Drawings

In Ingres’s Study for the Figure of Stratonice, the dress seems heavier than the woman wearing it. Deep shadows gather inside the folds; the cloth drops in long columns, then turns into curling edges near the feet. Above it, the head and raised hand remain lightly drawn. The difference directs attention to a particular problem: how a body, a gesture and a mass of fabric can hold together. The Met dates the sheet to 1834–40 and records graphite, black chalk, rubbed charcoal and incised transfer contours. ^[1]

Ingres’s drawings invite several kinds of attention. Some are complete portraits, made to be kept and looked at. Others test a pose, preserve a borrowed invention, organise a composition or move a carefully judged contour onto canvas. Still others return to an earlier painting after decades of reconsideration. Following these different purposes reveals a more varied practice than the familiar image of a master producing an immaculate outline.

What Ingres meant by drawing

“Drawing does not simply mean reproducing contours.” This is our translation of a statement published in Henri Delaborde’s Ingres, sa vie, ses travaux, sa doctrine in 1870. The surrounding passage includes expression, inner form, planes and modelling within drawing. A plane is a surface turning towards or away from the viewer; modelling makes that turn legible through light and shade. For Ingres, drawing encompassed the organisation of solid form as well as the line enclosing it. ^[2]

That breadth matters when looking at his work. An outline can establish the sweep of a sleeve, yet the sleeve also needs weight, its folds must turn, and its position must make sense in relation to an arm. In a portrait, the smallest shifts between the mouth, nose and eyes can change the whole likeness. These relationships belong to drawing even where no separate pencil line is visible in the finished painting.

The famous maxim “Drawing is the integrity of art” also appears on Delaborde’s page 123; the translation here renders probité as integrity, with its sense of honesty and sound judgement. It is a demanding claim about the foundations of the work. Read beside the longer passage, it places considerable responsibility on looking, selection and the construction of form. It cannot fairly be reduced to praise for a clean edge. ^[2]

Delaborde published these notes and thoughts after Ingres’s death, arranging material from his papers and correspondence. They are therefore valuable testimony mediated by an editor, rather than chapters of a finished treatise that Ingres prepared for publication. Their emphatic language also needs to be set beside the objects. The drawings contain searching marks, changes of mind, broad rubbed shadows and practical aids to transfer. The statements give us a vocabulary for considering those decisions; the sheets show how varied the decisions could be. ^[2] A reader interested in the wider relationship between drawing and paint can continue with Style and Technique. Here, the individual sheet remains the starting point.

A portrait could be complete on paper

The Biography follows Ingres’s Roman years. Here, the portrait sheet itself is the starting point: how do the face, clothing and surrounding paper divide our attention? In the Lethière portrait, the shaded face and curled hair occupy a small part of a much larger sheet. The child’s head sits beside the adult’s hands, while the broad skirt spreads below them. Close detail and open paper are both present. Following their distribution is a way into the image before deciding what its degree of finish might mean.

The Paintings guide discusses finish in painted portraits. The drawings here pose a narrower question: which passages carry the likeness, and what does the paper leave open? Looking for those differences gives the eye a route around the sheet before considering its history. The Met identifies the sitters as Rosa Meli, wife of Alexandre Lethière, and their infant Letizia. It dates the graphite sheet to about 1815. ^[3]

Ingres, Madame Alexandre Lethière and Her Daughter Letizia, c. 1815. Graphite. The Metropolitan Museum of Art, 43.85.7. ^[4]

Drawing was well suited to preserving those connections. It could unite likeness, costume and a seated or standing pose without requiring the material extent of an oil portrait. A family group could give several people distinct positions within one shared space. Such images have their own scale and rhythms; calling them preliminary sketches would assume a future painting for which there may be no evidence.

Ingres, The Kaunitz Sisters: Leopoldine, Caroline and Ferdinandine, 1818. Graphite. The Metropolitan Museum of Art, 1998.21. ^[5]

The Met identifies Nathalie Gallois as a cousin of Ingres’s second wife, Delphine Ramel, and describes the 1852 sheet as graphite with gold in oil. Its inscription records a gift to Monsieur Gallois. ^[6] The distinction between independent portrait and preparatory study should therefore come before an assessment of finish. A sheet’s inscription, provenance, related works and physical signs of use may help establish its purpose. Its degree of detail alone cannot settle the matter. The broader social and artistic range of these likenesses is explored in Portraits.

Stratonice: working through the weight of cloth

Chantilly dates Antiochus and Stratonice to 1840, when it was completed for the Duke of Orléans. ^[7] On the separate figure study shown here, the drapery has room to spread across the paper without the surrounding narrative. Look first at the shoulder. Several dark bands sweep diagonally across the upper body. Below the narrow belt, the direction changes: long folds descend towards the floor, broken by broad passages of shadow. Near the lower centre, one edge turns back on itself, creating a succession of pale openings and dark interiors. These are substantial volumes. The figure’s silhouette gives the cloth its boundary; the shifts within that boundary make it appear thick, flexible and affected by gravity.

The head is much less emphatic. Its inclined position is established, with the features and hair indicated by fine lines. The hand near the face is also lightly described. This distribution of effort gives the study an unusual visual balance: a relatively delicate upper passage sits above a dense body of fabric. It would be misleading to call the whole sheet equally finished, but equally misleading to judge it by the finish expected of an independent portrait. The artist has concentrated effort where the drawing most visibly requires it.

There are also places where the figure’s edge is explored through more than one line, especially around the lower drapery. Those alternatives matter. A contour can be tested, moved and settled; an apparently effortless shape may preserve the results of several decisions. The lightly suggested forms beside the figure make the concentration of the darker passages still more apparent. Blank or nearly blank paper helps distinguish the area of sustained investigation from everything left secondary.

The Met records contours incised for transfer. ^[1] The dark and light passages in the reproduction are visible, but an incision is a physical property of the paper. A photograph alone cannot tell us which of its many lines was pressed into the surface.

The combination of materials helps explain why “pencil drawing” can be too loose a description. A fine graphite line can establish a delicate edge, while softer dark media allow broader tonal work. Here the visible result ranges from faint linear description to shadows wide enough to give the cloth a sculptural presence. The medium is part of the thinking: line establishes relations, tone tests volume, and transfer preserves selected decisions for further use.

A pale standing figure occupies the left of the painting, while a large dark canopy encloses the bed. The floor’s patterned bands pull the eye across the space. In the isolated drawing, those surrounding patterns disappear: the large mass of cloth can be followed from the shoulder down to the feet.

Ingres, Study for the Figure of Stratonice, 1834–40. The Metropolitan Museum of Art, 63.66. ^[8]

From a study to a painted portrait

A drawing, a finished painting and an image made during technical examination preserve different kinds of evidence. The following comparisons distinguish what can be seen on the reproduced surface from findings reported by a conservator. Keeping that distinction clear makes it possible to look closely without claiming to see through paint.

The grid beneath Madame de Senonnes

Hale found a partial transfer grid beneath Madame de Senonnes, spaced at 17 cm in the lower part. ^[9] In the finished portrait, the face sits above a pale ruff and a necklace. The red velvet spreads into broad folds across the body; the hand resting on the cushion interrupts that larger mass with a cluster of small, sharply distinguished shapes. Behind the sitter, the mirror gives another view of her head. These visible relationships offer several scales of attention, from a tiny ornament to the broad sweep of a sleeve.

This is a comparison of appearances on the finished surface. A dark painted stroke, a shadow beside an edge and a groove in a support can look similar in a reproduction. Before calling any of them a transfer mark, the reader needs evidence about the particular object. Close looking can formulate that question; it cannot supply an unseen physical examination.

Why make a nude study for a dressed sitter?

For the Princesse de Broglie, Hale identified matching grids in the nude and dressed studies and beneath the painting. ^[9]

Ingres, Nude Study for The Princesse de Broglie, c. 1852–53. Graphite on squared paper. Musée Bonnat-Helleu, Bayonne; reproduced as figure 277 in the 1999 catalogue. The model’s identity is not established. ^[10]

Ingres, Dressed Study for The Princesse de Broglie, c. 1852–53. Graphite and red chalk on paper. Reproduced as figure 278 in Portraits by Ingres, 1999; location then unknown. Complete small reproduction. ^[11]

Look first at the crossing arms in the nude study. One forearm rises diagonally while the other crosses the body, making the hands part of the pose rather than detachable details. The dressed sheet surrounds a similar upright figure with a broad skirt. Comparing the two directs attention to the space that clothing occupies around a body. Neither picture, by itself, identifies the person who posed for an unclothed study.

The separate arms sheet concentrates that arrangement into a much smaller field of attention. Long fingers extend beyond a broad wrist; the cuff and bracelet create bands across the forearm. The meeting of the arms is a junction of curves and overlaps. Looking back from this sheet to the full figure makes those connections easier to follow, without requiring every mark to be a final decision.

Ingres, Study of Arms for The Princesse de Broglie. Charcoal on tracing paper. Reproduced as catalogue 146 in Portraits by Ingres, 1999. Complete small reproduction. ^[12]

The sheets can be compared through a sequence of questions: where does a contour recur, where does its direction change, and how does cloth enlarge the figure’s silhouette? Similarity is a starting point for comparison. Establishing the precise means by which one contour reached another surface requires further evidence. The surviving images give the reader something specific to ask about.

Copying, tracing and the work of other hands

Ingres’s advice about studying the masters helps connect observation to these practical operations. In the notes published by Delaborde, he presents the study of antique art and earlier painters as a way of learning to see nature. The examples supply ways of understanding form that the artist can make his own. Copying, in this account, is an active part of education and invention. ^[2] Vigne records Charles Gounod tracing engravings under Ingres’s supervision. ^[13] Chantilly identifies Calamatta’s Prince Royal drawing as a sheet retouched by Ingres after the prince’s death in 1842. ^[7]

Paganini: a portrait becomes a template

The Met catalogues Niccolò Paganini, about 1830, as a counterproof or tracing strengthened with graphite and gouache on translucent paper. Its commentary favours tracing from the Louvre drawing, reversed and reworked for Calamatta’s engraving. ^[14]

Ingres, Niccolò Paganini, c. 1830. Counterproof or tracing strengthened with graphite and gouache. The Metropolitan Museum of Art, 43.85.10. ^[15]

The white neckcloth makes a small, bright triangle below the face. Around it, the coat is lightly outlined. Lower down, long straight edges from the bow cut across the curves of the hands and violin. Those different kinds of mark make the lower half feel more open than the closely shaded head. Follow one continuous edge through the otherwise pale clothing before returning to the face.

Tracings could also serve much smaller revisions. The Met’s Study of Hands for “Christ Among the Doctors”, dated 1852–62, is graphite on tracing paper and only 11.1 × 11.3 cm. The museum relates it to the addition of a cropped figure in the lower right of the painting during its late development. A small working sheet can preserve a change in a large composition that would otherwise be difficult to isolate. ^[16]

Ingres, Study of Hands for Christ Among the Doctors, 1852–62. Graphite on tracing paper. The Metropolitan Museum of Art, 2022.25.11. ^[17]

These examples complicate the idea that tracing was a substitute for knowing how to draw. It could maintain scale, reverse an image, coordinate a detail with a larger design or allow a collaborator to work from established forms. Its significance depends on the job it performed. For the subsequent circulation of those designs, see Prints and Reproductions.

Drawings made for stained glass

Chantilly records seventeen cartoons for Saint-Ferdinand and eight further designs for Dreux after the prince’s death in 1842. ^[7] Drawing here extends an artist’s reach through the work of other makers. Paper establishes the figure, pose and arrangement, while the final window requires decisions in glass and its assembly. Preserving the distinctions between sketch, cartoon and completed window helps the reader understand both Ingres’s contribution and the collaboration required to give it material form.

Homer: a drawing after a painting, still changing

The Louvre dates Homère déifié, RF 5273, to 1865 and records graphite, brushwork, gouache and grey wash. It measures 64.5 × 85.5 cm. ^[18] The related 1827 Apotheosis of Homer is a different object. In the reproduced drawing, a seated central figure is raised above two reclining figures. Steps divide this upper group from the crowded lower foreground. Vertical columns and the triangular roof mark the middle, while lines of heads and shoulders extend sideways. Even at this small scale, the eye can follow a few large divisions before attempting to identify individual people.

Delaborde’s historical catalogue, entry 180, gives a particularly revealing account of the drawing’s instability. He describes figures introduced, removed and reconsidered, followed by further changes after a photograph had already been made. In his account, Mozart and André Chénier entered at a late stage, with Chénier replacing Dryden. A photograph circulating as a record of the drawing could therefore preserve an earlier state. The apparent precision of reproduction did not guarantee that it showed the latest version. ^[2]

This is a useful lesson when comparing images today. Differences between a drawing and its reproduction may belong to the work’s history, as well as to the quality of the photograph. Equally, a late drawing is a hazardous key to the cast of an earlier painting: the figures have been reconsidered across decades. The date and identity of the particular object must precede the interpretation. Rabinow records Haro showing the unsold drawing in 1865. ^[19] The Chantilly publication documents Aumale’s later attempt to acquire it, between 1868 and 1870. ^[7]

Ingres, Study for the Figure of the Iliad in The Apotheosis of Homer, c. 1827. The Metropolitan Museum of Art, 1972.118.219. ^[20]

Ingres, Homer Deified, begun in 1840 and signed in 1865. Graphite, brush, gouache and grey wash. Musée du Louvre, RF 5273 recto. The separate later drawing. ^[21]

How a drawing recovered its sitter’s name

A portrait’s title can also change through research. In a 1971 article, Hans Naef investigated an 1817 Roman drawing that had successively been associated with the names Falzacappa and Fiano. Parish household records placed Luisa and Giovanna, daughters of the Duke of Fiano, in the relevant family. Marriage records then established their marriages and supplied a possible explanation for the stray name Falzacappa: an archbishop of that name officiated at Giovanna’s wedding. ^[22]

Naef’s eventual identification of Luisa Papafava rested on a combination of archival investigation, family history and comparison with two other likenesses: a marble bust by Rinaldo Rinaldi and a Saint Catherine painted by Ferdinando Cavallieri using Luisa’s features. Family information also helped him reconstruct the drawing’s descent. Yet one question remained: although the portrait was made in her wedding year, he could not determine whether it showed a bride or a newly married woman. A recovered identity could be persuasive while the precise occasion remained uncertain. ^[22]

What technical examination can tell us

When reading a technical report, begin with the identity of the object, the area examined and the type of record being discussed. A normal photograph, a working sheet and an examination image answer different questions. A conclusion about one small passage should not silently become an account of how an entire picture was made.

A useful comparison also keeps the wording of a catalogue separate from the reader’s observations. The nude Broglie study has a visible grid; its lines can be followed across the reproduced sheet. A claim about marks concealed inside a painting has another evidential basis. Naming that basis allows a reader to distinguish an observation they can repeat from a reported finding.

The same discipline helps with questions of authorship. A measured dimension, a material description and an attribution do different jobs in a catalogue. None should be made to stand for all the others. Read the stated maker, the particular contribution and any qualification together before describing an object as a collaboration. In 2000, Tinterow treated Hockney’s camera-lucida proposal as a hypothesis requiring demonstration. ^[23]

Delaborde places the admonition against reliance on a compass among advice about drawing and judgement. ^[2] That printed advice and the marks on a particular sheet invite different questions: what did the artist recommend, and what survives on this object? The Quotes page considers the published sayings and their contexts.

Collections, records and the life of the sheet

The Musée Ingres Bourdelle in Montauban is central to the survival of this material. The museum describes a gift in 1851 followed by the bequest of 1867, bringing roughly thirty paintings and several thousand drawings to the artist’s native town. Its resources encompass drawings, other artists’ work, photographs and manuscripts. This breadth preserves the environment of a working artist as well as an ensemble of admired finished images. ^[24]

That distinction matters when reading collection totals. A museum’s complete holdings are not a count of autograph Ingres drawings. A sheet associated with his studio may be by a pupil; a photograph may record a lost or altered state; an engraving may transmit a design through another maker’s interpretation. The value of such material depends partly on retaining those differences. Bertin records Ingres donating fifty-six prints or photographs after his works to the Bibliothèque Impériale, 1859–65. ^[25] Useful labels name maker, material, date and relationship. “Study for”, “copy after”, “tracing” and “replica” describe different things to look for in a catalogue.

When those distinctions are kept in view, Ingres’s drawings become easier to see closely. A pale head above dark drapery, a separately drawn arm, or an old composition opened to new figures records a particular act of judgement. The paper is the place where that act remains available to us.