# How Paintings Are Authenticated
> Three investigations have to agree, and the scientific one can only exclude. The lead scholar of the largest technical project ever run on one artist says connoisseurship decided every case.
> The Study - 2026-08-02
> Canonical HTML: https://iconotheca.com/study/how-paintings-are-authenticated/
## Key points
- The Bosch Research and Conservation Project states that connoisseurship played a part in every single attribution to Bosch as matters stand today.
- The accepted Rembrandt corpus has been counted at 611, 420, 280 and 340 works by four authorities between 1935 and 2014.
- The Andy Warhol authentication board was dissolved on 19 October 2011 after legal costs its president put at millions of dollars.
- A New York bill to protect authenticators passed the State Senate twice and has died in Assembly committee every session since.
## Three investigations, and none of them is sufficient
## Science excludes, and the exclusion is where forgers fail
## The eye still decides, and the largest technical project says so
## Corpora move, and the Rembrandt numbers show how far
## The real question is who will sign, and for how long that signature holds
## What consensus looks like when it fails in public
Authentication is not a verdict handed down by an instrument. It is three lines of argument that have to point the same way, and the honest position when they do not is to say so and to publish the disagreement. The most expensive picture ever sold at auction carries a National Gallery heading with no qualifier, a Prado listing among attributed and workshop works, a Louvre study asserting full autograph status, and three named scholars who reject Leonardo's hand in most of it. That is the actual state of the art, and no laboratory report on earth resolves it.
## Questions
**Can a scientific test prove a painting is genuine?**
Almost never. Tests establish that materials and structure are consistent with the claimed date and artist, and consistency is the absence of a specific disproof rather than proof. The Bosch project's own conclusion, after years of instrumentation, was that connoisseurship played a part in every attribution it made.
**What is the single most damaging finding?**
An anachronistic material. A pigment, ground, canvas or binder that did not exist at the claimed date ends the discussion, subject only to questions about later restoration. Titanium white, unavailable as an artists' oil colour before about 1921, is the classic example and is what exposed Beltracchi's Campendonk.
**Why do authentication boards keep closing?**
Litigation and its cost. The Warhol board dissolved in October 2011, Haring and Basquiat in 2012, Lichtenstein in 2011, Pollock-Krasner as early as 1996. In Thome v Calder Foundation the New York courts held there is no legal duty to authenticate at all, which leaves the decision voluntary and exposed.
**If the boards have gone, who decides now?**
Individual scholars, the compilers of catalogues raisonnes, and the auction house departments that must publish a heading. All three are more cautious about written opinions than they were twenty years ago, so the practical question for a buyer is who will put an opinion in writing and whether it will still be respected in ten years.
## Sources
- [1] Matthijs Ilsink, 'Jheronimus Bosch and Connoisseurship: Remarks Prompted by the Bosch Research and Conservation Project', CODART, December 2015.
- [2] Jane Kallir, 'Art authentication is not an exact science', The Art Newspaper, 23 November 2018.
- [3] WebExhibits, 'Pigments through the Ages: Titanium Dioxide Whites', on 1916 commercial production and the 1921 artists' oil colour. Tertiary reference source; should be replaced with Marilyn Laver, 'Titanium Dioxide Whites', Artists' Pigments vol. 3, Oxford University Press, 1997.
- [4] Peter Bert, 'Art Law: Auction House Liable in Sale of Fake Expressionist Painting', Dispute Resolution Germany, 28 September 2012, on the Landgericht Koeln ruling against Lempertz.
- [5] CODART, 'A Corpus of Rembrandt Paintings Volume VI presented', 8 October 2014, giving van de Wetering's count of 340 against Tuempel's 280.
- [6] Mariet Westermann, 'Back to Basics', Perspective 2/2011, giving the Bredius 1935 and Gerson 1968 counts.
- [7] Noah Charney, 'Can AI shed new light on how much of The Polish Rider was painted by Rembrandt?', The Art Newspaper, 1 December 2024.
- [8] Artforum, 'Warhol Authentication Board to Shut Down', 28 October 2011, with Joel Wachs's statement.
- [9] Center for Art Law, 'Will the real Andy Warhol please stand up: the Authentication Board to shut down', 24 October 2011.
- [10] Thome v Alexander and Louisa Calder Foundation, 70 A.D.3d 88, New York Appellate Division, First Department, 1 December 2009.
- [11] New York State Senate, bill S1229 (2015-2016), record of Senate passage and Assembly committee referral.
- [12] Nate Freeman, 'Leonardo da Vinci's Salvator Mundi Sells for $450.3 M. at Christie's in New York', ARTnews, 15 November 2017, separating the USD 400 million hammer from the fees.
- [13] Martin Bailey, 'Prado museum downgrades Leonardo's $450m Salvator Mundi in exhibition catalogue', The Art Newspaper, 11 November 2021.
- [14] Alison Cole, 'What the Louvre's scientific examinations of the Salvator Mundi really revealed, according to the museum's own book', The Art Newspaper, 13 April 2021.