MetaMuseumVersioned research report · v3
A recovered Braque can be followed across official records—but the expected Nazi inventory entry is absent. Two official object images now strengthen identification, while photograph 3492 and final handover proof remain unlocated.
The short version
Several records report that a Braque painting was returned to Paul Rosenberg in 1947, even though it is absent from the Nazi inventory researchers might expect to contain it. Primary letters document recovery and intended restitution—but stop short of proving delivery.
| Why it matters | Historical databases are incomplete and often dependent on one another. This case shows why researchers must connect catalogues, claim files, museum records, and archival photographs before treating a missing database entry as a missing historical event. |
|---|---|
| What changed in our understanding | Primary correspondence connects the generic title Nature Morte to Mandoline, fruits et bouteille, confirms the 97 × 130 cm dimensions, and documents the Brussels recovery process. A complete review of one promising returned-paintings folder ruled out its strongest Braque candidate, while an updated official record now supplies two images of the registered object. |
| The question that remains | Can photograph 3492 or a signed September 1947 receipt complete the chain from the recovered painting to its physical return? |
The historical trail
A diplomatic letter reports a Braque still life called Nature Morte, measuring 97 × 130 cm, found with a Brussels picture dealer and claimed by Paul Rosenberg.
Evidence state: documented. The letter describes a recovery effort, not a completed restitution.
Inspect this sourceA second letter calls the recovered work Mandoline, fruits et bouteille and says it is to be restituted to Rosenberg after formalities.
Evidence state: documented. Intended restitution is not evidence of physical delivery.
Inspect this sourceThe official Répertoire des biens spoliés lists entry *5034: a Braque with mandoline, fruit and bottle, the same 97 × 130 cm dimensions, and Paul Rosenberg as owner.
Evidence state: documented. The catalogue supports identity and named ownership, not the route through Belgium.
Inspect this sourceERR and Rose Valland database records report that the painting was restituted on this date—even though it was absent from the pertinent ERR inventories.
Evidence state: reported later. The February and May letters do not prove this final handover; a signed receipt or photograph 3492 could close the gap.
Inspect this sourceMoMA's V.A.8 returned-paintings folder was reviewed page by page. Its strongest apparent Braque sequence is identified by an accompanying notarized declaration as L'intérieur au vase noir—not the Rosenberg work researchers are seeking.
Evidence state: documented. The inspected folder did not reveal photograph 3492. That does not prove the photograph is absent from the wider Paul Rosenberg Archives.
Inspect this sourceThe updated official JDP-0056 record embeds two black-and-white image files of the registered still life and binds them to the 1936 date, title family, and 97 × 130 cm dimensions.
Evidence state: documented. The record does not identify either image as photograph 3492, and the images do not prove the September 1947 physical handover or establish image-republication rights.
Inspect this sourceYour response
Save it for later, share it with someone who can check the evidence, or tell us whether you learned something new.
Audience signals are recorded only if you grant analytics consent. Saving works locally in this browser either way.
awaiting independent human review. This report remains an invitation to reproduce, correct, or reject the proposed answer.
Research question
Across Getty, ERR, Rose Valland, Lost Art, the official RBS, primary correspondence in Archives diplomatiques dossier 45.15, and a bounded review of MoMA folder V.A.8, the system can distinguish actors, locate specific displaced works, preserve aliases, and expose two useful negative facts: the Braque is reported as absent from the relevant ERR inventories despite recovery in Brussels, and the inspected V.A.8 surrogate does not supply the referenced photograph 3492. RBS *5034 corroborates the description, 97 x 130 cm dimensions, OBIP number, and Paul Rosenberg as named owner. The 1947 letters connect the generic title Nature Morte to Mandoline, fruits et bouteille, document the Brussels recovery process, and describe restitution to Rosenberg as intended after formalities. The updated official JDP-0056 record now binds two image files to that registered title-and-dimension identity, providing relevant image-based registration evidence without establishing that either file is photograph 3492. The letters and images do not prove the final September handover, complete seizure history, present ownership, legal title, or scholarly novelty. The combined sources support a stronger reproducible research answer, not a legal conclusion.
Public-benefit research
If you work with provenance records, archives, Braque catalogues, or the Paul Rosenberg papers, try to reproduce or disprove the answer. Corrections are part of the record.