How a Museum Actually Acquires a Work

An acquisition begins as an argument, not an offer

A museum does not buy a picture the way a collector buys one. The transaction that ends in an inventory number begins, in almost every accredited institution, with a written case made by a curator to a committee, and that case has to survive four separate tests before money moves: is the work right for this collection, is the title clean, will the governing body vote for it, and can the sum be assembled. Most proposals fail the second or the fourth test, and they fail quietly, which is why the acquisitions the public hears about are a heavily filtered sample.

The Louvre's purchase of Cimabue's La Derision du Christ is unusually legible at every stage, because the museum publishes its own catalogue record with the procedural dates inside it. The record for inventory number RFML.PE.2023.33.1 shows the acquisition commission approving the purchase on 12 April 2023 and the work entering the inventory on 25 October 2023. The public announcement, made by the culture minister Rima Abdul Malak and the Louvre's president-director Laurence des Cars, followed on 3 November. The object is small: tempera and gold ground on poplar, 25.8 by 20.3 centimetres, catalogued around 1280, entered in the Department of Paintings.

What the record does not show, because no museum publishes it, is the internal document that started the process. The curatorial case for a work like this has to argue three things at once: that the panel fills a demonstrable gap, that the department can conserve and display it, and that no comparable opportunity is likely to recur. For a thirteenth-century Florentine panel of this rarity, the third argument writes itself. For most proposals it does not, and that is the argument committees reject.

The provenance check is the step that kills proposals

Between the curatorial case and the vote sits the part of the process that has changed most in thirty years. The American Alliance of Museums' Code of Ethics for Museums, adopted in 1993 and amended in 2000, requires that acquisition and disposal both serve the museum's mission and that title be established. The Association of Art Museum Directors sets the same expectation in its Professional Practices in Art Museums, revised on 19 December 2022. In the United Kingdom, museums seeking approval to borrow under the immunity-from-seizure regime must show that their due diligence follows the 2005 guidance issued by the Department for Culture, Media and Sport, and the same standard is applied to what they buy.

Two dates do most of the work in a modern provenance check. The first is 17 November 1970, the date of the UNESCO Convention, which functions as the practical threshold for archaeological material and ancient art: a museum is expected to establish that the object was outside its country of probable modern discovery before that date, or lawfully exported after it. The second window is 1933 to 1945. British loan regulations require a borrowing institution to state, in public and in advance, whether it holds a complete ownership history for those twelve years. No equivalent statutory requirement governs purchase, but the disclosure standard set for loans has become the working standard for acquisition committees on both sides of the Atlantic.

The Cimabue is instructive here because its provenance is short and unglamorous. The Louvre's record places it in a private collection at Compiegne from at least 1871 until 2019. A single continuous domestic ownership across the whole of the twentieth century is the cleanest possible answer to a 1933 to 1945 question, and it is one reason the acquisition proceeded without a restitution dispute attached to it.

The trustees sign, and in France the state signs

Curators do not acquire. Governing bodies do. In the American model, an accessions or acquisitions committee weighs the case and the board of trustees gives final approval, and the museum owns the object. In France the structure is different in a way that matters for anyone reading a credit line. The Cimabue was approved by the Louvre's acquisition commission, but the record states that the work is the property of the French state, held by the Musee du Louvre. The museum is the custodian of a national collection, not its owner.

That distinction is not decorative. It determines what can later be sold, lent or transferred, and it explains why the deaccession debates that periodically convulse American museums have no direct French equivalent. It also determines who negotiates. When France refuses an export certificate and declares a work a national treasure, the purchase offer that follows is made by the administrative authority, not by the museum, and it may be made on behalf of any public body.

The statutory machinery is set out in the Code du patrimoine. Refusal of an export certificate under article L111-4 opens a window of thirty months during which the state may offer to buy, at a price that must take account of prices on the international market. If the owner rejects that offer within three months, an expert procedure begins: each side appoints an expert, the two produce a joint report within three months, and if they disagree a third expert is chosen jointly or appointed by a court. The authority then has two months to make a final offer. The Cimabue was designated a national treasure on 20 November 2019 and acquired in 2023, which is the thirty-month clock plus the time the negotiation took.

The money is assembled, and the assembly is public

The least understood part of a museum acquisition is that the institution rarely has the money. It has a fund, usually restricted, usually small relative to the target, and it builds the rest from named sources whose names then appear on the label. On the Cimabue the Louvre used income from the brand licence for Louvre Abu Dhabi as its own contribution. The Societe des Amis du Louvre, the museum's membership body, supplied the final million euros. Harry and Linda Fath, American property developers, contributed through the American Friends of the Louvre.

France adds a fiscal instrument that has no British or American parallel. Under article 238 bis-0 A of the general tax code, a company that contributes to the state's purchase of a national treasure receives a tax reduction of ninety per cent of the sum given. Under article 238 bis-0 AB, a company that buys a national treasure for itself receives forty per cent, provided the state has not already made an offer. That ninety per cent rate is the reason French corporate patronage clusters around export-barred works rather than around general acquisition funds.

The British route runs through grant-making bodies instead. The Art Fund, a membership charity, makes acquisition grants to museums across the United Kingdom; the National Heritage Memorial Fund acts as the fund of last resort for objects at risk of export; and museums with American donor bases raise through bodies such as the American Friends of the Louvre or their institutional equivalents. When the National Portrait Gallery bought Van Dyck's last self-portrait in 2014, the total was ten million pounds and the Art Fund's own record puts its grant at 500,000 pounds, the remainder coming from a public appeal and private trusts. A grant of five per cent bought the campaign its credibility, which is the function such grants usually perform.

The price a museum pays is not the price in the headline

The Cimabue was sold at Senlis by Acteon on 27 October 2019. The Louvre's catalogue record gives the sale price as EUR 19.5m to the Alana Collection. The figure that travelled around the world, and that still appears in most accounts, is EUR 24.18m. Both are correct and they are not the same number: 19.5 million is the hammer, and 24.18 million is the premium-inclusive total. Almost every report of this acquisition, including reputable ones, uses the second figure without saying which it is.

The distinction matters more than usual here, because the reported purchase price paid by the Louvre in 2023, EUR 24.1m, is indistinguishable from the 2019 premium-inclusive total. The Louvre has not published a line-item breakdown of what it paid against what the underbidder's total had been, so the coincidence should be read as a coincidence of reporting rather than as evidence of how the negotiated price was fixed. A researcher citing this acquisition should cite the Louvre's own record for the hammer and the trade press for the reported purchase total, and should say which is which.

The same habit applies to every acquisition figure a museum announces. A campaign target is a gross sum including fees, tax and sometimes conservation; a purchase price is net of them; a valuation used to set an export bar is neither. Reading an acquisition properly means asking which of the three you have been given.