# The Longest Link in a Provenance Is the One Nobody Wrote Down > Measured across 223 records, an inheritance link spans a median of 27 years against 14 for a purchase and 7 for a purchase at auction. Descent generates no document unless something outside the family compels an inventory. > The Study - 2026-09-02 > Canonical HTML: https://iconotheca.com/study/the-longest-link-is-the-one-nobody-wrote-down/ ## Key points - Measured across the 223 records published here, an inheritance link spans a median of 27 years against 14 for a purchase and 7 for a purchase at auction. Descent is the longest documented link a provenance carries, and the market's links are the shortest. - The reason is structural rather than archival. A sale needs a document because two parties with opposed interests require one, and a death inside a family produces nothing unless something outside the family compels a list. - The Rothschild Prayerbook is documented at four moments and every one of them is a compelled inventory: a collection catalogue in 1872, a probate valuation in February 1906, two later palace inventories, and a confiscation in 1938. - Christie's own lot record uses the word presumably for a father-to-son inheritance in one of the best-recorded private collections in Europe, which is the correct grade for a transfer nobody minuted. - The numbers assigned by those compelled inventories are physically on the object: 927 on a label at the corner of the lower cover and AR3390 in pencil on the front flyleaf. - The Codex Leicester spent roughly 112 years in one family with nothing recorded, and more than two and a half centuries in another, at Holkham, where it acquired both its name and its scholarly reputation. ## A sale writes itself down and a death does not ## What the corpus measures ## The Rothschild Prayerbook is documented four times, and always by compulsion ## Christie's says presumably, and it is right to ## The inventory numbers are on the object ## A century in one family, and two and a half in another ## How to read an inheritance link ## Why this archive is free to read The market is loud and the family is quiet, and a provenance line records the noise rather than the object's life. Every figure in this corpus points the same way: the shortest, best-evidenced links are the ones where money changed hands between strangers, and the longest are the ones where it did not. So when a chain runs smoothly through three generations without a document behind it, that smoothness is the warning. Find the outsider who had a reason to make a list, name them and date them, and where no such person existed, grade the link as what it is and let the silence stand on the page where a reader can see it. ## Questions **Why are inherited objects harder to trace than bought ones?** Because a sale has two parties with opposed interests who both need a document, and a death inside a family has neither. Across 223 records in this archive an inheritance link spans a median of 27 years against 14 for a purchase and 7 for a purchase at auction. **What kind of document actually records an inheritance?** Almost always one compelled by someone outside the family: a probate or estate inventory, a collection catalogue, an insurance schedule, a confiscation inventory or a restitution file. The Rothschild Prayerbook is fixed in place by a catalogue of 1872, an estate inventory of February 1906, two palace inventories and the 1938 appropriation. **How should a provenance line reading by descent be graded?** As disputed or unknown unless a document names the object and the recipient. Christie's own lot record for the Rothschild Prayerbook says Alphonse von Rothschild inherited the palace and, presumably, the manuscript with it, which is the correct grade for a transfer nobody minuted. **Can an inventory number on an object be evidence?** Yes, and it frequently outlives the ledger it came from. The Rothschild Prayerbook carries 927 on a label at the upper corner of the lower cover and AR3390 in pencil on the front flyleaf, both from palace inventories. **Does a long family tenure always mean a gap in the record?** No. It depends on whether the object was public. The Codex Leicester passed roughly 112 years with the della Porta heirs unrecorded, then more than two and a half centuries with the Coke family at Holkham while being copied, cited and studied throughout. ## Sources - [1] Christie's, lot record and catalogue note for lot 157, 'THE ROTHSCHILD PRAYERBOOK, a Book of Hours, use of Rome, in Latin, illuminated manuscript on vellum', London, 2014. Read 2 September 2026 for the provenance: Schestag's 1872 catalogue of Anselm von Rothschild's collection (no 597); the February 1906 inventory of Nathaniel von Rothschild's estate (no 452, in the Galerie of the palace at Theresianumgasse, valued at 150,000 Kronen against 80,000 for the Brandenburg prayerbook, no 453); the statement that Alphonse von Rothschild 'inherited Nathaniel's palace and, presumably, the manuscript along with it'; the inventory numbers 927 on a label at the upper corner of the lower cover and AR3390 in pencil on the front flyleaf; the 1938 appropriation; and restitution to the family in 1999. - [2] Museo Galileo, 'The European circulation of the Codex Leicester', in the online exhibition 'Water as Microscope of Nature. Leonardo da Vinci's Codex Leicester'. Read 2 September 2026 for della Porta's ownership in 1537, Ghezzi's by 1689, the acquisition by the future Earl of Leicester and the copy made at Florence about 1717, the manuscript's keeping in the Earl's London house from 1719 to 1759, Lalande's 1769 reference to the earthshine studies, and the copy acquired by the Grand-Ducal Library at Weimar on Goethe's advice about 1818. - [3] Museo Galileo, 'Water as Microscope of Nature. Leonardo da Vinci's Codex Leicester', introduction to the online exhibition. Read 2 September 2026 as the framing account of the manuscript and its study.