages close study of their construction and the readings they present, although they would normally be used out of doors and there is no obvious point of entry within the composition for the sun to cast shadows necessary for some of those readings; nor does it appear that they can be read to produce meaningful dates or times. The skull itself casts a shadow, but in the opposite direction to those cast by the other objects, in a way that could be seen as ironic or mocking, reminiscent of the personifications of Death in Holbein’s woodcut, which endures while human life falters.

Holbein’s inclusion of the anamorphic skull gives a third dimension to the composition of The Ambassadors, further ensuring that the visual experience of the painting is intentionally complex and ambivalent, constantly shifting rather than precisely ordered. As with Holbein’s rendering of details, what appears at first sight to be rational and ordered is distorted and ambiguous. Space is irrationally contorted or falls away from conventional perspective.³²⁴ It is in the extraordinary nature of this painting that it conveys in extremely detailed and specific form something ambiguous and ungraspable – the skull dissolves in front of our eyes and perhaps even the sight of the crucifix is uncertain and elusive.

Select Bibliography

Hervey 1900; Heise 1959; Levey 1959, pp. 46–54; Charlton 1960; Samuel 1963; Pope‐Hennessy 1966, pp. 245–52; Rowlands 1985; Baltrusaitis 1984; Foister, Roy and Wyld 1997; Wyld 1998; Dekker and Lippincott 1999; Roy and Wyld in Roskill and Hand 2001, pp. 108–23; Bomford 2004; Foister 2004b, pp. 214–22; Roberts 2009; Nuechterlein, forthcoming.

Notes