# Where the Silence Falls: The Gap Is Almost Always at the Beginning > The profession defines provenance as a chain running from creation to the present. Across 198 researched records, 119 carry a gap, and in 76 of them the silence is the first thing in the object's recorded life. > The Study - 2026-08-24 > Canonical HTML: https://iconotheca.com/study/where-the-silence-falls-in-a-provenance/ ## Key points - Of 198 records published here, 119 carry at least one span of ownership history recorded explicitly as a gap, across 128 separate gap events. - In 76 of those 119 records the earliest silence is the first thing in the object's recorded life: 70 open immediately after the object is made, and 6 open the chain outright. - Where both ends of that first gap carry a readable date, the median runs to just over 123 years and the mean to 284; 25 of 66 run two centuries or more. - The silence tracks the archive, not the object. All 8 Islamic metalwork records carry a gap, with a median first silence of 596 years; musical instruments, documented by makers' ledgers and players' names, run to a median of 11. - The AAMD does not ask its members to reconstruct a chain back to origin for ancient art. It asks for a single date, 17 November 1970, which is an admission that the origin end is usually unrecoverable. ## What the standard actually asks for ## What 198 researched records actually contain ## The silence is not in the middle of the story ## How long the first silence runs ## Which classes of object go dark, and for how long ## The shape of the silence is the shape of the archive ## What the market uses instead of a chain ## How to read a record whose chain begins in 1828 The received picture of provenance research is a detective closing a hole in the middle of a known story. The data in this archive says the work is something else: 76 of the 119 gapped records begin in silence, and the researcher's real task is to reach backwards from the earliest surviving document towards a making that no document witnessed. That is why the length of a gap is nearly useless as a warning sign and its position is nearly everything. A four-century silence that opens at the workshop and closes on an inventory is the normal condition of an object older than the archive. An eight-year silence that opens in 1943 is not. Anyone doing diligence who screens on the size of the hole rather than on its edges has sorted the corpus by the age of the object and learned nothing about its title. ## Questions **Does a gap in a provenance mean something is wrong with the object?** Usually not. Sixty per cent of the records in this archive carry at least one gap, and in most cases the silence is the ordinary result of the object predating the inventories, sale catalogues and stock books that provenance research relies on. What matters is where the gap sits and what closes it, not that one exists. **Why do Islamic metalwork and ancient glass have such long silences?** Because the surviving documentary apparatus is European and begins in the sixteenth century at the earliest. The Getty Provenance Index covers European sales catalogues from 1650, European private inventories from 1520 and public collections from 1500. An object made in Mosul in 1232 enters the record when it enters a European collection, which is why the median first gap for Islamic metalwork in this corpus is 596 years. **What is the difference between a gap and an unknown owner?** In this archive a gap is a span of years for which no owner is recoverable, recorded as its own dated event in the chain. An unknown is a single link whose holder is not established. The corpus carries 128 gap events and 196 links graded unknown, and both are published rather than bridged, because a recorded gap is worth more than an invented chain. **If a chain cannot reach back to the object's making, what do institutions require instead?** For archaeological material and ancient art the AAMD substitutes a date for a chain. Its guidelines ask that provenance research substantiate the work was outside its probable country of modern discovery before 17 November 1970, or legally exported after it, and they state that many works will lack a complete documented ownership history even after extensive research. ## Sources - [1] International Council of Museums, 'ICOM Code of Ethics for Museums', as adopted by the 41st Ordinary General Assembly, Paris, 25 June 2026. Glossary definition of provenance and the collections clause on provenance research. Read in full from the ICOM PDF. - [2] Association of Art Museum Directors, 'Guidelines on the Acquisition of Archaeological Material and Ancient Art (revised 2013)', final text adopted by AAMD membership 29 January 2013. Sections III.E and III.F, and the introduction recording that the 2008 revision deleted the ten-year provenance provision and introduced the 17 November 1970 threshold. - [3] Association of Art Museum Directors, press release, 'Strengthened Guidelines on the Acquisition of Archaeological Material and Ancient Art Issued by Association of Art Museum Directors'. Confirms the 1970 threshold wording in the AAMD's own summary. - [4] Getty Research Institute, 'Getty Provenance Index', coverage statements for Sales Catalogs (Europe, 1650 to 1945), Archival Inventories (European private collections, 1520 to 1880), Dealer Stock Books (Goupil, Paris, from 1846; Knoedler, New York, from 1872), Payments to Artists (Rome, 1576 to 1711) and Public Collections (Great Britain and the United States, 1500 to 1990). - [5] Iconotheca, records.json, the published corpus of 198 researched object records with graded provenance chains. All counts, positions and gap spans in this article are computed from the published provenance events of that corpus.