Jean-Auguste-Dominique Ingres, La Grande Odalisque, 1814, Musée du Louvre

Jean-Auguste-Dominique Ingres

1780–1867 · France

Researched by Daniel Bilmes, painter and educator.

Ingres built a painting from drawing outward. He made hundreds of preparatory studies, engineered the figure from tracing-paper composites, transferred the contour to canvas with a blind stylus or a squared grid, blocked in thin over a bright white ground, then glazed colour and blended until the brush left no trace. Line decided everything; colour came last and served.

ProcessLayered TemperamentMeasured Lineage19th-Century Studio

Signature moves

Decide the painting in drawing, not in paint

Treated drawing as seven-eighths of painting and resolved a picture on paper first (nude studies, drapery studies, hand studies) before colour was allowed in. He made more than five hundred preparatory drawings for the Dampierre murals alone.

Why it matters · The structure of the picture gets settled in graphite, where a wrong line costs nothing, before any colour risk. By the time oil goes down, every decision is already made. The drawing is the painting; the paint only dresses it.

Henri Delaborde, Ingres: sa vie, ses travaux, sa doctrine, 1870

Engineer the figure from tracing-paper composites

Drew an arm, a leg, or a fold of drapery three or four times on one sheet, hunting the contour through short exploratory strokes, then laid tracing paper over the best versions to stack, shift, and refine them into a single master drawing.

Why it matters · It breaks the impossible task of drawing a perfect figure in one pass into parts that can each be solved on their own, then assembled. The final contour is engineered piece by piece, not caught in a single lucky attempt.

Technical studies of La Grande Odalisque; the Morgan Library & Museum

Judge the masses through half-closed eyes

Told his students to look at the live model through half-closed lids, squinting so the distracting local detail dropped away and only the primary masses of light and shadow remained.

Why it matters · You cannot place the structure of a head while you are busy with an eyelash. Squinting throws away the small information and leaves the big relationships of light and dark, which is what the drawing has to get right first.

Amaury-Duval, L'Atelier d'Ingres, 1878

Transfer the contour with a blind stylus

For a finished portrait he laid the resolved drawing over a fresh sheet and traced the contour with a hard point, inscribing colourless grooves into the paper below, then drew over those blind grooves in graphite. The graphite skips the low points, which proves the groove came first.

Why it matters · A contour you spent weeks resolving gets moved to the next surface without being re-drawn, and re-degraded, by hand. The transfer protects the decision. The cushioned drawing board under the sheet is what keeps the sharp point from tearing the paper.

The Morgan Library & Museum, "Ingres at the Morgan: Materials and Methods" (RTI study)

Square the grid to scale a composition

Divided a sheet into a grid of small and large squares to transfer or rescale a composition onto another sheet, square by square.

Why it matters · A small, fully resolved drawing becomes a large canvas with its proportions intact. The hard compositional decisions are made once, at the scale where they are easy to see, and then carried up faithfully.

The Morgan Library & Museum, "Ingres at the Morgan: Materials and Methods"

Scrape the wrong line off and burnish it smooth

Removed a wrong contour by scraping the paper fibres and graphite off the surface, then burnishing the spot flat again. The changes cluster in collars, cuffs, and the outlines of the torso.

Why it matters · An error is physically lifted out, not drawn over. The corrected line lands on clean paper, so the final contour stays as crisp as a first attempt. Nothing accumulates as a smudge under the answer.

The Morgan Library & Museum, "Ingres at the Morgan: Materials and Methods"

Hide the brush entirely

Abhorred the visible brushstroke and blended the oil with soft sable until no evidence of the brush was left: the porcelain, enamel-smooth surface, with local colours only faintly modelled in light by half-tones.

Why it matters · A smooth surface stops the paint from competing with the drawing. Where a Sargent wants you to see the stroke, Ingres wants you to see only the line and the form. The handling is engineered to disappear.

Ingres's stated detestation of visible brushwork; standard description of his finish

Bend the anatomy to serve the line

Elongated limbs and added vertebrae where a smoother, more harmonious contour demanded it. The back of La Grande Odalisque (1814) famously carries more vertebrae than a spine has.

Why it matters · He held the ideal contour to be more true than the literal skeleton. Line is the master and anatomy the servant. When the two conflict, the painter decides in favour of the curve. The "error" is a choice.

Contemporary critical record on La Grande Odalisque, 1819

Hold a painting for decades until the line is right

Reworked canvases across whole careers: Venus Anadyomène begun 1807 and finished 1848 (about forty years); The Source begun in the 1820s and completed 1856; Madame Moitessier begun 1844 and finished 1856.

Why it matters · Finish is governed by whether the drawing satisfies him, not by a deadline. The measurer never signs off until every relationship resolves. A painting is allowed to wait years for the painter to see what is still wrong.

Standard chronology; Amaury-Duval, L'Atelier d'Ingres (the master "never satisfied"), 1878

Studio

Light: Worked entirely indoors under steady natural studio light, never out of doors. His pupil Raymond Balze recorded the exact rhythm of a finished portrait drawing: four hours of the sitter's time, split into an hour and a half in the morning, a shared lunch with the model, and two and a half hours in the afternoon. He used the midday break to watch the sitter relaxed and off guard, and rarely retouched a drawing the next day.
Position: Worked both standing and sitting, depending on the scale of the canvas and the stage of the work. For oils the rhythm was fraught: long gaps between sittings while he agonised over a pose. He wept over the pose for the publisher Louis-François Bertin until he found the definitive seated one by watching Bertin sit and argue politics off-duty. He trained in David's Paris studio, spent eighteen years in Italy (Rome from 1806, Florence 1820 to 1824), returned to Rome as Director of the Académie de France (1835 to 1841), and ended as a professor at the École des Beaux-Arts.
Tools: Medium-hard graphite (his primary medium for portraits) · Fine imported English wove paper, smooth (preferred over textured laid paper) · Prefabricated cushioned drawing boards: wove paper over cardboard on a laid-paper cushion · A blind stylus for inscribing transfer grooves, also used to incise detail through wet paint · Tracing paper for stacking and refining studies into a master contour · A squared grid for scaling and transfer · Fine red sable brushes, soft enough to lay paint without leaving a stroke
Notes: He took his students to the Louvre to copy Raphael and the antique. A biographical aside that became a French idiom: Ingres played the violin seriously (a friend's 1818 painting of his studio shows him pausing to play while his wife posed), and "violon d'Ingres" is now French for a serious second pursuit beside one's main work.

Source: Raymond Balze (recounted); Amaury-Duval, L'Atelier d'Ingres; the Morgan Library & Museum; standard biography

Palette

Ground: A bright lead-white ground, to push light back up through the thin glazes laid over it. He then blocked in with dilute washes of warm tone: on the Portrait of Paul Lemoyne, conservators found the background put in with a dilute green wash, the bristle striations of a three-quarter-inch brush still visible under a later gray-green scumble.
Whites: Lead white (flake white)
Earths: Yellow ochre · Raw umber
Colors: Naples yellow · Vermilion · Light red · Ultramarine (lapis lazuli) · Prussian blue
Blacks: Vine black · Bone black
Medium: Oil, indirect: thin, fluid, slow-drying layers and glazes over the underpainting, with little medium, blended with soft sable until the brush left no visible mark. Warm overall tonality, so the few cool notes read. The exact medium formulation (linseed, walnut, or a resinous megilp) is not recorded.

Source: Nelson-Atkins Museum of Art, technical analysis of the Portrait of Paul Lemoyne; Ingres palette studies — Pigments reconstructed from technical analysis and palette studies, not a verbatim primary inventory. [link]

Workflow, from blank canvas

1. Draw the figure from many studies
Separate graphite studies for the nude, the drapery, and the hands, on smooth English wove paper, each part drawn several times until its contour is fully resolved.
Why: The figure is solved in the medium where mistakes are cheap. Each part is decided on its own terms before it has to live inside a painting.
2. Assemble the composition on tracing paper
The best versions of each part are stacked and shifted under tracing paper into a single settled composition, through more drawings (sometimes hundreds for a major work).
Why: Nothing about the picture is left to be discovered in oil. The whole image exists as a finished, engineered drawing before a canvas is prepared.
3. Transfer the resolved design to canvas
The contour is carried over by blind stylus, tracing paper, or a squared grid, depending on the job, not re-drawn freehand.
Why: The decision that took weeks to reach is moved without being degraded by a fresh hand. The transfer is a way of protecting the drawing.
4. Block in with dilute washes
Over the bright white ground he laid dilute washes of warm tone to establish the structure. On the Paul Lemoyne portrait conservators found the background put in with a dilute green wash, the marks of a wide bristle brush still showing.
Why: The thin washes set the value and temperature map without burying the ground. The white keeps shining up through everything laid over it, which is where the luminosity comes from.
5. Glaze colour in thin layers
Thin layers and glazes of colour go over the block-in, with little medium, keeping a warm overall tonality with cool accents placed against it. Fine detail is sometimes incised through the wet paint with a stylus.
Why: Colour is added last and kept subordinate. It dresses a structure that is already complete, rather than building the structure itself.
6. Blend the brush out of the surface
The paint is worked with soft sable until no brushmark survives, producing the enamel finish with half-tone modelling in the lights.
Why: The handling is engineered to disappear so that nothing on the surface competes with the line. The drawing stays the subject.
7. Rework until the line is right
On paper, errors are scraped off and burnished smooth; in oil, a canvas may be reopened for years. He painted out the bust of Napoleon's son behind Monsieur de Norvins after the empire fell, and reworked Madame Moitessier across twelve years, banishing her restless daughter from the picture and changing her dress to keep up with fashion.
Why: There is no schedule. The painting is finished when the draughtsman is satisfied with the contour, and not one session before.

Refusals — what they would not do

Refused to let colour lead; he called colour the "lady-in-waiting" to drawing, there only to make the perfections of art more agreeable.
Refused the visible brushstroke; blended until no mark of the brush remained.
Refused anatomical accuracy when it fought the line; added vertebrae and elongated limbs for a smoother contour.
Refused to be hurried to a finish; held canvases for years, even decades, until the drawing satisfied him.
Refused photography and the novelty argument; built his figures from life drawing, classical models, and tracing-paper composites, not from mechanical copies of reality.

Reference

Primary source: The live model for figures, but seen through the lens of classical antiquity. He trained himself and his students to correct living people toward Greek vases, Roman statuary, and Raphael. Nature supplied the information; the antique and the Old Masters supplied the standard. Drawing, not optical truth, was the goal.
Photography: Rejected. He relied entirely on life drawing, classical models, memory, and tracing-paper composites. No primary source places photographs in his method, and his whole practice runs against mechanical reproductions of reality.

Exceptions

Copied Raphael and antique sculpture constantly, as the corrective standard against which the live model was judged.
Worked from the natural, observed pose rather than an imposed one: he found Monsieur Bertin's definitive seated pose only by watching him sit and argue politics off-duty.

Lineage

Teachers

Jacques-Louis David · c. 1797 to 1801 — Trained in David's Paris atelier, the centre of French Neoclassicism, before winning the Prix de Rome in 1801.

Influences

Raphael (his deepest model, copied throughout his life)
Greek vase painting and antique sculpture

Students

His Paris atelier ran about nine years and trained roughly eighty pupils on a strict line-over-colour doctrine
Théodore Chassériau, reportedly called his truest pupil and predicted to become "the Napoleon of painting"
Amaury-Duval, an early pupil and author of the firsthand atelier memoir L'Atelier d'Ingres (1878)
Hippolyte and Paul Flandrin; Paul and Raymond Balze
Henri Lehmann
His drawing-first doctrine carried on to the young Degas, who sought Ingres out in 1855.

In their own words

“Drawing is the probity of art. To draw does not mean simply to reproduce contours; drawing does not consist merely of line: drawing is also expression, the inner form, the plane, the modelling.”

— Ingres, Collected in Henri Delaborde, Ingres: sa vie, ses travaux, sa doctrine, 1870

From Ingres's own notes, gathered after his death. Translated from French.

“Drawing is seven-eighths of what makes up painting.”

— Ingres, Collected in Delaborde, Ingres: sa vie, ses travaux, sa doctrine, 1870

“The masterpieces of antiquity were made from nude models such as we have before us in Paris today. You must find the secret of beauty in truth.”

— Ingres, to his students, Recounted by his pupil Louis Janmot, 1878

The live model is the route to the ideal, not an alternative to it. Translated from French.

“Colour adds ornament to painting; but it is no more than the lady-in-waiting, because all it does is make the true perfections of art more agreeable.”

— Ingres, Attributed saying, widely anthologised from his pensées

The clearest statement of his rank order: drawing first, colour in service.

“Draw lines, young man, many lines, from memory or from nature; it is in this way that you will become a good artist.”

— Ingres, to the young Degas, Recounted by Degas, 1855

Second-hand, but from the recipient. The drawing-first doctrine passed forward to the next generation.

Techniques and practices

Squaring Up from Studies

Transferring a small master sketch to a large canvas via a grid, preserving proportion across scale.

Tonal Imprimatura

A thin, neutral-colored wash applied over the full canvas before painting begins, killing the white and establishing a middle value.

blind-stylus-transfer

enamel-smooth-finish

What was Ingres's painting technique?

An indirect oil method built on drawing. He resolved a composition in hundreds of graphite studies and tracing-paper composites, transferred the contour to canvas, blocked in thin dilute washes over a bright white lead ground, then glazed colour in thin layers and blended with soft sable until no brushstroke showed, leaving the porcelain, enamel-smooth finish.

Did Ingres value drawing or colour more?

Drawing, by a wide margin. He called drawing "the probity of art" and "seven-eighths of what makes up painting," and described colour as the "lady-in-waiting," there only to make the picture more agreeable. Colour was added last and kept subordinate to line.

How did Ingres transfer his drawings to canvas?

Three ways, documented by the Morgan Library: tracing paper for preliminary designs, a blind stylus that inscribed colourless grooves he then drew over in graphite, and a squared grid for scaling a composition up. The point was to move a resolved contour without re-drawing and degrading it.

Did Ingres use photography?

No. Although he lived into the photographic age, no primary source places photographs in his method, and he rejected mechanical copies of reality. He built his figures entirely from life drawing, classical models, memory, and tracing-paper composites. (His rival Delacroix, by contrast, posed models for the camera.)

Why does Ingres's Grande Odalisque have too many vertebrae?

On purpose. Ingres elongated the figure and added vertebrae to lengthen and smooth the line of the back. He held the ideal contour to be more true than literal anatomy. When the curve and the skeleton conflicted, he chose the curve.

If this painter is your match

You decide the picture before you commit the paint. The drawing is where the real work happens, and you would rather hold a canvas for a year than sign off on a contour that is still a millimetre wrong. Colour serves the structure; it does not invent it.

Borrow this: Resolve the whole composition as a finished drawing before you prepare a canvas, building the figure from tracing-paper composites so each part is solved on its own. Transfer it rather than re-drawing it freehand, because a grid or a traced contour protects the decision you already made. Block in thin over a bright white ground so the light comes back up through the glazes. Then keep colour subordinate: thin, warm, and quiet enough that the line stays the subject.

Adjacent painters

Isaac Levitan1860–1900

The Peredvizhniki lyricist who invented the Russian mood landscape by trusting memory over direct observation and finishing paintings by knowing when not to touch them.

Ivan Kramskoy1837–1887

The intellectual strategist of the Peredvizhniki, whose studio ran on analytical silence, early photographic reference, and the conviction that a portrait was a biography rather than a likeness.

William-Adolphe Bouguereau1825–1905

The Parisian academic master who ran his studio on a factory schedule—7 AM until dark, no lunch break—and resolved every figure, every fold, and every leaf in preparatory studies before a single brushstroke landed on the final canvas.

Lawrence Alma-Tadema1836–1912

The Dutch-born Victorian archaeologist-painter who built a private library of five thousand photographs of Roman ruins, reconstructed marble and bronze from the actual excavations at Pompeii, and resolved every canvas as if he were producing forensic evidence that the ancient world looked exactly the way it did.

Shared the workbench

Other researched painters who used at least one of Ingres’s techniques.

Ilya Repin1844–1930

The Peredvizhniki history painter and portraitist who worked from zenith-lit studios, standing, from long social sittings, and painted monumental scenes from years of field observation.

Ivan Shishkin1832–1898

The Peredvizhniki landscape master who lived in the forest in summer and reconstructed its anatomy in the studio in winter, using photography and projection as tools of discipline rather than shortcuts.

Vasily Surikov1848–1916

The Peredvizhniki monumental reconstructionist, who built history paintings like buildings—over years, from authentic artifacts, trained crowds of real faces, and a structural drawing logic inherited from Pavel Chistyakov.

Giovanni Battista Tiepolo1696–1770

The Venetian Rococo master who planned monumental ceilings through small, fully resolved oil modelli and executed them in wet plaster at the speed a buon fresco giornata demanded.

Jan Matejko1838–1893

The Polish history painter who built monumental canvases over Van Dyck brown underpaintings, aggressively adopted new industrial pigments the year they became commercially available, and filled his Kraków studio with authentic seventeenth-century armor and textiles.

William-Adolphe Bouguereau1825–1905

Primary sources

Amaury-Duval, L'Atelier d'Ingres: souvenirs, 1878. Firsthand memoir by a pupil. Source for Ingres's constant scraping-out, the half-closed-eye instruction, and decades-long dissatisfaction. French.
Henri Delaborde, Ingres: sa vie, ses travaux, sa doctrine, 1870. Collects Ingres's own pensées and doctrine on drawing and colour. French.
The Morgan Library & Museum, "Ingres at the Morgan: Materials and Methods". RTI imaging study of the Morgan's Ingres drawings: papers and watermarks, cushioned drawing boards, blind-stylus transfer, gridding, and scrape-and-burnish corrections. [link]
Nelson-Atkins Museum of Art, technical analysis of Ingres's Portrait of the Sculptor Paul Lemoyne. Conservation study: off-white ground, dilute green-wash block-in, and detail incised through wet paint with a stylus. [link]
Raymond Balze, recounted account of Ingres's working hours. First-hand source for the four-hour portrait-drawing sitting and its morning/afternoon split.
Degas, recounted advice from Ingres (1855). Preserved in the Degas literature (Paul Valéry, Degas Danse Dessin, and others).

See how every master in the atlas worked, indexed by method →