The House That Cannot Be Asked: Who Keeps the Price After the Sale

A price outlives its sale only if somebody keeps it

A hammer price is not a property of an object. It is a fact about an afternoon, written down by whoever ran the room, and it survives exactly as long as that party's records do. This archive has written elsewhere about how old the last price usually is and about prices denominated in money that no longer exists. The question here is narrower and more practical: is there anybody left to ask.

Across 293 published records there are 217 auction rows, and 136 of them name a seller. The remaining 81 are mostly entries recording that an object has never been to auction at all, which is a different kind of silence. Those 136 sellers carry 56 distinct names, and the shape of that distribution decides how much of this archive's price record is still checkable at source.

Two firms hold just under half the record

Sotheby's appears 37 times and Christie's 28, which is 65 of 136 rows between them under their own names. Add the variant spellings and the two entries recording a private sale arranged by Sotheby's and the pair account for 69, a little over half. Both firms still trade, both keep their historical sale archives and both answer enquiries about them, so for half this archive's prices there is an address to write to.

The concentration is not surprising and it is not the finding. The finding is the tail, because that is where the prices sit that nobody can confirm.

Eight guineas, then five, for a Caravaggio

The Taking of Christ passed through Dowell's of Edinburgh twice, on 25 June 1921 at eight guineas and on 3 May 1922 at five, catalogued both times as a Honthorst. Eight guineas is eight pounds and eight shillings, five guineas is five pounds and five shillings, and the second sale was eleven months after the first at a lower figure. The picture was recognised as a Caravaggio in Dublin in the early 1990s.

Those two numbers are among the most consequential in the archive, because they are what a masterpiece fetched while wearing the wrong name. They are also the least verifiable, because Dowell's is gone. Every later account of them traces back to the same research done on the same surviving catalogues, and no living institution holds the firm's books in a form that would let a reader confirm the buyer, the lot number or whether the two lots were the same picture without that research being redone.

A lot number, and no price, from a house that closed in 1933

Galerie Georges Petit was among the most powerful firms in the French market: founded as Galerie Francois Petit in 1846, taken over by Georges Petit in 1877, host of the Expositions internationales from 1882, the venue for the Degas estate sale of 1918 to 1919 and for Picasso's first retrospective in June 1932. It closed in 1933 and its assets were sold.

Two prices in this archive were set in that room. The Burgundian cabinet now at the Getty appears in the posthumous Seilliere sale of 5 to 10 May 1890, at pages 80 to 81, lot 540, illustrated. The catalogue survives and gives the lot and the illustration. It does not give the hammer, and there is no longer a firm whose ledgers could. The Metropolitan's mosque lamp has the same shape of record from the Mannheim estate sales of 1910: the museum's own provenance cites the house and the lot and gives no price, no premium and no buyer of record.

One hundred and sixty guilders, Amsterdam, October 1940

Kandinsky's Bild mit Hausern was sold at Frederik Muller and Company in Amsterdam on 9 October 1940 for 160 guilders, five months into the German occupation, and bought by the director of the Stedelijk Museum for the City of Amsterdam. The collector who consigned it had paid 500 guilders for the same painting in 1923. The gap between those two figures is the substance of the case that followed.

Frederik Muller and Company no longer trades under that name. In a restitution argument the identity of the recording house matters more than anywhere else, because the question is not only what was paid but under what conditions, and the conditions are recorded, if at all, in a firm's internal papers rather than in its printed catalogue. Where those papers went when the firm went is itself a research question, and it is one that has to be answered before the price can be interpreted rather than merely quoted.

Eleven entries in this column were never auction houses

The seller column also holds things that are not firms. John Harrison's marine timekeepers carry three entries: an award voted by the Commissioners of Longitude on 30 June 1737 after the Lisbon trial, and two Acts of Parliament, in 1765 and 1773. No sale took place in any of them, and the 1765 Act required Harrison to surrender the timekeepers, which is the opposite of a purchase. His lifetime receipts came to 23,065 pounds across about forty years, and the 20,000 pound prize of the 1714 Act was never awarded to anyone in the form the Act described.

The Dunhuang Diamond Sutra carries a payment rather than a price: four horse-shoe-shaped silver ingots handed in May 1907 to Wang Yuanlu, the Daoist priest who had appointed himself custodian of the sealed library cave, for a large body of material of which the scroll was one item. Canova's Napoleon was a direct purchase by the British government in 1816. The Royal Gold Cup went to the British Museum in 1892 by private treaty through a Bond Street dealer for 8,000 pounds, met by a Treasury grant and named subscribers.

Each of those is correctly recorded and none of them is an auction result. They are in the column because the column is where money goes, which is a field design problem rather than a market one, and the label on each entry says so in words.

What a surviving house is actually worth to a reader

The contrast is easiest to see at the top of the market. Rubens's Massacre of the Innocents sold at Sotheby's in London on 10 July 2002, and the trade press reported the hammer on the day at 45 million pounds while the figure that travelled afterwards was the premium-inclusive total. Both numbers are recoverable a generation later because the firm is still there, the sale is still in its archive and the reporting that covered it is still indexed.

The Faberge egg found in the American midwest makes the same point from the other end. It was lot 259 at Parke-Bernet on 7 March 1964, in the sale of the property of Mrs Rena Clark, catalogued as a gold watch in an egg-form case and sold for 2,450 dollars. That lot is traceable because Sotheby's bought Parke-Bernet in the same year and the records went with the purchase. A firm that is absorbed keeps its past; a firm that is wound up usually does not.

How to read a price whose house is gone

Ask first who wrote the number down, then whether that party still exists, and only then what the number was. A figure from a surviving house is a claim you can test. A figure from a wound-up firm is a claim about a document, and the document is usually a printed catalogue that gives a lot and a description and frequently no price at all.

Where the house is gone, find out where its papers went: to a successor firm, to a national library, to a dealer's archive, or nowhere. Treat a price with no surviving custodian as provisional however often it is repeated, because repetition in that situation is not corroboration, it is one piece of research being cited by everybody. And when recording a sale, write down the house as carefully as the figure. The name of the room is the part that decides whether anyone can ever go back.

Why this archive is free to read

We do this for everyone who loves art. The people who have studied it for years, and the people who just stopped to look. Iconotheca exists so that what we learn belongs to all of them.

Nobody pays us for this. No ads, no sponsors, no commercial interest. These works belong to all of us, and we would rather share them with as many people as we can than keep them to ourselves.

If it gave you something today, tell us to keep going. Follow us, leave a like, or write a positive comment. We read every one, and they are what keeps us going.