The Astronomer

A small canvas of 1668 showing a man in a gown turning a celestial globe, painted in Delft, sold and resold through Amsterdam salerooms for a century under four different titles, bought for a Paris collector in 1866 by the critic who invented Vermeer's modern reputation, and acquired by Baron Alphonse de Rothschild before 1888. In 1940 the German authorities seized it from the Rothschild family in Paris for the museum Hitler planned at Linz. A small swastika was stamped on the back. The Allies recovered it in 1945 and it was restituted to Baron Edouard de Rothschild in 1946.

It entered the Louvre in 1983, and not by purchase. The French state took it from Guy de Rothschild in settlement of inheritance duties under the dation en paiement, which is why no price for it has ever been published. This is the record in the corpus where the machinery worked: a documented seizure, a documented recovery, a documented return to the family within a year of the war ending, and a transfer to the state four decades later on the family's own terms.

The one that came back

Most Nazi-era provenance records in this archive are about failure: gaps nobody can close, claims that ran for twenty years, panels that refused, objects that stayed where they were while money moved instead. This one is the control case.

The painting was seized from a named Jewish family in Paris in 1940 by a named German organisation for a named purpose, was found by the Allies in 1945, and was given back to its owner in 1946. There was no litigation, no advisory committee and no settlement. The chain runs unbroken through the war because the object was recovered before anyone had time to sell it, and because the family that owned it was still there to receive it.

What happened next is the part worth the reader's attention. Thirty-seven years later the same picture became French state property, and it did so through a tax mechanism rather than a sale. The Louvre did not buy the most famous Vermeer in France. It accepted it in payment of a death duty.

What it is

Oil on canvas, 51 by 45 centimetres, dated 1668. A man in a long gown leans forward at a table covered with a rucked carpet and turns a celestial globe with his fingertips. Light comes through a leaded window at the left.

The instruments are identifiable, and the Louvre identifies them. The globe is Jodocus Hondius's celestial globe of 1600. The book open on the table is Adriaan Metius's manual of astronomy of 1621, open at the chapter on the stars, at a page carrying an engraving of the wheel astrolabe Metius invented. There is a pair of compasses and a sixteenth-century astrolabe.

On the wall behind hangs another picture, a Moses saved from the waters. The Louvre gives it to Peter Lely, from his Dutch period before 1643, and raises the possibility that it was in Vermeer's own collection. Vermeer traded pictures as well as painting them; a canvas he owned appearing on the wall of a canvas he painted is the kind of thing that is provable in a probate inventory and almost nowhere else.

The frame is not original and is not modern either. It is a Dutch seventeenth-century frame, possibly a mirror frame, given specifically for this painting by the De Boer gallery of Amsterdam in 1983, the year the picture entered the Louvre.

What the man is doing, and what the picture on the wall is for

The subject is the scholar at work, a Dutch pictorial tradition running through Rembrandt and Gerrit Dou, and adjacent to the older image of Saint Jerome in his study and of hermits at prayer. A seventeenth-century viewer would have read the room as a version of the study, and the man as a version of the contemplative.

The Moses on the wall is the instruction for how to read him. The Louvre's own catalogue puts it as an allusion to divine Providence, set as a counterpoint to the astronomer's activity, and then declines to say whether that activity is meant as vain or as wise. That refusal is correct and it is worth copying. The picture holds a man measuring the heavens against a scene of a child saved by God's hand, and it does not tell you which way to take it.

There is a further question the identification of the instruments raises and does not answer. Metius's manual is open at the stars. The globe is celestial, not terrestrial. Astronomy and astrology were not cleanly separated in 1668, and the picture has been catalogued as L'Astrologue as often as L'Astronome, including by the Louvre, which lists both titles. What the man is doing depends on which word you use, and the painting supplies no evidence either way.

The sitter, and why the usual answer is not one

The figure is often said to be Antoni van Leeuwenhoek, the Delft draper who built microscopes, who was born in the same year as Vermeer and who was appointed to administer his estate. The Louvre states the identification and then states its status: sometimes identified, without proof, with the scientist Antoni van Leeuwenhoek, who in 1668 to 1669 was still too little known.

That second clause is the useful one. The argument for Leeuwenhoek runs on biography rather than likeness, and it needs him to have been a plausible subject for a portrait in 1668. He was not yet. His first communications to the Royal Society came in 1673.

The man in this picture also appears in The Geographer at Frankfurt. Whoever he was, Vermeer painted him twice.

The Geographer, and what the two pictures are to each other

The Geographer, in the Staedel Museum in Frankfurt since 1885 as inventory 1149, measures 51.6 by 45.4 centimetres against this picture's 51 by 45, and is signed and dated on the wall at upper right, I.Ver-Meer MDCLXVIIII, the date retouched but probably right. The two canvases show the same room with the furniture slightly moved and, on the usual reading, the same man.

They travelled together for most of their recorded history. The Louvre's provenance shows them sold as a pair or as adjacent lots at Rotterdam in 1713, at Amsterdam in 1720, 1729, 1778 and 1797, and separated only at the Gildemeester sale of 1800, where this one appears alone as A philosopher in solitude. They were not shown together again until 1997, when the Louvre and the Staedel reunited them, first in Paris in January and then in Frankfurt in May, a hundred and ninety-seven years after they were parted.

Technical work has since tied them harder together: analysis of the canvas established that both were cut from the same bolt of cloth.

The Louvre is careful about what that makes them, and this record follows it. They are not exactly a pendant pair. They are a very close variation on one iconographic theme, and the antithetical reading often imposed on them, earth against sky, action against thought, has probably been overdone, not least because it does not show up in the way the two pictures are actually composed.

A century of Amsterdam salerooms, and four titles

The recorded chain begins on 27 April 1713 at Rotterdam, where the picture was lot 10, catalogued as a learned mathematician, with its companion as lot 11. The sale may have been part of the cabinet of Adriaen Paets, the patron of Adriaen van der Werff, most of which had gone under the hammer the day before.

It was bought by H. Sorgh, an Amsterdam dealer and grandson of the painter of that name. His estate sale in 1720 offered it as lot 3 with the companion as lot 4, and from that point the two pictures become genuinely hard to tell apart in catalogues: at the Looten sale of 1729 they appear simply as two Astrologues, and the Louvre records that distinguishing them is impossible. At the Crammer Simonsz sale of 1778 the description of lot 18 matches this canvas.

Then a title migration, which is the mechanism by which pictures get lost. Savant mathematicien in 1713. Astrologist in 1720. Astrologue in 1729. Un philosophe in 1778 and 1797. Un philosophe en solitude in 1800. Astronome by 1881. At a London sale in 1804 a picture that may be this one is catalogued as an Astronome by Van der Neer, and the entry survives only in a second edition of the catalogue.

A researcher tracing this object through the eighteenth century is not tracing a title. There is no stable title to trace.

The picture Louis XVI did not buy

In 1784 the canvas was with J. E. Fizeaux in Amsterdam when it was engraved at the request of the Paris dealer Jean-Baptiste Pierre Lebrun, who did not publish the plate until 1792. The engraving omits the signature and the date.

Fizeaux's widow then put the picture, with its companion, into the hands of A. J. Paillet, a Paris dealer and Lebrun's rival, working in partnership with P. Fouquet the younger in Amsterdam. In 1785 Paillet offered it to Louis XVI. The purchase came to nothing and Paillet returned the pictures to Fouquet, who had advanced the money.

Paillet left a judgment on the pair which is the only period opinion of their relative quality on the record: the astronomer, he wrote, is as well painted, but the other does it much harm. Two hundred years later the state he tried to sell it to acquired it for nothing, in settlement of a tax.

Thore-Burger buys it, and Vermeer is invented

The picture passes through English hands for four decades in the nineteenth century: a Gibbons sale, probably at Bristol, in 1820, where it went unsold or was bought in, since Gibbons lent it to an exhibition at Birmingham in 1828 as an Astrologue; a second Gibbons estate sale in London in 1857, as a Philosophe; a collector named Tate, whom the Louvre thinks was more likely Thomas Moss Tate of Liverpool than Henry Tate of the Tate Gallery; and a Christie's sale in 1863 whose catalogue has not been found.

At the end of 1866 or in the first days of January 1867 it was bought for the Paris collector Leopold Double by Theophile Thore, who wrote as William Burger.

That is not an incidental name. Thore-Burger published the catalogue raisonne of 1866 that recovered Vermeer for the nineteenth century and made him an international figure, having initially given him more than seventy works. The man who assembled the modern idea of Vermeer bought this canvas in the same season that he published the book. There is no cleaner example anywhere of a critic making a market and then buying into it.

Double's estate sale of 30 May 1881 offered it as lot 17, as the Astronome. The Paris dealer Leon Gauchez bought it, and sold it to Baron Alphonse de Rothschild, in whose possession it is recorded by 1888.

1940: seized for Linz

Alphonse de Rothschild died in 1905 and the picture went to his son Edouard, born 1868. In 1940 it was seized by the German authorities in Paris. The Louvre's own record states the destination in as many words: for the Hitler museum at Linz.

That is the Fuehrermuseum, the collection Hitler intended for his home town, for which the Einsatzstab Reichsleiter Rosenberg stripped French Jewish collections through the Jeu de Paume from 1940. The Rothschild collections were among the first and largest taken. A small swastika was stamped in black ink on the back of this canvas.

The mark matters out of proportion to its size, and it is the reason this object belongs in a provenance archive rather than only in a Vermeer catalogue. A provenance gap is an absence and has to be argued about. A stamp is a presence. It is the seizure written on the object in the seizing authority's own hand, and it converts the 1940 to 1945 link from a claim into a physical fact.

1945 and 1946: recovered and returned

The Allies recovered the painting in 1945 and it was restituted to Baron Edouard de Rothschild in 1946.

Two things about that sentence are worth stating plainly, because the rest of this archive's Nazi-era records make them look remarkable. First, it took one year. Second, the object went back to the person it was taken from, not to his heirs, not to a compensation fund, and not to a settlement in which somebody kept it and paid for the privilege.

The reasons are unglamorous. The picture was taken late, kept intact for a museum that was never built rather than sold on the market, found by the Allied recovery effort with its records legible, and claimed by a family that had survived, had the means to pursue it and could prove ownership from a published nineteenth-century catalogue. Change any one of those and this record would read like the others in this archive.

Edouard de Rothschild died in 1949. The picture passed to his son Guy.

1983: a dation, not a purchase

The Louvre acquired the painting from Guy de Rothschild in 1982 to 1983 by dation en paiement of transfer duties. It carries the inventory number RF 1983 28 and it is owned by the State.

A dation is not a sale and should never be reported as one. Under the French mechanism, an heir liable to death duties may discharge the liability by handing over a work of art of agreed value instead of money. The state receives the object; the family receives a credit against a tax bill. No cash changes hands and no market price is established.

So there is no acquisition price for this painting and there never was one. The Louvre publishes none, and any figure quoted for it would be either the agreed valuation for the tax, which has not been published in the sources consulted, or an invention. This is the point the archive makes repeatedly about museum acquisitions and it is at its sharpest here: the most important Vermeer in France entered the national collection at a price of zero and a cost of whatever the Treasury forwent.

It is also the reason the object is where it is. A family holding a picture of this stature and facing a death duty has two options, and one of them is a sale abroad. The dation is the instrument that keeps such objects in the country, and it does so by making the tax bill the lever rather than an export ban.

What is happening to it now

It hangs in the Richelieu wing of the Louvre, level 2, room 837, in a room built around Jacob van Ruisdael.

It travels. It was in the Louvre's Vermeer and the Masters of Genre Painting in 2017, which went on to Dublin and to the National Gallery of Art in Washington. It was at the Louvre Abu Dhabi through 2021 to 2023. Most recently it was lent to Lire le ciel at the Mucem in Marseille, from 9 July 2025 to 5 January 2026, hung as an object about reading the sky rather than as a Dutch genre picture.

The Louvre's own record was last updated on 2 December 2025 and carries a warning on its face that its contents do not necessarily take account of the latest data. That warning is on every entry in that database, and it is a more honest statement than most museum labels manage.