IconothecaThe Provenance Record
MenuClose

Paintings Draw the Attribution Fights, but Glass Is Disputed Far More Often

Seven of 45 paintings carry a disputed attribution and seven of 16 glasses do. No print does. A table of 330 records by object type.

Records in this piece: Under the Wave off Kanagawa (Kanagawa oki nami ura), also known as The Great Wave, from the series Thirty-Six Views of Mount Fuji, Melencolia I, The Rhinoceros, Christ Preaching, Healing the Sick, and Blessing the Children ("The Hundred Guilder Print"), The Great Dish from the Mildenhall Treasure, The Water Newton Treasure, The Luck of Edenhall, The Corning Ewer, Hedwig Beaker, The Last Judgment Triptych

Hokusai, Under the Wave off Kanagawa, colour woodblock print, 25.4 by 37.6 cm, signed and published. Prints are the one object type in the archive where no record is disputed: all 18 are accepted.
Hokusai, Under the Wave off Kanagawa, colour woodblock print, 25.4 by 37.6 cm, signed and published. Prints are the one object type in the archive where no record is disputed: all 18 are accepted.Art Institute of Chicago, Clarence Buckingham Collection, 1925.3245. Public domain., CC0
On this page 4 sections
ContentsClose

In short

  • In the 330 published records, 63 carry a disputed attribution. Painting has the most disputed records by count, 7 of 45 (16 per cent), but glass has the highest share, 7 of 16 (44 per cent).
  • No print is disputed. All 18 are accepted as to maker, and the records grade the individual impression instead.
  • Silver is the type with the most records that name no maker: 9 of 17. All nine are ancient or early medieval silver, most of it from hoards and church treasures, and none carries a surviving maker's mark.
  • The disputes in glass are mostly about where and when, while the disputes in painting are about whose hand.

Seven of 45 paintings are disputed and seven of 16 glasses are

In the 330 published records, 63 carry a disputed attribution, and glass has the highest share of them, not painting. Painting has 7 disputed records of 45 (16 per cent) plus one rejected. Glass has 7 of 16 (44 per cent), Islamic metalwork 6 of 16, and textiles and arms and armour 5 of 16 each. The word disputed in these records means a named source or sources disagree with the catalogued attribution, and the record says who.

Attribution status of 330 published records by object type, highest share disputed first; "Other" is workshop, attributed, documented or rejected
Object typeRecordsAcceptedDisputedDisputed %No maker knownOther
Glass16674430
Islamic metalwork16963810
Textile16453152
Arms and armour16553142
Manuscript19852624
Map19852660
Printed book161142501
Antiquity17442454
Clocks and instruments171042421
Goldsmith's work17831860
Painting453771601
Musical instrument161021240
Furniture171121222
Photograph171421210
Sculpture16111622
Silver1771690
Print18180000
Ceramic15140010

The table counts records and the types are not equal in how they were chosen, so the shares carry some noise. A single record moves a type by six points. The size of the gap between print and glass is not noise.

A print is never disputed because the dispute moves to the sheet

All 18 print records are accepted as to maker, because a print is a multiple, so the argument moves to the individual impression. The record for Duerer's Melencolia I says that no attribution dispute exists over the authorship of the print, and that what it grades is the individual impression. The record for the 1515 Rhinoceros says the same of the individual sheet. Rembrandt's Hundred Guilder Print is universally accepted as his own, and since he did not sign or date it, the date rests on style, technique and watermark evidence.

This is not a sign that printmaking is easier to know. It is a sign that in most of these records a signature, monogram or printed imprint fixes the maker and leaves the open questions to the state, the impression and the date.

Albrecht Duerer, Melencolia I, 1514, the impression in the National Gallery of Art, Washington (Rosenwald Collection). The design is signed in the plate; the record grades this impression.
Albrecht Duerer, Melencolia I, 1514, the impression in the National Gallery of Art, Washington (Rosenwald Collection). The design is signed in the plate; the record grades this impression.National Gallery of Art, Washington, D.C., Public domain

Silver has no maker in nine of 17 records because nobody signed it

Nine of the 17 silver records name no maker, and all nine are ancient or early medieval silver, most of it from hoards and church treasures. They are the Antioch Chalice, the Morgantina silver, the Kaper Koraon treasure, the Mildenhall Great Dish, the Projecta Casket, the Corbridge Lanx, the Traprain Law treasure, the Trewhiddle hoard and the Water Newton treasure. For the Mildenhall dish the records state that no maker's mark or workshop stamp survives, and that the British Museum says only that manufacture somewhere in the general Mediterranean region is a safe surmise.

The Water Newton record puts it best. What survives is a record of donors and not of makers: four individuals are named in the votive inscriptions, and nobody who made the silver is. These are not disputed attributions. They are attributions that were never recorded, so the archive grades them unknown and not disputed.

The Great Dish from the Mildenhall treasure, British Museum. No maker's mark or workshop stamp survives on it.
The Great Dish from the Mildenhall treasure, British Museum. No maker's mark or workshop stamp survives on it.Photograph by JMiall, own work, 4 June 2010, via Wikimedia Commons, CC BY-SA 3.0

In glass the argument is where and when, and in painting it is whose hand

The seven disputed glass records are mostly arguments about where and when the piece was made. The Luck of Edenhall carries no maker's name or mark, scholars agree it is Near Eastern enamelled and gilt glass of the Mamluk period from about the middle of the 14th century, and they disagree on the city: Aleppo, Damascus and Cairo have all been proposed. The Corning ewer has no consensus on whether it was made in Iran or in Fatimid Egypt, and an ultraviolet test matched neither reference group. The Hedwig beaker was placed in a Byzantine or Fatimid milieu for a century, and in 2005 Rosemarie Lierke argued for Palermo.

The paintings are disputed on a different question. The Memling Last Judgment in Gdansk is unsigned, accepted as Memling's since the nineteenth century, and since 2014 a team has argued from dendrochronology, reflectography and XRF that Rogier van der Weyden made the design and underdrawing and Memling finished it. The record notes that no independent test of that proposal was found. A painting argument names a rival hand. A glass argument names a rival city.

The Luck of Edenhall, Victoria and Albert Museum, with its leather case. The museum's own formula for the place of making is Syria or Egypt.
The Luck of Edenhall, Victoria and Albert Museum, with its leather case. The museum's own formula for the place of making is Syria or Egypt.Photograph by Marie-Lan Nguyen (2012), Victoria and Albert Museum, CC BY 2.5
Ask what kind of disagreement a disputed label hides before you weigh it. In the records here it is a rival hand in a painting, a rival city in glass, a different impression in a print and no maker at all in ancient silver.

Questions

Which objects are disputed most often?

By share, glass: 7 of 16 records (44 per cent) carry a disputed attribution. By count, painting has 7 of 45 (16 per cent), and Islamic metalwork has 6 of 16.

Why are no prints disputed?

All 18 print records are accepted as to maker, because a signature, monogram or printed imprint fixes the author in most of the records. The records grade the individual impression or sheet instead, and the dispute moves there.

What does no maker known mean for silver?

It means no maker's mark, name or workshop survives. Nine of 17 silver records are in that group, all ancient or early medieval silver, mostly from hoards and church treasures, and the archive grades them unknown, not disputed.

Is a 7 of 16 rate reliable?

It is a count in a sample chosen for researchable objects, and one record moves a type by about six points. Read the gap between print and glass, which is large, and not the exact percentage.

Sources

  1. 1National Gallery of Art, Washington, object page for Melencolia I, accession 1943.3.3523 (Rosenwald Collection). nga.gov/artworks/...
  2. 2Rijksmuseum, Amsterdam, catalogue entry for Rhinocerus (Neushoorn), object number RP-P-OB-1473. rijksmuseum.nl/en/...
  3. 3National Gallery of Art, Washington, accession 1943.3.7173, 'Christ Preaching (The Hundred Guilder Print),' Rosenwald Collection (identified via the object's Wikimedia Commons file description and corroborating search-indexed National Gallery of Art catalogue text; the Gallery's own object page returned a server error on every direct access attempt during this research pass). nga.gov/artworks/...
  4. 4British Museum, object description text for the Great Dish (iconography of Oceanus and the Bacchic frieze, low relief and engraved line technique, and the statement on manufacture location), as reproduced via web search of britishmuseum.org content. britishmuseum.org/collection/...
  5. 5Samaretz, Nicolas. "Altar Space in Late Antiquity: The Water Newton Treasure and the Construction of the Sacred Space." Eikon Imago 15, e98767 (2026), Masaryk University / University of Fribourg. dx.doi.org/10.5209/...
  6. 6Victoria and Albert Museum, 'The Luck of Edenhall: history and legend of a 14th-century glass beaker.' vam.ac.uk/articles/...
  7. 7Victoria and Albert Museum, online collections catalogue, 'The Luck of Edenhall (beaker and case),' museum no. C.1 to B-1959. collections.vam.ac.uk/item/...
  8. 8British Museum, collection online, 'beaker', museum number 1959,0414.1. britishmuseum.org/collection/...
  9. 9Whitehouse, David. "The Corning Ewer: A Masterpiece of Islamic Cameo Glass." Journal of Glass Studies 35 (1993): 48-56. jstor.org/stable/...
  10. 10Hanserecesse, Zweite Abtheilung (1431-1476), Siebenter (Schluss-)Band, ed. Goswin Frhr. von der Ropp (Leipzig: Duncker & Humblot, 1892): editor's introduction on the taking of Thomas Portinari's galley, Utrecht diet of 1473, no. 41 (Portinari's specification of losses, pp. 115-117, with the editor's footnote identifying one of the two altar tables as Memling's Last Judgment), and nos. 134, 270, 272. Page images read from the Internet Archive scan. archive.org/details/...

More in Attribution and Authenticity

The part

What the Name al-Mawsili Tells You About Where It Was Made

9 min

Two signed ewers, a pyxis and a candlestick carry or borrow the name al-Mawsili. What the name proves about Mosul, and where the museums place each...

The Name on an Instrument Covers More Than One Maker

7 min

A harpsichord with a false Ruckers rose, a Couchet rebuilt to pass, the oldest piano with a 1938 soundboard and a harp with a legend.

The Name on the Map Is an Argument: Four Early Maps

6 min

The Catalan Atlas is catalogued as 'Abraham Cresques ?'. The Carta Marina survives once, the Tabula Rogeriana only in later copies, the Paris...

The Iconotheca Letter

One artwork and its story every Tuesday, with the picture: who owned it, what it made, and what closes soon in the exhibition calendar. No promotion, and nothing we cannot cite.

We send one email to confirm the address and keep nothing else. Every letter carries a one-click unsubscribe. How we handle your data.