What the Opera Paid Marie van Goethem

The most quoted sentence about her face was written about two other pictures

The phrase that has followed Marie van Goethem for more than a century, the low foreheads and the jutting jaws, was written by Joris-Karl Huysmans about two other pictures. In his account of the sixth impressionist exhibition, collected in L'Art Moderne in 1883, Huysmans works through what Degas had sent: a view of the wings, singers on stage, the dancer Becat. Then he writes, and the words are his own list-making, add two more sketches, criminal physiognomies, animal muzzles, with low foreheads, jutting jaws, short chins, shifty evasive eyes and a very astonishing nude woman at the end of a room, and you will have the list of the drawn or painted works brought by this artist.

The sentence is a full stop. Everything before it is drawing and paint. Only after it does he turn to the object, and he turns with an announcement: the interest of the exhibition this year is not in the drawn or painted work, it is entirely in a wax statue called Petite Danseuse de quatorze ans, in front of which the public, bewildered and as if embarrassed, flees. Two pages later he gives his verdict on it. This statuette, he writes, is the only truly modern attempt I know of, in sculpture.

The two works he had just listed are identifiable. The National Gallery's own scholarly edition names them as a pair of pastels, Criminal Physiognomies of 1880 and 1881, one in a private collection and one in the National Museum in Belgrade, which some critics of the day connected with recent murder trials. So the criminal vocabulary in that sentence was spent on two portraits of actual accused people before Huysmans reached the dancer at all.

None of this means the wax escaped a physiognomic reading. It did not, and the honest version is more interesting than the misquotation. The critic who read vice into this girl's face is Paul Mantz, in Le Temps of 23 April 1881, and the National Gallery's edition calls his the most comprehensive physiognomic criminal account of the figure, quoting him on the arms whose blotches are truly horrible to see and on the face, or more appropriately, as he put it, her little muzzle, which projects with bestial effrontery, a face where all the vices imprint their detestable promise. Huysmans himself was not kind either: he describes her face as sickly and grey, drawn and old before its time.

The link between the wax and those two criminal pastels is a modern scholarly argument rather than a contemporary one. The National Gallery's edition attributes it to Anthea Callen and to Douglas Druick, initially with Peter Zegers. It says their arguments are rarely questioned and then questions them: apart from the forehead hidden under her bangs, the figure lacks the standard atavistic signs, and the same edition concludes that Degas apparently chose against such physiognomic signs in the finished figure. The name Cesare Lombroso, so often attached to this object in passing, does not occur once in Huysmans's book.

The only continuous record of her life is her employer's bookkeeping

The only continuous record of Marie van Goethem's life is not an artist's studio, it is the monthly bookkeeping of the Paris Opera, held at the Archives nationales in the AJ 13 series. Martine Kahane went through it, and what she brought out is a wage record. Marie and her younger sister entered the dance school in May and November 1878 respectively. As students they earned bonuses for appearing in some shows, Marie between 20 and 30 francs and her sister up to 20. In 1880 Marie was recruited into the ballet at the bottom of the scale, second set of quadrille, second section, second class, on an annual contract of 900 francs, which is 75 francs a month before bonuses.

Their mother appears in the same books, briefly, as an additional dresser: one evening performance in 1879 and six in 1880, paid one franc a performance. The contract of 1880 names her as Marie's guardian and lists her as Widow Van Goethem.

Set against that scale, what Degas paid for a sitting is not trivial and not generous either. Kahane takes her comparison from the notes Emile Zola gathered in 1885 for L'Oeuvre, where a model earns five to six francs for a four-hour sitting, and she concludes that Degas was paying a fair price. Five francs is five evenings of her mother's dressing work. It is also about two days of Marie's salaried month. The costs ran the other way too. Kahane lists what the examinations took out of the same money: tarlatan skirts at 15 francs, silk ribbons at 6 and artificial flowers at 6, with customary New Year gifts to the Opera's supporting staff reaching 47.80 francs for the quadrille sets, against a monthly salary that ran between 75 and 87 francs for quadrille dancers generally. Those last figures come from a newspaper of December 1881 rather than from the payroll.

One caution about the last line. Zola's figure was written down in 1885 and the sittings for this sculpture fall between 1878 and 1881, so it is a near comparison and not the same year. Zola himself notes in the same breath that the prices had gone up and that by then a model was paid 10 francs.

Her name survives twice in Degas's own handwriting

Two objects carry her name in the hand of the man who made the sculpture, and one of them is the reason anybody knows who she was. The first is a sheet in the Louvre, Quatre etudes d'une danseuse, RF 4646 in the Departement des Arts graphiques, charcoal with white heightening on pink paper, 0.49 by 0.32 m, dated about 1878 to 1879. At the top left, in pencil, it reads 36 rue de Douai Marie. The Louvre's own entry states plainly that the inscription in Degas's hand is what allowed the model to be identified.

The second is an address list. Kahane records that she was given permission to consult the two Degas notebooks then kept in the storage room of the Departement des Estampes at the Bibliotheque nationale that mention the Van Goethem girls, and her notes place the relevant one at the shelfmark Dc 327d. That is the whole of it. A street number on a drawing and a line in a working notebook, against an employer's ledger. Nothing in her own hand is known.

The street number is worth a line, because Kahane's essay prints two of them and then settles the matter itself. Her summary list of the family's successive addresses gives 35 rue de Douai for 1880. Everything else says 36: the 1880 contract, which she says allows the residence to be verified, the notebook entry and the inscription on the drawing. She writes that the contract agrees with the notebook. Three independent records against one line in a summary list, so 36 it is, and the Paris cadastre for that address describes a seven-storey house of cut stone with a single very dark staircase, an innkeeper and a hairdresser at street level, and among the tenants laundresses, a beer retailer and a Dame Van Gotten.

The dismissal is a document and the reason for it is an inference

Her dismissal from the Opera in 1882 exists as paper, and the cause of it does not. Kahane quotes the form of the record: in April, in Pluque's handwriting and countersigned by Vaucorbeil, a note states that Marie has been dismissed from the Opera. In June her name was crossed off the list. Effective in August she was no longer on the roster for a quadrille group. The shelfmark is AJ 13 905A at the Archives nationales.

What the note does not carry, on the evidence of the scholar who read it, is a stated cause. Immediately after quoting it Kahane turns to the employment rule for the ballet corps, which indicates penalties for mistakes, tardiness and absences during rehearsals and shows. That is the rule, not a finding about this dancer, and the two are easy to run together.

The same payroll that records what she earned also argues against the reason usually given. Kahane's own sentence is that the payroll records for either sister do not show any disciplinary measures for days missed or illnesses, and elsewhere she writes that at the Opera Marie and Charlotte were good employees who did not call attention to themselves. That converts the absenteeism story from a gap in the evidence into a claim the evidence contradicts.

The other candidate cause is a gossip paragraph. In L'Evenement of 10 February 1882, quoted by Kahane, a notice reads that Miss Van Goeuthen, fifteen years old, has an older sister who is supernumerary at the Opera and a younger sister in the Opera dance school, and consequently she frequents the Martyrs Tavern and the Rat Mort. The load-bearing word is consequently. The paragraph was written for the Opera's presentation of the ballet Namouna, in which each dancer got a few lines, and Kahane notes that Marie is the only one in it described as an artist's model. Kahane also records, a few sentences later, that Degas himself was a customer at both places.

Our own record for this object said she was dismissed for missing too many classes. That stated a cause the document does not give, and the line has been corrected in the same push as this article.

The Louvre still prints an 1864 birth date, and 1864 belongs to a sister who lived eighteen days

The Louvre's entry for the drawing that identified her gives her birth as 17 February 1864, and 1864 is the year of a different child. Kahane's reconstruction of the family from the civil registry is unambiguous: of five children born to the Van Goethem couple, three girls survived. A first Marie was born in Paris in 1864 and died eighteen days after her birth. The Marie of this story is her sister, Marie Genevieve, born on 7 June 1865. Kahane gives no month or day for the first child, so the year is what the two records have in common and the precise date on the Louvre's page matches nothing published. The page carries its own last-updated stamp of 24 May 2023, with a note that its contents do not necessarily take account of the latest data.

The consequence is not only genealogical. A birth in February 1864 makes the sitter more than a year older at every point in the story, which moves the reading of a sculpture whose title is an age. Our own record carries 7 June 1865 and is right, which is worth stating plainly because this archive is usually the thing being corrected.

The rest of the family, from the same source: a brother, Jean-Baptiste, born in Brussels in 1859 and dead in Paris in 1862; Antoinette, the eldest girl, born in Brussels in 1861; and the youngest, Louise Josephine, who quickly took the name Charlotte, born in Paris on 18 July 1870 and dead there on 17 January 1945. Charlotte stayed at the Opera and became a teacher there. Antoinette was an extra without a contract, paid by the performance.

What is printed about her that no document supports

The most damaging claim in circulation about Marie van Goethem rests, in the source it comes from, on a novel. Kahane's essay reaches a section headed Prostitution, and its opening move is a comparison: in 1882 the life of these young girls begins to resemble the novel about the Cardinals, and just like Mrs Cardinal, Mrs Van Goethem prostitutes her daughters. Mrs Cardinal is a character. Kahane has introduced the Cardinal novels in an earlier section as the work of Ludovic Halevy, a well-to-do theatre lover and backstage regular who took Degas along with him, and she says of those books that poverty, exploitation and prostitution are not mentioned in such an elegant novel.

What follows the comparison, as evidence, is the L'Evenement paragraph with its consequently. No document about Marie is cited in support of it. The claim has since hardened into structured data. Wikidata's entry for her, Q6763113, lists three occupations: ballet dancer, model and prostitute. All three are recorded the same way, as fields with no qualifier and no reference of any kind. The same entry gives her death as 7 December 1935, a date our own record declines to treat as established, and the only thing behind it there is a personal website. The girl these fields describe entered the Opera school at twelve and was fifteen in the year the claim is attached to.

Kahane is also the source of the reason no better evidence exists, and she gives it as a matter of record about the archives rather than about the family. She began at the civil registry, continued at the Opera museum library and spent her Saturdays in the reading room of the Archives nationales. Her account of the limit she hit is worth quoting as she put it: legal records for that particular period were never stored in the National Archives and instead immediately destroyed at the courthouse, and police records had long vanished. What the newspapers reported of Antoinette, that she was sentenced in 1882 to three months for stealing 700 francs from a patron at the Chat Noir and held at Saint-Lazare, therefore cannot be checked against a judgment. Neither can anything else of this kind, in either direction.

Everyone names her flatly, and the one hedge is the clause her citers drop

Every institution that holds one of these objects names Marie van Goethem as the model without qualification. The Art Institute of Chicago's description of its 1962.703 sheet calls her the model Marie van Goethem, seen in three views. The National Gallery of Art, which owns the sculpture, states in its own object text that Marie van Goethem, the model for the figure, was the daughter of a Belgian tailor and a laundress, then repeats the identification in its own published article on the wax. Our record does the same. So does Martine Kahane, who established nearly everything documented about her, in the introduction to her own essay: thanks to Degas's notes and his surviving studies, she writes, we know that the model for the little dancer was Marie Van Goethem, born in 1865.

The hedge exists, and it is one clause, later in the same essay. Introducing the family, Kahane calls Marie Genevieve the model, or one of the models, for the Little Dancer. Her reason is visible in her own closing pages, where she takes seriously the younger sister Charlotte's claim to have modelled for Degas too. That is a careful scholar leaving a door open in the middle of a piece of work whose whole purpose was to shut doors, and it is the clause that gets dropped when she is cited.

Set this beside the archive's other famous young woman. For the Vermeer in our records there is no name, no payroll, no wage and no inscription naming her, and the record's own point is sharper than an unsolved identification: it is a tronie rather than a portrait, a study of a head in exotic costume made for the open market, so the category itself means there may never have been a sitter to name. For Marie van Goethem there is a contract, a monthly figure, a dismissal note, a street number in the maker's pencil and a birth certificate, and the best scholarship on her still says one of the models. The difference between the two is not really a difference of knowledge about a person. It is a difference in how much of an employer's paperwork happened to survive.